Ivermectina y Abemectina

William C.
Campbell
Editor
Ivermectin and
Abamectin
With 66 Illustrations
Springer-Verlag
New York Berlin Heidelberg
London Paris Tokyo
William C. Campbell
Merck Institute for Therapeutic Research
Rahway, N.J. 07065
Library of Congress Cataloging-in-Publication Data

Ivermectin and abamectin I edited by William C. Campbell.
p. cm.
Includes index.
ISBN-13:978-1-4612-8184-9 e-ISBN-13:978-1-4612-3626-9
DOl: 10.1007/978-1-4612-3626-9
I. Ivermectin. 2. Abamectin. I. Campbell, William C. (William

Cecil), 1930-
SF918.183I94 1989
636.089'6964061--dc20 89-6254
CIP
© 1989 by Springer-Verlag New York Inc.

Softcover reprint of the hardcover 1st edition 1989
All rights reserved. This work may not be translated or copied in whole or in part
without the written permission of the publisher (Springer-Verlag, 175 Fifth
Avenue, New York, New York 10010. U.S.A.), except for brief excerpts in
connection with reviews or scholarly analysis. Use in connection with any form of
information storage and retrieval, electronic adaptation, computer software, or by
similar or dissimilar methodology now known or hereafter developed is forbidden.
The use of general descriptive names, trade names, trademarks, etc. in this
publication, even if the former are not especially identified, is not to be taken as a
sign that such names, as understood by the Trade Marks and Merchandise Marks
Act, may accordingly be used freely by anyone. While the advise and information
in this book are believed to be true and accurate at the date of going to press,
neither the authors nor the editors nor the publisher can accept any legal
responsibility for any errors or omissions that may be made. The publisher makes
no warranty, express or implied, with respect to the material contained herein.
Typeset by TCSystems, Inc., Shippensburg, PA.
98765432 1
Dedicated
to the memory of the late Mr. Robert K.

Hartman and the late Mr. Robert E.
Ormond, who contributed to the discovery
of the avermectins, and to the late Dr.
Mohammed A. Aziz, who directed the
clinical evaluation of ivermectin in
humans.
Preface
Ivermectin was introduced to the marketplace as an antiparasitic drug in

1981, and abamectin was introduced as an agricultural pesticide (and as an
antiparasitic drug) in 1985. They are both members of the avermectin
family of compounds, and their impact has been enormous. The efficacy
of ivermectin against nematode and arthropod parasites is unprecedented
i{l potency and breadth of spectrum. Its worldwide acceptance in live-
stock production and in the health care of companion animals has made it
a major commercial success. Its efficacy in human onchocerciasis (river
blindness) has made it a promising candidate for the control of one of the
most insidious and intractable of tropical diseases. Abamectin is in
commercial use as an agricultural pesticide, and its applications are
continuing to expand. Most striking of all, perhaps, is the intense
scientific activity triggered by the discovery of the avermectin com-
pounds. In the fields of chemistry, microbiology, parasitology, entomol-
ogy, biochemistry, pharmacology, and the clinical and agricultural scien-
ces, interest in the avermectins has evolved rapidly, and a considerable
bibliography has already been amassed. Chemicals that are widely used
raise correspondingly wide concerns about safety-and the ongoing
evaluation of the avermectins thus far has included not only studies on the
safety of the compounds for the animals, plants, and people in which they
are used but also studies on their safety for the people who make and
apply them, safety for the people who eat the plants and animals to which
these agents have been applied, and safety for the environment which is
exposed, one way or another, to their effects. Given the intense industrial
and scientific interest in ivermectin and abamectin, and the likelihood that
they will be the forerunners of an expanding family of relatives and
descendants, it seems timely to take stock of what we have learned about
them so far.
With few exceptions, the authors of this book are members of the
scientific staff of the Merck Sharp & Dohme Research Laboratories; and
there were reasons for this. In certain disciplines, such as toxicology and
fermentation technology, most of the available information on the aver-
viii Preface
mectins has been obtained by MSDRL scientists, and they are therefore
in a good position to review the data. The toxicological and environmental
safety data that have been accumulated within our laboratories have been
made available to governmental regulatory agencies and other interested
parties in the form of official reports, but have seldom appeared in the
scientific literature. In other disciplines, much information already has
appeared in the scientific literature; but here, too, a large proportion of
the research has been carried out by, or in collaboration with, our own
scientists-and the names of those individuals inevitably came to mind as
prospective authors of review chapters. There was, moreover, a practical
advantage in the preponderance of MSDRL authors: the chapters could
be prepared synchronously. The editor knows, from hard-won experi-
ence, the difficulty of getting manuscripts from far-flung scribes with no
obligation other than a cheerfully and hastily given promise to write a
chapter for a particular volume. In some multiauthored books, a chapter
written by a conscientious contributor may be older by many months, or
even by a few years, than one written by some hopelessly delinquent
contributor. In the present case, the editor has been blessed with
colleagues who have been willing to make an extraordinary effort to bring
their labors to fruition at the same time. This book, then, comes close to
the objective of capturing a likeness of its subject at one instant in time.
Except in the context of synthetic and microbiological chemistry
(which must of necessity deal with structural diversity), this volume deals
almost exclusively with ivermectin and abamectin-not with the ex-
tended family of the avermectins or with the related family known as
milbemycins.
The expression of dosage in terms of micrograms (f.Lg) per kilogram of
body weight has become commonplace in accounts of the efficacy of
these compounds, and this form is generally used in this book. Toxicolog-
ical studies, however, usually deal with mUltiples of milligrams, rather
than fractions, and in that context dosages are expressed in terms of
milligrams (mg) per kilogram of body weight.
The chapters of the book have been arranged in three groups. The first
deals with the basic chemical, biochemical, and microbiological aspects;
the second deals with pharmacological, safety, and environmental
aspects; and the third covers practical use of the compounds as antipar-
asitic and pesticidal agents. In the case of human application, however,
the available safety data have emerged from the clinical experience, and
are to be found in Part III.
William C. Campbell
Acknowledgments
The editor thanks Dr. Harold D. Hafs, Vice-President of Animal Science

Research, Merck, Sharp & Dohme Research Laboratories, and Dr. Mark
Licker, Executive Editor, Springer-Verlag, for their support and counsel
throughout the preparation of this volume. He is indebted to Dr. M.H.
Fisher for preparation of the introductory statement on Nomenclature
and Physical Properties and to Dr. Dennis J. Underwood for the
computer-generated structural models that appear on the cover.
Contents
Preface vii
Part I
Chapter 1. Chemistry
M.H. Fisher and H. Mrozik 1
Chapter 2. Isolation and Characterization of

the Producing Organism
R. W. Burg and E.O. Stapley 24
Chapter 3. Fermentation Development and Process Improvement

Mary Nallin Omstead, Louis Kaplan, and
Barry C. Buckland 33
Chapter 4. Biosynthesis
S.T. Chen, O.D. Hensens, and M.D. Schulman 55
Chapter 5. Mode of Action of Ivermectin

M.J. Turner and J.M. Schaeffer 73
Part II
Chapter 6. Toxicology
George R. Lankas and Lea R. Gordon 89
Chapter 7. Pharmacokinetics of I vermectin in Animals

and Humans
David W. Fink and Arturo G. Porras 113
Chapter 8. Metabolism and Tissue Residues

Shuet-Hing Lee Chiu and A. Y.H. Lu 131
xii Contents
Chapter 9. Chemical Assay for Ivermectin in Edible Tissues

George V. Downing 144
Chapter 10. Safety of I vermectin in Target Animals

J.D. Pulliam and J.M. Preston 149
Chapter 11. Environmental Aspects of I vermectin Usage

in Livestock: General Considerations
B.A. Halley, R.J. Nessel, and A.Y.H. Lu 162
Chapter 12. Environmental Aspects of Use of Ivermectin and

Abamectin in Livestock: Effects on
Cattle Dung Fauna
R.A. Roncalli 173
Chapter 13. Environmental Aspects of Abamectin Use in

Crop Protection
P.G. Wislocki, L.S. Grosso, and R.A. Dybas 182
Chapter 14. Worker Safety Aspects of Abamectin Use in

Crop Protection
L.S. Grosso, R.A. Dybas, and S.F. Rickard 201
Part III
Chapter 15. Use of Ivermectin in Cattle, Sheep, Goats,

and Swine
G. W. Benz, R.A. Roncalli, and S.J. Gross 215
Chapter 16. Use of Abamectin in Cattle.

G. W. Benz and J.L. Cox 230
Chapter 17. Use ofIvermectin in Horses

W.C. Campbell, W.H.D. Leaning, and R.L. Seward 234
Chapter 18. Use ofIvermectin in Dogs and Cats

W.C. Campbell 245
Chapter 19. Use of Ivermectin in Laboratory and

Exotic Mammals and in Birds, Fish, and Reptiles
M.D. Soli 260
Chapter 20. Abamectin Use in Crop Protection

R.A. Dybas 287
Contents xiii
Chapter 21. Use of Ivermectin in Humans

B.M. Greene, K.R. Brown, and H.R. Taylor 311
Appendixes
I: The Determinative Method for Assaying Ivermectin

Residues in Tissue and Plasma
II: Confirmatory Assay for Assaying Ivermectin Residues

in Liver
III: List of Registrations

J. Di Netta 344
Subject Index 347
Contributors
G.W. BENZ, Ph.D.

Department of Animal Science Research, Merck Sharp & Dohme
Research Laboratories, Rahway, New Jersey 07065
K.R. BROWN, M.D.

Department of Clinical Research, Merck Sharp & Dohme Research
Laboratories, West Point, Pennsylvania 19486
BARRY C. BUCKLAND, Ph.D.

Department of Biochemical Process Research and Development, Merck
Sharp & Dohme Research Laboratories, Rahway, New Jersey 07065
R.W. BURG, Ph.D.

Department of Basic Microbiology, Merck Institute for Therapeutic
Research, Rahway, New Jersey 07065
WILLIAM C. CAMPBELL, Ph.D.

Department of New Drug Discovery, Merck Institute for Therapeutic
S.T. CHEN, Ph.D.

Department of Fermentation Microbiology, Merck Sharp & Dohme
SHUET-HING LEE CHIU, Ph.D.

Department of Animal Drug Metabolism, Merck Sharp & Dohme
J.L. Cox, Ph.D.

xvi Contributors
J. DI NETTA, J.D.
Department of Regulatory Affairs, Merck Sharp & Dohme Research
Laboratories, Rahway, New Jersey 07065
GEORGE V. DOWNING, Ph.D.

Department of Analytical Research, Merck Sharp & Dohme Research
R.A. DYBAS, Ph.D.

Department of Agricultural Research, Merck Sharp & Dohme Research
Laboratories, Three Bridges, New Jersey 08887
DAVID W. FINK, Ph.D.

Department of Animal Formulation Development, Merck Sharp &
Dohme Research Laboratories, Rahway, New Jersey 07065
M.H. FISHER, Ph.D.

Department of Synthetic Chemistry Research, Merck Sharp & Dohme
LEA R. GORDON, D.V.M., Ph.D.

Department of Safety Assessment, Merck Sharp & Dohme Research
B.M. GREENE, M.D.

Division of Geographic Medicine, School of Medicine, University of
Alabama at Birmingham, Birmingham, Alabama 35205
S.J. GROSS, Ph.D.

L.S. GROSSO, Ph.D.

B.A. HALLEY, Ph.D.

O.D. HENSENS, Ph.D.

Department of Natural Product Chemistry, Merck Sharp & Dohme
Contributors xvii
LOVIS KAPLAN, Ph.D.

GEORGE R. LANKAS, Ph.D.

Department of Safety Assessment, Merck Sharp & Dohme Research
W.H.D. LEANING, B.V.Sc.

Department of Technical Services, MSD AGVET, Rahway, New
Jersey, 07065
A.Y.H. Lu, Ph.D.

H. MROZIK, Ph.D.
Department of Synthetic Chemistry Research, Merck Sharp & Dohme
M. NALLIN OMSTEAD, Ph.D.

R.J. NESSEL, Ph.D.

Department of Regulatory Affairs, Merck Sharp & Dohme Research
ARTURO G. PORRAS, Ph.D.

Department of Drug Metabolism, MerSharp & Dihme Research
J.M. PRESTON, B.V.M.S., Ph.D.

J.D. PULLIAM, D.V.M., Ph.D.

S.F. RICKARD, Ph.D.

xviii Contributors
R.A. RONCALLI, D.V.M.

J.M. SCHAEFFER, Ph.D.

Department of Biochemical Parasitology, Merck Sharp & Dohme
M.D. SCHULMAN, Ph.D.

R.L. SEWARD, D.V.M.

Department of Animal Drug Evaluation, Merck Sharp & Dohme
M.D. SOLL, B.V.Sc.

E.O. STAPLEY, Ph.D.

Department of Basic Microbiology, Merck Institute for Therapeutic
H.R. TAYLOR, M.D.

Dana Center for Preventive Ophthalmology, The Wilmer Institute, The
Johns Hopkins University, Baltimore, Maryland 21205
M.J. TURNER, Ph.D.

Department of Biochemical Parasitology, Merck Sharp & Dohme
P.G. WISLOCKI, Ph.D.

NOMENCLATURE AND PHYSICAL PROPERTIES
NATURAL A VERMECDNS
ORS
components A: R5 = CH 3 components 1: X = -CH=CH-

OH
components B : Rs = H components 2: X = -CH 2 tH-
components a : R26= C2 H5
components b : R26= CH 3
ARAMECDN
Synonyms: Avermectin B, ; AVM; C-076; MK-936.
Trade names: AVID® AGRI-MEK® VERTIMEC® AVOMEC®
Abamectin contains at least 80% of avermectin B 'a and not more than 20% of avermectin B,b.
Avermedin R..
C48Hn O ,4 ; mol wt: 872. otT-white powder; [alo + 55.7 ± 2° (c = 1.06 in chloroform); uv max
(methanol): 237, 243, 252 nm (E 29, 120; 31,850; 20,510).
Avermedin Rib
mol. wt. 858.

IVERMECON
OH
Synonyms: 22.23-Dihydroavermectin B I ; 22.23-dihydro C-076B I ; MK-933;
Trade names: HEARTGARD30®;CARDOMEC®:EQYALAN®;

IYOMEC.®; ZIMECTERIN®; MECTIZAN®
Semisynthetic derivative of the avermectins. q.v. Ivermectin contains at least 80% of 22.23-
dihydroavermectin Bla and less than 20% of 22.23-dihydroavermectin BIb. Off-white powder;
[aID + 71.5 ± 3° (c = 0.755 in chloroform); uv max (methanol): 238. 245 nm (f; 27.100; 30.100).
Component 810 (MK-932)
mol. wt. 874
5-0-Demethyl-22.23-dihydroavermectin Ala; 22.23-dihydroavermectin Bla ; 22.23-dihydro

C-076B la
Crystals from ethanol/water. m.p. 1545 - 157°C
Composeat 8110
mol. wt. 860
5-O-Demethyl-25-deO-methylpropyl)-22.23-dihydro-25-( I-methylethyl)avermectin Ala; 22.23-

dihydroavermectin BIb
CHAPTER 1
Chemistry
M.H. Fisher and H. Mrozik
I. Structures of the Naturally Occurring Avermectins

II. Chemical Properties
III. Chemistry
A. Hydroxy group protection, acylation, elimination to fluorescent
aromatic derivative, and analytical assays
B. Monosaccharides and aglycones
C. Oxidations
D. Reductions including ivermectin
E. Radiolabeled derivatives
F. Alkylations
G. Interconversions of avermectins
H. Glycoside syntheses
IV. Structure-Activity
I. Structures of the Naturally Occurring Avermectins

The avermectin structures are closely related complex 16-membered
macrocyclic lactones. Although they share structural features with the
antibacterial macrolides and the antifungal macrocyclic polyenes, the
avermectins are not usually grouped with those compounds. Avermectins
have neither antibacterial nor antifungal activities and do not inhibit
protein or chitin syntheses as do the other two groups. In the avermectins
the macrocycle is a backbone for further substitution with a spiroketal
unit (C-17 to C-28), a hexabydrobenzofuran unit (C-2 to C-8a), and a
disaccharide substituent at C-13 (Fisher and Mrozik 1984, Davies and
Green 1986). The milbemycins are structurally related but are missing the
C-13 disaccharide substituent (Takiguchi et al. 1980). Although the
milbemycins and their potent miticidal and acaricidal properties were
discovered shortly before the avermectins, their anthelmintic properties
became known only after the discovery of the avermectins.
Fermentation of the actinomycete Streptomyces avermitilis produces 4
2 M.H. Fisher and H. Mrozik
homologous pairs of closely related compounds: avermectin AI. A 2 , B\,

and B2, originally also called compounds C-076 AI through B2. The
A-compounds have a 5-methoxy, whereas the B-compounds have a
5-hydroxy substituent; and the I-compounds have a 22,23-double bond
which is formally obtained by dehydration of the axial 23-hydroxy group
of the 2-compounds. The 4 pairs are further divided into the major
components Ala, A2a , B la , B2a-with a secondary butyl sidechain at
carbon position 25-and the minor components (usually present in
amounts of 1% to 20%) Alb, A2b , BIb, B2b-with an isopropyl substi-
""
DRS
Avermectin Rs R26 Cn -X-C23

AI. CH3 C2HS -CH=CH-
Alb CH3 CH 3 -CH=CH-
BI• H C2Hs -CH=CH-
Bib H CH 3
,
-CH=CH-
OH
A20
A2b
CH3
CH3
C2HS
CH3
,
-CH2-CH-
OH
-CH:rCH-
OH
B20
B2b
H
H
C2HS
CH 3
,
-CH2-CH-
OH
-CH 2-CH-
Ivermectin H >80% C2Hs -CH2-CH2-

<20% CH3
FIGURE 1.1. A vermectin structures.

1. Chemistry 3
tuent at C-25 (Figure 1.1). Three of these-A2a , Bla , and B2a-are

produced in quantity in the fermentation. Avermectin B la is the most
important because it has high potency against a broad spectrum of endo-
and ectoparasites of farm animals and many agricultural mite and insect
pests. It also serves as starting material for the semisynthetic 22,23-
dihydro analog, which is used almost exclusively under the generic name
ivermectin for both prevention and cure of parasitic infections of animals.
Because the separation of the major 25-secondary-butyl components
Ala through B2a from the minor 25-iso-propyl components Alb through
Bib is very tedious and impractical on a large scale, and because the
biological activities of the two homologs are practically identical, the
avermectin components are often used as a mixture of these two
homologs. In such cases they are just defined as At. A 2 , Bt. and B2,
which means a mixture of not less than 80% of components "a" (and
not more than 20% of components "b"), and the antiparasitic agent
ivermectin (22,23-dihydroavermectin B I ) is marketed as such a mix-
ture.
Avermectin B la contains 5 double bonds, 2 of which are conjugated as
an 8,9,10,1l-diene function. This results in a strong UV absorption at 245
nm, an advantageous property for analytical detection. Of the 3 remaining
double bonds, the two 3,4- and 14,15-enes are trisubstituted and the
22,23-ene (occurring only in the I-components) is cisdisubstituted. Aver-
mectin B la has 3 free hydroxy groups, a very reactive secondary allylic
C-5, a somewhat less reactive secondary C-4", and a relatively unreactive
tertiary allylic C-7 hydroxygroup. The compound also has many ether
bonds, of which the 2 glycosydic ones at C-l' and C-l" of the 2-deoxy
sugars are particularly interesting: they are responsible for the- ease of
acidic alcoholyses and hydrolyses of the disaccharide to the monosaccha-
rides and aglycones.
The avermectin structures were determined by spectral means (Albers-
SchOnberg et al. 1981) and shown to be closely related to the milbemycins
(Mishima et al. 1975). Final confirmation of the structure as well as
relative and absolute stereochemistry was achieved by X-ray crystallo-
graphy of avermectin B2a aglycone and of avermectin B la (Springer et al.
1981). Isolation of more than 1 mole of L-oleandrose served as reference
for the absolute stereochemistry and proved that the disaccharide was
made up of 2 identical sugars. Extensive mass spectral fragmentation
patterns (Figure 1.2) and high field IH_ and \3C-NMR assignments (Tables
1.1 and 1.2) for avermectin B la , 22,23-dihydroavermectin B la (main
component of ivermectin) and many derivatives have been described and
serve as ready references for the identification of newly obtained deriv-
atives. A detailed IH NMR study suggests that the conformation of
avermectin B la in solution is very closely related to its conformation in
the crystal structure.
~3
o H
87
,/" OCH3
-CH~H
127
CH~+
1 COOH
<71
o CH
OR 3
275/261
~
CH 3 ~CH3 3
HO0 CHz
HOWCHs- Oto......H +H
CH3
169 111 179
FIGURE 1.2. Mass fragments ofavermectins. (G. Albers-SchOnberg, B.H. Arison,
J.e. Chabala, A.W. Douglas, P. Eskola, M.H. Fisher, A. Lusi, H. Mrozik, J.L.
Smith and R.L. Tolman, J. Am. Chern. Soc. 103, 4216 [1981]).
1. Chemistry 5
TABLE 1.1. IH NMR Data for ivermectin and avermectin

B la •
Ivermectin Avermectin B 1a
H I) (ppm) I) (ppm)
C2H 3.30 (q, 2.0) 3.33

C 3H 5.44 (q, -I) 5.46
C4CH3 1.88 (s) 1.90
C,OH unassigned 2.38
C,H 4.31 (t, 7.0) 4.34 (t, 6.0)
CJI 3.99 (d, 6.0) 4.08 (d, 6.0)
C,oH 4.16 (s) 4.01
CSaH2 4.69 (d,d, -14.2, -2.0) 4.74 (d,d, 16.0, 2.0)
4.71 (d,d, -14.2, -2.0) 4.72 (d,d, 16.0,2.0)
CgH -5.88 (m) 5.91 (m)
CIOH -576 (m) 5.78 (m)
CIIH -5.76 (m) 5.78 (m)
C I2CH 3 1.17 (d) 1.18 (d)
C I2H 2.54 (m) 2.55
CI3H 3.96 (broad s) 3.98
C I4CH3 1.50 (s) 1.50
C1,H 5.00 (d, 10.0) 5.03 (d,d, 10.0, 4.0)
C I6CH 2 2.35 (m) 2.2-2.4
C 17H -3.65 (m) 3.90 (m)
C 1sH 2 ax -0.85 (q) 0.88 (q 12)
eq -1.78 (d) 1.80 (br.d 12)
C1gH 5.38 (t,t,5.0,5.0) 5.44 (m)
C20H2 unassigned eq. 2.09 (dd 12, 4)
ax 1.50-1.65
C n H 2/C 22H unassigned 5.82 (dd 10, 1.7)
C n H 2/C 23H unassigned 5.60 (dd 10, 2)
C 24H unassigned unassigned
C 24CH 3 0.79 (d) 0.93
C2SH 3.59 (d,11.0) 3.52 (d, 9)
C26H unassigned 1.50-1.65
C26CH3 0.86 (d) 0.92
C 27CH2 unassigned 1.50-1.65
C 28CH 3 0.94 (t) 0.95
C1H 4.8 (d, -3.0) 4.82 (d,3)
C/H2 unassigned 2.2-2.4
C/H -3.3 or 3.6 3.66 (ddd 12,7,4)
C/OCH3 3.44 or 3.45 3.46 or 3.44
C/H 3.26 (t, 9.0) 3.28 (t, 9.0)
Cs'H 3.84 (d,q,9.2,9.0) 3.88 (d, q, 9, 6)
Cs'CH3 1.28 or 1.25 (d) 1.27
Ct"H 5.42 (d, 3.0) 5.44 (d, 3.5)
CtCH 2 unassigned 2.2-2.4
C 3"H -3.6 or 3.81 3.52 (m)
CtCH 3 3.44 or 3.45 3.46 or 3.44
C/OH 2.52 (d,1.9) 2.52
C/OH 3.18 (d,t,1.9,9.0) 3.20 (t,9.5)
Cs"H 3.78 (d,q,9.5,7.0) 3.81 (d, q, 9.5, 6.5)
Cs"CH3 1.28 or 1.25 (d) 1.30
Courtesy Dr. Byron Arison, Merck & Co.
TABLE 1.2. 13C NMR Data for avermectin

B 1a and ivermectin.
Shifts (ppm)
C .!!!!. IVM
28 12.0 12.1
26a 12.9 12.4
14a 15.1 15.1
24a 16.4 17.4
6' 17.7 17.7
6" 18.4 18.4
4a 19.9 19.9
12a 20.2 20.2
ICHJ 8
27 27.5 27.3
24 30.6 31.2
16 -34.3 34.5
2' -34.3 34.2
2" -34.3 34.2
26 35.2 35.5
18 36.6 36.9
12 39.8 39.7
20 40.5 41.2
2 45.7 45.7
-CHz- 6
>CH- 4
I(CH 2 /CH) 10
3'a 56.4 56.4

3"a 56.4 56.5
I(OCHJ ) 2
5' 67.3 67.2

5 67.8 67.7
19 -68.2 68.7
8a -68.2 68.4
17 -68.4 67.2
5" -68.4 68.1
25 75.0 76.7
4" 76.1 76.0
3" 78.3 78.2
3' 79.4 79.2
6 79.4 79.4
4' 80.5 80.4
7 80.5 (s) 80.4
13 82.0 81.8
CH 20 1
CH-o 12
C-o 1
R(CHP) 14
1' 95.0 94.8
21 95.8 97.5
1" 98.5 98.5
1. Chemistry 7
TABLE 1.2. Continued.

Shifts (ppm)
C .!!!!. IVM
0-
-CH/ 2
'0-
, /0-
C
/ '0-
I(CHx°:z) 3
15 1ll8.1 118.1
3 118.4 118.3
9 120.4 120.4
10 124.8 124.7
23 127.9 28.1
14 135.2 135.0
22 136.2 35.7
4 137.9 137.8
11 -138.0 138.0
8 139.7 139.6
-cH = C 7
>C=C 3
I(CHx = C) 10
~o-
-C 173.6 173.8
'0-
I(C02) 1
overall I 48
II. Chemical Properties

Avermectins are highly lipophilic substances and dissolve in most organic
solvents, such as chloroform, methylene chloride, acetone, alcohols,
toluene, cydohexane, dimethylformamide, dimethylsulfoxide, and tetra-
hydrofuran. Their solubility in water is correspondingly low-only .006 to
.009 ppm (= mgll).
Chemical stability studies are easy to monitor by silicagel thin-layer
chromatography (TLC) or by high-performance liquid chromatography
(HPLC), most advantageously using a reverse-phase CIS coated column
(Miller et al. 1979). HPLC peaks or TLC spots are visualized by their UV
absorption at 245 nm; the TLC spots can also be detected by ceric sulfate
or phosphomolybdic acid staining.
The well-known acid lability of 2-deoxy-glycosides predicts that the

avermectins are acid-sensitive. They are stable in anhydrous glacial acetic
acid solution at room temperature, but aqueous AcOH and more rapidly
stronger acids such as dilute HC1, H 2S04 , and p-toluenesulfonic acid in
methanol solution cleave off the first sugar. Much stronger concentrations
of methanolic or aqueous acids are necessary to hydrolyze off the second
sugar to give the aglycone. Although the 5,7-dihydroxy-3,4-cyclohexene
part of the molecule might be expected to be subject to dehydration to an
aromatic derivative, this does not occur very readily, thus allowing the
preparation of monosaccharides and aglycones in good yields.
The avermectins contain an acidic proton at the asymmetric C-2 next to
the lactone carbonyl group. Treatment with a strong base such as
methanolic potassium hydroxide, sodium methoxide, or 1,8-diazabicyclo
[5.4.0] undec -7-ene (DBU) causes epimerization at that center and shifts
the beta-gamma double bond into conjugation with the lactone carbonyl
(Figure 1.3; from Pivnichny et al. 1983, 1988). Prolonged base treatment
leads to aromatization of the 6-membered ring, opening of the lactone
ring, and further degradation to a gross mixture of unidentified products.
Anhydrous pyridine, however, even at temperatures up to 100° C, is
without effect. Strongly basic conditions must be avoided during reac-
tions of avermectins, but, consequently, strong alkali can be used to
destroy avermectin residues for safe disposal.
UV light below 280 nm rapidly isomerizes the E (trans) 8,9 and 10,11
double bonds to the 8,9- and 1O,1l-Z isomers. Prolonged irradiation leads
to a large number of unidentified decomposition products which lack any
UV absorption (Figure 1.4; from Mrozik et al. 1988). Solutions of
FIGURE 1.3. Treatment of ivermectin with methanolic potassium hydroxide

solution. Adapted from J. Pivnichny et al. Copyright 1983 by The American
Pharmaceutical Association. (Reprinted by permission of the copyright owner.)
1. Chemistry 9
8,9 - Z
H:'x5
· ."0x5
/
3
3
3 1,_
He·"
3 0".
UV LIGHT
~
10,11 - Z
FIGURE 1.4. Photoisomerization of Avermectins. (Adapted from Mrozik et aI.

1988.)
avermectins in pyrex flasks are generally stable because the glass

excludes UV light.
III. Chemistry
A. HYDROXY GROUP CHI\RACTERISTICS
Avermectin BJ and ivermectin each contain 2 secondary and 1 tertiary

hydroxy group. The 2 secondary alcohols are readily acetylated with
acetic anhydride in pyridine, although at 0° C, the reaction yields a fair
amount of the 4"- but not the 5-monoacetate. With more reactive acid
chlorides, however, it is not possible to obtain the 4"-monoacyl deriv-
ative; moreover, it is necessary to protect the 5-0H as a 5-0-tertiary
butyldimethysilyl (TBDMS) derivative, which can be prepared selectively
due to the hindered C-4" position (Mrozik et al. 1982a). Removal of the
5-protecting group is accomplished with p-toluenesulfonic acid monohy-
drate (p-TsOH x H 20) in MeOH (1%, room temperature for 20 to 30
minutes; longer reaction times generate substantial amounts of monosac-
charide) or with an HF-pyridine-THF mixture (Bliard et al. 1987). It is
sometimes possible to prepare 5-monoacyl derivatives directly from
unprotected ivermectin by careful use of 1 equivalent of acylation
reagent, particularly when the latter is bulky (e.g., bis-trichloroethoxy
phosphoryl chloride). Many 4"-O-substituted derivatives have been pre-

pared, as they retain considerable bioactivities; whereas 5-0-substitution
products lose most anthelmintic activities. Other protecting groups used
advantageously include phenoxyacetyl (reversed by NaOMe in MeOH or
NH3 in MeOH) (Gallo et al. 1983) or trichioroethyIcarbonate (reversed by
zinc-AcOH reduction).
Because the avermectins exhibit unprecedented potency, they are used
at unusually low doses of 6 to 300 I'g/kg, which makes the detection and
isolation of residues and metabolites from animal tissue a new challenge.
For this reason a very sensitive analytical assay required a derivative
suitable for detection at levels down to 1/10 or 1/100 of one ppm.
Ivermectin and avermectin B, are therefore converted into an aromatic
derivative which allows detection via fluorescence absorbance. For this
purpose avermectin B" ivermectin, or their derivatives were heated with
acetic anhydride (Ac 20) in pyridine at 100° C for 24 hours (Tolan et al.
1980). The reaction time can be reduced to 1 hour by using N-methyl
imidazole as a catalyst (Tway, Wood, and Downing 1981). The resultant
4"-O-acetyl-2,5,6,7-bisdehydro analog is detected after HPLC separation
and characterize<l by a distinct retention time. Using a Crextraction
cartridge, with 22,23-dihydro avermectin B'a monosaccharide as an
internal standard, it is possible to detect as little as 1 ng/ml of ivermectin
in plasma or milk (Chiou, Stubbs, and Bayne 1987). This procedure was
adapted as a confirmatory assay by hydrolysis of the avermectin deriv-
atives to aglycone, monosaccharide, and parent disaccharide (vide infra),
subsequent aromatization, and, finally, detection of 3 distinct HPLC
peaks corresponding to the acetylated bisdehydro aglycone, monosaccha-
ride, and parent compound peaks via fluorescence detection. A procedure
using chemical ionization mass spectrometry/mass spectrometry
(MS/MS) was developed for a confirmatory assay of ivermectin tissue
residues in cattle (Tway et al. 1984). It is also possible to determine
avermectins directly after extraction, using a synthetically isomerized
avermectin derivative as internal standard for HPLC with UV detection
(Pivnichny, Shim, and Zimmerman 1983). A HPLC-reverse isotope
dilution assay is available for the determination of ivermectin residues in
animal tissue (Chiu et al. 1985). A similar fluorescent derivative contain-
ing a 5-phenol group obtained by dehydration of the 5-keto derivative of
ivermectin with ammonium acetate was also reported (Stong 1987).
B. MONOSACCHARIDES AND AGL YCONES
The macrocyclic lactone of all avermectins has at carbon 13 an a-L-

oleandrosyl-a-L-oleandrosyloxy substituent which is a 2-deoxy sugar
glycoside; and as such it is relatively sensitive to acid hydrolysis or
alcoholysis. A solution of ivermectin in methanol containing a strong acid
such as 1% sulfuric acid readily gives a good yield of the aglycone after 16
1. Chemistry 11
to 24 hours at room temperature. When isopropanol is substituted for

methanol in this reaction, a good yield of the monosaccharide is obtained,
as the bulkier isopropanol preferentially attacks the sterically less hin-
dered 1"-position over the 1'-position, which is attached directly to the
C-13 ofthe macrocycle, which in turn is flanked on either side by the C-12
and C-14 methyl groups respectively (Figure 1.5; from Chabala et al.
1980). These procedures readily yield the monosaccharides of ivermectin
and avermectin BJ and the aglycone of ivermectin. The preparation of the
aglycone of avermectin B), however, is complicated: under the reaction
conditions, addition of methanol to the 22,23-double bond occurs, yield-
ing mainly the 2 epimeric 22,23-dihydro-23-methoxy monosaccharides
and/or aglycones. The aglycone of avermectin BJ therefore must be
prepared with aqueous acid; it can be obtained as a I: 1 mixture of
aglycone and monosaccharide using 10% sulfuric acid in aqueous THF
solution (Mrozik et al. 1982b).
Removal of the 13-hydroxy group from avermectin aglycones gives
13-deoxyavermectin aglycones which are closely related to certain milbe-
O~H
Clb7""--o~tb~Oi CIb
HO~" 0"
ClbO
CHI'
I,
HtSO
(CHs'ICHOH
l%
18"c.J6h
CIHs H
"
Rs %HtS04
~
CHsOH
lS"C)6h
H ):--CHs
Rn' s
18"C,16h
Rs Rs
Monosocchoride Aglycone
RS'OCHs 01 OH
Ru=OHOIH
FIGURE 1.5. Avermectin Monosaccharides and Aglycones. (Adapted from Cha-

baIa et aI. 1980.)
mycins. For example, ivermectin was converted into its aglycones

(22,23-dihydroavermectin B la and Bib aglycones): the 5-hydroxy group
was then protected as an O-tert-butyldimethylsilyl group; then the
13-hydroxy group was substituted by chIoro; the 13-chIoro group was
removed by reduction; C-5-0H deprotected; and, finally, the C-25 B la
and Bib homologs were separated into 13-deoxy-23,23-dihydroaver-
mectin B la and Bib aglycones (Mrozik et al. 1983). Interestingly, the
latter was subsequently also obtained from a milbemycin fermentation
and given the name milbemycin B41-D.
C. OXIDATIONS
Avermectin BI and ivermectin have 2 secondary hydroxy groups suscep-

tible to oxidation to the corresponding ketones. Due to the allylic nature
of the C-5 hydroxy group, its selective oxidation to 5-oxoavermectin is
readily accomplished with manganese dioxide (Chabala, Rosegay, and
Walsh 1981). Although tautomerisation to the enol and elimination of the
remaining tertiary C-7 hydroxy group leads to a phenol (Stong 1987), the
5-oxo derivatives are nonetheless stable enough to allow further chemical
reactions. Particularly important is the stereospecific reduction with
NaBH4 to the 5-f3 alcohol with natural configuration, which is used to
prepare tritiated avermectin derivatives (vide infra). Reaction with tri-
ftuoroacetic anhydride, however, gives the 2,5,6,7-tetradehydro phenol
analog. Use of mild oxidation procedures such as oxalyl chloride-DMF
(Swem oxidation) gives 4",5-dioxo avermectin derivatives or-after
protection of the C-5 hydroxy group as a tert-butyldimethylsilyl
derivative-5-0-tert-butyldimethylsilyl-4"-oxoavermectins, which are
valuable intermediates for further 4"-modified analogs. Sodium boro-
hydrate reduction of the 4"-keto function gives a mixture of 4"-epimeric
hydroxy compounds, with the unnatural axiaI4"-hydroxy group the major
product and the natural epimer the minor one. Reductive amination with
NaCNBH3 and NH40Ac likewise yields an epimeric mixture of 4"-amino
derivatives with interesting biological properties (Mrozik 1984a).
The avermectins also possess a number of allylic positions that are
susceptible to oxidative modification. In particular the 8a methylene
group-which is both allylic and a to an ether oxygen-is
susceptible to radical oxidation. The primary product is the 8a-
hydroperoxide, which has been isolated occasionally as an impurity of an
avermectin BI reaction (such as the catalytic hydrogenation of avermec-
tin BI with Wilkinson's rhodium chloride-triphenylphosphine catalyst to
obtain ivermectin). An 8a-hydroxy derivative can also be detected
occasionally as a metabolite (Bull et al. 1984) or as an impurity arising
presumably by air oxidation. An 8a-oxo-derivative can be obtained by
oxidizing 5-0-protected avermectins with pyridinium dichromate (Mrozik
and Waksmunski 1985). This also can arise by treating the 8a-hydroperox-
ide with base.
1. Chemistry 13
The allylic 4a-methyl group can be oxidized with SeOz-catalyzed

tert-butylhydroperoxide to a 4a-hydroxy analog (Mrozik 1984b).
The double bonds of avermectins react with m-chloroperbenzoic acid
to give 3,4-, 8,9-, and 14, 15-epoxides. The 8,9-epoxide is the major
product and can be isolated in good yield (Mrozik 1985a). The 8,9-oxide
was opened by aqueous acids to the 8,9-diol (Figure 1.6; from Smith and
Thompson 1985). The 3,4-diol can be obtained readily and regiospecifi-
cally by osmium tetroxide oxidation. Neither peracids nor OS04 will
attack the 22,23-double bond. The 10,11-double bond can be oxidized
selectively with N -bromacetamide to the 11,1 O-bromohydrin.
D. REDUCTIONS INCLUDING IVERMECTIN
Reductions of avermectin B\ containing 5 double bonds with Pd or Pt

catalysts proceed at almost comparable rates at the 2 disubstituted 10,11-
and 22,23-double bonds to give a mixture of 10,11- and 22,23-dihydro
derivatives. Further hydrogenation leads to 10,11,22,23-tetrahydro and
3,4,10,11,22,23-hexahydro derivatives; exhaustive hydrogenation leads to
3,4,8,9,10,1l,22,23-octahydro analogs. Under no circumstances is the
14,15-double bond hydrogenated. For the preparation of ivermectin
(22,23-dihydro avermectin B\), Wilkinson's homogeneous (Ph3P)3RhCl
catalyst was employed for the regiospecific hydrogenation of avermectin
B\'s 22,23-double bond, which is the only disubstituted cis double bond
(Figure 1.7; from Chabala et al. 1980). To prepare 1O,1l-dihydro avermec-
tin Bt, the regioselective reaction with N-bromacetamide was used to
make a 1l,10-bromohydrin, which can then be dehalogenated and deoxy-
genated to the desired product.
E. RADIOLABELED DERIVATIVES
Because ivermectin (= 22,23-dihydroavermectin B\) is obtained by
catalytic reduction of avermectin B\ (vide infra), the same procedure
using tritium gas conveniently affords tritiated ivermectin (22,23- 3[H]-
xS'
H:" "'0xS'
J
He
I"
-'I
RO",
3 0",
8,9 - OXIDE 8,9 - DIHYDROXY
FIGURE 1.6. Avermectin B\ 8,9-oxide and its hydrolysis to the 8,9-diol. (Adapted
from Smith 1985.)
4,OCH3 C2H!lJi 4'OCH3
H "C:--CH3
~-H .,' .CH3
OCH3~~1 CH3 0~HCH3
CH3~ OJ:H 0'''fi3I4~ 'H CH3~ oJ.-: H 0''''r;3f4~
HO~ Iii HO~· li2
CH30 CH30 CHi
H2 (1 ATM)(Ph3PbRhCI
OH
~\
0,\
CH3
C-4S: H!l 83.37dd C-4 R: H!l 83.89t

J4.!l=10Hz, Ju=3Hz J4,!lKJ!l,'. 4.5 Hz
FIGURE 1.7. Synthesis of ivermectin by selective hydrogenation of avermectin B 1 • (Adapted from Chabala et aI. 1980.)
1. Chemistry 15
22,23-dihydroavermectin B\). The preparation of a tritiated derivative

containing a 22,23-double bond starts with the readily available 5-ketone,
which is reduced with 3[H]-sodiumborohydride stereospecifically to a
5-3[H]-derivative (Chabala, Rosegay, and Walsh 1981). Carbon-14-
labeled avermectins can be obtained by a biosynthetic process using
sodium (l-\4C)propionate as labeled precurser (Ku and Hwang 1985).
F. ALKYLATIONS
Because avermectins are sensitive to strong base, alkylations are difficult.

In the presence of silver oxide, however, the more reactive alkylating
reagents (such as methyl iodide) react readily at the 5-hydroxy group, and
more slowly at the remaining hydroxy groups (including the tertiary
7-hydroxy) (Fisher, Lusi, and Tolman 1980).
G. INTERCONVERSIONS OF A VERMECTINS
Methylation of avermectins B\ and B2 leads to the corresponding

derivatives of the A series. A procedure involving the oxidation of the
5-methoxy group with mercuric acetate and NaBH4 reduction of the
5-keto-intermediate allows the conversion of the A to the B components
(Mrozik, Eskola, and Fisher 1986). The 23-hydroxy group of the "2"
components-after selective protection of the other secondary hydroxy
groups-is converted to a thionocarbonate, which can be eliminated to
give the 22,23-double bond of the "1" components; alternatively, it can
be reduced with tributyltin hydride to the 22,23-dihydro derivatives
(= ivermectins) (Mrozik, Eskola, and Fisher 1982).
H. GLYCOSIDE SYNTHESES
Avermectin aglycones, monosaccharides, and the naturally occurring
disaccharides themselves have been further modified by attaching various
sugars to the different hydroxy groups. Most of these methods have used
I-bromo sugars via the Konigs-Knorr procedure (Fisher and Tolman
1979, 1980). More recently, the use of I-phenylthio and I-fluoro sugars
has resulted in better yields of those glycosides (Nicolaou et al. 1984).
There are also derivatives of avermectin B\ containing fluoro groups in
the sugars (Bliard et al. 1987). A mixture of the two anomeric methyl
glycosides of oleandrose is obtained by acid methanolysis of avermectins,
and these have been converted to L-oleandrose (Els, Celmer, and Murai
1958) and the I-phenylthio glycoside (Nicolaou et al. 1984). The 4-a-L-
oleandrosyl-L-oleandrose disaccharide was obtained through an oxida-
tive degradation of avermectin B\ (Hanes sian et al. 1987). L-Oleandrose
and the 4-alfa-L-oleandrosyl-L-oleandrose disaccharide was also ob-
tained by total synthesis (Wuts and Bigelow 1983; Danishefsky et al.
1987).
IV. Structure-Activity
The avermectins, when administered orally to sheep, demonstrate activ-
ity against gastrointestinal parasites. A study was done to compare the
anthelminitic activities of the principal naturally occurring avermectins,
the monosaccharides and aglycones, and some reduced derivatives
(Chabala et al. 1980). The data are shown in Table 1.3.
In general, compounds of the B-series, containing a 5-hydroxy group,
are more potent than those of the A-series, containing a 5-methoxy group.
The I-series and 2-series have similar potency against many parasites, but
the 2-series compounds showed some important deficiencies which made
them less interesting. For example, avermectin B2 was inactive against
adult Haemonchus contortus in sheep at 0.1 mg/kg per os. Reduction of
the 22,23-0Iefin had only a small effect on potency but conferred sufficient
overall improvement in spectrum and safety that 22,23-dihydro avermec-
tin B) was selected for commercial development under the nonproprie-
tary name ivermectin. The monosaccharides were two to fourfold less
active than the disaccharide, and dihydroavermectin B) aglycone was
thirtyfold less potent.
A study of the activities of various acylated avermectins against
Trichostronglyus colubriformis in gerbils is shown in Tables 1.4 and 1.5
(Mrozik et al. 1982a).
The 4"-O-acetates have the same potencies as the un substituted com-
TABLE 1.3. Activity of avermectin derivatives against adult gastrointestinal

helminths of experimentally infected sheep on oral administration.
Efficacy'
Structureb Dose, mg/kg H.c. c O.c. T.a. T.c.' C.spp O.c.
Al 0.1 2 2 0 0 2 0
A2 0.1 3 3 3 3 0 3
BI 0.05 3 3 3 3 3 3
B2 0.1 0 3 3 3 3 3
H2AI 0.3 3 2 0 1 0 3
H2BI 0.1 3 3 3 3 3 3
BIMS 0.15 2 2 3 3 3 0
B2MS 0.2 I I 3 3 3 3
H 2B IMS 0.3 3 3 3 3 2 3
H 2B IAG 3.0 I 2 3 3 I 3
H4BI 0.2 0 0 I 0 0 3
, 0 = <50%, I = 50-74%, 2 = 75-90%, 3 = > 90% efficacy. Abbreviations used: H.c.,
Haemonchus contortus; O.c., Ostertagia circumcinta; T.a., Trichostrongylus axei; T.c.,
Trichostrongylus colubriformis; C.spp., Cooperia spp; O.c., Oesophagostomum colum-
bianum.
b MS = monosaccharide, AG = aglycone, H2 = 22,23-dihydro derivative, H4 = 3,4,22,23-
tetrahydro derivative. C Benzimidazole resistant.
Reprinted from Chabala et al. 1980, by permission of The American Chemical Society.
1. Chemistry 17
TABLE 1.4. Derivatives of avermectin A2a and B2a and anthelmintic activity
against trichostrongylus colubriformis in gerbils.
Anthelmintic
No. ~" R2 Rn act.'
1 H CH l H 0.005
3 H H H 0.0125
4 CHlCO CH3 H 0.0625
5 H CH 3 CH 3CO 0.25
6 CH 3CO CH3 CH3CO 0.5
10 CH3CO CH3CO CH 3CO 0.5
Reprinted from H. Mrozik et al. 1982a, by permission of The American Chemical Society.
pounds, whereas 5- and 23-acylation are generally detrimental to activity.

In particular the 5-hydroxy group is essential for good anthelmintic
activity, as illustrated by the diminished activity of the 5-methoxy and
5-0-t-butyldimethylsilyl derivatives.
The efficacy of acylated avermectins was also investigated against a
broad spectrum of parasites in sheep as shown in Table 1.6 (Mrozik et al.
1982a).
The T. colubriformis sheep results closely parallel the relative poten-
cies in gerbils.
The activities of avermectin Bb its 2-epimer, and the conjugatively
stabilized 6. 2 isomer were compared against the two-spotted spider mite
(Pivnichny, Shim, and Zimmerman 1983). The data are shown in Ta-
ble 1.7.
The 2-epimer is approximately Ytoo as potent as avermectin Bb whereas
the 6. 2 isomer is Ys as potent.
TABLE 1.5. Derivatives of avermectin B'a and anthelmintic activity against

Trichostrongylus colubriformis in gerbils a •
Anthelmintic
No. ~" Rs actb
2 H H 0.025
7 CH3CO H 0.031
8 H CH3CO 0.125
9 CH3CO CH3CO 0.25
12 H SI(CH3hC(CH 3h >2.5
13 Si(CH3hC(CH3h Si(CH3hC(CH3)3 >2.5
14 (CH3hCCO Si(CH3hC(CH3)3
15 (CH3)3 CCO H 0.5
16 CH3(CH2)6CO Si(CH3hC(CH3)3
17 CH3(CH2 )6CO H 0.125
18 CCl3CH2OOCCH2CH2CO Si(CH3hC(CH3)3
19 CCl3CH2OOCCH2CH2CO H
20 HOOCCH2CH2CO H 0.03
21 (4-N02C6H 4O)CO Si(CH3hC(CH3)3
22 H 2NCO Si(CH3hC(CH3)3
23 H 2NCO H 0.025
24 (CH3hNCO Si(CH3hC(CH 3h
25 (CH3hNCO H 0.06
26 CH 3CONHCH2CO Si(CH3hC(CH 3h
27 CH 3CONHCH2CO H 0.025
a From Mrozik et al. 1982a.
b Minimal doses (mg/kg needed to remove > 83% of the worm burden.
TABLE 1.6. Anthelmintic efficacy by oral administration in experimentally infected sheepa.
H.c. O.c. C.o.
See tables 4 & 5 Dose No. of
treatment mg/kg sheep EL4 Adult EL4 Adult T.a. T.c. EL4 Adult Oe.c.
none 6 (7W (442) (1421) (1137) (2852) (4110) (193) (1761) (61)
1 0.1 3 3c 3 0 3 3 3 1 0 3
0.05 3 2 2 0 2 2 2 0 2
0.025 3 0 0 2 0 0 0 2
4 0.25 3 3 3 3 3 3
5 0.25 1 0 0 0 2 1
6 0.25 1 3 2 1 2 2 3
2 0.1 3 3 3 3 3 3 3 3 3 3
0.05 3 3 3 3 3 3 3 3 3 3
0.025 3 3 3 1 2 2 2 0 1 3
7 0.1 3 3 2 3 3 3 3 2 3
0.05 3 3 0 3 3 3 2 0 3
0.025 3 3 0 3 1 3 2 0 3
9 0.25 1 3 3 3 3 3 3
0.15 2 3 3 3 3 3 3
20 O.lsc 3 0 2 3 3d 3d 3
a H.c. = Haemonchus contortus; O.c. = Ostertagia circumcincta; T.a. = Trichostrongylus axei; T.c. = Trichostrongylus colubriformis; C.o. = Coop-
eria oncophora; Oe.c. = Oseophagostomum columbianum.
b Geometric mean of the number of worms per untreated experimentally infected lamb, representative of' 'typical" infections encountered under standard
experimental procedures.
c 3 = > 90% efficacy; 2 = 60-89% efficacy; I = 20-59% efficacy; 0 = 0-19% efficacy.
d C. curticei.
TABLE 1.7. Efficacy of avermectin isomers against

Tetranichus urticae in a foliage assay on bean leaves.
Mortality, %
6.25 1.25 0.25 0.05 0.01 LC90
Compound ppm ppm ppm ppm ppm ppm
1 100 66 0.038
2 100 78 17 3 4.0
3 100 100 97 35 0.23
1: Avermectin B I; 2: the 2-epimer, 3: the conjugatively stabilized /:::,.2
isomer
Adapted from Pivnichny et al. 1988.
It has been reported elsewhere (Fisher 1985) that the aromatic deriv-
ative of avermectin B] is devoid of biological activity.
A comparison of the activities ofavermectins in anthelmintic, miticidal,
and insecticidal assays has been reported Mrozik 1985b). The results are
shown in Table 1.8.
These data illustrate the complex structure-activity relationships
among the avermectins when measured over several species. For exam-
ple, while 22,23-dihydroavermectin B] monosaccharide is three to four-
fold less active than avermectin B] against helminths, it is sixteen fold
more active than avermectin B] against the southern armyworm.
TABLE 1.8. Biological activities of abamectin and ivermectin monosaccharide

and aglycone derivatives.
In vivo Two-spotted Southern
anthelmintic spider mite armyworm
assay assay assay
A vermectins ED90 (mg/kg)' EC90 (ppm)b EC90 (ppm)C
BI (abamectin) 0.03 0.03 8.0
BI-monosaccharide 0.1 <0.01 8.0
13-deoxy-B I-aglycone 0.06 <0.005
22,23-dihYdro-B I(ivermectin) 0.03 0.05 8.0
22,23-dihydro-B I monosaccharide 0.1 0.5
22,23-dihydro-B I aglycone 0.2 >0.1
13-deoxy-22,23-dihydro-B 1- 0.06 0.05 0.5
aglycone
Adapted from Mrozik 1985b.
REFERENCES
Albers-SchOnberg G, Arison BH, Chabala IC, Douglas, AW, Eskola P, Fisher

MH, Lusi A, Mrozik H, Smith IL, Tolman RL (1981) Avermectins. Structure
determination. J. Am. Chern. Soc. 103:4216-4221
Bliard C, Escribano FC, Lukacs G, Olesker A, Sarda P (1987) Synthesis of
1. Chemistry 21
C-2"beta- and C-2"alfa-fluoro avermectin B 1a . J. Chern. Soc., Chern. Cornrnun.

386-370
Bull DL, Ivie GW, MacConnell JG, Gruber VF, Ku CC, Arison BH, Stevenson
JM, VandenHeuvel WJA (1984) Fate of avermectin B1a in soil and plants.
J. Agric. Food Chern. 32:94-102
Chabala JC, Mrozik H, Tolman RL, Eskola P, Lusi A, Peterson LH, Woods MF,
Fisher MH (1980) Ivermectin, a new broad-spectrum antiparasitic agent.
J. Med. Chern. 23:1134-1136
Chabala JC, Rosegay A, Walsh MAR (1981) Tritium labeling of avermectins B1a
and B2a via stereoselective reduction of 5-ketoavermectins. J. Agric. Food
Chern 29:881-884
Chiou R, Stubbs RJ, Bayne WF (1987) Determination of ivermectin in human
plasma and milk by high-performance liquid chromatography with fluorescence
detection. J. Chrornatogr. 416:196-202
Chiu SHL, Buhs RP, Sestokas E, Taub R, Jacob TA (1985) Determination of
ivermectin residue in animal tissue by high-performance liquid chroma-
tography-reverse isotope dilution assay. J. Agric. Food Chern. 33:99-102
Danishefsky SJ, Selnick HG, Armistead DM, Wincott FE (1987) The total
synthesis of avermectin Ala. New protocols for the synthesis of novel 2-
deoxypyranose systems and their axial glycosides. J. Arn. Chern. Soc.
109:8119-8120
Davies HG, Green RH (1986) Avermectins and milbemycins. Nat. Prod. Rep.
3:87-121
Els H, Celmer WD, Murai K (1958) Oleandomycin (PA-105). II. Chemical
characterization (I). J. Arn. Chern. Soc. 80:3777-3782
Fisher MH (1985) Recent Advances in the Chernistry of Insect Control. Janes NF
(ed), Royal Soc. of Chern. Special Publication No. 53, p. 53
Fisher MH, Lusi A, Tolman RL (1980) Alkyl derivatives of C-076 compounds.
U.S. Patent 4,200,581
Fisher MH, Mrozik H (1984) The avermectin family of macrolide-like antibiotics.
In Omura S (ed), Macrolide Antibiotics, Academic Press, New York, pp
553-606
Fisher MH, Tolman RL (1979) Carbohydrate derivatives of milbemycins and
processes therefor. U.S. Patent 4,134,973
Fisher MH, Tolman RL (1980) Sugar derivatives of C-076 compounds. U.S.
Patent 4,203,976
Gallo VP, Kempf AJ, MacConnell J, Mrozik H, Arison B, Putter I (1983) The
microbial formation of the 23-keto derivative from avermectin B2a in soil.
Pestic. Sci. 14:153-157
Hanessian S, Ugolini A, Hodges PJ, Beaulieu P, Dube D, Andre C (1987) Progress
in natural product chemistry by the chiron and related approaches-synthesis
of avermectin B 1a . Pure & Appl. Chern. 59:299-316
Ku CC, Hwang SC (1985) The preparation of carbon-14 labeled avermectin Bla.
J. Labelled Cornpd. Radiopharrn. 22:451-459
Miller TW, Chaiet L, Cole DJ, Cole LJ, Flor JE, Goegelman RT, Gullo VP,
Joshua H, Kempf AJ, Krellwitz WR, Monaghan RL, Ormond RE, Wilson KE,
Albers-SchOnberg G, Putter I (1979) Avermectins, new family of potent
antihelminthic agents: isolation and chromatographic properties. Antirnicrob.
Agents Chernother. 15:368-371
Mishima H, Kurabayashi M, Tamura C, Sato S, Kuwano H, Saito A (1975)

Structures of Milbemycin betal, beta2, and beta3. Tetrahedron Leu. 711-714
Mrozik H (1984a) 4"-keto- and 4"-amino-4"-deoxy avermectin compounds and
substituted amino derivatives thereof. U.S. Patent 4,427,663
Mrozik H (1984b) 4a-Substituted avermectin compounds. U.S. Patent 4,457,920
Mrozik H (1985a) Avermectin epoxide derivatives and method of use. U.S. Patent
4,530,921
Mrozik H, (1985b) Chemistry and Biological Activities of Avermectin Derivatives
in Biotechnology and Its Application to Agriculture. Copping L (ed), British
Crop Protection Council (BCPC) Monograph No. 32, pp. 133-143.
Mrozik H, Chabala JC, Eskola P, Matzuk A, Waksmunski F, Woods M, Fisher
MH (1983) Synthesis of milbemycins from avermectins. Tetrahedron Lett.
24:5333-5336
Mrozik H, Eskola P, Arison BH, Albers-SchOnberg G, Fisher MH (1982b)
Avermectin aglycones. J. Org. Chem. 47:489-492
Mrozik H, Eskola P, Fisher MH (1982) Partial syntheses of avermectin B la and
ivermectin from avermectin B2a . Tetrahedron Lett. 23:2377-2378
Mrozik H, Eskola P, Fisher MH (1986) Mercuric acetate oxidation of avermectin
A2a as a route to the selective cleavage of the allylic C-5-methoxy group. J. Org.
Syn. 51:3058-3059
Mrozik H, Eskola P, Fisher MH, Egerton JR, Cifelli S, Ostlind DA (1982a)
Avermectin acyl derivatives with antihelminthic activity. J. Med. Chem.
25:658-663
Mrozik H, Eskola P, Reynolds GF, Arison BH, Smith GM, Fisher MH (1988)
Photoisomers of avermectins. J. Org. Chem. 53:1820-1823
Mrozik H, Waksmunski FS (1985) C-8a-oxo-avermectin and milbemycin deriv-
atives, pharmaceutical compositions and method os use. U.S. Patent 4,547,491
Nicolaou KC, Dolle RE, Papahatjis DP, Randall JL (1984) Practical synthesis of
oligosaccharides. Partial synthesis of avermectin B l •. J. Am. Chem. Soc.
106:4189-4192
Pivnichny JV, Arison BH, Preiser FA, Shim JSK, Mrozik H (1988) Base-
catalyzed isomerisation of avermectins. J. Agric. Food Chem. 36:826-828.
Pivnichny JV, Shim JSK, Zimmerman LA (1983) Direct determination of aver-
mectins in plasma at nanogram levels by high-performance liquid chromatogra-
phy. J. Pharm. Sci. 72:1447-1450
Smith AB III, Thompson AS (1985) Avermectin-milbemycin studies. 2. An
efficient chemical degradation of avermectin B la • Tetrahedron Lett. 26:4279-
4282
Springer JP, Arison BH, Hirshfield JM, Hoogsteen K (1981) The absolute
stereochemistry and configuration of avermectin B2• aglycon and avermectin
Bla • J. Am. Chem. Soc. 103:4221-4224
Stong JD (1987) Determination of ivermectin by Fluorescence Derivatization.
Anal. Chem. 59:266-270
Takiguchi Y, Mishima H, Okuda M, Terao M, Aoki A, Fukuda R (1980)
Milbemycins, a new family of macrolide antibiotics: Fermentation, isolation
and physico-chemical properties. J. Antibiot. 33:1120-1127
Tolan JW, Eskola P, Fink DW, Mrozik H, Zimmerman LA (1980) Determination
of avermectins in plasma at nanogram levels using high-performance liquid
chromatography with fluorescence detection. J. Chromatogr. 190:367-376
1. Chemistry 23
Tway PC, Downing GV, Slayoack JRB, Rahn GS, Isensee RK (1984) Confirma-
tory assay for ivermectin in cattle tissue using chemical ionization
mass spectrometry/mass spectrometry (MS/MS). Biorned. Mass Spectrorn.
11:172-176
Tway PC, Wood JS, Downing GV (1981) Determination of ivermectin in cattle and
sheep tissues using high-performance liquid chromatography with fluorescence
detection. J. Agric. Food Chern. 29:1049-1063
Wuts PGM, Bigelow SS (1983) Total synthesis of oleandrose and the avermectin
disaccharide, benzyl alfa-L-oleandrosyl-alfa-L-4-acetoxyoleandroside. J. Org.
Chern. 48:3489-3493
CHAPTER 2
Isolation and Characterization of

the Producing Organism
R.W. Burg and E.O. Stapley
I. Source of Streptomyces avermitilis and discovery of the avermectins

II. Taxonomy of Streptomyces avermitilis
III. Isolation of the avermectins
IV. Antibacterial and antifungal activity
Of the several thousand fermentation products that had been described
through 1978, only a few were reported to have anthelmintic activity.
Among these were anthelmycin, anthelvencin, the antibiotic complex
S-15-1, aspiculomycin, the axenomycins, G-418, hygromycin B, the
destomycins, myxin, paromomycin, and the thaimycins. Only hygromy-
cin B has been used commercially; and, until the last decade, synthetic
compounds dominated the anthelmintic market. The discovery of the
avermectins changed that dramatically.
I. Source of Streptomyces avermitilis and

Discovery of the Avermectins
The microorganism that produces the avermectins was isolated from a
soil sample collected near a golf course at Kawana, Ito City, Shizuoka
Prefecture, Japan. Research scientists at the Kitasato Institute in Tokyo,
under the direction of Dr. Satoshi Omura, isolated this culture. This group
is noted for its skill in isolating exceptional microorganisms from soil;
and, as part of ajoint research project between the Kitasato Institute and
Merck Sharp & Dohme Research Laboratories (MSDRL), the cultures
isolated at the Kitasato Institute are tested in MSDRL's screening
programs. These cultures quickly gained a reputation for being a rich
source of new activities, and they are used to test many new screening
assays. One of the new assays being developed at MSDRL was based on
the administration of test substances to mice infected with Nematospi-
roides dubius, and it was decided to test all of the Kitasato cultures in that
assay.
2. The Producing Organism 25
The avermectin-producing strain was among one group of 50 Kitasato

cultures grown at MSDRL and screened by the N. dubius assay. This
strain, originally designated OS-3153, was active but quite toxic in the
first assay and in a confirmatory assay. In subsequent tests, however, this
culture exhibited good activity with little or no toxicity. The unknown
activity received the code designation C-076, and the culture was entered
into the MSDRL culture collection as MA-4680. Several accounts of the
discovery of the avermectins have been published (Burg et al. 1979;
Campbell et al. 1984; Stapley and Woodruff 1982; and Woodruff and Burg
1986).
II. Taxonomy of Streptomyces avermitilis

The taxonomy of the producing organism, MA-4680, was established
using the classical method of Shirling and Gottlieb (1966) and a number of
published taxonomic keys (Buchanan and Gibbons 1974; Shirling and
Gottlieb 1968a, 1968b, 1969, 1972; Waksman 1961). The taxonomic
description has been published (Burg et al. 1979) as follows :
Taxonomic studies of culture MA-4680 revealed it to be a Streptomyces species
with sporophore forming spirals as side branches on aerial mycelia. The spirals
are compact but become open as the culture ages (Figure 2.1). Spores were in
chains of more than 15 and were usually spherical to oval. The spore surface was
smooth as revealed by electron microscopy (Figure 2.2). Sporulation occurred on
oatmeal agar, glycerol-asparagine agar, inorganic salts-starch agar, and egg
albumin agar. The culture grew well at 28° and 37° C but not at 50° C. These
characteristics, plus those summarized in Tables 2.1 and 2.2, were compared with
those of known species in the classical references. . . . No Streptomyces species
FIGURE 2.1. Scanning electron micrograph

of S. avermitilis MA-4680 showing spo-
rophores forming compact to open spirals.
26 R.W. Burg and E.O. Stapley
FIGURE 2.2. Transmission electron micro-

graph of S. avermitilis MA-4680 showing a
sporophore with an open spiral and the
smooth surface of the spherical to oval
spores.
is described as having the brownish-gray spore mass color, smooth spore surface,
spiral sporophore structure, production of melanoid pigments and cultural
and carbon utilization patterns that together comprise the distinctive species
S. avermitilis.
The culture was named Streptomyces avermitilis because of its ability to
produce anthelmintic activity.
III. Isolation of the Avermectins

The avermectins were isolated by solvent extraction from the mycelia,
solvent partition, and column chromatography on Sephadex LH-20
(Miller et al. 1979). Early work on the isolation of the avermectins
proceeded slowly since it had to be guided by the mouse assay. During the
isolation, the active extracts were monitored by specrophotometry; and
the isolation chemists soon realized that the anthelmintic activity was
correlated with a characteristic ultraviolet absorption spectrum (Miller et
al. 1979). The examination of highly concentrated preparations by
thin-layer chromatography (TLC) revealed the presence of four active
components that quenched under ultraviolet light. This TLC assay guided
the subsequent isolation and early work on improvement of the fermenta-
tion (Burg et al. 1979). Rapid progress in both areas followed the
development of this physical assay. Further improvements in the assay by
the use of high-performance liquid chromatography (HPLC) led to the
identification and ultimate isolation of the 8 different compounds that
comprise the original avermectins (Miller et al. 1979).
TABLE 2.1. Cultural characteristics of S. avermitilis MA-4680 (NRRL 8165)a [Burg et al. 1979].
Medium Vegetative growth Aerial mycelia Soluble pigment
Oatmeal agar (ISP medium 3)h Reverse-very dark brown Powdery, brownish gray (4Ii)C Brown
mixed with white
Czapek-Dox agar Poor, colorless Scant, grayish Light grayish tan
Egg albumin agar Tan Moderate, light grayish yellowish Light yellowish brown
brown (3ge)C mixed with white
Glycerol-asparagine agar (ISP Reverse-yellowish brown Powdery, brownish gray (4li)C Light yellowish brown
medium 5)h mixed with white
Inorganic salts-starch agar (ISP Reverse-grayish-yellowish brown Powdery, light brownish gray Light yellowish brown
medium 4)b (4ig)C edged with darker
brownish gray (4Ii)C
Yeast extract-dextrose-salts agar Reverse-dark brown Moderate, brownish white Brown
Yeast extract-malt extract agar Reverse-dark brown Moderate, brownish white Brown
(ISP medium 2)b
Peptone-iron-yeast extract agar Dark brown None Dark brown to black
Nutrient agar Tan Sparse, grayish Light brown
Nutrient starch agar Tan Sparse, grayish white Light brown
Nutrient gelatin agar Tan Sparse, grayish white Light brown
Gelatin stab Brown ring None Greenish brown
Skim milk agar Dark brown None Dark brown
Litmus milk Dark brown growth ring None Dark brown
Skim milk Dark brown growth ring None Dark brown
Potato plug Tan Brown mixed with grayish white Grayish brown
Loeffler blood serum Grayish tan None Some browning of medium
Nutrient tyrosine agar Reverse-dark brown to black Sparse, grayish Dark brown
a All readings taken after 3 weeks at 280 C.
b See Shirling and Gottlieb 1966
CColor number designations were taken from Color Harmony Manual, 4th ed., Container Corp. of America, Chicago, 1958.
~
TABLE 2.2. Physiological properties of S. avermitilis MA-4680 (NRRL 8165)

[Burg et al. 1979].
Characteristic Observation a
Carbohydrate utilizationb Good growth: glucose, arabinose, fructose,
inositol, lactose, maltose, mannitol, mannose,
raffinose, rhamnose, sucrose, xylose
No growth: cellulose
Melanin production C Positive
H 2S production C Positive
Hydrolysis of starch d Good
Liquefaction of gelatine Good
Hydrolysis of caseinf Good
Peptonization of milkg Complete, becoming alkaline (pH 8)
Liquefaction of Loeffler blood serum None
Decomposition of tyrosine h None
a All readings were taken after 3 weeks at 28 0 C.
b Pridham-Gottlieb basal medium plus 1% carbon source (Shirling and Gottlieb 1966)
C Peptone-iron-yeast extract agar (Shirling and Gottlieb 1966)
d Nutrient starch agar (Waksman 1961)

e Nutrient gelatin agar; gelatin stab (Waksman 1961)
f Skim milk agar (Waksman 1961)
g Litmus milk; skim milk (Waksman 1961)
h Nutrient tyrosine agar (Waksman 1961)
IV. Antibacterial and Antifungal Activity

The antibacterial spectrum of whole broth was assayed early in the study
of S. avermitilis. The results of these first assays of the culture grown on
two media (KH and JH), along with a similar assay of abamectin
performed at a later date, are shown in Table 2.3 (SB Zimmerman 1975,
1984: private communication). The whole broth exhibited no detectable
antibacterial activity except as a zone against Bordetella bronchiseptica
(MB-965), an extremely sensitive bacterium. From the anthelmintic
activity, the KH broth was estimated to contain at least 9 ILg/ml of
avermectins (Burg et al. 1979). The JH broth had only slight anthelmintic
activity. At 500 ILg/ml, abamectin gave small zones against Bacillus sp.
633 and Micrococcus luteus. At 1 mg/ml, a few other bacterial strains
began to exhibit sensitivity. It is safe to conclude that the avermectins do
not have significant antibacterial activity except at extremely high con-
centrations. Extensive studies have failed to discover any microorganism
that could be used to assay for the avermectins.
The antifungal spectrum of whole broth was not determined as early
in the studies of S. avermitilis (MA-4680), although a later isolate,
S. avermitilis MA-5856, produced considerable activity against fila-
mentous fungi. Table 2.4 illustrates the antifungal activity of S. avermitilis
MA-5856 and the insignificant antifungal activity of pure abamectin (G
TABLE 2.3. Antibacterial activity of OS-3153 and abamectin.

Zone diameter (mm)
OS-3153 b Abamectin C
Bacterium" KH med. JH med. 0.5 mg/ml 1.0 mg/ml
Alcaligenes faecalis MB-lO 0 0
Bacillus sp. MB-633 0 0 9 0
Bacillus subtilis MB-964 0 0 0 0
Bordetella bronchiseptica MB-965 13W 20 0 0
Comamonas terrrigena MB-1272 0 0 0 0
Comamonas terrigena MB-2566 0 0 0 0
Corynebacterium 0 0 0 0
pseudodiphtheriticum MB-261
Enterobacter aerogenes MB-835 0 0 0 0
Erwinia carotovora subsp. 0 0 0 0
atroseptica MB-1159
Escherichia coli MB-1418 0 0 0 0
Klebsiella pneumoniae MB-1264 0 0 0 0
Micrococcus lute us MB-lOll 0 0 10 10
Proteus mirabilis MB-3126 0 0
Proteus vulgaris MB-1012 0 0 0
Proteus vulgaris MB-2112 0 0 0 0
Proteus vulgaris MB-838 0 0 0 0
Pseudomonas aeruginosa MB-2824 0 0 0 0
Pseudomonas aeruginosa MB-979 0 0 0 0
Pseudomonas stutzeri MB-1231 0 0 0 0
Salmonella gallinarum MB-1287 0 0 0 0
Serratia marcescens MB-252 0 0 0 0
Staphylococcus aureus MB-1909 0 0
Staphylococcus aureus MB-2756 0 0 0 13
Streptococcus agalactiae MB-2875 0 0 0 0
Streptococcus faecalis MB-753 0 0
Streptococcus faecium MB-2820 0 0 0 0
Streptomyces sp. MA-4798 0 0
Xanthomonas campestris pv. 0 0 0 0
vesicatoria MB-815
"All cultures were grown on nutrient agar + 0.2% yeast extract except Streptococcus
faecium MB-2820 and Streptococcus agalactiae MB-2875, which were grown on brain heart
infusion agar. Incubation was at 25° or 37° C.
b Filter paper discs (6.2 mm) were dipped into whole broth and applied to the assay plates.
C Filter paper discs (6.2 mm) were dipped into acetone solutions of abamectin. The disks
were air dried before applying them to the assay plates.

d Not determined.
e H = hazy zone
TABLE 2.4. Antifungal activity of MA-5865 and abamectin.

Zone diameter (mm)a
Abamectin C
Organism MA5856 b 1.0 mg/ml
Filamentous fungi
Alternaria solani MF-3550 20F llH
Aspergillus niger MF-442 19F 0
Aspergillus niger MF-ll 15F 0
Botrytis allii MF-3587 llH 0
Cephalosporium sp. MF-4641 13F 0
Ceratocystis ulmi MF-4042 19F 0
Cladosporium sp. MF-5015 0 0
Cochliobolus miyabeanus MF-4626 23F 16F
Fusarium oxysporum MF-4014 13H 0
Penicillium sp. MF-5020 10H 0
Penicillium sp. MF-5016 15F 0
Penicillium sp. MF-5014 17F 0
Phoma sp. MF-4332 20F 0
Scopulariopsis communis MF-3769 0 0
Trichoderma lignorum MF-3560 17H 0
Trichoderma sp. MF-4064 18F 0
Ustilago maydis MF-1996 14 0
Verticillium serrae MF-3794 0 0
Yeasts
Brettanomyces bruxellensis MY-315 0
Candida albicans MY-992 15H 0
Candida albicans MY-I099 0
Candida albicans MY-I028 9V
Candida guilliermondi MY-IOI9 0
Candida pseudotropicalis MY-lloo 0
Candida rugosa MY-I022 0
Candida tropicalis MY-IOl2 0
Cryptococcus albidus MY-I070 0
Cryptococcus laurentii MY-I073 0
Cryptococcus laurentii MY-I077 0
Cryptococcus laurentii MY-I074 8V
Kluyvermyces fragilis MY -1l13 0
Saccharomyces cerevisiae MY-34 0
Saccharomyces cerevisiae MY-410 0
Torulopsis glabrata MY-I062 0
Torulaspora hansenii MY-321 0
a F = Fuzzy edge, S = Slightly hazy, H = Hazy, V = Very hazy,
- = Not determined.
bWhole broth was applied to 6.2 mm filter paper discs.
CAbamectin in 25% aqueous DMSO was applied to 6.2 mm filter
paper discs (25 ul/disc) and air dried before applying to the seeded
assay plates.
Garrity 1988: private communication}. A report that avermectin had

antifungal activity (Caicott and Fatig 1984) proved erroneous, probably
due to the authors' use of a crude avermectin preparation. In refuting this
work, Onishi and Miller (1985) reported that pure avermectin B 1a lacks
any detectable antifungal activity at 400 ILg/ml and that S. avermitilis
produces both a polyene and oligomycin. In fact, the first paper on the
avermectins (Figure 2 in Burg et al. 1979) described a "Peak D" on a
densitometer tracing of a thin-layer chromatogram of a mycelial extract of
S. avermitilis MA-4680 as "a fermentation product unrelated to the
avermectins." This product had already been identified as oligomycin (T
Miller 1976: personal communication).
REFERENCES
Buchanan RE, Gibbons NE (eds) (1974) Bergey's Manual of Determinative

Bacteriology, 8th ed, The Williams & Wilkins Co., Baltimore
Burg RW, Miller BM, Baker EE, Birnbaum J, Currie SA, Hartman R, Kong Y-L,
Monaghan RL, Olson G, Putter I, Tunac JB, Wallick H, Stapley EO, Oiwa R,
Omura S (1979) Avermectins, new family of potent anthelmintic agents:
producing organism and fermentation. Antimicrob. Agents Chemother. 15:361-
367
Calcott PH, Fatig RO III (1984) Inhibition of chitin metabolism by avermectin in
susceptible organisms. J. Antibiot. 37:253-259
Campbell WC, Burg RW, Fisher MH, Dybas RA (1984) The discovery of
ivermectin and other avermectins. In Magee PS, Kohn GK, Menn JJ (eds),
Pesticide Synthesis through Rational Approaches, American Chemical Society,
Washington, DC, pp 5-20
Miller TW, Chaiet L, Cole DJ, Cole LJ, Flor JE, Goegelman RT, Gullo VP,
Joshua H, Kempf AJ, Krellwitz WR, Monaghan RL, Ormaond RE, Wilson KE,
Albers-SchOnberg G, Putter I (1979) Avermectins, new family of potent
anthelmintic agents: isolation and chromatographic properties. Antimicrob.
Agents Chemother. 15:368-371
Onishi JC, Miller TW (1985) The lack of antifungal activity by avermectin B1a •
J. Antibiotics 38:1568-1572
Shirling EB, Gottlieb D (1966) Methods for characterization of Streptomyces
species. Int. J. Syst. Bacteriol. 16:313-340
Shirling EB, Gottlieb D (1968a) Cooperative description of type cultures of
Streptomyces II. Species descriptions from first study. Int. J. Syst. Bacteriol.
18:69-189
Shirling EB, Gottlieb D (1968b) Cooperative description of type cultures of
Streptomyces III. Additional descriptions from first and second studies. Int.
J. Syst. Bacteriol. 18:279-392
Shirling EB, Gottlieb D (1969) Cooperative description of type cultures of
Streptomyces IV. Additional descriptions from second, third and fourth stud-
ies. Int. J. Syst. Bacteriol. 19:391-512
Shirling EB, Gottlieb D (1972) Cooperative description of type cultures of
Streptomyces V. Additional descriptions. Int. J. Syst. Bacteriol. 22:265-394
Stapley EO, Woodruff HB (1982) Avermectins, antiparasitic lactones produced
by Streptomyces avermitilis isolated from a soil in Japan. In Umezawa H,

Demain AL, Hata T, Hutchinson CR (eds), Trends in Antibiotic Research,
Japan Antibiotics Research Association, Tokyo, pp 154-170
Waksman SA (1961) The Actinomycetes, Vol 2, The Williams & Wilkins Co.,
Baltimore
Woodruff HB, Burg RW (1986) The antibiotic explosion. In Parnham MJ,
Bruinvels J (eds) , Discoveries in Pharmacology, Vol 3: Pharmacological
Methods, Receptors & Chemotherapy, Elsevier Science Publishers, Amster-
dam, pp 303-351
CHAPTER 3
Fermentation Development and

Process Improvement
Mary Nallin Omstead, Louis Kaplan, and
Barry C. Buckland
I. Introduction
II. Process Improvement-Shake Flask Scale
A. Culture Development
1. Mutational Procedures
2. Isolation/Detection of Mutants
a. morphological variants
b. nonproducers
c. auxotrophs
d. altered structural mutants
e. compositional mutants
f. superior producers
3. Cross-Feeding Studies
4. Cultural Genealogy
B. Seed Media Development
C. Production Media Development
1. Medium Constitutent Substitutions
2. Medium Optimization Studies
a. titration studies
b. surface response methodology
III. Process Scale-Up
A. Fermentation Kinetics: Stirred Tanks
B. Model to Predict Cell Mass On-Line
C. Oxygen Transfer
1. Correlation of Power Input to Oxygen Transfer Coefficient
2. Use of Hydrofoil Impellers to Improve Oxygen Transfer
Coefficient
D. Strategy Used for Successful Scale-Up
IV. Downstream Extraction
34 Mary Nallin Omstead, Louis Kaplan, and Barry C. Buckland
I. Introduction
Streptomyces avermitilis was originally isolated from soil as described in
Chapter 2. The first fermentations of the culture used a standard screening
approach: a loopful of slant growth was inoculated into a seed flask which
was incubated and then transferred into two types of production flasks.
Seed cultures were grown for 48 to 72 hours in baffled flasks containing
50 ml seed medium I (Burg et al. 1979). Production flasks were inoculated
by transferring 2.0 ml of seed growth into 40 ml production medium A
(Burg et al. 1979) and incubated for 4 days. The production cultures were
then harvested and assayed for activity as indicated in Chapter 2.
This chapter describes the development of an optimized process for the
large-scale fermentation of Streptomyces avermitilis, beginning with the
original strain. Improvements in the process involved mUltiple pa-
rameters-including strain improvement, seed and production media
development, and modification of physical fermentation parameters (such
as aeration, temperature, pH, and length of incubation). Furthermore,
scaling up the avermectin process to large fermentation tanks demanded
the optimization of yet another set of parameters-such as oxygen
transfer and media sterilization.
II. Process Improvement-Shake Flask Scale

A. CULTURE DEVELOPMENT
After the discovery that the avermectins possess anthelmintic activity-
and the intense interest that fact elicited-it became an obvious goal to
develop and improve the avermectin fermentation process. Process
improvement can and usually does proceed along several different
avenues of investigation. In the case of the avermectins, various fermen-
tation parameters were studied simultaneously. These parameters in-
cluded developing the S. avermitilis strain itself, modifying and/or
altering the composition of the seed and production media constituents,
and determining the nutritional and physiological requirements of the
organism. To perform the latter studies, one must develop either an
adequate synthetic medium or a resting cell system. In either case, the
system must be capable of supporting a basal level of product formation
against which gains or losses can be measured.
To develop the S. avermitilis strain, several mutational procedures
were examined and adapted to obtain a number of different mutant
cultures. As a rule, these procedures were performed on spores sus-
pended in O.lM HEPES buffer, pH 7.4. However, several ofthe methods
also were used to treat washed vegetative mycelia suspended in various
media. Suspensions of treated cultures were plated on complex media to
guarantee optimal recovery of survivors. Survival curves were deter-
3. Process Development and Improvement 35
mined for all mutations. The frequency of mutation was determined by

calculating the degree of morphological variation induced by the muta-
tion, the frequency of induced auxotrophy, or the resistance to growth
inhibitory levels of various compounds.
The first method used to induce mutations involved irradiating spores
with ultraviolet (UV) light, varying exposure times (Clarke and Hopwood
1976). The irradiated suspensions were plated and grown in the dark to
prevent possible photoreactivation. The resulting decrease in viability
was directly related to the exposure time. Irradiation also induced
obvious mutations in some of the survivors, as evidenced by their altered
morphologies (Table 3.1). These morphological changes proved to be
stable through several reisolation (generational) events.
The second type of mutation involved the use of various alkylating
agents-including nitrosoguanidine (NTG) (Awerbuch and Stark 1979;
Godfrey 1974; Kirkpatrick and Godfrey 1973), nitrosomethylurethane
(NMU) (Awerbuch and Stark 1979), and ethylmethane sulfonate (EMS)
(Awerbuch and Stark 1979; Sega 1984). Suspensions of these mutagens
were prepared as follows: the NMU and EMS were dissolved in the same
buffer as that in which the spores were suspended, O.IM HEPES buffer,
pH 7.4; initially, the NTG was dissolved in 0.05 or O.lM Tris-maleate
buffer, pH 6.0. After it was discovered that the Tris itself was adversely
affecting the culture, subsequent experiments used HEPES-buffered
NTG solutions. As with the irradiation method, exposure times for NTG
and NMU were varied. EMS exposure times had to be prolonged because
of the apparent resistance of the S. avermitilis to this mutagen as
TABLE 3.1. Types of Streptomyces avermitilis mutants.

Relative avermectin
Types of mutants production titers'
A. Morphological variants
1 altered spore pigmentation I, R, NP
2 absence of aerial mycelia ("balds") NP
3 loss of melanin production NP
B. Nonproducing mutants NP
C. Auxotrophs (various nutritional requirements) R,NP
D. Modified avermectin-producing mutants
1 des-methyl macrolide ring I, R
2 des-methyl oleandrose R
3 open furan ring R
4 aglycone R
5 altered ratio of components I, R
• : Titers obtained from mutant strains compared to the titers of their specific
parental strain
I: improved, increased
R: reduced
NP: nonproducer
demonstrated by the low level of kill and the low frequency of morpholog-
ical variants. EMS is also a less effective mutagen for other organisms
(Sega 1984). In all procedures using alkylating agents the mutagen either
was removed from the suspension by centrifugation or was inactivated by
adding 10% sodium thiosulfate to the reaction mixture.
Treating S. avermitilis with these alkylating agents resulted in various
types of mutants. Morphological variants were easily detected. These
survivors consisted of the following types: non sporulating (i.e., "bald")
colonies, sporulating colonies which displayed altered spore pigmenta-
tion, colonies which failed to produce the melanin pigment typical of the
S. avermitilis parent, and colonies which displayed various combinations
of these characteristics. Mutants categorized within each type of variant
class exhibited a range in the ability to produce avermectin: many
exhibited lowered production as compared to the parental type, while
some had become complete nonproducers. Auxotrophic strains were also
detected and purified. They, too, varied in the amount of avermectin
produced, although generally their titers were significantly reduced.
Other mutants produced structurally modified avermectin components,
including the aglycone or monosaccharide avermectins (Ikeda, Kotaki,
and Omura 1987; Schulman et al. 1985); components which no longer
contained completely closed furan ring structures (Goegelman, Gullo, and
Kaplan 1983); and des-methyl avermectins (either on the macrolide ring
or on the oleandrose units) (Schulman etal. 1985; Schulman et al. 1986).
Once again, these mutants produced reduced total avermectin com-
ponents. In addition, other mutants were isolated which synthesized all
the avermectin components (a total of 8 structural types) (Miller et at.
1979) but produced them in different ratios than the original parental
strain.
Treating various parental strains by the procedures described resulted
in the accumulation of many mutants of each type. They were subse-
quently tested as to whether any of them were capable of cross-feeding
each other. These tests were done either by cofermenting the mutant
strains in a mixed fermentation or by feeding to one fermentation an
extract prepared from another fermentation broth. The cofermentations
were generally difficult to interpret, although they were adequate in
evaluating whether strains that had independently lost the ability to
produce melanin could regain that function if cocultured. Experiments
that involved feeding one strain with an extract from another strain
proved to be more useful. This type of feeding study enabled us to
categorize our nonproducing mutants into several classes based on their
ability to convert a given substrate (e.g., an aglycone avermectin) into an
easily distinguishable product (in the example cited, a standard avermec-
tin). In this fashion, it was possible to differentiate nonproducing mutants
even further.
There remains one final type of mutant, that which is most important to
any strain improvement program: a strain that produces significantly

more total avermectin (and more of the active BI components) than its
parent. Table 3.2 depicts the partial genealogy of improved avermectin
producers, both in terms of BI(a + b) components and in relation to the
total avermectin produced. Several features of this table warrant further
attention. Culture 1 is the original soil isolate, the first to manifest
avermectin activity (Burg et al. 1979). There was a clear gain in activity
between this and the next culture (2): a twenty-onefold increase in BI titer
and a twenty-fivefold improvement in total avermectin produced. The
levels cited in Table 3.2 represent accumulated gains in titer achieved
through strain improvement, media development, and fermentation con-
dition changes. Prior to discussing subsequent cultures, it should be
emphasized that S. avermitilis was first described as having a brownish-
gray spore mass color (Burg et al. 1979). Culture 4, and all cultures
obtained subsequently, displayed altered spore pigmentation: isolates
were blue-green. Moreover, this change in morphological appearance was
correlated with an increase in avermectin production. Through culture 7
all improved producers showed a coupled response between improved BI
titer and improved total avermectin production. Cultures 8b and 12,
however, exhibited improved BI titers but a concomitant reduction in the
other components, resulting in a smaller increase in total avermectin
levels (103X vs 91X and 108X vs 92X, respectively). Culture 12 was
particularly important because it demonstrated that one could continue to
enhance B I production while maintaining the other components at
reduced levels.
B. SEED MEDIA DEVELOPMENT

Along with developing improved avermectin-producing strains, we were
studying and modifying the fermentation process itself. A particular set of
90nditions had been defined for the original parental strain (Burg et al.
TABLE 3.2. Streptomyces avermitilis genealogy.

Fold
improvements'
Culture BIb Total Comments
I (1.0) (1.0) Original soil isolate; brownish-gray spore coloration
2 2lX 25X brownish-gray spore coloration
4 34X 38X spore pigmentation mutant (blue-green coloration)
7 56X 69X blue-green spore coloration
8b I03X 9lX blue-green spore coloration; reduced "A" components
12 108X 92X blue-green spore coloration; reduced "A" components
': As shown in this table, the fold improvement includes increases in titer due to the
following factors: strain improvement, seed and production media modifications, and
changes in physical fermentation parameters
b: BI refers to the combined titers for B I, + BIb
1979). This set specified a seed medium (II) grown for 24 to 48 hrs, with
subsequent transfer to production medium (B) which fermented at 28° C
for 5 to 6 days. As the S. avermitilis strain was being developed, we
wanted to be sure that the fermentation conditions (including media
composition, temperature, aeration, and length of fermentation) would
not become limiting to the new, improved strains. To start, we examined
the initial seed medium. As reported (Burg et al. 1979), seed medium II
contained distiller's solubles (a particulate raw material), lactose, and
Ardamine PH. Our first manipUlation of this medium involved testing
the effect of various carbon sources. Of those, cerelose is optimal for
S. avermitilis in this medium. Seed medium II is one in which cerelose
was the primary source of carbon. As shown in Table 3.3, this single seed
medium change, combined with superior mutants (note culture improve-
ment in Table 3.3), resulted in five- to elevenfold improvement in
avermectin production. The original seed medium (II) demonstrated
about the same degree of improvement (six to tenfold) based on improved
cultures alone. However, simultaneously modifying the seed and the
production media produced even better results. Seed medium III, com-
bined with production medium C, resulted in sixteen- to twenty-onefold
improved avermectin production, whereas seed medium II with produc-
tion medium C netted only thirteen- to fourteenfold improvement.
Therefore, the combined effects of culture improvement and modification
TABLE 3.3. Seed medium development.

Production media"
Seed medium Culture B C D E
II I (1.0) NDb ND ND
3 6X I4X ND ND
4 lOX 13X ND ND
5 8X ND ND ND
III 3 7X I8X ND ND
4 5X I6X ND ND
5 llX 2IX 23X ND
IV 3 IX 12X ND ND
4 IX I8X ND ND
5 2X 24X 37X 40X
7 ND ND ND 56X
8 ND ND ND 66X
V 5 ND 22X ND ND
VI 5 ND ND 29X ND
VII 5 ND ND ND 32X
7 ND ND ND 44X
8 ND ND ND 54X
a: Values shown are total fold improvements in B1(a+b) titers, including increases
due to the following factors: strain improvement, seed and production media
modifications, and changes in the physical fermentation parameters
b: Not determined
of both the seed and production media resulted in an optimal improve-

ment in avermectin formation of twenty-onefold. The degree of improve-
ment cited in Table 3.3 is relative to the avermectin titer produced by the
original strain grown in seed medium II and fermented in production
medium B.
Despite these impressive yield improvements, a problem occurred
concomitant with the change made in the seed medium. Cultures which
had been superior avermectin producers in the original fermentation
process were no longer so. Culture 4 was no longer producing higher titers
than culture 3 in production medium C, whether it was grown in seed
medium II or III (l3X vs 14X from seed II, and 16X vs 18X from seed III).
Interestingly, culture 5 demonstrated superiority only when grown in seed
III. Similarly, culture 4 displayed superior production when seed II was
fermented in production medium B but not when fermented in production
medium C (lOX vs 6X for culture 3, compared to 13X vs 14X, respec-
tively). Apparently, the ability to detect superior mutants with higher
avermectin titers was strongly dependent on the specific fermentation
conditions (seed medium, production medium combinations) being used.
Clearly, this situation demanded a seed medium that would be neither
specific to a given culture nor so strongly dependent on the subsequent
production medium. To formulate a reasonable mixture of ingredients, we
compared the formulations used in several published Streptomycete
fermentations (Aharonowitz and Demain 1977; Godfrey 1973; Hirsch and
Ensign 1976; Kirkpatrick and Godfrey 1973; Stark, Hoehn, and Knox
1967). As a consequence, we developed medium IV, and coupled this
formulation to the specific nutritional requirements of S. avermitilis.
Results from the original tests of this medium did not meet expectations.
Cultures 3, 4, and 5-grown in seed IV and transferred to production
medium B-produced quite poorly (IX or 2X levels). However, when
these same cultures (grown in seed IV) were transferred into production
C, quite impressive results were obtained. Not only had avermectin
production increased twenty-fourfold, but the mutant cultures maintained
the superiority they had originally shown. Seed medium IV had an
additional benefit: unlike its predecessors, it was completely soluble and
essentially defined.
Seed medium development continued with the acquisition of new
mutant strains but was constantly checked against seed medium IV (the
best at that time). Three additional seed media were devised and tested:
V, VI, and VII. As shown in Table 3.3, cells grown in seed medium V
failed to synthesize as much avermectin from production medium C as
they had when grown in seed IV (22X vs 24X, respectively). Using culture
5, seed medium VI transferred into production medium D resulted in the
production of more avermectin (29X) than when seed medium III was
used (23X). Changing to seed medium IV and keeping the other parame-
ters (culture and production medium) the same, avermectin production
was improved to thirty-sevenfold. Once again, seed medium IV proved to

be the most flexible in terms of demonstrating culture superiority and in
relation to use of the various production media. Seed medium VII, which
was developed later, was tested only in production medium E. Seed
medium VII clearly demonstrated the superiority of improved mutants
(32X, 44X, and 54X for cultures 5, 7, and 8, respectively) fermented in
production medium E. However, when cultures 5, 7, or 8 were grown in
seed medium IV rather than seed medium VII, avermectin production
was higher still: 40X versus 32X for culture 5, 56X versus 44X for culture
7, and 66X versus 54X for culture 8. A total evaluation of the data
contained in Table 3.3 indicates that maximum avermectin production
was obtained from superior cultures when they had been grown in seed
medium IV and were fermented in production medium E. Overall, the
accumulated results demonstrate that it is essential to monitor the
fermentation process continually to guarantee optimal product formation.
c. PRODUCTION MEDIA DEVELOPMENT
As the seed medium was being developed, so too was the production
medium. Early studies involved classical titration experiments: the
concentration of a given component of production medium B (Burg et al.
1979) was varied while the remaining ingredients were held constant. In
this approach, one looks for the occurrence of maxima (in this case,
highest avermectin titers) as they are collected from tests of each
ingredient assessed independently (Box, Hunter, and Hunter 1978). It
was exactly this approach that led to the development of production
medium C from production medium B (Table 3.3). Individual ingredients
of production medium B were titrated while the remaining ingredients
were maintained at constant levels. Concentrations of each individual
component that resulted in maximum avermectin production were then
simply combined to yield the composition of production medium C. As
seen in Table 3.3, combining any of the seed media and any of the cultures
fermented in production medium C increased the avermectin titer com-
pared to that obtained under the identical conditions in production
medium B. Focusing on culture 5 (Table 3.4), avermectin titers improved
by a factor of 24 when fermented in production medium C. Obviously,
this approach does have merit; its use can have a definite positive impact
on process development.
We then pursued a somewhat different but equally "classical" ap-
proach to medium development. These tests substituted various nitrogen
sources for the peptonized milk ingredient in medium C. Protein sources
such as soy, pork, cottonseed, and fish, as well as soybean and cot-
tonseed meals, were substituted based on their nitrogen equivalence. The
best avermectin titer obtained from any of these substituted media was
only 42% of that obtained from the control medium. The avermectin
TABLE 3.4. Production medium development.

Production mediaa
Seed
Culture medium B C D E F G J
1 II (1.0) NDb ND ND ND ND ND
5 IV 2X 24X 37X 40X ND ND ND
7 IV ND ND ND 56X ND 41X 36X
8b IV ND ND ND ND 87X 103X 117X
12 IV ND ND ND ND ND l08X 128X
a: Values shown are total fold improvements in B 1(a+b) titers including increases due to the
following factors: strain improvement, seed and production media modifications, and
changes in the physical fermentation parameters
b: Not determined
production levels in the other media ranged from 0 to 25% of control

values. When the yeast ingredient was varied, production values varied
within only 3% to 4%. Therefore, of the various media ingredients we
tested, none improved avermectin production relative to those ingredi-
ents currently in the production medium.
In summary, we used the classical approach to medium development to
devise production medium C. Substituting the nitrogen source and yeast
component in medium C did not improve avermectin titers. There are,
however, other approaches to medium/process development that could
be pursued (Box, Hunter, and Hunter 1978; Greasham and Inamine 1986).
Despite the improvement in avermectin synthesis, it should be recognized
that the classical approach has an inherent weakness. It fails to consider
the possibility that individual ingredients may interact with one another in
some way, altering their role in the overall medium composition. In
addition, factors other than medium formulation also influence the
fermentation's outcome-factors such as culture, temperature, aeration,
and pH. Although experiments can be devised to consider all these
parameters, they would be extremely laborious and time-consuming.
Alternatively, it is possible to use various statistical methods that allow
testing of multiple variables concurrently (Box, Hunter, and Hunter 1978;
Greasham and Inamine 1986).
Frequently the development of production media follows a sequential
path, beginning with Plackett-Burman studies designed to identify inde-
pendent variables (e.g., media ingredients) of interest (Greasham and
Inamine 1986). Once the variables that impact product accumulation have
been determined, the next step is to optimize the level of each identified
variable. Generally, this optimization procedure uses response surface
methodology (Box, Hunter, and Hunter 1978), which allows estimation of
up to 5-variable interactions. Response surface methodology involves
generating contour curves (Box, Hunter, and Hunter 1978; Greasham and
Inamine 1986). It is precisely along these curves where the optima occur.
To adequately (mathematically) identify each maximum point, a variable
must be tested at 3 levels. A 5-variable interaction study therefore

requires 247 trials to identify the optimum for each of the 5 variables.
Obviously, this many trials become unwieldy and impractical to perform.
As a consequence, partial factorial designs have been constructed
(Greasham and Inamine 1986). Partial factorials consist of selected points
extracted from a full factorial design; those points are used to obtain an
estimated response. A partial factorial result may only indicate where an
optimum may be, and a second set of trials may be needed to confirm that
optimum. As a general rule, the use of partial factorials followed by
independent experiments to confirm identification of optima results in
better and ultimately faster process improvement than alternative ap-
proaches.
Tables 3.3 and 3.4 demonstrate the consequences of using response
surface methodology to develop production media. As previously stated,
production medium C was composed using the classical approach.
Production medium D, however, was developed using partial factorial
experiments, followed by secondary trials to confirm acquisition and
location of optimum values. As shown in both Tables 3.3 and 3.4, culture
5 fermented in production medium D produced thirty-sevenfold more
avermectin than the original strain. Furthermore, it produced higher
avermectin titers in D than in C (37X vs 24X, respectively). Medium
modification as a consequence of observations made on statically growing
cultures (and applied to those in liquid media) resulted in the formulation
of production medium E. Culture 5, fermented in production medium E,
yielded fortyfold improved avermectin production compared to that in
medium D. Clearly, culture 5 demonstrated improved avermectin synthe-
sis as it progressed through the various production media (Table 3.4). The
compositions of production media E and F are identical. The differences
in titer obtained from these media demonstrate the importance of
examining modifications in physical fermentation parameters (such as
aeration, pH, and length of incubation) as the media are developed and as
newly acquired strains are obtained. Production media G and J were
derived once again as a consequence of partial factorial experiments with
subsequent confirmation trials. As Table 3.4 indicates, improved cultures
(7, 8b, and 12) demonstrated improved ability to synthesize avermectin in
the new production media: 56X in production E, 87X in production F, and
108X in production G, respectively. Furthermore, both cultures 8b and 12
showed improvement in avermectin production based only on the devel-
oped production media. For culture 8b, titers improved from 87X to I03X
to 117X as the production changed from medium F to G to J. Similarly,
culture 12 displayed improvement to 128X in medium J compared to I08X
in medium G. Interestingly, the data in Table 3.4 also indicate that not all
cultures will overproduce product in all media. For example, culture 7
produced significantly less avermectin in medium J (36X) and in medium
G (41X) than in medium E (56X). Obviously, media G and J had not been
optimized for use with this culture. In general, these data definitely prove
that production medium development, coupled with culture improvement
and modifications in the physical fermentation parameters, can and does
result in improved processes for the fermentation of secondary metabo-
lites.
Figure 3.1 illustrates the shake flask fermentation of culture 8 in
production medium J. Avermectin production began sometime between
the second and third day of the fermentation, after logarithmic growth had
ceased. Although avermectin production has been reported to be catabo-
lite repressible (Ikeda, Kotaki, and Omura 1987; McCann-McCormick et
al. 1981), using the cultures and conditions described herein avermectin
production was unaffected by glucose. Neither ammonia nor pH levels
rose until relatively late in the fermentation cycle, indicating that the
culture had not lysed. Now that we had developed superior cultures and
25 I
"- 20
-. "'- ......
~ DCW .".",. .......
-------- ....
- - .............. .,..,..,....
7
.c ... ~-----
!
",.
,!!t15
..
C
~
_ 10
I;:)
:z:Do
U 5 5
~
Q
0
r-----____________________________________ ~4
100
80
..........................
80
~
c
;:)
•... ·······~Bl
40 ...........
20 ......................
....
0
............
0 20 40 60 80 100
Percent of Cycle
FIGURE 3.1. Avermectin formation and growth of Streptomyces avermitilis in
shake flasks.
optimized the media and fermentation parameters in shake flasks, it

remained to transfer and translate the whole system to larger vessels.
III. Process Scale-Up

The objective was to scale up a process developed in shake flasks (20 ml
in a 250 ml flask) to production-scale fermentors (50,000L operating
volume), i.e., a scale-up factor of 2.5 million. In practice, the scale-up was
accomplished in two steps, first by developing a stirred-tank process at
the 500L scale (pilot plant), followed by a hundredfold scale-up to the
factory.
A. FERMENTATION KINETICS: STIRRED TANKS
In a typical avermectin fermentation in stirred tanks, rapid cell growth is

followed by a prolonged period of product synthesis with no evidence of
further cell mass accumulation. Under the range of conditions investi-
gated within our pilot plant, the formation of avermectin occurs only
when cell growth is extremely low, perhaps zero. A typical fermentation
profile is shown in Figure 3.2. This is similar to that obtained in shake
flasks (Figure 3.1) and also to that previously described by Burg et al.
(1979).
The relative mixture of different avermectin components produced in
20
Dry Cell Weight g / I 15
/ -
[0
Cii 10 $:
--
OJ
Titer 8 1
~ ()
0
0
5
L-__~~~____L-__~__~____L -_ _~_ _~ 0
25 50 75 100
Percent of Cycle (%) --..
FIGURE 3.2. Avermectin formation and growth of Strepromyces avermitilis in

stirred tank formentor.
stirred tanks is similar to that obtained in the shake flask process. This
presents a special challenge for downstream extraction; the desired B'a
and BIb components must be separated during the purification process.
An additional challenge to the purification is that S. avermitilis also
synthesizes lipids in large amounts. These consist primarily of a mixture
of triglycerides (Metz et al. 1988) possessing C14-C17 free fatty acids.
B. PREDICTION OF CELL MASS ON-LINE
Using a combination of mass spectrometry, mass flow meters, and

minicomputer (Buckland et al. 1985), it is possible to obtain a very
accurate on-line measurement of oxygen consumption rates (expressed as
mmoles of oxygen consumed per liter of fermentation broth per hour) for
any of our stirred tank fermentations. For S. avermitilis in a variety of
different media, there was an excellent correlation between total oxygen
consumed-obtained on-line by integrating the sum of the oxygen uptake
rates-and cell mass accumulation during the active growth phase of the
fermentation (Figure 3.3) according to Equation 1. This correlation
cannot be used after the period of rapid growth.
t
X = Y o/x LOUR· At
t = 0
Correlation of Cell Growth to 02 Uptake

20r--------------------------------------,
y= 3.38 E -02 *x -0.083
R**2=0.999
o 100 200 300 400
mMoles/L Oxygen Used

FIGURE 3.3. Correlation between cell growth and oxygen uptake (from Buckland
et al. 1985).
Where t time
OUR oxygen uptake rate (mmoles/lIh)
Y o/x ratio of total oxygen consumed to cell concentration
X = cell concentration (g dry cell weightll)
Therefore the function OUR· t::.. t denotes the total oxygen consumed,
easily calculated by on-line computer summation of the oxygen uptake
rate with respect to time.
Assuming that there is a constant yield coefficient between cell mass
and total oxygen consumed during this time period, it is possible to extend
this model further by dividing the oxygen uptake rate by the total oxygen
consumed; this provides an estimation of the specific growth rate of the
culture (gram increase in cell mass/ total cell mass in grams/hour). Using
the power of the minicomputer, this estimation can be done on-line.
1 dx OUR
f.L=--=
x dt ±
t=O
OUR. At
Where = specific growth rate, reciprocal hours

f.L
x = cell concentration [g (dry cell weight)/l]
The on-line estimation of both cell mass and cell growth rate (Figure 3.4)
(Buckland et al. 1985) has proven to be accurate, reliable, and a very
Avermectin 800 L Production Stage
:£
-
.5
~ o Tolol 02 (Moles/L) (J d.c.w.
.! o Specific Growth Rote (lIH) 15

c
0:: .4
.s::.
~
c:::::
C>
<!)
1:
'*
.3 10 .2'
cC j
::::r
.....til
.2 ~
&
CP
-
(5
~
C\I
0 .1
if 5
(5 d~
~ (y/
00 10 20 30 40
Hours
FIGURE 3.4. On-line estimation of cell concentration and specific growth rate
(from Buckland et al. 1985).
powetful tool for process development. An example is given [Figure 3.5

(Brix, Drew, and Buckland 1988)] in which two fermentations were run at
different temperatures, one 4° C higher than the other. The resulting
difference in specific growth rate is very pronounced.
c. OXYGEN TRANSFER
As the cell mass increases, the cells' mycelia begin to tangle, making the
broth viscous (Figure 3.6). This increase in viscosity adversely affects
oxygen transfer. It has been determined experimentally (Buckland et al.
1988) that the oxygen transfer coefficient is approximately proportional to
the reciprocal of the square root of the viscosity. This effect limits the
maximum achievable cell concentration and therefore the reactor's
volumetric productivity. Consequently, the study of oxygen transfer
efficiency was especially important for this process.
Computer-controlled, highly instrumented fermentors can calculate
on-line a number of different parameters for a single fermentation batch.
In one series of experiments (Buckland et al. 1988), the performance of a
large-diameter axial flow hydrofoil impeller was compared to that of a
conventional Rushton radial flow impeller. Two fermentations were run
side by side and the oxygen uptake rate profiles for the 2 different agitator
Avermectfn; Effect of Temperatu,. on Speclflo Growth Rafe
~,,
,,
"
~\ •• A. .. Itt
i .6 '.....
.~ \
.,.
\,.," \
A
"
i .4
·0
....'
...........
)a. ...,
T+4C
'A..........lt'·... -.........• ...............~..
ict .2
\ T
·o·· .. ··~ .. ~ ......~ .. ~...... ~ ..•.. +.·~ .. ~ .. +.~ ..
o~------~------~--------------~------~------~
o 2 6 8 10 12
Hours
FIGURE 3.5. Specific growth rate (estimated on-line) for two fermentations at
different temperatures. Standard temperature (T) and standard temperature plus
4° C (T + 4C) (from Brix, Drew, and Buckland 1988).
Viscosity Profiles in Avermectin Fermentation

2000~------------------------------------~
Brookfield Viscometer
--
G-
o
Spindle LVT#2
1000
20 40 50 60 70
Rotation Speed, rpm

FIGURE 3.6. Viscosity profile for the avermectin fermentation. 400mL samples of
broth, taken at different fermentation ages, were measured on Brookfield Viscom-
eter. The rotation speed of the spindle LVT #2 was varied from 6, 12, 30, and
60 rpm to determine torque deflection. The % torque deflection was converted to
apparent viscosity (cP) units using calibration charts provided by Brookfield
instruments.
types are compared (Figure 3.7). The dissolved oxygen profiles are also
similar (Figure 3.8). The on-line calculation of the oxygen transfer
coefficient (KLa) is made using the data from Figures 3.7 and 3.8; not
surprisingly, the profiles are also similar (Figure 3.9). The most striking
difference is in the relative power consumption of the 2 different agitator
types (Figure 3.10). The results of this study (and others) demonstrated
that-for a viscous mycelial fermentation of this type-a large-diameter
axial flow impeller is much more efficient at oxygen transfer than the more
conventional Rushton radial flow impeller.
D. STRATEGY USED FOR SUCCESSFUL SCALE-UP
There were a number of key issues related to successful scale-up from

250 ml shake flasks to 50,OOOL production fermentors. These include the
following:
80 r------------r----------------------------~
.....<>.....
PROCHEM
:J: 70 -0- RUSHTON
3
"-
~ 60
...J
o
~
~ 50
w
~ 40
<
ct:
w
~ 30
~
Q.
:::> 20
z
w
(!)
>- 10
x
0
0
0 10 20 30 40
HOURS
FIGURE 3.7. Oxygen uptake rate profiles for two avermectin fermentations: one
using Rushton impellers and the other using Prochem impellers (from Buckland et
al. 1988).
150
.... <).... PROCHEM
140 -a- RUSHTON
130
120
z 110
0
i= 100
<
ct:
:::> 90
~
< 80
(I)
ct:
70
< 60
~ 50
ci 40
c:i 30
20
10
0
0 10 20 30 40
HOURS
FIGURE 3.8. Dissolved oxygen profiles for same two fermentation described in
Figure 3.7 (from Buckland et af. 1988).
400
o Prochem
o Rushton
300
....
:::l
0
~
::::: 200
0
..J
::.::::
100
OL-----~----------~-----------J
20 30 40
Hours
FIGURE 3.9. On-line calculation ofthe oxygen transfer coefficient for the same two
avermectin fermentations described in 3.7 (from Buckland et al. 1988).
Prochem
2
O~----------~----------~----------~--------~
o 10 20 30 40
Hours
FIGURE 3.10. Agitator power draw for the same two avermectin fermentations
described in Figure 3.7 (from Buckland et al. 1988).
Environmental Containment
Streptomyces avermitilis is not known to be unusually harmful; neverthe-
less, it was necessary to maintain a high level of containment because the
product-avermectin-is very potent and could have a potential environ-
mental impact on aquatic life forms. Containment concerns extended to
both the fermentation operations and to the downstream purification
facility. These considerations were incorporated into the factory design
from the beginning and were resolved successfully: all possible exit
streams from the process are captured and the avermectin therein
degraded chemically.
Media Sterilization
As with many processes of this type, media sterilization is an important
variable (Jain and Buckland 1988) for both the seed tank stages and the
production stage. By necessity, more stringent conditions are required
as scale is increased to achieve the same assurance of complete steriliza-
tion (Buckland 1985). With a batch procedure, this is normally accomp-
lished by increasing time at sterilization temperature; with a continuous
sterilizer either the time or the temperature can be easily increased.
Superimposed on the requirement for sterility is the fact that a number
of chemical changes occur in the medium during sterilization, some of
which can have a great impact on culture productivity. This therefore
becomes an important area for optimization and, if possible, chemical
analysis.
Oxygen Transfer
Because of the viscous characteristics of the broth, the ability to transfer
sufficient oxygen to the individual microbial cell ultimately limits the cell
mass and hence productivity of the fermentation, assuming that the other
critical physical and chemical parameters that influence metabolism have
been chosen correctly to achieve high levels of product expression. When
evaluating a proposed process change in the pilot plant, it is necessary to
be sure that the oxygen requirements of the new process can be met
adequately in the larger-scale equipment. One simple method for doing
this is to run the pilot-scale fermentor in a way that mimics the limitations
ofthe factory equipment. An example is given in Figure 3.11, in which a
new medium formulation was evaluated in the pilot plant (medium 2). It
was immediately apparent that the oxygen demand of the new process
was in fact less than that for the standard process; given that the broth
viscosities were similar, introducing this process into the factory would
not result in starving the cells for oxygen.
Scale-Up of Avermectin
40.---------------------------------------~
...
::J
2
~
......
~ 30
o
~
E
-
~
Q)
20
~
o
~
a
&i 10
g
01
30 40 50
Hours
FIGURE 3.11. Scale-up information for new avermectin process (from Buckland
1985).
IV. Downstream Extraction

The challenge to develop an economic process for the purification of the
desired BI component of avermectin from this fermentation broth was
formidable. The harvested fermentation broth contains the following
components:
• various unused substrates from a complex fermentation medium
• other closely related avermectin compounds (Ala, Alb, B2a , B2b ; see
Chapter 1)
• a mixture oftriglycerides possessing C14-C17 acyl groups (Metz et al.
1988), made by the culture
• low levels of C15-C17 free fatty acids, squalene and diglycerides
(Metz et al. 1988), made by the culture
• significant amounts of chemical defoamer added to the medium
• other unidentified metabolites produced by the culture
An early process scheme for achieving this purification was described
by Miller et al. (1979). It is interesting to contrast this very elaborate
procedure with an improved version described in a recent patent by
Bagner and Wildman (1983). In the revised process, the fermentation
broth is first acidified and then mixed with an extractant such as toluene at
high temperature. The extractant is then decanted away from the aqueous
phase and, after just a few more processing steps, transferred to the final
crystallization tank.
REFERENCES
Aharonowitz Y, Demain AL (1977) Influence of inorganic phosphate and organic

buffers on cephalosporin production by Streptomyces clavuligerus. Arch.
Microb. 115:169-173
Awerbuch TE, Stark AA (1979) Plate diffusion assay as a rapid method for
dosimetry of mutagens. Appl. & Envir. Microb. 38:1127-1131
Bagner C, Wildman AS (1983) Process for the whole broth extraction of
avermectin. U.S. Patent #4,423,211
Box GE, Hunter WG, Hunter JS (1978) Response surface methods. Statistics for
Experimenters, John Wiley & Sons, Inc., New York, pp 510-539
Brix T, Drew SW, Buckland BC (1988) Fermentation process development within
a computer controlled pilot plant. In Bushell ME (ed), Progress in Industrial
Microbiology, Vol 25, Elsevier, New York, pp 151-193
Buckland BC (1985) The translation of scale in fermentation processes: the impact
of computer process control. Bio/Technology 2:875-883
Buckland BC, Brix T, Fastert H, Gbewonyo K, Hunt G, and Jain D (1985)
Fermentation exhaust gas analysis using mass spectrometry. Bio/Technology
3:982-988
Buckland BC, Gbewonyo K, DiMasi D, Hunt G, Westerfield E, Nienow AW
(1988) Improved performance in viscous mycelial fermentations by agitator
retrofitting. Biotech. & Bioeng. 31:737-742
Burg RW, Miller BM, Baker EE, Birnbaum J, Currie SA, Hartman R, Kong YL,
Monaghan RL, Olson G, Putter I, Tunac JB, Wallick H, Stapley EO, Oiwa R,
Omura S (1979) Avermectins, new family of potent anthelmintic agents:
producing organism and fermentation. Antimicrob. Agents Chemother. 15:361-
367
Clarke CH, Hopwood DA (1976) Ultraviolet mutagenesis in Streptomyces coe/i-
color: induction of reversions in a polyauxotrophic strain. Mutat. Res. 41:201-
208
Godfrey OW (1973) Isolation of regulatory mutants of the aspartic and pyruvic
acid families and their effect on antibiotic production in Streptomyces lipmanii.
Antimicrob. Agents Chemother. 4:73-79
Godfrey OW (1974) Directed mutation in Streptomyces lipmanii. Canad. J.
Microb.20:1479-1485
Goegelman RT, Gullo VP, Kaplan L (1983) Novel C-076 compounds. U.S. Patent
#4,378,353
Greasham R, Inamine E (1986) Nutritional improvement of processes. In Demain
AL, Solomon NA (eds), Manual of Industrial Microbiology and Biotechnology,
American Society For Microbiology, Washington DC, pp 41-48
Hirsch CF, Ensign JC (1976) Nutritionally defined conditions for germination of
Streptomyces viridochromogenes spores. 1. Bacter. 126: 13-23
Ikeda H, Kotaki H, Omura S (1987) Genetic studies of avermectin biosynthesis in

Streptomyces avermitilis. J. Bacter. 169:5615-5621
Ikeda H, Kotaki H, Tanaka H, Omura S (1988) Involvement of glucose catabolism
in avermectin production by Streptomyces avermitilis. Antimicrob. Agents
Chemother. 32:282-284
Jain D, Buckland BC (1988) Scale-up of the efrotomycin fermentation using a
computer controlled pilot plant. Bioproc. Eng. 3:31-36
Kirkpatrick JR, Godfrey OW (1973) The isolation and characterization of auxo-
trophs of the aspartic acid family from Streptomyces lipmanii. Folia Microb.
18:90-101
McCann-McCormick PA, Monaghan RL, Baker EE, Goegelman RT, Stapley EO
(1981) Studies on the avermectin fermentation. In Moo-Young M, Robinson
CW, Vezina C (eds), Advances in Biotechnology,Pergamon Press, New York,
pp 69-74
Metz P, Omstead DR, Kaplan L, Liesch JM, Stearns RA, VandenHeuvel WJA
(1988) Characterization of a lipid-rich fraction synthesized by Streptomyces
avermitilis. J. Chromatog. 441:31-44
Miller TW, Chaiet L, Cole DJ, Cole LJ, Plor JE, Goegelman RT, Gullo VP,
Joshua H, Kempf AJ, Krellwitz WR, Monaghan RL, Ormond RE, Wilson KE,
Albers-Schonberg G, Putter I (1979) Avermectins, new family of potent
anthelmintic agents: isolation and chromatographic properties. Antimicrob.
Agents Chemother. 15:368-371
Schulman MD, Valentino D, Hensens OD, Zink D, Nallin M, Kaplan L, Ostlind
DA (1985) Demethylavermectins: biosynthesis, isolation and characterization.
J. Antibiot. 38:1494-1498
Schulman MD, Valentino D, Nallin M, Kaplan L (1986) Avermectin B2
O-methyltransferase activity in Streptomyces avermitilis mutants that produce
increased amounts ofthe avermectins. Antimicrob. Agents Chemother. 29:620-
624
Sega GA (1984) A review of the genetic effects of ethyl methanesulfonate. Mutat.
Res. 134:113-142
Stark WM, Hoehn MM, Knox NG (1967) Mebramycin, a new broad-spectrum
antibiotic complex. I. Detection and biosynthesis. Antimicrob. Agents Che-
mother. pp 314-323
Stark WM, Knox NG, Wilgus R, DuBus R (1976) The nebramycin fermentation:
culture and fermentation development. In Developments in Industrial Micro-
biology, American Society for Microbiology, Washington DC, pp 61-77
CHAPTER 4
Biosynthesis
S. T. Chen, O. D. Hensens, and M. D. Schulman
I. Introduction
II. Incorporation of Labeled Precursors
III. Bioconversion of Intermediates
IV. Enzyme Studies
V. Effect of Specific Inhibitors
VI. Conclusions
I. Introduction
The avermectins are a family of oleandrose disaccharide derivatives of
pentacyclic lactones. Figure 4.1 presents a generalized structure of the
avermectins produced by Streptomyces avermitilis. In addition, a number
of key intermediates in the biosynthesis of the avermectins have been
identified. These structures are as follows: aglycones lack the oleandrose
disaccharide and possess a hydroxyl at C13; demethylavermectins lack
one or more methoxyls on the oleandrose disaccharide and possess a
hydroxyl at either or both the C3" and C3'; desfurano avermectins lack
the oxygen at C8a and C6 and possess a CH3 at C8a and an H at C6; and
5-keto avermectins lack the hydroxyl (or methoxyl) at C5 and possess a
keto group at C5.
Presently, the pathway for avermectin biosynthesis is not completely
understood. The pathway proposed in this chapter is based on evidence
obtained from four types of studies: the incorporation of labeled pre-
cursors into the avermectins; the conversion of proposed intermediates
into avermectins by producing strains and blocked mutants; the in vitro
measurement of enzymes involved in the biosynthesis of the avermectins;
and the production of incomplete avermectins by administration of
specific enzyme inhibitors.
56 T.S. Chen, O.D. Hensens, and M.D. Schulman
Oleandrose di saccharide Avermectin aglycone

~ ______~__________~A~__________~
OR,
FIGURE 4.1. Avermectin terminology is as follows: X = CH = CH for "1"
components; X = CH 2CHOH for "2" components; RI = H for "B" com-
ponents, RI = CH3 for "A" components; R2 = CH 2CH3 for "a" components;
R2 = CH 3 for "b" components (from Schulman, Ruby 1987).
II. Incorporation of Labelled Precursors

Figure 4.2 summarizes results obtained from incorporation studies with
[I-I3C] acetate, [I-I3C] propionate (Cane et al. 1983); [2-I3C] acetate,
[I,2-I3C] acetate, [3-13C] propionate, [2,3-13C] propionate, [I-I3C]
2-methylbutyrate, [1-13C] isobutyrate (Chen et al. in press); [U- 14C]
valine, [U- 14C] isoleucine (G. Albers-Schonberg, A.W. Douglas, R.T.
Goegelman, L. Kaplan, A. Kempf, and J.B. Tanac, unpublished observa-
tions); [1_14C] acetate, [2)4C] acetate, [l_14C] propionate, [2)4C] pro-
pionate, [methyl 14C] methionine, and [methyl 13C] methionine (Schul-
man, Valentino, and Hensens 1986). Tables 4.1 through 4.4 present
incorporation data from the most important of these studies. The data
show that the aglycone is derived from a head to tail condensation of
seven acetates and five propionates. The 2-methylbutyryl group (C25-
C28) of the "a" components and the isobutyl group (C25-C27) of the "b"
components were not labeled by either precursor; instead, they were
derived from isoleucine and valine, respectively. This is supported by the
efficient incorporation of [14C] isoleucine into the "a" components and
4. Biosynthesis 57
Precursors
(CH 3 ) Methionine methyl
---- Acetate
o (CH 3 )
Y
28
HaC
26 25
26a
0
L Isoleucine
or
27
\62~O L Valine
~
FIGURE 4.2. Incorporation of labeled precursors 10 the avermect1Os.
(CH3) = S-CH3 of methionine; ~ = acetate unit; .-~ = propionate unit;
• = carbons derived from isoleucine found in "a" components; • = carbons
derived from valine found in "b" components.
[14C] valine into the "b" components (Table 4.2). In addition, there was
high enrichment (twenty-fivefold) at C25 of avermectin B la by [I_I3C]
2-methylbutyrate (Table 4.1) and of avermectin BIb [l-I3C] by isobu-
tyrate. The metabolism of [1-13C] 2-methylbutyrate to [1-13C] propionate
accounts for the enrichment at C3, C7, Cll, C13, C23 of B la .
The methoxyl substituents at C5 of the macrolide moiety and C3' and
C3" of the oleandrose disaccharide are derived from the methyl of
methionine. It was shown using both 14C and 13C-methyl methionine and
[2_14C] methionine that the S-methyl and not the carbon backbone is
incorporated into avermectin (Tables 2.3, 2.4, and 2.5). The methyl is
incorporated equally into all 3 methoxyl groups.
The origin of the oxygen atoms in the macrolide moiety was investi-
gated by measuring the incorporation of [I_ 13 C, 1- 180] acetate and [I_ 13C,
I_ISO] propionate (Cane et al. 1983). The oxygens at C1, C5, C7, C13,
C17, C19, and C23 (for the "2" components) retained their isotope
TABLE 4.1. Incorporation of [Be] precursors in avermectins by S. avermitilis.
Relative 13C abundance in
Carbon avermectin B I • biosynthesized from ICC in avermectin B I • from
No. [2- 13] acetate [3- 13C]propionate [1- 13C]2-methylbutyrate [l,2- 13C 2]acetate [2,3- 13C 2]propionate
Col 1.4 1.0 1.1 59.3Hz N.D.2
C-2 3.2 1.2 0.9 59.3Hz N.D.
C-3 I.S 1.4 6.7 N.D.' N.D.
C-4 3.0 1.2 1.3 N.D. 43.4 Hz
C-4a 2.2 5.S 1.2 N.D. 43.4 Hz
C-5 1.0 1.3 O.S 40.7 Hz N.D.
C-6 3.S 1.2 1.0 40.7 Hz N.D.
C-7 1.6 1.4 7.7 N.D. 43.2 Hz
C-S 2.S 1.3 1.2 N.D. 43.2 Hz
C-Sa 2.4 6.2 1.0 N.D. N.D.
C-9 1.1 1.0 1.2 55.9 Hz N.D.
ColO 3.2 1.3 0.9 55.9 Hz N.D.
C-1I I.S 1.5 6.4 N.D. N.D.
C-12 2.6 1.5 1.0 N.D. 35.5 Hz
C-I2a 2.4 6.S 1.1 N.D. 35.5 Hz
C-l3 1.2 1.4 5.9 N.D. N.D.
C-14 2.9 0.9 0.9 N.D. 44.0 Hz
C-14a 2.3 6.6 1.1 N.D. 44.0 Hz
C-15 1.1 1.2 1.0 43.7 Hz N.D.
C-16 3.0 1.1 1.1 43.7 Hz N.D.
C-17 1.2 1.1 1.3 36.3 Hz N.D.
C-lS 4.4 1.3 1.0 36.5 Hz N.D.
C-19 1.2 1.1 1.3 36.3 Hz N.D.
C-20 3.8 1.3 0.9 36.3 Hz N.D.
C-21 1.3 0.9 1.2 54.9 Hz N.D.
C-22 3.7 1.3 1.0 54.9 Hz N.D.
C-23 1.8 1.5 5.6 N.D. N.D.
C-24 2.7 1.7 0.9 N.D. 35.2 Hz
C-24a 2.8 6.7 1.2 N.D. 35.2 Hz
C-25 2.2 1.0 24.5 N.D. N.D.
C-26 1.3 1.1 0.8 N.D. N.D.
C-26a 1.3 1.0 0.9 N.D. N.D.
C-27 1.4 1.0 1.1 N.D. N.D.
C-28 1.3 1.0 0.9 N.D. N.D.
C-l' 1.2 1.0 0.9
C-l" 1.1 1.0 1.0
C-2' 1.3 1.0 1.0
C-2" 1.2 1.0 1.0
C-3' 1.2 1.0 1.0
C-3" 1.1 1.1 1.1
C-4/ 1.1 1.1 1.1
C-4" 1.1 1.1 1.0
C-5' 1.1 1.3 0.8
C-5" 1.0 1.1 1.0
C-6' 1.0 1.1 1.0
C-6" 1.0 1.1 1.0
3/,3"O-CH 3 1.7 1.2 1.3
': Not determined.
60 T.S. Chen, D.D. Hensens, and M.D. Schulman
TABLE 4.2. The incorporation of [U- 14C] valine, isoleucine, and

sodium isobutyrate into the C-25 sidechain of avermectin B la and BIb'
Labeled avermectin
Spec.
Spec. incorporation
Spec. radioact. into C25
radioact. (p.Ci/mmole) sidechain (%)
Substrate Conc (uM) ("Ci/mmole) B la BIb B la BIb
Valine 4.1 5.85 0.54 1.57 12.3 21.3
Isoleucine 7.3 3.28 2.32 0.40 70.4 14.7
Isobutyrate l.1 11.00 0.17 0.57 1.5 5.0
2.2 6.36 0.14 0.69 6.4 10.9
3.3 3.62 0.71 0.82 4.7 22.7
4.4 2.58 0.27 0.77 10.5 29.8
TABLE 4.3. [14C]Metbionine incorporation into avermectins (from Schulman,

Valentino, and Hensens 1986).
14C in avermectins
Specific Specific
activity activity
of A ofB 14C in cell
Incorporation components components Ratio constituents
Precursor (%) (dpml "mol) (dpml "mol) AlB (%)
[methyl- 14C]methionine 60-63 179,155 107,521 1.67 23-27
[2- 14C]methionine 0 0 0 0 16-20
TABLE 4.4. [Methyl- 14C]methionine incorporation in

avermectin aglycones (from Schulman, Valentino, and
Hensens 1986).
Avermectin Specific activitya Original
component (dpml "mol) specific activity (%)
Ala 30,075 100
Deglycosylated Ala 11,910 40
A2a 45,409 100
Deglycosylated A2a 17,787 39
a: Determined following isolation of the pure component.
TABLE 4.5. Incorporation of [methyl- 13 C]methionine into
avermectins (from Schulman, Valentino, and Hensens 1986).
Isotopic enrichmenta
Avermectin component e3' methoxyl e3" methoxyl e5 methoxyl
A2a 5.5 6.5 4.2
Ala 4.3 4.2 6.2
Bla 2.9 2.5
a: The assignments of the e3' and e3" methoxyl carbons are not unequivocal.
The e3' methoxyl carbon is arbitrarily chosen as the high field resonance.
TABLE 4.6. Incorporation of [1_ 1802> 1- 13 C]acetate and [1_ 180 2 ,

1- 13C]propionate into avermectins (from Cane et al. 1983).
13e shift,
e 4 ppma [1_ 180 Z ' 13e]Ac [t-t 80 Z , 13e]Pr 0: 180
16
Al 173.96 0.035 N.D! 50:40: loa

Az 173.76 0.035 N.D. 60: 35: 5a
BI 173.69 0.035 N.D. 65: 35
Bz 173.54 0.035 N.D. 70:30
5 Al 77.03 0.023 N.D. 60:40
Az 77.03 0.017 N.D. 60:40
BI 67.76 0.023 N.D. 70:30
Bz 67.72 0.023 N.D. 70:30
7 Al 80.64 N.D. 0.023 60:40
Az 80.66 N.D. 0.023 60:40
BI 80.44 N.D. 0.023 60:40
Bz 80.52 N.D. 0.023 70:30
13 Al 82.06 N.D. 0.023 55:45
Az 81.82 N.D. 0.023 60:40
BI 81.99 N.D. 0.023 60:40
Bz 81.77 N.D. 0.023 60:40
17 Al 68.49 b N.D.
Az 68.42 0.029 N.D. 60:40
BI 68.42 0.029 N.D. 60:40
Bz 68.37 0.029 N.D. 60:40
19 Al 68.44 0.023 N.D. 60:40
Az 67.72 0.035 N.D. 60:40
BI 68.32 b N.D.
Bz 67.61 b N.D.
21 Al 95.84 0.0 N.D. 100:0
Az 99.73 0.0 N.D. 100:0
BI 95.83 0.0 N.D. 100:0
Bz 99.70 0.0 N.D. 100:0
23 Az 69.93 N.D. 0.017 60:40
Bz 69.93 N.D. 0.023 60:40
a: e 13 0 18 peak due to excess intramolecular multiple labeling by 90% acetate
precursor.
b: Signal obscured by overlap with other peaks.
c: Not determined.
TABLE 4.7. Incorporation of specifically labeled glucose into

avermectin A2a and methyl oleandrose.
Position of label 3H1 4C of A vennectin Ala Methyl oleandrose
in glucose fed glucose 3HJl 4C 3H retention % 3H/ 14C 3H retention
6- 14C, 1-3H 0.91 0.14 15.4 0.67 74
6- 14C,6_ 3H 0.46 0.10 21.7 0.46 100
content (Table 4.6). Several conclusions were drawn from these results:
(1) the avermectin "1" components which contain a double bond at
C22-23 arise from the "2" components by dehydration of the hydroxyl at
C23; (2) the oxygen at C21 is probably derived from the 2-methylbutyryl
or isobutyryl sidechain (C25); (3) the avermectins are not derived from a
milbemycin type intermediate by late-stage oxidation at C13 (Ono et al.
1983).
To investigate the origin of the oleandrose backbone, studies with
3H/ 14C and 13C labeled glucose were conducted. Comparison of the
3H/ 14C ratio in the labeled avermectin and the methyl oleandrose obtained
by methanolysis (Table 4.7) indicated a direct conversion of glucose to
oleandrose (Chen et al. in press). This was confirmed by studies with
[1_I3C] glucose and [U-I3CJ glucose which showed high enrichment of the
oleandrose units (Schulman, Valentino, and Hensens 1986).
Studies with labeled glucose provided some insight into the metabolic
origin of the avermectin precursors (Schulman, Valentino, and Hensens
1986). Equal incorporation of [1- 14CJ, [2_ 14CJ and [6_ 14C] glucose into the
avermectin aglycone provided strong evidence for the symmetrical cleav-
age of glucose via the Embden-MyerhofIpathway. A parallel study with
[U-I3C] glucose demonstrated that the entire avermectin molecule is
synthesized from glucose (Figure 4.3). The oleandrose units derived
directly from glucose had the highest incorporation rate. The methoxyls
of the oleandrose units which are derived from the methyl of methionine
were also enriched. Methyl-labeled methionine can be synthesized from
glucose via a known microbial pathway that involves serine and serine
transhydroxymethylase. The carbons derived from acetate were enriched
3.3-fold over those derived from propionate. This is due to a dilution
of acetate by the passage through the tricarboxylic acid cycle to
yield propionate (via succinate). The asymmetric labeling of the
2-methylbutyrate-derived sidechain is explained by assuming isoleucine is
formed from oxalacetate via the known pathway in the presence of
threonine aldolase.
III. Bioconversion of Intermediates

To study the biosynthetic interrelationships among the avermectin "a"
components, 14C-Iabeled Ala, A 2a , B la , B2a were individually fed to
cultures of S. avermitilis for 5.5 days (Chen et al. in press). The "a"
4. Biosynthesis 63
CH
FIGURE 4.3. Incorporation of [U13C] glucose into avermectin B la.

CH3e-e-e = propionate carbons; - = acetate carbons; (CH 3) = S-methylof
methionine. Numbers in parentheses are % 13C incorporated (from Schulman,
Valentino, and Hensens 1986).
components were then isolated and the distribution of radioactivity was

determined. The results, summarized in Table 4.8, show that 30% of the
B2a was converted to A2a and 3% of the B la was converted to Ala' Neither
Ala nor A2a was converted into any other avermectin, suggesting that they
are the end products of the pathway. Further feeding studies with
l4C-labeled B2a aglycone and B2a monosaccharide indicate the biosynthe-
tic reaction sequence B2a aglycone ~ B2a monosaccharide ~ B2a ~ A2a .
Avermectins Ala and B la were not labeled by B2a aglycone and B2a
monosaccharide, indicating that neither compound is the precursor of the
"I" components. Avermectin A2a aglycone was converted to A2a by a
nonproducing blocked mutant of S. avermitilis. Feeding of desfurano
avermectins (Figure 4.4) to this blocked mutant revealed that 6,8a-seco-
6,8a-deoxy-5 ketola avermectin aglycone (Figure 4.4, I) was converted
exclusively to avermectins Ala and B la ; 6,8a-seco-6,8a-deoxy-5 ket02a
avermectin aglycone (Figure 4.4, II) was converted exclusively to A2a and
B2a . These results demonstrate that the 5-keto desfurano "2" com-
ponents are not precursors of the" 1" series avermectins and indicate that
dehydration at C22-23 must occur before formation of the desfurano
avermectins. 6,8a-seco-6,8a-deoxy Bla aglycone (Figure 4.4, III) and
6,8a-seco-6,8a-deoxy A2a aglycone (Figure 4.4, IV) were not biocon-
verted into complete avermectins; instead they were converted to their
respective desfurano mono- and disaccharides. This suggested that
formation of the furan ring cannot occur in the presence of a C5 hydroxy
or methoxyl and requires a keto group at C5. Avermectin 5-keto aglycone
in the 2 series was bioconverted into B2a and A2a in agreement with the
TABLE 4.8. Radioactivity distribution in the avermectins after feeding 14C-labeled avermectins Ala' A2a, B la , B2a , B2a
monosaccharide (B 2aMS) and B2a aglycone (B2aAG) to S. avermitilis.
Compounds Radioactivity Incubation Total radioactivity recovered in avermectins, dpm
fed fed, dpm time hr B2aAG B2aMS Ala A2a Bla B2a
Ala 3.99 X lOS 156 NOb NO 4.02 X lOS NO NO NO
(100)"
A2a 1.92 X lOS 156 NO NO NO 1.78 X 1QS NO NO
(93)
B la 8.3 X lOS 156 NO NO 0.22 X 1QS NO 7.77 X 1QS NO
(2.7) (93.6)
B2a 5.59 X lOS 156 NO NO NO 1.66 X lOS NO 3.82 X 1QS
(29.7) (68.3)
B2aAG 5.4 X lOS 144 2.50 X !OS 0.3 X lOS NO 0.5 X lOS NO 1.72 X lOs
(46.3) (5.6) (9.3) (31.6)
B2aMS 4.16 X lOS 144 NO 1.34 X 1QS NO 0.78 X 1QS NO 1.88 X 1QS
(32.2) (18.6) (45.2)
a: Values in parenthesis are given in percent.

b: Not detected.
4. Biosynthesis 65
R X
I. 6,8a-seco-6,8a-deoxy-5-keto 1a aglycone ==0 -CH=CH-
II. 6,8a-seco-6,8a-dexoy-5-keto 2a aglycone ==0 -CH 2 -CHOH-
III. 6,8a-seco-6,8a-deoxy Bla aglycone -OH -CH=CH-
IV. 6,8a-seco-6,8a-deoxy A2a aglycone - OCH 3 -CH 2 -CHOH-
FIGURE 4.4. Structure of 6,8a-seco-6,8a-deoxy-5-keto avermectin aglycone.
hypothesis that reduction of the C5 keto occurs after the formation of the
furan ring.
Glycosylation to the mono- and disaccharide was observed with a
variety of intermediates, indicating that the glycosyltransferases lacked
specificity. When [5-methoxyl-14C] A2a bisdemethylavermectin and un-
labeled methionine were fed to wild-type S. avermitilis, the unaltered
substrate was the only radioactive compound recovered (Schulman and
Ruby 1987). Its specific radioactivity was virtually identical to that added.
This demonstrates that methylation of the oleandrose units is not a
terminal step in avermectin biosynthesis and that it occurs before
attachment of the sugars to the macrolide ring. The addition of 6,8a-
deoxy-5,13,23-triketo avermectin to a nonproducing blocked mutant did
not yield any natural avermectin compound, indicating this compound is
not a biosynthetic intermediate.
IV. Enzyme Studies
A VERMECTIN B O-METHYL TRANSFERASE

Avermectin B O-methyltransferase catalyzes the conversion of avermec-
tin "A" components to "B" components by transferring the methyl of
S-adenosylmethionine to the C5 hydroxyl of the "B" component, yield-
ing the "A" component and S-adenosylhomocysteine (Figure 4.5). This
enzyme has been purified and characterized (Schulman, Valentino, and
Ruby 1985). It has a molecular weight of 70,000 and is composed of two
identical subunits. The enzyme's substrate specificity for the avermec-
tins, in order of decreasing activity, is as follows: B2a aglycone> B2a
monosaccharide> B'a aglycone> B'a monosaccharide> B2a > B'a. Aver-
mectin "2" components react faster than "1" components; within each
group the order of activity is aglycones> monosaccharides> disac-
charides. There was virtually no activity detected with Bla as substrate.
The time course of avermectin B O-methyltransferase activity and
avermectin production was measured in several high-producing mutants
of S. avermitilis (Schulman et al. 1986). Table 4.9 presents the average
specific activity of the avermectin B O-methyltransferase, the maximum
observed specific activity of the O-methyltransferase, and the avermectin
titers of 4 mutant strains. Both average specific activity and the maximum
observed specific activity increased in direct proportion to the quantity of
avermectin formed. These data support the hypothesis that genes coding
for enzymes involved in the secondary metabolism biosynthesis may be
coordinately expressed.
A VERMECTIN 5-KETOREDUCTASE
Avermectin 5-ketoreductase catalyzes the NADPH specific reduction of

5-keto avermectins to B components as shown in Figure 4.6. The time
course of enzyme activity and avermectin production was determined in a
number of high-producing strains of S. avermitilis (M.D. Schulman,
unpublished observations). The maximal specific activity of 5-ketoreduc-
tase was proportional to avermectin production in these strains, as shown
in Figure 4.7. These data further support the hypothesis that genes coding
for enzymes involved in secondary metabolism is coordinately regulated.
A VERMECTIN AGLYCONE GL YCOSYL TRANSFERASE
Avermectin aglycone glycosyltransferase, which catalyzes the stepwise

addition of oleandrose units from the nucleotide sugar thymidine diphos-
phate oleandrose to the avermectin aglycone (Figure 4.8), has been
demonstrated in cell-free extracts of S. avermitilis (M.D. Schulman,
D. Valentino, S. Acton, S., and B.H. Arison, unpublished observations).
OH OH
CH 2CH 3 CH 2 CH 3
H3 C •
S-Adenosyl S-Adenosyl
methionine homocysteine
OH OCH 3
Avermectin B component Avermectin A component
FIGURE 4.5. Reaction of avermectin B O-methyltransferase. R = oleandrose disaccharide.

TABLE 4.9. Avermectin B O-methyltransferase activity and avermectin production in strains A, B, C, and D (From Schulman et ai.
1986).
Sp. act. of O-methyltransferase Total
(nmol/h per mg of protein) Fold increase in:
avermectin
Maximum Avg at 144 h Sp. act. of O-methyltransferase
Strain observed (48 to 144 h) (relative units) Maximum Avg Avermectin
A 0.37 0.26 0.85 1 1 1
B 0.86 0.54 2.10 2.3 2.1 2.5
C 1.06 0.63 2.4 2.9 2.5 2.8
D 1.35 0.85 3.0 3.6 3.3 3.5
OH OH
CH 2CH 3 CH 2CH 3
H3 C
NADPH NADP+
+
H+
o OH
5-Keto avermectin Avermectin B component
FIGURE 4.6. Reaction of avermectin 5-ketoreductase. R = oleandrose disaccharide.

20
19
18
0'
co
17
~ 16
< 15
~ 14
13
l!!
:J
c
=:J 12
2. 11
f 10
9
~ 8
Ql
1/1
III
7
g 6
'tI
l!! 5
4
~
lI:: 3
.n 2
1
0
3 5 7
Ratio Avermectin Yield (Culture X/MA 5080)
FIGURE 4.7. Avermectin production versus avermectin 5-ketoreductase activity

in high-producing strains of S. avermitilis. - MA5080, • MA5502, • MA5508,
o MA5856, 0 MA6016, t:,. MA6233. Line represents a linear regression.
It has not yet been determined if both sugars are transferred by a single
enzyme or if two enzymes are required. It is clear, however, that the
sugars are transferred sequentially.
V. Effect of Specific Inhibitors

Sinefungin, an analogue of S-adenosylmethionine, is a potent inhibitor
of methyltransferases and has been shown to inhibit avermectin B
O-methyltransferase in vitro (Schulman, Valentino, and Ruby 1985).
When fed to a culture that produces virtually only avermectin aglycone
"A" components (91%-99%), sinefungin inhibited the formation of "A"
components and caused a concomitant accumulation of "B" components
(Schulman et al. 1985). Sinefungin thus inhibited the conversion of "B"
components to "A" components in vivo. When sinefungin was added to
wild-type S. avermitilis, new avermectin components lacking methoxyl
groups on the C-3' and C-3" of the oleandrose accumulated (Schulman et
al. 1985b). In this case, sinefungin inhibited methylation of the oleandrose
4. Biosynthesis 71
Avermectin Avermectin Avermectin

Aglycone Monosaccharide Disaccharide
Oleandrosyl
Diphosphonucleotide
FIGURE 4.8. Reaction of avermectin aglycone glycosyltransferase.
SCHEME 1 PROPOSED BIOSYNTHETIC PATHWAY LEADING

TO AVERMECTIN "a" COMPOUNDS
Isoleucine - - - - - . . 2·methylbutyryl-CoA ....f - - - - - - 2-Methylbutyrate
7 acetate --+ 7 acetyl-CoA

5 propionate --+ 5 propionyl·CoA
I
6, 8a-Seco-6, 8a·deoxy·
~
6, 8a-Seco-6, 8a-deoxy-
5 Keto "2a" AG 5 Keto "1a" AG
~
5-Keto "2a'· AG
~
5-Keto "1 a" AG
SAM ~
I---~).::-.--
A2.AG .... 8 2.AG
~
8 1aAG - - - - -•• A1.AG
SAM
SCHEME 1. Proposed pathway for biosynthesis of the avermectins in S. avermi-
tiUs.
units, indicating that an S-adenosylmethionine-dependent methyltransfer-

ase is involved.
VI. Conclusions
The proposed pathway for biosynthesis of the avermectin "a" com-
ponents is presented in Figure 4.9. An exactly analagous pathway
is envisioned for the "b" components, with valine, isobutyrate, and
isobutyryl-CoA replacing isoleucine, 2-methylbutyrate, and 2-methyl-
butyryl-CoA. Isoleucine undergoes transamination and decarboxylation
to yield 2-methylbutyryl-CoA, which serves as the primer for chain
elongation. 2-methylbutyryl-CoA can also arise from fed 2-methylbuty-
rate by activation via a thiokinase. Chain elongation proceeds by addition
of acetate and propionate via their CoA esters in a fashion analagous to
fatty acid synthesis. At some point during the process, a precursor of the
"2" components is dehydrated at C22-23 to yield a precursor of the" 1"
compounds. From this point on, biosyntheses of the "1" and "2"
components proceed in parallel to produce the respective aglycones.
Because the methyltransferase and the glycosyltransferase display little
specificity, the aglycones are either methylated or glycosylated to pro-
duce "A" and "B" disaccharides. At the disaccharide level, B2a can be
methylated to A2a but B la cannot be methylated to Ala.
REFERENCES
Cane DE, Liang T-C, Kaplan LK, Nallin MK, Schulman MD, Hensens OD,
Douglas AW, Albers-Schonberg G (1983) Biosynthetic origin of the carbon
skeleton and oxygen atoms of the avermectins. J. Am. Chem. Soc. 105:4110-
4112
Chen TS, Arison BH, Gullo VP, Inamine ES (in press) Further studies on the
biosynthesis of the avermectins. J. Indust. Microbiol.
Chen TS, Inamine ES (in press) Studies on the biosynthesis of avermectins. Arch.
Biochem. & Biophys.
Ono M, Mishima M, Takiguchi Y, Teroa M (1983) Milbemycins, a new family of
macrolide antibiotics. Studies on the biosynthesis of milbemycins a2, a4 and D
using BC labelled precursors. J. Antibiot. 36:991-1000
Schulman MD, Ruby C (1987) Methylation of demethylavermectins. Antimicrob.
Agents Chemther. 31:964-965
Schulman MD, Valentino D, Hensens OD (1986) Biosynthesis of the avermectins
by Streptomyces avermitilis. J. Antibiot. 39:541-549
Schulman MD, Valentino D, Hensens OD, Zink D, Nallin MK, Kaplan LK,
Ostlind DA (1985) Demethylavermectins. Biosynthesis, isolation and charac-
terization. J. Antibiot. 38:1494-1498
Schulman MD, Valentino D, Nallin MK, Kaplan LK (1986) Avermectin B2
O-methyltransferase activity in Streptomyces avermitilis mutants that produce
increased amounts of the avermectins. Antimicrob. Agents Chemother. 29:620-
624
Schulman MD, Valentino D, Ruby C (1985) Avermectin B O-methyltransferase of
Streptomyces avermitilis. Fed. Proc. 44:931
CHAPTER 5
Mode of Action of Ivermectin

M.J. Turner and J.M. Schaeffer
I. Introduction
II. Invertebrates
A. Electrophysiological Studies
B. Biochemical Studies
1. IVM Binding Sites
2. Chloride Uptake
3. Acetylcholine Release
4. Interaction with Retinal Binding Proteins
5. Inhibition of Chitin Synthesis
III. Vertebrates
A. A VM Stimulated Neurotransmitter Release
B. A VM Binding Sites
C. A VM Effect on [3H]GABA Binding
D. A VM Effect on Benzodiazepine Binding Sites
E. Chloride Uptake
F. Other Actions of AVM
IV. Conclusions
I. Introduction
The anthelmintic activity of the avermectins was first described in 1979
(Burg et al. 1979; Egerton et al. 1979; Miller et al. 1979). The mode of
action of these compounds, however, has remained elusive. Identifying
the mode of action is all the more difficult because the avermectins have
been studied in so many different model systems with an array of
experimental protocols. For example, direct injection of avermectin
(AVM) into Ascaris suum results in a rapid paralysis that is neither flaccid
74 M.J. Turner and J.M. Schaeffer
nor rigid; incubation of the free living nematode, Caenorhabditis elegans,

with AVM results in slow-onset rigid paralysis; and incubation of
Haemonchous contortus ex vivo with AVM has no observable effect.
Exactly why AVM affects these 3 AVM-sensitive nematodes differently
is unknown, but it may reflect in part the drug's ability to reach its site of
action. The problems in identifying the mechanism(s) by which the
avermectins work have been further confounded by several additional
factors: the drug acts at mUltiple sites; various target species have
different sensitivities to the drug; and avermectins have poor solubility in
aqueous solutions.
The actions of avermectins on vertebrate and invertebrate systems are
distinct and will be dealt with separately. Although the most parsimonious
explanation for how avermectins work is that they specifically increase
membrane chloride ion permeability, there are clearly other sites of action
at which avermectins affect either the host or target organism.
It is likely that the entire series of avermectins share a common mode of
action; therefore, we will refer to all of the avermectins with the generic
abbreviation, AVM. In actuality, nearly all of the studies discussed in this
review used either avermectin B 1a or 22,23-dihydroavermectin B1a (iver-
mectin, IVM).
II. Invertebrates
A. ELECTROPHYSIOLOGICAL STUDIES
The earliest studies designed to elucidate the mode of action of A VM
were reported by Fritz, Wang, and Gorio (1979). They demonstrated that
when lobster stretcher muscle is perfused with 1O-5-10-6M AVM, the
inhibitory postsynaptic potentials are rapidly eliminated, followed by a
more gradual reduction in the amplitude of excitatory postsynaptic
potentials. In these preparations, AVM reduces the input resistance of the
muscle fiber by increasing its permeability to chloride (Cl-) ions. These
effects were not reversed by washing. The reduction of both excitatory
potentials and input resistance, however, was reversed by picrotoxin (a
GAB A antagonist active at the chloride channel). It was hypothesized
that these responses to AVM were caused by an increase of membrane
permeability to chloride ions, perhaps due to its interaction with GAB A
binding sites or by regulating the release of endogenous GABA. These
findings have subsequently been confirmed (Mellin, Busch, and Wang
1983; Albert et al. 1986).
The motor nervous system of Ascaris lumbricoides is an excellent
experimental system to study avermectin's mode of action. An elegant
description of the nervous system is available (Stretton et al. 1978) which
outlines the connections and electrical characteristics of the neurons. The
neurons, moreover, are large enough to be penetrated with microelec-
5. Mode of Action of Ivermectin 75
trodes. Using this model system, Kass and his collaborators (1980, 1984)
demonstrated that A VM (=5 x 1O-6M) inhibits transmission between
interneurons and excitatory motoneurons in the ventral nerve cord
(Figure 5.1), as well as inhibiting transmission between inhibitory
motoneurons and muscle; it has little effect, however, on excitatory
neuromuscular transmission. Picrotoxin can reverse the AVM-induced
block of interneuron-excitatory motoneuron transmission but has no
effect on the inhibitory motoneuronal synapse in either the presence or
absence of AVM. These findings suggest that AVM can function either as
a GABA agonist or as a stimulator of GABA release from presynaptic
inhibitory terminals.
Duce and Scott (1983, 1985) demonstrated that within the extensor
tibiae muscle of the locust (Schistocerca gregaria) , specific muscle
bundles are sensitive to GABA, whereas a separate population of muscle
bundles, which receive only excitatory innervation, are GABA-
insensitive. This specificity within a single preparation of muscle fibers
was exploited to investigate the interaction of AVM with the GABA
receptor-ionophore complex in insect muscle. The muscle bundle that is
sensitive to GABA receives inhibitory innervation and exhibits both
reversible and irreversible responses to AVM. There is a reversible
dose-dependent increase in CI- permeability in response to very low
doses of AVM (= 10-IOM); these effects appear to be due to an interaction
of AVM with the GABA receptor-CI- ion channel complex. Higher AVM
concentrations (=10- 8M) evoke an irreversible increase in Cl- conduc-
tance that continues to rise after removal of AVM. GABA-insensitive
neurons with no inhibitory innervation were also examined; in this case
AVM
8 5 p,g/ml
> 6
E
a) I I tI It Picrotoxin Wash
If
L!
en 10 p,g/ml
c: 4
I
,
0
a.
en
Q)
a: 2 I ! I
0
•I
0 20 40 60 80 100 120 140 160
Time, min.
FIGURE 5.1. Responses to indirect stimulation of DEL Each point represents the
mean of five responses. All recordings are from the same dorsal muscle cell.
Drugs were added at arrows. (From Kass et al. 1980).
76 M.J. Turner and J .M. Schaeffer
AVM (=lO-I~) induces only irreversible increases in Cl- conductance.

These important results clearly demonstrate a site of action for AVM
distinct from the GABA receptor chloride channel complex. Mellin,
Busch, and Wang (1983) studied the effect of AVM on the extensor tibialis
muscle from cockroach (Periplaneta americanus) walking leg. This
preparation consists of numerous fibers which lack an inhibitory GABA
innervation. AVM had no effect on the "fast" axon excitatory electrical
responses (glutamatergic synapses). The reason for the contradictory
effect of A VM on the non-GAB A neurons in locust and cockroach has not
been resolved.
Recently, single Cl- channels have been studied by patch-clamping
techniques. Martin and Pennington (1988) examined the effect of AVM on
single channels in Ascaris suum muscle membranes. A progressive
stepwise opening of chloride ion channels at AVM concentrations of
2 x lO- 12M and above was observed, with a concentration-dependent
delay in onset. The amplitude of the channel was about 1 pA, with a
conductance of 12 pS. The duration of opening was very long (> 1 second)
and was irreversible. This chloride channel is not voltage dependent, it is
not GABA-gated, and it is not sensitive to any other neurotransmitters
examined. Using the same preparation, GABA-gated channels were also
studied. GABA produced a 35-msec opening of a 22-pS conductance
channel. Effects of AVM on this channel could be seen only at concentra-
tions of > lO- 8M, at which levels antagonism of the GABA effect was
observed.
6. BIOCHEMICAL STUDIES
1. IVM Binding Sites

Specific AVM binding sites have been identified in membranes prepared
from the nonparasitic nematode, Caenorhabditis elegans (Schaeffer and
Haines, in press). Specific binding is saturable with an apparent disso-
ciation constant of 2.6 x lO-IOM (Figure 5.2). Kinetic analysis of the
binding showed that the reaction proceeds by a 2-step mechanism.
Initially, a rapidly reversible complex is formed; after additional incuba-
tion this complex is transformed to a much more slowly reversible
complex (see Equation 5.1).
k+1 k+2
IVM + R ~ IVM· R IVM· R*
Stereospecificity of AVM binding was demonstrated by competition with

a series of avermectin derivatives. As shown in Figure 5.3, the AVM
analogs that are the most potent anthelmintics are also the analogs with
the highest affinity for the AVM binding site. Several neurotransmitters,
including GABA, were found to have no effect on AVM binding. Specific
3.0
0
~O-
0> /0 :
/
E Ko=0.26nM
.......
(5 2.0 :,; 7
E
0..
.,en
c
6
BMa.=3.53
"0
E
0
/
c: (5 5
:J E
0-
m cD
4
II)
~ !l: 3
>
...... "-
0
"C
.-----. c
1.0
•
::J 2
I 0
'\
In
~
0
3.0
Bound. pmol/mg
o 0.5 1.0 1.5 2.0
[IVM], nM
FIGURE 5.2. Equilibrium binding of [3H] IVM to C. elegans membranes. Increas-
ing concentrations of PH] IVM were incubated with C. elegans membranes and
specific binding was determined as described in the text. Each point is the average
of 4 determinations. Replicate experiments gave similar results. A scatchard
analysis (inset) of the saturation data is shown. (From Schaeffer and Haines, in
press).
AVM binding sites identified in rat brain have an approximately one

hundredfold lower affinity than that observed in C. elegans, and the
stereospecificity is also different (see below).
Specific, high-affinity GABA binding sites (K D = 3.7 x 1O- 8M) were
also identified in C. elegans, and AVM (10- 10 to 1O-6M) had no effect on
[3H]-GABA binding (Schaeffer and Bergstrom, 1988). However, in other
invertebrate systems A VM-sensitive GABA binding sites have been
observed. Lummis and Sattelle (1985) identified a GABA binding site in
nerve cord extracts from the cockroach, Periplaneta americana, with a
KD value of 3.84 x 1O- 7M. AVM enhances the amount of binding of
eH]-GABA over a narrow range concentration (0.5-5.0 x 10- 6M). No
effects of AVM were seen at lower or higher concentrations. Unfortu-
nately, this study does not report whether AVM alters the affinity of
GABA for its binding site. Abalis and Eldefrawi (1986) have reported that
AVM acts as a partial agonist at the GABA binding site and competes
with [3H] muscimol for specific binding to membranes prepared from
honey bees.
250
8/
100
/
-7 -
:::E
c
~
0
w 50
25
10 ~--------~------~--------~--------~~
0.1 0.25 0.5 1.0 2.5
KI,nM
FIGURE 5.3. Correlation between binding affinities of AVM analogs and their
biological potencies of C. elegans motility in vivo. EDso is the concentration of
AVM needed to produce immotility in 50% of the worms. The results are the mean
values from 2 to 10 experiments (SEM less than 15%). A correlation coefficient (r)
of 0.975 was calculated by linear regression analysis. For the log-log plot,
r = 0.923. The compounds tested were (1) IVM; (2) AVM B2a ; (3) AVM Bla
4'-0-phosphate; (4) 22,23-dihydro-AVM B 1a aglycone; (5) 22,23-dihydro-AVM BI
monosaccharide; (6) AVM BI monosaccharide; (7) 2-dehydro-4-hydro-a-2,3-
AVM B 1; and (8) AVM B 1a-5-ketone. (From Schaeffer and Haines, in press).
2. Chloride Uptake
The action of AVM on chloride uptake was studied on the perfused leg
muscle of the American cockroach, Periplaneta americana (Tanaka and
Matsumura 1985). AVM stimulated chloride uptake at concentrations as
low as to- 8M. The AVM-stimulated chloride uptake is antagonized by
picrotoxin and to a lesser extent by bicuculline methiodide. At 10-7M
AVM the leg muscles fail to respond to external stimuli. This inhibition
was postulated to be due to an increase of chloride uptake by the leg
muscle. These results suggest that AVM is opening the chloride channel
on the plasma membrane. Since A VM had no stimulatory effect on eH]

muscimol or [3H] dihydropicrotoxinin binding to the muscle membranes,
the authors suggest that A VM is not active at the GAB A binding site but
directly at the site of the chloride ion channel. It should be pointed out
that the quantitative determination of chloride flux using 35Cl- is very
difficult because of the high background in these experiments, and the
results presented in this paper are equivocal. The basal rate of chloride
uptake (100.00 ± 5.42%) does not seem to be significantly different from
the maximal A VM-stimulated chloride uptake (114.56 ± 6.07%).
3. Acetylcholine Release
Nicholson and colleagues (1988) measured AVM-induced presynaptic
acetylcholine release in a cockroach CNS synaptosomal preparation. The
synaptosomes were preloaded with [3H]choline and pp-rfused under
various conditions. A VM causes a release of radioactivity presumed to be
acetylcholine (this was not verified). The threshold AVM concentration
for acetylcholine release was 10- 10M, the EDso value was lO- 8M, and
maximum stimulation required lO- 6M. The effect was chloride dependent
and could be antagonized by picrotoxin, trioxabicyclooctaines, and
hexachlorocyclohexane. In contrast, the ,a-isomer of hexachlorocy-
clohexane potentiated the A VM effect. The authors conclude that in the
insect CNS, AVM induces the opening of a presynaptic chloride ion
channel (not GABA-activated), leading to efflux of chloride ions, depolar-
ization of the nerve terminal, and hence, to neurotransmitter release.
4. Interaction with Retinol Binding Proteins

A novel mode of action for the A VMs was suggested by Sani and Vaid
(1988), who demonstrated that AVM specifically binds to retinol-binding
protein isolated from parasitic worms of the family Filarioidea. Interest-
ingly, A VM has no affinity toward retinol-binding proteins isolated from
the host organism, and A VM has no affinity for retinoic acid-binding
proteins from either parasite or host tissues. Binding studies using
radiolabeled A VM and retinol reveal that A VM has a higher affinity than
retinol for the parasite retinol-binding protein. In addition, a correlation
exists between the binding affinities of A VM analogs and their antiparasi-
tic activities. These experiments suggest a binding protein-mediated
interrelationship may exist between the actions of retinol and A VM in the
parasites, but not in the host. This hypothesis will require further
experimentation.
5. Inhibition of Chitin Synthesis

A VM was reported to inhibit fungal growth by interfering with chitin
metabolism (Calcott and Fatig 1984a). The evidence for this conclusion
was based on data obtained with a methanol extract of a Streptomyces

avermitilis culture. The sample inhibited N-acetylglucoseamine incorpo-
ration into chitin, reduced chitin turnover in brine shrimp, and inhibited
purified chitinase. These results could not be confirmed using purified
AVM (Onishi and Miller 1985), and it was subsequently reported that the
antifungal activity was due to oligomycin and a polyene in the extract
(Gordnier, Brezner, and Tanenbaum 1987).
III. Vertebrates
Avermectins have been tremendously successful anthelmintic agents
because of their ability to kill parasites without affecting the host
organism. Many of the reports describing AVM effects in vertebrates
require concentrations of A VM far in excess of those that will ever be
attained outside of the laboratory. For this reason, the effects of AVM on
nontarget vertebrate species cannot necessarily be related to the mode of
action against invertebrate target species.
A. AVM STIMULATED NEUROTRANSMITTER RELEASE
AVM stimulates the release of endogenous GAB A from rat cerebral

cortex synaptosomes with an EC 50 of approximately 2 x lO- 6M (Pong,
Wang, and Fritz 1980). This response appears relatively specific because
endogenous glutamate was not released. Using rat brain slices from the
region of the caudate nucleus, Ishiko, Inagake, and Takaon (1985)
perfused the tissue with AVM (10- 3 to lO- 2M) and demonstrated an
inhibitory effect of calcium-dependent, potassium-stimulated dopamine
release. The inhibitory effect of AVM was completely antagonized by
lO- 5M picrotoxin and lO- 4M bicuculline. The superpharmacological
concentrations of AVM required to elicit these responses suggests that
they may not be of importance in understanding the mode of action of
AVM in target species.
B. AVM BINDING SITES
At least 4 different laboratories have investigated the specific binding

of [3H]AVM to membranes prepared from dog or rat brains. Pong and
Wang (1982) used synaptosomes prepared from dog brain and identified
specific AVM binding with an apparent dissociation constant, KD of
1.2 x lO- 9M. This high-affinity binding was saturable, and the density of
binding sites was estimated to be 1.54 pmollmg protein. Results from
competition experiments suggest that the binding is stereospecific and the
affinity to the binding sites correlates well with anthelmintic activity. The
binding sites are unevenly distributed in the dog brain; the highest density
is in the cerebellum (this is also the region of the brain with the highest
concentration of GABA binding sites). Similar binding sites were ob-

served in rat brain tissue. Significantly, these authors did not observe any
effect of GABA on AVM binding sites. Drexler and Sieghart (1984a,
1984b, 1984c) confirmed the existence of specific AVM binding sites in rat
brain membrane preparations. They also report an affinity of approxi-
mately 2 x 1O-9M. In contrast to Pong and Wang (1982), however, these
investigators observed partial inhibition of AVM binding in the presence
of GABA (approximately 40% inhibition of specific AVM binding in the
presence of 3 x 1O- 4M GABA). Other GABA agonists had similar
effects, and this could be blocked by GABA antagonists and also by
chloride ions. The presence of chloride ions in the incubation solution
may explain why Pong and Wang (1982) observed no effect of GAB A on
eH]A VM binding. Specific high-affinity binding of [3H]AVM was also
stimulated by the convulsant picrotoxin and inhibited by the anticon-
vulsant pentobarbital and the anxiolytic etazolate. These results support
an interaction between the AVM binding site and the GABA-
benzodiazepine receptor chloride channel complex. Two other laborato-
ries have examined [3H]A VM binding to rat brain membranes and report
KD values in the range of 20 nM (Schaeffer and Haines, in press; Huang
and Burg 1986). The reason for the discrepancy in the binding affinity is
not clear but may be explained by differences in nonspecific binding: in
the more recent reports (Huang and Burg 1986; Schaeffer and Haines, in
press) the glass fiber filters used in the binding assay were pretreated with
polyethyimine and Triton X-loo, resulting in a 90% decrease of nonspe-
cific binding.
C. A VM EFFECT ON PH]GABA BINDING

Because the data obtained in invertebrate systems suggest a link between
the action of A VM and GAB A binding sites, the effect of AVM on
[3H]GABA binding has been extensively investigated in rat brain tissue.
Pong and Wang (1982) reported that AVM stimulates [3H]GABA binding
to washed rat brain membranes (Figure 5.4). Scatchard analysis demon-
strated that AVM does not alter the affinity of GAB A for its binding site
but, rather, increases the number of binding sites by approximately 60%.
The AVM effect on GABA binding was antagonized by both bicuculline
and picrotoxin and is chloride-dependent. Similar experiments were
reported by Olsen and Snowman (1985). However, they report that AVM
(10-7M) decreases GABA binding to rat brain membranes (approximately
50% inhibition) and increases the binding of the GAB A antagonist
[3H]bicuculline, an effect that was potentiated by picrotoxin. These
results are a direct contradiction to those of Pong and Wang (1982), and
the explanation for this discrepancy is not known. Calcott and Fatig
(1984b) also report that AVM increases [3H]GABA binding to rat brain
membranes. The maximal stimulation was approximately 50% in the
2.0r-------r-------~------~------~
-
c-
.Qj
a
0.. 1.5
Ol
--'0Een
Q)
E
.e,
"0
C
::l 1.0
a
[D
~
Cii
~
·0
Q)
Q.
w 0.5
«
[D
«
<D
O~------~------~-------L------~
o 2 4 6 8
AVM (~M)
FIGURE 5.4. Dose-response enhancement ofNa+ -independent [3H] GAB A binding

by A VM in washed rat brain membrane preparations. The concentration of [3H]
GABA was 10 nM. Each point is the mean ± S.D. of 3 experiments performed in
triplicate (from Pong and Wang 1982).
presence of 4 x to- 7M AVM. At higher and lower concentrations of

AVM, the stimulatory effect was less pronounced.
D. A VM EFFECT ON BENZODIAZEPINE BINDING SITES
It is well documented that benzodiazepine agonists increase the affinity of

GABA A recognition sites associated with the GAB A receptor/chloride
ion channel complex (Guidotti, Tottano, and Costa 1978; Meiners and
Salama 1982). Several laboratories have reported that AVM potentiates
benzodiazepine receptor binding in rat, mouse, and human brain mem-
brane preparations (Figure 5.5; Williams and Yarbrough 1979; Paul,
Skolnick, and Zatz 1980; Pong, DeHaven, and Wang 1981; Supavilai and
Karobath 1981; Williams and Risley 1984; Sieghart et al. 1985). These
studies clearly demonstrate that AVM causes a concentration-dependent
13
0>
c::
=SOP
c::.c:
'CO.-
- 0> 11
EQ)
",3=
a.Ci)
Q)3=
~Q)
-::::I
'2 ~ 9
.2 ..."
-0>
IE 0--0 + NaCI
C')U5 A--A + NaCI + Picrotoxinin
-
.0
Q)
_ -
0
7 ....... - NaCI
13.5
Q) -
A--A - NaCI + Picrotoxinin
a.
(J)
5 i I
10 100
Avermectin (nM)
FIGURE 5.5. Effect of AVM on [3H]flunitrazepam binding and modulating'of
AVM-stimulated [3H]flunitrazepam binding by chloride ions and picrotoxinin.
Membranes from rat cerebellum were incubated with 1 nM [3H]flunitrazepam and
various concentrations of AVM in the absence or presence of p.M diazepam in
a solution containing 50 mM Tris-citrate, pH 7.3, in the absence or presence of
150 mM NaCI and/or 30 p.M picrotoxinin. Incubations were performed at 23°C for
30 minutes, and membranes were then filtered through Whatman GF/B filters and
washed as described in Materials and Methods. Results are from a typical
experiment which was repeated three times with identical results (from Williams
and Yarbrough 1979).
increase in the in vitro binding of eH]benzodiazepine to rat brain

membranes. AVM elicits both an increase in the affinity and number of
binding sites for eH]benzodiazepine. The concentration of AVM required
to increase the benzodiazepine binding varies in these reports from ECso
values of 40 to 250 x 1O-9M. In addition, the magnitude of the increase in
binding varies from 15% to 100% of the control.
AVM does not inhibit the binding of the benzodiapine receptor inverse
agonist eH],B-carboline-carboxylate methyl ester, and it enhances the
binding of the cage convulsant [3s S]t-butyl cyclophosphorothionate to
picrotoxin receptor sites. Interestingly, Williams and Risley (1982) exam-
ined a similar system and report that AVM does enhance [3H],B-carboline
carboxylate binding to rat brain membranes. The increase in binding
reflects an increase in the Bmax with no change in the Kn value. This
change is bicuculline insensitive.
84 M.J. Turner and J .M. Schaeffer
[35S]tert-butylbicyclophophorothionate (TBPT, a potent convulsant

that is thought to bind to the chloride channel associated with the GABA
benzodiazepine receptor complex) binding is stimulated by AVM in rat
brain membrane preparations (Drexler and Sieghart 1984c). Conversely,
TBPT either weakly stimulates or does not significantly influence the
specific high-affinity binding of [3H]AVM to the same membranes in the
absence or presence of chloride ions. These results indicate that AVM
and TBPT bind to different but closely associated binding sites.
E. CHLORIDE UPTAKE
Neurosynaptosomal preparations provide a model system for the study of

chloride influx into neuronal tissue (Harris and Allen 1985; Schwartz et al.
1986). It has been clearly established that GABA stimulates Cl- influx
into neuronal tissues. Soderlund and colleagues (1987) reported that in rat
neurosynaptosomes AVM is a weak activator of chloride uptake in the
absence of GABA as well as an inhibitor of GABA-stimulated chloride
uptake. Conversely, when these authors used mouse brain tissue, AVM
acted only as a noncompetitive antagonist of GABA-stimulated uptake.
The authors conclude that AVM binds with high affinity to a site
associated with, but not identical to, the GABA receptor. It is unclear
why mice and rats would have such different responses to A VM. In
addition, these results are difficult to reconcile with the fact that, in other
systems, AVM increases membrane permeability to chloride ions and is
acting as a GABA agonist, not a GABA antagonist. Our laboratory
(Schaeffer and Bergstrom, in press) recently demonstrated that GAB A
increases Cl- uptake into rat cerebral cortex neurosynaptosomes 125%
with an EC50 value of3 x 1O- 4M. AVM has no effect on chloride influx; it
does, however, potentiate the GABA-stimulated chloride influx, without
increasing the maximal response to GABA (the EC50 value is shifted to
9 x 1O- 5M). The reasons for the difference between these results and
those presented in earlier reports (Soderlund et al. 1987) are not under-
stood. However, the suggestion that AVM stimulates chloride uptake into
neurosynaptosomes correlates very well with the results reported by Sigel
and Baur (1987). They injected chick brain mRNA into Xenopus oocytes
and demonstrated the expression of functional GABA receptors. The
oocyte was voltage clamped and the effects of various compounds on the
GABA-induced chloride current were monitored. AVM strongly en-
hanced the GAB A-induced chloride current in a dose-dependent manner
(Figure 5.6). The half-maximal dose of AVM was 10-7M, which may be
physiologically significant in the mammalian system (however, not in
invertebrate systems). Interestingly, AVM did not affect the reversal
potential of the current or the maximal response elicited by GABA, nor
did it alter membrane permeability in the absence of GABA. The major
effects of AVM were a shift of the Ka for GABA from 21 to 2 x 1O- 6M,
and a decrease of the apparent Hill coefficient for GAB A from 1.7 to 1.1.
10
8
'E
....
Ql
....
::J
u 6
Ql
.~
Cij
(j) 4
a:
l
0 I I I I
10-8 10-7 10-6
Avermectin concentration (M)
FIGURE 5.6. Concentration dependence of the stimulatory effect by avermectin
B'a. The measurements were standardized by assigning the value of I to the
current amplitude elicited by a control application of 5 JLM GABA. Different
concentrations of avermectin B'a were applied for 3 minutes alone and then in
combination with 5 JLM GABA. For each point the Mean ± standard deviation is
given for at least three determinations performed on different oocytes (from Sigel
and Baur 1987).
F. OTHER ACTIONS OF A VM
Graham, Pfeiffer, and Betz (1982) identified and characterized glycine

binding sites associated with membranes isolated from rat spinal cord.
They report that A VM inhibits the binding of the glycine antagonist,
strychnine, with a KI value of 1.3 x 1O-6M. The high concentration of
A VM required for this action suggests that this may not be of physiologi-
cal significance.
There has been at least one report that A VM is a specific inhibitor of
protein kinase C activity isolated from rat brain (Ellis et al. 1987); in these
experiments, however, 10-6M AVM was required to inhibit 50% of the
enzyme activity. Once again, this high dose may not be physiologically
significant.
IV. Conclusions
It is not possible to assign a single mechanism of action for A VM in the
various systems that have been studied. In target organisms, the effect of
AVM is mediated via a specific, high-affinity (=10- IOM) binding site. The
physiological response to A VM binding is an increase in membrane
permeability to chloride ions, which is independent of GABA-mediated

chloride ion channels. GABA-gated chloride ions may also interact with
AVM, but at significantly greater AVM concentrations (=10- 7M). In
addition to their interaction with chloride channels, several other actions
of the AVMs have been proposed; the importance of these remains to be
resolved.
REFERENCES
Abalis 1M, Eldefrawi AT (1986) [3H]Muscimol binding to a putative GABA

receptor in honey bee brain and its interaction with avermectin B la • Pestic.
Biochem. Physiol. 25:279-287
Albert J, Lingle DH, Marder E, O'Neil MB (1986) A GAB A-activated chloride-
conductance not blocked by picrotoxin on spiny lobster neuromuscular prepa-
rations. Brit. J. Pharmacol. 87:771-779
Burg RW, Miller BM, Baker EE, Birnbaum J, Currie JA, Harman R, Kong VL,
Monaghan RL, Olson G, Putter I, Tunac JP, Wallick H, Stapley EO, Oiwa R,
Omura S (1979) The action of avermectin on identified central neurons from
Helix and its interaction with acetylcholine and gamma-aminobutyric acid
responses. Antimicrob. Agents Chemother. 15:361-367
Calcott PH, Fatig RO (1984a) Inhibition of chitin metabolism by avermectin in
susceptible organisms. J. Antibiot. 37:253-259
Calcott PH, Fatig RO (1984b) Avermectin modulation of gamma-aminobutyric
acid binding to membranes of rat brain, brine shrimp and a fungus, mucor
miehei. J. Antibiot. 37:797-801
Drexler G, Sieghart W (1984a) Properties of high affinity binding site for
tritium-labeled avermectin B la • Eur. J. Pharmacol. 99:269-277
Drexler G. Sieghart W (1984b) Evidence for association of a high affinity
avermectin binding site with benzodiazepine receptor. Eur. J. Pharmacol.
101:201-207
Drexler G. Sieghart W (1984c) Sulfur-35-labeled tert-butyl-bicyclophosphorothio-
nate and avermectin bind to different sites associated with the gamma-
aminobutyric acid-benzodiazepine receptor complex. Neurosci. Let. 50: 273-
277
Duce IR, Scott RH (1983) GABA sensitivity in the distal bundles of the locust
extensor tibiae muscle. J. Physiol. 343:32P
Duce IR, Scott RH (1985) Actions of dihydroavermectin B la on insect muscle.
Brit. J. Pharmacol. 85:395-401
Egerton JR, Ostlind DA, Blair LS, Eary DH, Suhayda D, Cifelli S, Riek RF,
Campbell WC (1979) Avermectins, new family of potent anthelmintic agents:
efficacy of the B la component. Antimicrob. Agents Chemother. 15:372-378
Ellis C, Nathwani B, Morrice N, Parker P, Evans FJ, Aitken A (1987) Ivermectin:
an inhibitor of protein kinase C -: A potential target enzyme for onchocerciases
chemotherapy. Brit. J. Pharmacol. 91:22P
Fritz LC, Wang CC, Gorio A (1979) Avermectin Bla irreversibly blocks pre-
synaptic potentials at the lobster neuromuscular junction by reducing muscle
membrane resistance. Proc. Natl. Acad. Sci. USA 76:2062-2066
Gordnier PM, Brezner J, Tanenbaum, SW (1987) Chitin metabolism: not a target
of avermectin/milbemycin activity in insects. J. Antibiot. 40:110-112
Graham D, Pfeiffer F, Betz H (1982) Avermectin Bla inhibits the binding of

strychnine to the glycine receptor of rat spinal cord. Neurosci. Let. 29:173-176
Guidotti A, Tottano G, Costa, E (1978) An endogenous protein modulates the
affinity of GAB A and benzodiazepine receptors in rat brain. Nature 275:553-
555
Harris RA, Allen AM (1985) Functional coupling of GABA receptors to chloride
channels in brain membranes. Science 228:1108-1110.
Huang L, Burg R (1986) The avermectin receptor of the American cockroach.
Soc. Neurosci. Abstr. #9.9
Ishiko J, Inagake C, Takaori S (1985) Effects of avermectin B la and picrotoxin on
striatal release of dopamine with reference to replacement of extracellular
chloride with nitrate. Neuropharm. 24:1147-1154
Kass IS, Stretton AOW, Wand CC (1984) The effects of avermectin and drugs
related to acetylcholine and 4-aminobutyric acid on neurotransmission in
Ascaris suum. Molec. & Biochem. Parasit. 13:213-225
Kass IS, Wang CC, Walrond JP, Stretton AOW (1980) Avermectin B la , a
paralyzing anthelmintic that affects interneurons and inhibitory motoneurons in
Ascaris. Proc. Natl. Acad. Sci. USA 77:6211-6215
Lummis CR, Sattelle DB (1985) GABA and benzodiazepine binding sites in insect
CNS. Pestic. Sci. 16:61-65.
Martin R, Pennington AJ (1988) Effect of dihydroavermectin Bla on CI single-
channel currents in Ascaris muscle. Neurotox. '88 Abstr. 141
Meiners BA, Salama AI (1982) Enhancement of benzodiazepine and GABA
binding by the novel anxiolytic, tracazolate. Euro. J. Pharmacol. 78:315-322
Mellin TN, Busch RD, Wang CC (1983) Postsynaptic inhibition of invertebrate
neuromuscular transmission by avermectin Bla • Neuropharm. 22:89-96
Miller TW, Chaiet L, Cole DJ, Flor JE, Goegelman RT, Gullo VP, Joshua H,
Kempf AJ, Krellwitz WR, Monaghan RL, Ormond RE, Wilson KE, Albers-
Schonberg G, Putter I (1979) Avermectins, a new family of potent anthelmintic
agents: isolation and chromatographic properties. Antimicrob. Agents Che-
mother. 15:368-371
Nicholson R, Robinson PS, Palmer PJ, Casida JE (1988) Ivermectin-stimulated
release of neurotransmitter in the insect central nervous system: modulation by
external chloride and inhibition by a novel trioxabicyclooctane and two
polychlorocyloalkane insecticides. Neurotox '88 Abstr. #96
Olsen RW, Snowman AM (1985) Avermectin Bla modulation of gamma-
aminobutyric acid-benzodiazepine receptor binding in mammalian brain.
J. Neurochem. 44:1074-1082
Onishi JC, Miller TW (1985) The lack of antifungal activity by avermectin Bla •
J. Antibiot. 38:1568-1572
Paul SM, Skolnick P, Zatz M (1980) Avermectin Bla : an irreversible activator of
the gamma-aminobutyric acid-benzodiazepine-chloride-ionophore receptor
complex. Biochem. Biophys. Res. Commun. 96:632-638
Pong S-S, DeHaven R, Wang CC (1981) Stimulation ofbenzodiazepine binding to
rat brain membranes and solubized receptor complex by avermectin Bla and
GABA. Biochem. Biophys. Acta 646:143-149
Pong S-S, Wang CC (1982) Avermectin B la modulation of gamma-aminobutyric
acid receptors in rat brain membranes. J. Neurochem. 38:375-379
Pong S-S, Wang CC, Fritz LC (1980) Studies on the mechanism of action of
avermectin B 1a : stimulation of release of gamma-aminobutyric acid from brain

synaptosomes. J. Neurochem. 34:351-358
Sani BP, Vaid A (1988) Specific interaction of ivermectin with retinol-binding
protein from filarial parasites. Biochem. J. 249:929-932
Schaeffer JM, Bergstrom AR (1988) Identification of gamma-aminobutyric acid
binding sites in Caenorhabditis elegans. Life Sci. 43:1701-1706
Schaeffer JM, Bergstrom AR (in press) The effect of ivermectin on chloride
uptake by rat brain synaptosomes. Biochem. Biophys. Res. Comm.
Schaeffer JM, Haines H (in press) Avermectin binding in Caenorhabditis elegans:
a two-state model for the avermectin binding site. Biochem. Pharm.
Schwartz RD, Sudak PD, Paul SM (1986) GAB A- and barbituate-mediated
chloride uptake in rat brain synaptoneurosomes: Evidence for a rapid desensiti-
zation of the GAB A receptor-coupled chloride ion channel. Molec. Pharm.
30:419-426
Sieghart W, Eichinger A, Riederer P, Jellinger K (1985) Comparison of ben-
zodiazepine receptor binding in membranes from human or rat brain. Neuro-
pharm.24:751-760
Sigel E, Baur R (1987) Effect of avermectin B 1a on chick neuronal gamma-
aminobutyric acid receptor channels expressed in Xenopus oocytes. Molec.
Pharm. 32:749-752
Soderlaud DM, Adams PM, Bloomquist JR (1987) Differences in the action of
avermectin B 1a on the GABAA receptor complex of mouse and rat. Biochem.
Biophys. Res. Commun. 146:692-698
Stretton AOW, Fishpool RM, Southgate E, Donmoyer JE, Walrond JP, Moses
JER, Kass IS (1978) Structure and physiological activity of the motorneurons of
the nematode Ascaris. Proc. Natl. Acad. Sci. USA 75:3493-3497
Supavilai P, Karobath M (1981) In vitro modulation by avermectin B 1a of the
GABA/benzodiazepine receptor complex or fat cerebellum. J. Neurochem.
36:798-805
Tanaka K, Matsumura F (1985) Action of avermectin B 1a on the leg muscles and
nervous system of the American cockroach. Pest. Biochem. Physiol. 24: 124-
135
Williams M, Risley EA (1982) Interaction of avermectins with [3H],B-carboline-3
carboxylate ethyl ester and [3H] diazepam binding sites in rat brain cortical
membranes. Eur. J. Pharmacol. 77:307-312
Williams M, Risley EA (1984) Ivermectin interactions with benzodiazepine
receptors in rat cortex and cerebellum in vitro. J. Neurochem. 42:745-753
Williams M, Yarbrough GG (1979) Enhancement of in vitro binding and some of
the pharmacological properties of diazepam by a novel anthelmintic agent,
avermectin B 1a • Eur. J. Pharmacol. 56: 1273-276
CHAPTER 6
Toxicology
George R. Lankas and Lea R. Gordon
I. Introduction for Both Ivermectin and Abamectin: Similarities

II. Toxicity Studies with Ivermectin
A. Genotoxicity
B. Acute Toxicity
C. Subchronic Toxicity
D. Reproduction/Developmental Toxicity
III. Toxicity Studies with Abamectin
A. One-Year Dog Study
B. Chronic Toxicity and Carcinogenicity in Rats and Mice
C. Dermal Penetration in Monkeys
IV. Summary and Conclusions
I. Introduction
Abamectin (MK-0936) is a natural fermentation product of Streptomyces
avermitilis. Ivermectin (MK-0933) is a synthetic derivative of abamectin.
The chemical structure of abamectin differs from ivermectin only in the
bond between carbons 22 and 23; abamectin has a double bond where
ivermectin has a single bond and additional hydrogens on C-22 and C-23
(Figure 6.1). Both compounds are a mixture of homologous products with
B la and BIb components. The BIb component differs chemically from
the B la component by only 1 methylene (CH2) unit at the 26-carbon
position: the ethyl group (C2H5) is a methyl group (CH3) in the BIb form.
Abamectin and ivermectin are defined as containing a minimum of
80% B la and a maximum of 20% BIb components. Studies in our
laboratories have clearly demonstrated that the individual components
have very similar biological and toxicological properties and, for all
practical purposes, can be considered equivalent.
GAB A receptors. Abamectin and ivermectin increase calcium perme-
90 George R Lankas and Lea R Gordon
0 0
MK-0933 (Iverme~~!~)
(Bla component
093 6 (Abamectin) )
t1lC-
(B1a componen t shown
FIGURE 6 0 1 The chemical structures 0 fivermectm B 1a (above) an d abamectin B1ao

0 o o
6. Toxicology 91
ability by their interaction with GABA-gated chloride channels but it is

now thought that this is not the primary basis of their anthelmintic action
(see Chapter 5). The compounds' mechanism of toxicity in mammals is
unknown but GABA is a mammalian central nervous system neuro-
transmitter and effects on GAB A may be relevant to their safety in
mammals. The fact that much higher concentrations of these compounds
are needed in mammals as compared to nematodes to affect neurological
function may be due to the lack of a specific, high affinity site associated
with neuronal function or to the relatively poor penetration of these large
molecular weight compounds into the CNS.
In general ivermectin is slightly less toxic than abamectin in laboratory
animals. Clinical signs of toxicity for abamectin and ivermectin in
laboratory animals are identical, depending on the species: mydriasis
(pupillary dilatation) in dogs, emesis in monkeys, and convulsions and/or
tremors and coma at higher doses in most species.
II. Toxicity Studies Conducted with

Ivermectin (MK-0933)
A. GENOTOXICITY STUDIES
Ivermectin has been tested for genotoxic activity in a series of in vitro

bacterial and mammalian cell assays. In bacteria ivermectin was studied
for mutagenic activity by assessing its ability to revert Salmonella
typhimurium (strains TA1535, TA1537, TA98, and TAI00) from histidine
prototrophy to auxotrophy as described by Ames (Ames, McCann, and
Yamasaki 1975). Each component ofivermectin was tested separately in a
similar fashion, and all tests were done with and without metabolic
activation systems prepared from rat liver. None of the agents studied
produced significant increases in reversion to histidine prototrophy under
any of the test conditions, at concentrations up to 2000 ug/plate.
Ivermectin was studied in the mouse lymphoma assay (Clive et al.
1972; Clive and Spector 1975) to evaluate the compound's potential to
produce a forward mutation at the thymidine kinase (TK) locus of Fischer
L5178Y mouse lymphoma cells. The cytotoxicity dose range-finding
studies revealed that ivermectin apparently is detoxified in the presence
of S-9 rat liver homogenate, a microsomal activation system. Therefore,
higher doses of the test compound could be used with L5178Y cells in the
presence of S-9 than in its absence. The drug, dissolved in DMSO, was
added to cell cultures with and without S-9. Final concentrations of
ivermectin were 5 to 20 ug/ml in the absence of the S-9 system and 20 to
60 ug/ml in the presence of S-9. The cells were treated for 4 hours, and
then washed, fed, and cultured in growth medium for 3 days. At the end of
92 George R. Lankas and Lea R. Gordon
this expression period, TK'- mutants were detected by cloning the cells
in selection medium containing bromodeoxyuridine for 10 to 14 days. The
surviving cell population was determined by plating diluted aliquots in
nonselective growth medium.
The results showed that ivermectin at concentrations up to 60 ug/ml did
not produce any significant increase in mutation frequency (with or
without an S-9 metabolic activation system) when compared with appro-
priate negative controls.
Unscheduled DNA synthesis (UDS) occurs after cellular DNA is
damaged by agents that induce either base, short-patch, or long-patch
excision repair (Wilkins and Hart 1973; Setlow and Carrier 1964). The
damaged bases are enzymatically excised from their DNA strand, creat-
ing a gap; new bases are inserted and polymerized to fill the gap, using the
intact opposite strand as template. Since many carcinogenic and muta-
genic agents have been demonstrated to induce UDS, the assay is
considered to be a measure of genotoxicity (Williams, Laspia, and Dunkel
1982).
Ivermectin was assayed for its ability to induce UDS in an in vitro assay
using a normal human embryonic lung fibroblast cell strain, IMR-90, at
concentrations of 10 to 1000 ug/ml. The compound was tested with and
without the addition of a rat liver S-9 metabolic activation system.
Ivermectin did not induce UDS in human cells at any dose level.
In conclusion, ivermectin was tested in vitro for genotoxic activity in
microbial and mammalian cell mutagenesis assays and in human fibro-
blasts for DNA damage as measured by unscheduled DNA synthesis. In
all instances, ivermectin had no genotoxic activity.
B. ACUTE TOXICITY OF IVERMECTIN
As discussed in Chapter 5, ivermectin interacts with GABA receptors,

resulting in enhanced release and binding of GABA to its receptors. In
contrast to lower organisms, GABA is found only in the brain and spinal
cord of mammals, with no detectable levels in the peripheral nervous
system (Cooper 1982). Therefore, it was expected that the acute toxicity
of ivermectin would be due to its effects in the CNS. This hypothesis has
proved correct in acute toxicity studies conducted in a variety of
laboratory animal species, although the exact mechanism of neurotoxicity
remains unknown.
Table 6.1 shows the acute LDso values for ivermectin in a variety of
species using several different routes of administration. Rodents are
significantly more sensitive to ivermectin toxicity than are other species.
In mice the LDso values after oral and intraperitoneal administration were
approximately 25 and 30 mg/kg, respectively. The acute oral LDso values
in mice of the individual ivermectin components (ivermectin BJa and B Jb)
were comparable to each other and to ivermectin.
6. Toxicology 93
TABLE 6.1. Acute toxicity of ivermectin.

Species Route of administration LDSO (mg/kg)
Mouse Oral 25
Mouse Intraperitoneal 30
Rat Oral 50
Rat Intraperitoneal 55
Rat (infant) Oral 2 to 3
Rat Inhalation -'
Rat Dermal >660
Rabbit Dermal 406
Dog Oral - 80
Rhesus monkey Oral > 24
a Maximum attainable concentration of 5.11 mg/litter produced transient
irritation of mucous membranes but no deaths or other evidence of toxicity
after 1 hour exposure.
The drug's effects in mice were similar after either oral or intraperi-
toneal administration: they consisted of ataxia (uncoordinated move-
ment), bradypnea (abnormally slow respiration rate), and tremors at all
dosage levels within approximately 1 hour of treatment. Death, preceded
by loss of righting, occurred from 75 minutes to 6 days after dosing, with
most occurring within 24 hours. There were no significant sex-related
differences in acute toxicity in mice after either route of administration.
In rats treatment-related physical signs consisted of ataxia, ptosis
(drooping upper eyelids), and decreased activity; the symptoms appeared
from about 4 to 24 hours after oral or intraperitoneal administration.
Death, generally preceded by loss of righting, occurred overnight to the
second day. Body weight gains during the 2-week observation period
appeared unaffected by drug administration. There were no significant
sex-related differences.
Ivermectin was more toxic in infant rats (1-2 days old) than in young
adult rats. As discussed in more detail in Section D, this enhanced
sensitivity is believed due to postnatal development of the blood-brain
barrier in rats (Betz and Goldstein 1981) (in humans this structure is
formed prenatally; Bohr and Mollgard 1974). This conclusion is supported
by significantly higher brain/plasma drug concentration ratios in neonatal
rats compared to adults.
In rats and rabbits acute dermal LD50 values for ivermectin after 24
hours of occluded exposure were >660 and 406 mg/kg, respectively.
These values are significantly greater than those discussed previously for
oral exposure in rats and mice, suggesting that ivermectin is poorly
absorbed percutaneously. The poor dermal absorption of abamectin has
been confirmed in rhesus monkey (Macaca mulatta) studies, demonstrat-
ing that 0.5% or less of a dermally applied dose is absorbed (see section
III A below).
The acute oral toxicity of ivermectin in dogs was assessed by adminis-
tration of the drug dissolved in sesame oil at dosage levels of 5, 10,20,40,

or 80 mg/kg to groups of 2 males and 2 females each. Mydriasis was the
most sensitive indicator of toxicity in dogs, occurring at all dosage levels.
The incidence and duration of mydriasis were dose-related, with the
effect lasting up to several days in some dogs. More severe signs of CNS
intoxication, including ataxia and tremors, occurred at doses of 10 mg/kg
and above. Two animals in the 80 mg/kg group died between 12 and 24
hours posttreatment, while 1 dog in the 40 mg/kg group died 4 days after
treatment. These deaths were preceded by a comatoselike state. On the
basis of this data, the oral LD50 in dogs is approximately 80 mg/kg.
In view of the increased sensitivity of rodents, particularly mice, to the
acute toxicity ofivermectin compared to other species, including humans,
a comparative acute oral toxicity study in rhesus monkeys was conducted
with ivermectin and abamectin to determine the sensitivity of nonhuman
primates to these compounds. In this study single oral doses of either
ivermectin or abamectin dissolved in sesame oil were administered to
groups of 2 males and 2 females. The doses for each compound ranged
from 0.2 to 24 mg/kg, with intervals of 2 to 3 weeks between administra-
tion of the next higher dose to the same group of monkeys. Plasma
concentrations of both compounds were measured at frequent intervals
after treatment, with doses of 2, 8, and 24 mg/kg. After each treatment, all
animals were carefully examined for clinical signs of drug effect.
No drug-related physical signs were found with either drug at doses of
0.2,0.5, or 1.0 mg/kg. As shown in Table 6.2, the most sensitive indicator
of toxicity was emesis, which occurred in 1 of 4 monkeys with both
compounds at 2.0 mg/kg. Therefore, the no-observable-effect level
in rhesus monkeys with each compound is 1.0 mg/kg. At doses up to
12.0 mg/kg the only consistent effect was an increase in the incidence and
a decrease in the time of onset of emesis from about 8 to 4 hours after
treatment. At 24.0 mg/kg, the highest dose tested, marked mydriasis
occurred, as well as slight sedation and emesis. Recovery from these
effects was complete by 48 hours for both ivermectin and abamectin. No
tremors or convulsions were observed in any monkeys, and all survived
the highest dose tested of 24.0 mg/kg.
The minimum toxic dose (MTD) in rhesus monkeys is 2.0 mg/kg, about
tenfold higher than the human therapeutic dose for onchocerciasis
therapy (Greene et al. 1985). The peak plasma concentrations following
the MTD are shown in Table 6.2. The 110 ng/ml plasma level for
ivermectin at this dose is approximately 5.5-fold the mean peak plasma
level of 20 ng/ml in humans administered the 0.2 mg/kg therapeutic dose
of ivermectin (M. Dobrinska, personal communication; see Chapter 20).
Despite the approximately proportional increases in plasma levels with
increasing dose, the severity of clinical signs did not worsen appreciably.
Even at 24 mg/kg of ivermectin, a dose producing plasma levels about
thirty-fourfold above the human therapeutic level, all monkeys survived.
Therefore, the relatively steep dose-response curve in rodents for toxicity
6. Toxicology 95
TABLE 6.2. Acute toxicity and plasma concentrations of

ivermectin and abamectin in rhesus monkeys and humans.
Monkeys Humans
Ivermectin Abamectin Ivermectin
Therapeutic dose 0.2 mg/kg
Peak-plasma levels 20 ng/ml
Minimum effect level 2 mg/kg 2 mg/kg _a
Peak-plasma levels 110 ng/ml 76 ng/ml

Signs emesis emesis
Toxic effect level 8 mg/kg 8 mg/kg 6.6-8.6 mg/kg
Peak-plasma levels 270 ng/ml 150 ng/ml unknown
Signs emesis emesis emesis,
mydriasis,
sedation
Toxic effect level 24 mg/kg 24 mg/kg _a
Peak-plasma levels 680 ng/ml 390 ng/ml

Signs emesis, mydriasis,
sedation
Fatalities no deaths no deaths
a -Not determined
with the avermectins is not reproduced in monkeys, indicating that in

primates clinical evidence of toxicity may occur well in advance of
serious or life-threatening toxicity. This view is further supported by the
case of a child who survived a dose of ivermectin of about 7 to 8 mg/kg
(Table 6.2) with comparable signs of toxicity as observed in rhesus
monkey at similar dosage levels (R.J. Nessel, personal communication).
Monkeys can tolerate doses of ivermectin that produce more severe
forms of toxicity or death in other species, most notably the mouse but
also the dog and the rat. For example, acute oral doses of ivermectin at or
slightly above the 0.2 mg/kg therapeutic dose have produced toxicity and
even death in some mice. At this dose no adverse effects are produced in
rhesus monkey or humans, despite plasma drug levels in these species
that are equivalent to those found in mice. The oral LDso values
for ivermectin in dog, mouse, and rat vay from approximately 11.6 to
80.0 mg/kg, whereas in monkeys the minimum acute lethal dose is greater
than 24 mg/kg. Plasma-level data show that the tolerance of these high
doses by monkeys is not due to decreased absorption with increased
doses. It appears that primates, in general, possess a relatively lower
degree of sensitivity to the potential toxicity of ivermectin.
C. SUBCHRONIC TOXICITY
The toxicity ofivermectin was assessed in 3-month studies in Crl:CD(BR)

SD rats and beagle dogs and in a 2-week study in rhesus monkeys. In
addition, a 2-week oral study in neonatal rhesus monkeys was conducted,
since a previous study in neonatal rats (see Section II A above) suggested

that neonates may be more sensitive to ivermectin toxicity. The results of
these studies are summarized in Table 6.3.
A 3-month oral toxicity study in rats with in utero exposure at dosage
levels of 0.4, 0.8, and 1.6 mg/kg/day was conducted in which offspring
from dams treated with ivermectin for 2 weeks prior to mating and during
pregnancy were selected at weaning for continuation on treatment. There
were no abnormal physical signs or mortality attributable to treatment.
Also, there were no ocular abnormalities and no changes in body weight
gain, routine hematologic, or serum biochemical parameters which were
related to treatment.
The spleens of 4 rats (3 at 1.6 mg/kg/day and 1 at 0.8 mg/kg/day) were
enlarged. On microscopic examination there were varying degrees of
congestion of the red pulp and varying amounts of extramedullary
hematopoiesis. These rats also had iron-positive pigment in the renal
tubular epithelium. These changes suggest a possible drug-related intra-
vascular hemolysis, but the mechanism is unknown. There was reactive
hyperplasia of the bone marrow in these rats, probably related to
intravascular hemolysis. There were no other microscopic changes
related to treatment at these dosage levels. A 0.4 mg/kg/day no-observed-
TABLE 6.3. Summary of results of subchronic oral toxicity studies

with ivermectin.
Dose levels
Species Duration. (mg/kg/day) Results
Rats 3 months 0.4, 0.8, and 1.6 Splenic enlargement and
reactive hyperplasia
suggestive of
intravascular hemolysis
at 0.8 mg/kg/day and
above. NELa = 0.4
mg/kg/day
Dogs 3 months 0.5, 1.0, and 2.0 Mydriasis at 1.0
mg/kg/day and above.
Tremors, ataxia, and
anorexia at 2.0
mg/kg/day. NEL = 0.5
mg/kg/day
Rhesus monkeys (Macaca 2 weeks 0.3, 0.6, and 1.2 No treatment-related
mulatta) effects up to 1.2
mg/kg/day. NEL > 1.2
mg/kg/day
Neonatal rhesus monkeys 2 weeks 0.04 and 0.1 No treatment-related
(Macaca mulatta) effects up to 0.1
mg/kg/day. NEL > 0.1
mg/kg/day
• NEL: No-effect level
6. Toxicology 97
effect level was established in this study. Similar changes were not seen in
3-month, I-year, or 2-year studies with abamectin in rats at comparable
and higher dose levels, nor in other species given ivermectin.
In a 3-month oral toxicity study in beagle dogs the compound was
administered as a solution in sesame oil by gavage at doses of 0.5, 1.0, or
2.0 mg/kg/day. Mydriasis and slight weight loss were observed at doses of
1.0 and 2.0 mg/kg/day. Four of 8 dogs in the 2.0 mg/kg/day group
developed tremors, ataxia, anorexia, and dehydration, and were sacri-
ficed prior to the scheduled necropsy. No gross abnormalities were
observed in these dogs. Mydriasis was observed in dogs receiving 1.0 or
2.0 mg/kg/day. No other treatment-related effects were found.
Postmortem studies conducted on all animals disclosed no treatment-
related gross, microscopic, or organ weight changes. A no-effect level of
0.5 mg/kg/day was established for this study.
A 16-day oral toxicity study with ivermectin was conducted to deter-
mine its toxicity in immature domestic-bred rhesus monkeys (approxi-
mately 1 to 1.5 years old). Each of the 3 treatment groups were dosed
daily via nasogastric intubation at dosage levels of 0.3, 0.6, and
1.2 mg/kg/day. These dosage levels were arbitrarily chosen to provide an
approximate sixfold safety margin relative to the human clinical dose of
ivermectin; based on the acute toxicity in rhesus monkeys, these doses
would not be expected to produce clinical signs of toxicity. The control
group received the vehicle, sesame oil, at the same 1.0 mllkg dose volume
as the treated groups.
No drug-related physical signs were noted in any of the treated animals.
All animals survived the duration of the study with similar body weight
gains for all treated groups as compared to the controls. No treatment-
related ocular lesions or changes in hematology or serum biochemical
parameters were found during the study. There were no organ weight,
gross, or microscopic changes attributed to treatment. Therefore, treat-
ment of immature monkeys with 1.2 mg/kg/day for 14 days failed to
produce any treatment-related effects.
Because ivermectin is excreted in the milk of lactating animals, it was
necessary to generate appropriate preclinical data on the safety of
low-level exposure to nursing infants via maternal milk. Based on a
clinical study in lactating women treated with ivermectin, a maximum
milk level of the drug of 23 ng/ml was found on the day following
treatment, with levels depleting to below the 0.1 ng/ml detection limit by
approximately 1 week after treatment (M. Dobrinska, personal commu-
nication). This level of drug in human milk would result in an estimated
maximum-acute dosage of about 3 to 4 ug/kg in a nursing infant.
To assess the potential significance of a low-level neonatal exposure to
ivermectin, a study of neonatal rhesus monkeys was conducted. At
initiation of the study all animals were 7 to 13 days old. The animals were
divided into 2 treatment groups of 5 males and 3 females each; they
received ivermectin dissolved in sesame oil via nasogastric intubation at

dosage levels of either 40 or 100 ILg/kg/day for 14 days.
Results of the physical examinations revealed no treatment-related
physical signs. The results of the ophthalmic examinations and the daily
check for mydriasis and pupillary light response indicated no treatment-
related differences between the treated and control groups. Body weight
changes were similar between control and treated animals, and the results
of the hematologic and serum biochemical examinations indicated no
treatment-related effects. Similarly, there were no treatment-related
changes in organ weights and no gross postmortem or microscopic
changes related to ivermectin administration at doses up to 100 f.J.g/kg/
day. Based on the results of this study, exposure of nursing infants to low
levels of ivermectin is unlikely to produce toxicity.
D. DEVELOPMENTAL AND REPRODUCTIVE TOXICITY
The developmental toxicity of ivermectin was assessed by daily adminis-

tration throughout the major period of organogenesis in mice, rats, and
rabbits. The reproductive toxicity of the drug was determined by con-
ducting a rat multigeneration study. The results of these studies are
summarized in Table 6.4.
TABLE 6.4. Summary of results of developmental and reproductive toxicity studies

with ivermectin.
Dose NELa
Species (mg/kg) Observations (mg/kg)
Mouse 0.1,0.2,0.4,0.8 Mortality, tremors, Matemotoxicity 0.1
convulsions, coma developmental 0.2
at 0.2, 0.4, and 0.8 toxicity
mg/kg; cleft palate
at 0.4 and 0.8
mg/kg
Rat 2.5, 5.0, 10.0 Sedation, cleft palate Matemotoxicity and 5.0
at 10 mg/kg developmental
toxicity
Rabbit 1.5, 3.0, 6.0 Sedation and Matemotoxicity 3.0
decreased body developmental 1.5
weight at 6 mg/kg; toxicity
decreased fetal
weights, increased
number of fetal
deaths, cleft palate,
and clubbed
forepaws at 3.0 and
6.0 mg/kg
Rat Multi-generation 0.05,0.1,0.2, Neonatal mortality at Neonatal and 0.2
0.4 0.4 mg/kg/day developmental
toxicity
• NEL: No-effect level

6. Toxicology 99
To determine the potential for developmental toxicity in mice, iver-

mectin was administered orally in sesame oil to pregnant mice on days
6 through 15 of gestation. The dosage levels used were 0.1, 0.2,0.4, and
0.8 mg/kg/day. There was treatment-related mortality in each of the 3
highest dosage levels, as shown in Figure 6.2. Deaths occurred after 1 to 8
doses. An increased incidence of cleft palate was found in fetuses from
dams receiving either 0.4 or 0.8 mg/kg/day. There were no treatment-
related abnormalities among fetuses at 0.1 or 0.2 mg/kg/day, despite the
fact that doses of 0.2 mg/kg/day and above produced tremors, convul-
sions, coma, and death in some maternal animals.
The developmental toxicity of ivermectin in rats was determined by
oral drug administration as a sesame oil solution to dams on days 6
through 17 of gestation. Dosage levels were 2.5,5.0, and 10.0 mg/kg/day.
Maternotoxicity, characterized by sedation, was observed in rats in the
10 mg/kg/day group and a low incidence of cleft palate was found in
fetuses in the high dose group (Table 6.4). There were no other treatment-
related malformations or evidence of maternotoxicity in this or any other
treatment group.
In a developmental toxicity study pregnant rabbits were orally adminis-
tered invermectin on days 6 through 18 of gestation at dose levels of 1.5,
3.0, and 6.0 mg/kg/day. Slight to marked sedation was present in the
6.0 mg/kg/day females 24 hours after the seventh dose, and slight
2&
• Maternal death •
• Litter. "lth cleft palate
20
o 0.025 0.05 0.0'75 0.1 0.2 0.4 0.8

Dose ml/kl
FIGURE 6.2. Ivermectin: comparison of maternal mortality and incidence of cleft

palate in CF, mice.
sedation persisted in some females up to 8 days after cessation of dosing.

No clinical signs of toxicity were present in females at 1.5 or 3.0
mg/kg/day. In the 6.0 mg/kg/day group, there was a significant (P:5 0.05)
decrease in mean maternal body weight during the period of drug
administration; this was probably related to the sedation and likely lower
food consumption in dams in this group. There were no treatment-related
effects on mean maternal body weights in the 1.5 and 3.0 mg/kg/day
groups.
There was a low but dose-related incidence of cleft palate and clubbed
forepaws among fetuses in the 3.0 and 6.0 mg/kg/day dosage groups. At
3.0 mg/kg/day 1 fetus had cleft palate and 5 fetuses from 1 litter had
clubbed forepaws. Eight fetuses from 3 litters at 6.0 mg/kg/day had cleft
palate, and clubbed forepaw was observed in 6 fetuses from 3 litters.
There were no malformations observed among fetuses at 1.5 mg/kg/day.
The results of the developmental toxicity studies with ivermectin
demonstrate that a low incidence of abnormalities was produced only at
dosage levels at or near those producing severe signs of
maternotoxicity-including death-in some dams. Therefore, ivermectin
is not considered to be selectively embryotoxic. Years of widespread
clinical use of ivermectin in pregnant animals of a variety of species have
failed to produce any evidence of developmental toxicity, further sup-
porting the above conclusion.
Results of initial multigeneration studies of ivermectin in rats indicated
increased pup mortality several days after parturition and decreased pup
weights at maternal doses as low as 0.4 mg/kg/day. In an attempt to
determine the threshold dose for neonatal toxicity, a multigeneration
study was conducted in which ivermectin was administered to male
and female rats once daily throughout the production of 2 litters in
each of 3 successive generations at dose levels of 0.05, 0.1, 0.2, and 0.4
mg/kg/day.
There was no treatment-related mortality, nor were there physical signs
of toxicity or effects on reproduction among parents or offspring in any
dosage group throughout the production of2litters in each of the FO, FIb,
and F2b generations. The body weights of offspring (Fla through F3b)
during lactation (days 1 to 21 postpartum) were not adversely affected by
drug administration. Treatment-related effects on body weight gain were
limited to a slight but statistically significant (P :5 0.05) decrease in mean
body weight gain among FIb females in the 0.4 mg/kg/day group in the
postweaning period. Comparable effects were noted among weanling
male and female rats at the same dose level in a previous ivermectin
multigeneration study. On the basis of this study, and other multigen-
eration studies in rats at higher dosage levels, 0.4 mg/kg/day appears to
be near the threshold dose in dams for producing neonatal toxicity. Doses
:5 0.2 mg/kg/day produced no neonatal toxicity or other reproductive
effects.
6. Toxicology 101
The results of the multigeneration study and the acute toxicity study in
infant rats (see Section II B above) indicated that noenatal rats were more
sensitive to the drug than parental animals. To determine the reason for
this difference in sensitivity, female rats were administered tritium-
labeled ivermectin Bla at a dosage level of 2.5 mg/kg/day continuously
for 61 days prior to mating and during mating, gestation, and lactation.
Plasma and milk samples from dams and plasma samples from offspring
were analyzed for total radioactivity on days 1, 4, and 6 postpartum. On
day 10 postpartum all dams and offspring were sacrificed and total
radioactivity determined in plasma and brain samples.
The total drug residues in plasma and milk of dams and plasma from
their offspring are shown in Figure 6.3. The increase in maternal plasma
radioactivity at parturition relative to days 4 to 10 postpartum is likely due
to increased utilization of fat at parturition in the rat (Amano 1967; Scow
et al. 1964), resulting in release ofivermectin from adipose tissue (Chiu et
al. 1986). The concentrations of ivermectin in milk were consistently
three- to fourfold higher than those obtained from maternal plasma
samples on comparable days postpartum. Plasma drug levels in offspring
were relatively low on day 1 postpartum but increased rapidly from days 6
to 10 such that they equaled or exceeded the values in the dams. It should
be noted that the 0.8 ug/ml mean plasma concentration in offspring on day
10 postpartum is approximately 40 times the therapeutic plasma concen-
tration in humans and other species. Of greater importance is the finding
that the brain concentrations of ivermectin residues in offspring by day 6
postpartum were approximately tenfold the maternal brain concentrations
(Figure 6.4).
The results of this study indicated that the high drug concentrations in
milk from lactating dams chronically administered the drug were probably
responsible for the toxicity observed among neonatal offspring in multi-
generation studies. This conclusion is supported by the high drug levels
found in the plasma and brain of offspring relative to adult rats. In
particular there was a greater concentration of drug in the brain of
neonatal rats during the early neonatal period of enhanced sensitivity, and
this may explain the increased toxicity. This early neonatal period of
increased sensitivity to ivermectin toxicity in rats correlates with the
postnatal completion of the blood-brain barrier in this species (in humans
it is formed prenatally; Betz and Goldstein 1981; Bohr and Mollgard 1974;
Saunders 1977). Further support for the importance of postnatal exposure
in the toxicity observed in neonatal rats is derived from a cross-fostering
study, in which control pups cross-fostered to ivermectin-treated dams
exhibited the same degree of neonatal mortality and toxicity as rats
exposed in utero and fostered to drug-tested dams (R. T. Robertson,
personal communication). Therefore, postnatal exposure via milk-not a
combination of prenatal and postnatal exposure-explains the neonatal
toxicity observed in multigeneration studies conducted with ivermectin.
1.2
a
1
0.8
Plasma
cone 0.8
(p,g/ml)
0.4
0.2
5
b
Milk 3
cone
(p,g/ml) 2
1.4
C
1.2
1
F1
plasma 0.8
cone 08
(p,g/ml) .
0.4
0.2
0
1 4 6 10
Days postpartum
FIGURE 6.3. Mean concentrations of 3H-ivermectin in (a) plasma, (b) milk from
lactating rats, and (c) plasma from offspring of treated dams. Error
bars = ± standard deviation.
III. Toxicity Studies with Abamectin

Abamectin differs from ivermectin by a single double-bond at the C22-23
position. Abamectin is a pesticide with many different agricultural uses. It
has been tested in acute, subacute, and genotoxicity assays similar to
those for ivermectin with similar results, except that abamectin is slightly
more toxic than ivermectin (Tables 6.5 and 6.6). Neither compound
6. Toxicology 103
0.5
0.4
0.3
Brain
cone
(ppm)
0.2
0.1
- - -
0.0
1 4 6 10
Days postpartum
FIGURE 6.4. Mean 3H-ivermectin levels in brain from dams and offspring. Dashed
line represents steady state brain concentration of dams measured on day 10
postpartum. Error bars = ± standard deviation.
is genotoxic. The oral LDso in mice is 14-24 mg/kg for abamectin and
25-40 mg/kg for ivermectin; the minimal effect level for toxicity in dogs is
0.50 mg/kg/day for abamectin and 1.0 mg/kg/day for ivermectin. In
developmental toxicology studies on abamectin, as for ivermectin, cleft
palates were seen at or near matemotoxic doses. In multigeneration
TABLE 6.5. Comparative toxicity of abamectin and ivermectin in laboratory
animals.
Type of Abamectin Ivermectin
Study Species (mg/kg/day)
Oral LD50 CFt mouse 13.6-23.8 24.6-40
Oral LD50 CRCD rat 10.6-11.3 42.8-52.8
Oral LD50 CRCD 1.52 2.3
neonatal
rat
Acute toxicity Monkeys MEL = 2 mg/kg MEL = 2 mg/kg
Minimum Effect Level (MEL)
(mg/kg/day)
Abamectin Ivermectin
Suckling CRCD rat 0.12 0.4
neonatal
pup
Subacute oral Beagle dog 0.50 1.0
Chronic and CRCD rat 2.0' 0.8
subchronic
• 105-week study
TABLE 6.6. Results of teratology studies with abamectin and ivermectin in three species.
Mouse Rabbit Rat
Abamectin Ivermectin Abamectin Ivermectin Abamectin Ivermectin
Maternotoxicity
NEL" 0.05 0.1 1.0 3.0 1.6 5.0
MELb 0.075 0.2 2.0 6.0 2.0 to.O
Fetotoxicity
NEL 0.2 0.8 1.0 1.5 1.6 10.0
MEL 0.4 2.0 3.0
Developmental
toxicity
NEL 0.2 0.2 1.0 1.5 1.6 5.0
MEL 0.4 0.4 2.0 3.0 to.O
" NEL = No effect level (mg/kg/day)

b MEL = Minimal effect level (mg/kg/day)
C No adverse effects
d No cleft palates seen at the matemotoxic dose of 2.0 mg/kg/day
studies pup toxicity was the most sensitive parameter, with minimum
effect levels of 0.4 mg/kg/day for both compounds. Additional long-term
studies of abamectin are discussed below (Table 6.7).
A. ONE-YEAR ORAL TOXICITY STUDY IN DOGS
In a I-year oral toxicity study in dogs there was no cumulative toxicity.

Four groups of 6 male and 6 female 20 to 24-week-old beagle dogs were
given abamectin in the feed at doses of 0, 0.25, 0.50, and 1.0 mg/kg/day.
One dog in the high-dose group died and 2 were killed in moribund
condition. Clinical signs included mydriasis, weight loss, lethargy, trem-
ors, and recumbency. All of these deaths were attributed to abamectin
toxicity.
Mydriasis was seen in dogs in the middle (3% incidence) and high (15%
incidence) dose groups. Mydriasis has been a consistent finding in other
species-dogs, horses, monkeys-exposed to increased doses of iver-
mectin as well. No treatment-related changes in the structure of the eye
were seen at the routine ophthalmologic examination nor by histopa-
thology.
There were no gross or microscopic changes attributed to treatment
even in the 3 high-dose dogs that died or were killed with treatment-
related clinical signs. This has been true in other studies in dogs as well as
in mice and rats. The no-effect level for abamectin given in the diet to
dogs was 0.25 mg/kg/day.
B. CHRONIC TOXICITY AND CARCINOGENICITY DATA

IN RATS AND MICE
In a chronic toxicity and carcinogenicity study with a I-year intermin

necropsy, 5-week-old Crl:CD(SD) BR rats were assigned to 5 groups of 65
6. Toxicology 105
TABLE 6.7. Chronic studies with abamectin.

Dose
Species Duration mg/kg/day Findings
Rat 14 weeks 0.1, 0.2, and 0.4 (in sesame oil No treatment-related signs of
gavage) toxicity. NELa = 0.4 mg/kg/day
Rat 8 weeks 1.4-5.8 mg/kg/day: (diet) 75% deaths at 60 ppm; 50% deaths
at 40 ppm (about 5 mg/kg/day).
No effect at 5 and 10 ppm (1.4
mg/kg/day). At 2.1 mg/kg, 2/20
had tremors; at 2.7 mg/kg, 8/20
had tremors. NEL = 1.4
mg/kg/day. MTDb estimated at
2.0 mg/kg/day
Rat 53 weeks 0.75, 1.5, and 2.0/2.5/2.0 (diet) When high dose was increased to
2.5 mg/kg/day a few animals
developed tremors so the dose
was reduced to 2.0 mg/kg/day
No treatment-related gross or
microscopic changes. No
oncogenic effect seen.
NEL = 1.5 mg/kg/day
Mouse upto84d Initial study up to about 4 mg/kg/ d Body weight decrease seen at 40
CD-I caused no tox. Doses were ppm (about 8 mg/kg/day) and 60
increased in some groups and ppm (about 12 mg/kg/day)
more dose groups were added to (-11% and -29% respectively).
12 mg/kg/d (diet) No tremors or convulsions.
NEL = 4 mg/kg/day.
MTD = 8 mg/kg/day
Mouse 94 weeks 2, 4, 8 mg/kg/d (diet) Body weight gain depression at
CD-I high dose (males 7%, females
21%) (P < 0.05). No
treatment-related gross or
microscopic changes. No
oncogenic effect. NEL = 4
mg/kg/d
Dog 12 weeks 0.25, 0.5, 1.0, 2.0/4.0 (diet) Extreme weight loss at 2.0 and
4.0-discontinued dosing.
Mydriasis 2: 1.0 mg/kg/day.
NEL = 0.5 mg/kg/day
Dog 53 weeks 0.25,0.5, 1.0 (diet) Mydriasis seen at 1 mg/kg/day.
Intermittent mydriasis at 0.5
mg/kg/day. No treatment-related
gross or microscopic changes.
NEL = 0.25 mg/kg/day
a NEL = No Effect Level
b MTD = Maximum tolerated dose
male and 65 female rats each: Control I, Control II, and abamectin given
in the feed at 0.75, 1.50, and 2.0 mg/kg/day. The levels of abamectin in the
feed were adjusted and mixed biweekly to attain the targeted dose levels.
Treatment-related increases in body weight gain were seen in both
males and females in all dosage groups during the first year on study. At
drug week 60, when the peak difference in body weight gain was apparent
between treated and control rats, the treated females had 30%, 14%, and
13% increased body weight gain in the 0.75, 1.5, and 2.0 mg/kg/day
groups, respectively, when compared with controls. By the end of the
study (drug week 105) there were comparable mean body weight gain
values for all groups of females when averaged over the entire study
(P > 0.05); however, the increases in body weight gain over the com-
bined controls among the treated males were still statistically significant
(P s 0.05) (Robson 1959) at drug week 105 (21%, 9%, and 6% increases in
the 0.75, 1.5, and 2.0 mg/kg/day groups, respectively).
Clinical signs of toxicity were limited to tremors seen in a few rats in the
high-dose group and an occasional rat in the mid-dose group that
consumed enough feed to exceed the 2.0 mg/kg/day dose ofthe high-dose
group. Although some of the rats died with tremors, there were no gross
or microscopic changes seen in the nervous or muscular systems to
account for these tremors. This has been true for other species as well.
Abamectin was not carcinogenic in rats when given in the diet at doses
of 0.75, 1.5, and 2.0 mg/kg/day for 105 weeks. There was no statistically
significant (P > 0.05) increase in tumor incidence resulting from treatment
with abamectin (Mantel 1963, 1980; Mantel et al. 1982; Peto 1974; Peto et
al. 1980; Harter 1957; Cutler and Ederer 1958; Tukey, Ciminera, and
Heyse 1985). There were also no treatment-related nonneoplastic histopa-
thologic changes in this study. The no-effect level for toxicity was
1.5 mg/kg/day.
In a chronic toxicity and carcinogenicity study, Crl:CD-l (ICR) BR
mice, about 6 weeks old, were assigned to 5 groups: Control I, Control II,
2,4, and 8 mg/kg/day of abamectin in the feed.
Treatment-related tremors were seen in drug weeks 89 and 91 in 2
female mice in the high-dose group.
Increased mortality was seen among the high-dose males (P < 0.05) but
not the females (Mantel and Ciminera 1979). Dosing was stopped among
the high-dose males in drug week 90 due to the low number of survivors
(40%). All other mice were treated until sacrifice in drug week 94.
Treatment-related decreases in body weight gain (P < 0.05) were seen
among both the high-dose males and females (7% and 21%, respectively)
(Robson 1959). A dose-related trend toward increased food consumption
(g/kg) was seen among the females but not the males, and the high-dose
females had 20% decreased feed efficiency compared to controls (Tukey,
Ciminera, and Heyse 1985).
There were no treatment-related ophthalmic changes. There were no
effects of treatment on the hematologic or serum biochemical findings.
There were no treatment-related organ weight changes or gross lesions.
Abamectin was not carcinogenic to mice when given in the diet for 94
weeks at doses of 2, 4, and 8 mg/kg/day. There was no statistically
significant (P > 0.05) increase in tumor incidence reSUlting from abamec-
6. Toxicology 107
tin treatment (Mantel 1963, 1980; Mantel et al. 1982; Peto 1974; Peto et al.
1980; Harter 1957; Cutler and Ederar 1958; Tukey, Ciminera, and Heyse
1985). The no-effect level for clinical evidence of toxicity is 4 mg/kg/day.
These results demonstrate that abamectin, the close structural analog of
ivermectin, is not carcinogenic in rodents when administered at maximum
tolerated doses for 22 to 24 months.
C. DERMAL PENETRATION STUDY IN MONKEYS
Because the main potential route of exposure to abamectin's pesticidal

use on crops is dermal, a study was conducted to determine the amount of
dermal absorption of abamectin. The forearms of monkeys were exposed
to the concentrated emulsifiable concentrate (E.C.) of abamectin, the
diluted E.C. of abamectin, and a suspension of abamectin in alcohol.
Each of these 3 exposures represents a possible way humans working
with the product could be exposed. The mixer/loader could be exposed to
the concentrate, the applicator could be exposed to the dilute E.C., and
the harvester could be exposed to the powder (suspension). The monkey
was chosen as the most appropriate model to mimic skin absorption in
humans (Wester and Maibach 1975, 1983).
Tritium-labeled abamectin was used to quantitate the excretion of the
compound after i. v. and dermal administration. The tritium label is stable
in this biological matrix in the 5 position and does not significantly
exchange. By the i. v. route 95% of the 6-mg dose and 97% of the 300-mg
dose were excreted in the urine and feces. Greater than 90% of the
radioactivity was excreted in the feces within 7 days of the i. v. dosing.
After dermal exposure> 99% of the radioactivity was recovered from the
application site. Dermal penetration of abamectin was extremely low
(0.17%-0.55% of applied dose) based on plasma levels and recovery from
application site (see Table 6.8). Generally the dermal absorption of
abamectin was < 1% of applied dose. Because abamectin does not readily
penetrate the skin, perhaps because it is a high molecular-weight com-
pound, the hazard to agricultural workers is significantly less than it
would be otherwise. For the same reasons dermal penetration of iver-
mectin should also be very low.
IV. Summary and Conclusions

Acute oral toxicity studies with ivermectin in mice, rats, and monkeys
have shown clear species differences in sensitivity, with rodents being
uniquely sensitive to ivermectin's CNS toxicity. This conclusion is based
upon the observation that doses of 0.2 mg/kg in mice, and slightly higher
doses in rats, have produced clinical signs of drug effect (tremors and
ataxia); this same dose has been shown to be without adverse effect in
TABLE 6.8. Percent dermal absorption of abamectin in rhesus monkeys.

Average Average
percent absorbed percent absorbed
Treatment (excreta)" (plasma)b
6 ug EC d 1 hr diluted 0.23 0.26
6 ug EC 10 hr diluted 0.55 0.19
300 ug 1 hr technical suspension" 0.17 1.22
300 ug 10 hr technical suspension 0.29 ndc
300 ug EC 1 hr 0.23 0.19
300 ug EC 10 hr 0.52 ndc
a % excreted (topical)
% excreted (i.v.)
b normalized AUC (topical)
normalized AUC (i.v.)
The amounts of radioactivity excreted or in the plasma were normalized to the treatment
group since the amount of radioactivity dosed in each treatment was not identical.
AUC = area under the curve
C The percent could not be determined because the level of radioactivity in the plasma
samples was below the background level

d EC = emulsifiable concentrate of abamectin
e Technical suspension = abamectin powder in water
clinical studies in a variety of other species, including humans. In

addition, an acute oral study in monkeys demonstrated that the minimum
toxic dose in this species was 2.0 mg/kg or approximately tenfold the
human clinical dose, which resulted in plasma drug levels 5.5 times
greater than those in humans following treatment with 0.2 mg/kg of the
drug. Doses up to 24.0 mg/kg in monkeys produced only slight increases
in the observed toxic effects (emesis, mydriasis, and sedation), indicating
a much flatter dose-response curve in this species compared to rodents.
The finding that an acute overdose of ivermectin in a child exposed to
approximately 8.0 mg/kg orally produced clinical signs similar to those
produced in monkeys and that these effects rapidly reversed suggests that
monkeys, and not mice, are the most appropriate model for predicting the
acute toxic effects of ivermectin in humans.
Toxicity studies of 14 weeks duration in rats and dogs indicated no
adverse effects at doses of 0.4 and 0.5 mg/kg/day, respectively. At doses
of 0.8 mg/kg/day and above in rats a low incidence of splenic enlargement
was found. In dogs 1.0 mg/kg/day and above produced mydriasis, but
tremors, ataxia, and anorexia were produced only at the highest dose
level of 2.0 mg/kg/day.
A 2-week oral toxicity study in immature rhesus monkeys (approxi-
mately 1 to 1.5 years old) at doses up to 1.2 mg/kg/day demonstrated no
adverse effects, indicating no increase in sensitivity and little or no drug
accumulation in immature monkeys in view of the 1.0 mg/kg no-effect
level in the acute oral toxicity study in more mature monkeys.
6. Toxicology 109
A 2-week oral toxicity study in neonatal monkeys-with all animals

less than 2 weeks old at initiation of the study-produced no evidence of
toxicity with doses up to 100 ug/kg/day. Therefore, low-level human
neonatal exposure to < 100 ug/kg of ivermectin is unlikely to produce
toxicity.
In 3 laboratory species, developmental toxicity with ivermectin was
found only at dosage levels at or near those which also produced signs of
severe maternotoxicity. This indicates that ivermectin is not a selective
developmental toxin (Khera 1984; Brent 1986). In the mouse, the most
sensitive species tested, ivermectin was both maternotoxic and produced
fetal cleft palate at dosage levels of 0.4 mg/kg/day and above. Materno-
toxicity, characterized by tremors and death, without teratogenicity, was
also observed at dosage levels as low as 0.2 mg/kg/day, the safe human
therapeutic dose, clearly demonstrating that the mouse is an inappropri-
ate model for predicting the toxicity of ivermectin. In the rat develop-
mental toxicity study, dosages of 10 mg/kg/day, the highest dose tested,
produced maternotoxicity characterized by sedation and an unthrifty
appearance and cleft palates in affected offspring. At dosage levels of
5 mg/kg/day no maternotoxicity or fetal abnormalities were observed. In
rabbits, maternotoxicity, including sedation and weight loss, resulted at a
dosage level of 6.0 mg/kg/day. Cleft palate and clubbed forefeet were
found at dosages of 3.0 mg/kg/day and above. No evidence of materno-
toxicity or developmental toxicity was observed in rabbits receiving
1.5 mg/kg/day. The developmental toxicity data in laboratory animals
with ivermectin suggest that the drug is not selectively toxic to the
developing embryo and adverse effects are unlikely to occur at dose
levels which are nonmaternotoxic. This conclusion is further supported
by the widespread use of the drug in a variety of domestic animal species
with no adverse effects on development reported, even when used at
severalfold the 0.2 mg/kg therapeutic dose (Campbell and Benz 1984).
In an extensive series of multigeneration studies in rats no effects on
mating, fertility, pregnancy, or gestation length were found at doses up to
3.6 mg/kg/day, the highest dose tested. At doses ~ 1.2 mg/kg/day
decreases in body weight gain were found in treated dams compared to
controls. Postnatal toxicity characterized by decreased weight gain and
pup mortality was found in FI offspring at doses ~ 0.4 mg/kg/day. These
effects were found to be related not to in utero exposure but to postnatal
exposure via maternal milk. The three- to fourfold higher levels of
ivermectin in maternal milk compared to plasma resulted in significantly
higher brain and plasma levels of the drug in nursing offspring. The period
of enhanced neonatal sensitivity in rats (up to approximately day 10
postpartum) correlates with the completion of the blood-brain barrier in
this species. No adverse reproductive or postnatal effects were found at
doses ::s; 0.2 mg/kg/day.
Ivermectin has been used extensively in humans at an oral dose of
0.2 mg/kg for the treatment of onchocerciasis without any adverse

drug-related effects (Greene et al. 1985). In addition, more limited clinical
safety data are available in humans receiving 0.2 mg/kg/day on 2
consecutive days, demonstrating no adverse effects (K. Brown, personal
communication). In contrast to humans, single doses of 0.2 mg/kg in mice
have produced severe signs of toxicity and death in some animals.
Therefore, the mouse is clearly inappropriate for human risk assessment
of toxicity to ivermectin.
Comparative studies with abamectin and its semisynthetic analog,
ivermectin, in primates and other species have shown that the 2 com-
pounds produce virtually identical effects at the same or very similar
dosage levels. Carcinogenicity studies with abamectin at maximum
tolerated dosage levels for 22 and 24 months duration in mice and rats,
respectively, have demonstrated the lack of carcinogenicity of this
compound. The lack of genotQxic activity with ivermectin in a variety of
bacterial and mammalian genetic toxicity assays, along with its close
chemical structure and biological activity with abamectin, supports the
conclusion that ivermectin should also be noncarcinogenic.
In using these data for the establishment of acceptable levels of human
exposure to ivermectin in food residues, it must be remembered that there
are clear species differences regarding ivermectin toxicity. Primates,
including humans, are clearly less sensitive than rodents, particularly
mice. Therefore, the use of the lowest no-effect level (NEL) in the most
sensitive species (0.2 mg/kg in the mouse teratology study) provides an
even greater safety margin when calculating the acceptable daily intake of
ivermectin residues. Further safety margins are evident by recognizing
that ivermectin levels in muscle are extremely low compared to liver, the
target tissue used in establishing appropriate withdrawal periods in most
animal species (see Chapter 10). The extremely low therapeutic dose level
of 0.2 mg/kg in most species and limited number of treatments per year
result in extremely low total drug residues. Therefore, the very low
human exposure to ivermectin residues relative to the toxicity and safety
data accumulated in a variety of animal species, including humans treated
for onchocerciasis, supports the continued safe use of ivermectin in
veterinary practice.
REFERENCES
Amano Y (1967) Changes of the levels of blood glucose during pregnancy in the
rat. lap. l. Pharmacol. 17:105-114
Ames BN, McCann J, Yamasaki E (1975) Methods for detecting carcinogens and
mutagens with the Salmonella/mammalian microsome mutagenicity test.
Mutat. Res. 31:347-364
Betz L, Goldstein GN (1981) Developmental changes in metabolism and transport
properties of capillaries isolated from rat brain. l. Physiol. 312:365-376
Bohr V, Mollgard K (1974) Tight junctions in human fetal choroid plexus
visualized by freeze-etching. Brain Res. 81:314-318
6. Toxicology 111
Brent RL (1986) Definition of a teratogen and the relationship of teratogenicity to

carcinogenicity. Teratol. 34:359-360
Campbell WC, Benz GW (1984) Ivermectin: a review of efficacy and safety.
J. Vet. Pharm. & Therap. 7:1-16
Chiu SH, Sestokas E, Taub R, Buhs RP, Green M, Sestokas R, Vandenheuval
WJ, Arison BH, Jacob TA (1986) Metabolic disposition ofivermectin in tissues
of cattle, sheep, and rats. Drug Metab. & Dispos. 14:590-600
Clive D, Flamm W, Machesko M, Bernheim J (1972) A mutational assay system
using the thymidine binase locus in mouse lymphoma cells. Mutat. Res.
16:77-87
Clive D, Spector JASF (1975) Laboratory procedure for assessing specific locus
mutations at the TK locus in cultured L5178Y mouse lymphoma cells. Mutat.
Res. 31:17-29
Cooper JR (1982) Amino acids. In Cooper JR, Bloom FR, Roth RH (eds) ,
The Biochemical Basis of Neuropharmacology, 4th ed., Oxford University
Press, p 250
Cutler SJ, Ederer F (1958) Maximum utilization of the life table method in
analyzing survival. J. Chron. Dis. 8:699-712
Greene BM, Taylor HR, Cupp EW, Murphy RP, White AT, Aziz MA, Schulz-
Key H, D'Anna SA, Newland HS, Goldschmidt LP, Auer C, Hanson AP,
Freeman SV, Reber EW, Williams PN (1985) Comparison of ivermectin and
diethylcarbamazine in the treatment of onchocerciasis. New Eng. J. Med.
313:133-138
Harter HL (1957) Error rates and sample sizes for range tests in multiple
comparisons. Biometrics 13:511-36
Khera KS (1984) Maternal ToxicitY-A possible factor in fetal malformation in
mice. Teratol. 29:411-416
Mantel N (1963) Chi-square tests with one degree of freedom; extensions of the
Mantel-Haenszel procedure. J. Am. Stat. Assoc. 58:690-700
Mantel N (1980) Assessing laboratory evidence for neoplastic activity. Biometrics
36:381-99
Mantel N, Ciminera J (1979) Use of log-rank scores in the analysis of litter-
matched data on time to tumor appearance. Cane. Res. 39:4308-4315
Mantel N, Tukey JW, Ciminera JL, Heyse, JF (1982) Tumorigenicity assays,
including use of the jackknife. Biomet. J. 24:579-596
Peto R (1974) Guidelines on the analysis of tumor rates and death rates in
experimental animals. Brit. J. Cane. 29:101-105
Peto R, Pike MC, Day NE, Gray RC, Lee PN, Parish S, Peto J, Richards S,
Wahrendorf J (1980) Guidelines for simple, sensitive significance tests for
carcinogenic effects in long term animal experiments. In International Associa-
tion for Research on Cancer Monographs, Supplement 2, Lyon, France, pp.
365-367
Robson DS (1959) A simplified method for constructing orthogonal polynomials
when independent variable is unequally spaced. Biometrics 15:187-191
Saunders NR (1977) Ontogeny of the blood-brain barrier. Exper. Eye Res.
(Suppl.), pp 523-550
Scow RO, Chernick SS, Brinley MS (1964) Hyperlipemia and ketosis in the
pregnant rat. Am. J. Physiol. 206:796-804
Setlow RB, Carrier WL (1964) The disappearance of thymine dimers from DNA:
An error correcting mechanism. Proc. Natl. Acad. Sci. USA 51:226-231
Tukey JW, Ciminera JL, Heyse JF (1985) Testing the statistical certainty of a
response to increasing doses of a drug. Biometrics 41:295-301
Wester RC, Maibach HI (1975) Percutaneous absorption in the rhesus monkey
compared to man. Toxieol. & App. Pharm. 32:394-398
Wester RC, Maibach HI (1983) Cutaneous pharmacokinetics. 10 steps to percuta-
neous absorption. Drug Metab. Rev. 14(2): 169-205
Wilkins RJ, Hart RW (1973) Preferential DNA repair in human cells. Nature
247:35-36
Williams GM, Laspia MF, Dunkel, VC (1982) Reliability of the hepatocyte
primary culture/DNA repair test in testing coded carcinogens and non-
carcinogens. Mutat. Res. 97:359-370
CHAPTER 7
Pharmacokinetics of Ivermectin in
Animals and Humans
David W. Fink and Arturo G. Porras
I. Introduction
II. Cattle
III. Sheep
IV. Dogs
V. Swine
VI. Horses
VII. Humans
Introduction
The pharmacokinetic properties of ivermectin are a function of the
species in which the compound is studied. Ivermectin is effective against
parasites in a wide variety of hosts-including cattle, sheep, dogs, swine,
and horses. A previous review of ivermectin has described the effects of
formulation and route of administration on its pharmacokinetic properties
in animals (Lo et al. 1985). That publication included examples of
formulation modifications directed toward the development of oral and
parenteral dosage forms; it also illustrated the use of drug plasma
concentrations for characterizing the drug and for modifying formulations
for specific efficacy. The following summary presents the results of
representative pharmacokinetic and bioavailability studies of this drug in
different animal species and in humans. It includes descriptions of the
experimental procedures as well as the various measurement techniques
used in these studies to describe the pharmacokinetic properties of
ivermectin.
114 David W. Fink and Arturo G. Porras
Cattle
Commercially, ivermectin is most widely used for treating cattle. Bio-
availability studies have been run in this species using intravenous and
subcutaneous injectable formulations. To measure the intrinsic pharma-
cokinetic properties of ivermectin in cattle, the tritium-labeled compound
was formulated in a solution of 0.07 mCi/mg specific activity at a
concentration of 10.0 mg/ml; the solvent consisted of 60% (v/v) propylene
glycol and 40% (v/v) glycerol formal (Hibbert and Carter 1928) containing
5% polyvinylpyrrolidone. This solution was administered to cattle in a
single intravenous bolus at a dose of 300 ILg/kg body weight.
The pharmacokinetic results are shown in Figure 7.1. The data exhibit a
typical biexponential decay based upon the usual 2-compartment open
model.Figure 7.1 is a plot of the total concentration in plasma over time
following IV administration as determined by HPLC and radiochemical
analyses. The close agreement between the radioactivity data and the
HPLC data (the average relative difference is 5%) confirms that there are
no metabolites of ivermectin in the plasma over this time interval. This is
important in bioavailability studies, because the intact drug must be
discriminated in the plasma from possible degradates and metabolites of
similar structures. This selectivity is especially important for ivermectin,
because the structure is so complex that there are many structural
changes and degradation processes that can occur. For example, one
metabolic transformation of ivermectin (resulting from incubation at 37°C
with rat or steer liver microsomes) is the formation of the C(24)-
hydroxymethyl compounds and their respective monosaccharides (Miwa
et al. 1982).
The bioavailability analyses in this study used a sensitive high-pressure
liquid chromatographic (HPLC) analytical method. This method uses
fluorescence detection after preliminary sample preparation by liquid-
liquid partitioning and conventional gravity-fed column chromatography
(Tolan et al. 1980; Fink 1982). The potency ofivermectin requires the use
of extremely sensitive analytical procedures. This derivatization tech-
nique was developed to increase sensitivity for detection by forming an
intensely fluorescent compound based upon an analytical aromatization
reaction.
The results of the experiment depicted in Figure 7.1, in addition to the
biexponential decay, reveal a very short distributive phase in cattle, k = 6
day-I, a biological half-life t1l2 of 2.8 days, a very large volume of
distribution, 1.911kg (consistent with the lipophilicity of this compound),
and an area under the curve (AUC) of about 700 day-ng/ml. In an
independent study, application of a 3-compartment model to the data
obtained from 6 cattle also yielded an equal biological half-life t1l2 = 2.7
days. In this study, the 3-compartment (triexponential) model provided a
statistically significantly closer fit to the data than the 2-compartment
7. Pharmacokinetics ofIvermectin in Animals and Humans 115
1500 r----r----.---~----,----.r---._----._--~
o HPLC
1000
[J Radiotracer
700
E
.....
C>
c:
C 500
0
~
....
E
CI)
0
400
c:
8
c:
U
CI) 300
E
....
CI)
~
200
100 L...-_--'-_ _"'--_--'-_ _' - - _ - - ' -_ _' - - _ - - ' - _ - - - '
o 2 4 6 8 10 12 14
Time (hr)
FIGURE 7.1. Comparison of bioavailability analyses and radiotracer counts in
cattle plasma following intravenous dose. Data for 0.5 day following injection.
Dose rate: 300 ILg/kg body weight. Specific Activity: 0.07 mCi/mg.
(biexponential) model in 3 of the 6 animals (Wilkinson, Pope, and Baylis

1985). Schnitzerling and Nolan (1985) also found that the biological
half-life of ivermectin is 3 days in cattle, using a normal-phase chroma-
tographic system that does not resolve the 2 components of ivermectin.
These workers studied the binding of ivermectin to plasma protein: they
demonstrated that the drug is carried mainly in the plasma (80%), and that
this distribution equilibrium between plasma and red blood cells remains
relatively constant with time.
Figure 7.2 depicts the results of a replicate study in 2 cattle measured
by chemical analysis and extending out to 4 weeks following the intrave-
nous dose. The plasma samples again were collected at specific time
intervals and analyzed as described previously (Tolan et al. 1980). The
insert in this figure shows the rapid distribution phase in cattle: the
biexponential decay mode is more clearly evident in this expanded scale.
1000
~,0,0...
300 300
0.......,
0-----"0
-E01
C 60
C
0
-:0=
c
~
30
C
CI)
(.)
c
0
(.) 10
c 6
-:0=
(.)
CI)
E
~ 3
~
1.0
0.6
0.3
0
Time, days
FIGURE 7.2. Plasma concentration of ivermectin in cattle for 1 month following
intravenous administration. Mean of 2 cattle.
7. Pharmacokinetics of Ivermectin in Animals and Humans 117
The plasma profile of the commercially available ivermectin cattle

injectable formulation (IVOMEC) is shown in Figure 7.3. The dosage
form is a 1% (w/v) solution of the drug in a mixed solvent vehicle of
60% (v/v) propylene glycol and 40% (v/v) glycerol formal. As shown in
Figure 7.3, a subcutaneous injection, administered at a dose of 200 ltg/kg
body weight, attains a peak plasma concentration, Cp, of 44 ng/ml
ivermectin within a time tp of 1 day. It is striking that the biological
half-life afforded by this injectable formulation, 8.3 days, is significantly
longer than the intrinsic half-life of the drug, 2.8 days. This effect has been
attributed to a slow absorption (kJ process that dominates the pharmaco-
kinetics ofthe nonaqueous injectable product (Lo et al. 1985). The clinical
significance was noted by Campbell and Benz (1984) in a review of the
efficacy and safety of ivermectin: these workers reported that the
anthelmintic efficacy of this product persisted for 2 weeks after the
treatment of cattle, and that this prolonged effect was consistent with
these plasma levels of the drug (Figure 7.3).
Formulation modifications, based on the solubility properties of the
drug,. can have a significant effect on the bioavailability of ivermectin in
cattle following subcutaneous administration. For example, changes in
52
~t:Jl •
c:
C 39
('-\
0
+= • •
....
0
cCD
()
c: 26
0
()
c:
"-"-
+=
13r
()
CD
....E
CD
.2
•
00
0
I 6
I
12
----...-.-
18
.
24 • 30
Time, days
FIGURE 7.3. Bioavailability of ivermectin in cattle dosed subcutaneously with
IVOMEC propylene glycol-glycerol formal [60: 40 (v/v)] vehicle at 200 ~g/kg
body weight. Formulation concentration: 10 mg/ml ivermectin. Each point is the
average plasma concentration obtained from 10 cattle.
solvent composition can be used to achieve prolonged blood levels. As

illustrated in Figure 7.4, the bioavaiIability method was used to compare
two experimental formulations, labeled 4-1 and 4-2. Formulation 4-1 is an
aqueous micellar solution of ivermectin formulated to overcome its
solubility limitations in water (Lo and Williams 1983). A micelle is formed
with a surface-active agent, such as polyoxyethylene sorbitan monooleate
(polysorbate 80, Tween 80), and a cosolvent such as glycerol formal in the
presence of other substrates, such as benzyl alcohol. Formulation 4-2 is
modified from 4-1 by mixing it with glycerol formal solvent; Formulation
4-2 is designed for prolonged absorption following subcutaneous dosing.
The plasma concentrations achieved (Figure 7.4) with the aqueous
product (4-1) are significantly higher than with the modified formulation
(4-2); the former vehicle affords a much greater bioavaiIability (AUC)
than the latter. The aqueous formulation (4-1) is more rapidly and
90
0
80
70
~01 4-'
c: 60
--e
C
0
.2
50
\
cQ):
u
c:
8
0
40
c:
:;::
u
/0,\
Q)
E
~
30
~
0t2 '---
20
10
-
o~o ___
12 14
Time,doys
FIGURE 7.4. Effect of aqueous/nonaqueous solvent ratio on bioavailability of
ivermectin following subcutaneous administration to cattle at 200 ~g/kg body
weight. Formulation 4-1 is an aqueous (micelle) solution; 4-2 is an aqueous-
glycerol formal 50 : 50 (v/v) mixed solvent vehicle. (Modified from Lo et al. 1985.)
TABLE 7.1. Effect of injection vehicle on the bioavailability

of ivermectin in cattle dosed subcutaneously at 200 J-Lg/kg
body weight. (Adapted from Lo et al. 1985.)
Formulation"
Parameter 4-1 4-2
Mean plasma peak time (days) 2
Peak plasma concentration (ng/m!) 84 ± 7 25 ± 14
Biological half-life (days) 2.0 ± 0.3 3.7 ± 0.7
AUC (day-ng/m!) 246 ± 46 186 ± 81
Fraction of drug absorbed (%) 55 41
"4-1: aqueous (micelle): 4-2: Mixed aqueous (micelle)-glycerol
formal 50% (v/v)
extensively absorbed and, in the traditional sense, is actually the more

efficient delivery system of the two.
Table 7.1 is a summary of the pharmacokinetic data of Figure 7.4. A
significant difference in peak plasma concentration between the two
formulations is evident. The biological half-lives are also significantly
different, resulting from the slow absorption process that dominates the
pharmacokinetics in nonaqueous injectable products. The biological
half-life of the aqueous micellar product, 2.0 ± 0.3 days, is in the range of
the half-life calculated from the intravenous data (Figure 7.2), and results
from the rapid ka offered by this injection. Finally, it is seen that the area
under the curve (AVe) from the micellar solution is 246 day-ng/ml, which
is at least 25% greater than that obtained from the modified formulation. It
is apparent that the modified micellar solution (4-2)-with its intermediate
mixed-solvent system-exhibits pharmacokinetic properties that are in-
termediate between those of the aqueous micellar product (4-1) and the
commercial nonaqueous solution (Figure 7.3).
Sheep
For administration to sheep, ivermectin was dissolved in a mixed solvent
of 60% (v/v) propylene glycol-40% (v/v) glycerol formal containing 5%
polyvinylpyrrolidone; it was given as a single intravenous bolus. The
concentration of the solution for this species was 2 mg/ml ivermectin and
the dose rate was 300 JLg/kg body weight. Plasma samples were collected
at specific time intervals and analyzed for ivermectin.
The points shown in Figure 7.5 are the averages over 4 animals. At each
sampling point the range of results was ± 15% or less. The results in sheep
were analogous to those for cattle. Plasma concentration as a function of
time clearly exhibited a biexponential decay for sheep. The biological
half-life was again 2.7 days, very similar to that obtained in cattle.
However, there is a significant difference in the volume of distribution of
ivermectin between the 2 species. Volume of distribution, Vd, was
,
1000
700
•
4oo~
300
200 2 4 6 8 10
'C
E Time (hrl
.....Cl
-c:
100 0,
--e
c:
.2
c:
CD
80
70
60
40
Q
~ y-Cattletllz- 2.8 daYI

u o
c: 30
0
u
c: 20
:;:
u • Shttp
CD tllz-2.7da1l
E
~
CD 10 ~• •
> 8 0091 6
H 7 tllZ -1.8 daya A
6
4
3
2
•
10 14
Time (days)
FIGURE 7.5. Biexponential decay of ivermectin concentration in plasma follow-
ing intravenous administration to cattle (0), sheep (e), and dogs (.6.). Dose rates:
300 ~g/kg body weight for cattle and sheep; 200 ~g/kg for dogs.Insert depicts the
rapid distribution phase in cattle. (Points are average from: 2 cattle, 4 sheep, and
5 dogs, respectively.) (Reproduced, with permission, from Lo et af. 1985.)
calculated by dividing the total dose by the plasma concentration at

injection, which is obtained by extrapolation of the lines of Figure 7.5 to
time t = O. In sheep, the volume of distribution, Vd = 4.6 lIkg, is greater
than that in cattle. And, as in cattle, the distributive phase is very rapid,
~ = 10 day-I.
The bioavailability analyses in this species were obtained using the
fiuorescence-derivatization technique described for cattle. (Tolan et al.
1980). The responses of the known acid- and base-catalyzed hydrolysis
products of ivermectin (Fink, in press) were recorded to demonstrate that
the fiuorescence-derivatization reaction furnishes efficient resolution of
ivermectin in plasma from its possible hydrolysis products.
Other workers have also evaluated the pharmacokinetics of ivermectin
in sheep. In contrast to the data described previously,Prichard and
colleagues (1985) reported a longer ivermectin biological half-life in
sheep, of 7.42 days. For plasma analyses, these workers used a diace-
tylated derivative of ivermectin as an internal standard in a method based
upon direct ultraviolet absorption detection in lieu of fiuorogenic derivati-
zation. In another study, Marriner, McKinnon, and Bogan (1987) re-
ported that the biological half-life in sheep was 61 hours, which is in close
agreement with the measurement of 65 hours shown in Figure 7.5. These
authors also attribute the low plasma concentrations found in sheep to the
large volume of distribution in this species.
Further confirmation of the intrinsic pharmacokinetic properties of
ivermectin in sheep can be found following drug administration by other
routes of administration. Two groups of 5 sheep were each treated with an
oral formulation of ivermectin at a dose of 200 I'g/kg body weight using
a dosing syringe. One of the groups received a solution containing
0.8 mg/ml ivermectin in propylene glycol, and the other group was given
an aqueous micelle product (Lo and Williams 1983) containing the same
concentration of the anthelmintic. As usual, plasma samples were col-
lected from the sheep at appropriate time intervals for chemical analysis.
The results showed that when ivermectin was administered orally to
sheep (in either formulation), the biological half-life obtained from the
terminal portion of the bioavailability curve ranged from 3 to 5 days.
These data are in good agreement with the intravenous data mentioned
previously and the results of Marriner and colleagues (1987).
A replicate study by oral dosing in sheep has also demonstrated that the
aqueous micelle solution of ivermectin is bioequivalent to the propylene
glycol solution. Both formulations exhibited a peak plasma concentration
within 1 day and exhibited a terminal portion with a tl/2 in the range of 3 to
5 days. Compared to the subcutaneous injectable route, this tl/2 of the
nonaqueous oral solution refiects the elimination of injection-site absorp-
tion processes attendant with nonaqueous vehicles as previously de-
scribed (Lo et al. 1985). The fraction of the dose absorbed from each of
these formulations was 50%.
Dogs
The dosing solution for intravenous administration to dogs was a 4-mg/ml
solution of ivermectin in the aqueous micellar formulation, which uses a
surface active agent to dissolve the drug (Lo and Williams 1983). This
solution contained 8% (w/v) polyoxyethylene sorbitan monooleate sur-
factant, 20% (w/v) glycerol formal cosolvent, and 1% benzyl alcohol; pH
was regulated to 6.2 with IN He 1. The injection was given to 5 dogs, each
at a dose of 200 JJ.g/kg body weight.
The procedure used for these bioavailability studies in dogs was a direct
method, based upon ultraviolet photometric detection of the drug using a
chemical derivative (the a2 isomer) of ivermectin as an internal standard
(Pivnichny, Shim, and Zimmerman 1983). This procedure was adapted to
a laboratory robotic system to increase the number of data points
available in these pharmacokinetic studies (Pivnichny, Lawrence, and
Stong 1987).
The curve obtained for the plasma concentration in dogs over time
(Figure 7.5) shows a greater slope than that obtained from the ruminants.
The results in dogs reveal a significant species difference in that iver-
mectin is eliminated (biotransformation, distribution, and excretion) more
rapidly (t1l2 = 1.8 days) than in either cattle or sheep. This measurement
of the terminal half-life in dogs is in accord with results recently reported
by Dainippon in Japan, which yielded a t1/2 = 1.6 days in dogs (Kojima,
Yamamoto, and Nakanishi 1987). The pharmacokinetic properties be-
tween species are summarized for comparison in Table 7.2.
The widespread use of ivermectin in a tablet formulation
(HEARTGARD-30) to prevent heartworm disease in dogs in a monthly
dosage regimen of only 6 JJ.g/kg is a dramatic illustration of the potency of
this drug. Bioavailability studies of these tablets document the rapid oral
absorption achieved by this dosage form. The results, shown in Figure
7.6, demonstrate that the plasma level obtained from the tablets peaks
within 2 to 4 hours and decays exponentially beyond the peak. The dose
used for this study was 100 JJ.g/kg, to provide sufficient precision of the
replicates relative to the sensitivity of the measurement process.
TABLE 7.2. Intrinsic pharmacokinetic

properties of ivermectin in three animal
species. (Reproduced, with permission,
from Lo et al. 1985.)
Pharmacokinetic property
Species
Cattle 6 2.8 1.9
Sheep 10 2.7 4.6
Dog 8.1 1.8 2.4
50
45
40
E
0, 35
c:
r::
0 30
~
E
~ 25
c:
0
0
c: 20
ts
II)
...
E
II)
15
~
10
0
0 5 10 15 20 25 30 35 40 45 50
Time, (hours)
FIGURE 7.6. Bioavailability of ivermectin in dogs dosed with HEARTGARD
tablet formulation at 100 #Lg/kg body weight. Each point is the average of 8 dogs.
The oral bioavailability achieved by this tablet formulation in dogs has

been measured over a range of 3 dose rates using 3 different measurement
techniques, as shown in Table 7.3. The concentration at the peak (Cp)
increases directly with the dose, suggesting a linear relationship between
the amount of drug absorbed and the dose administered. Further, the rate
of absorption was equal in the 3 studies, as evidenced by the tp observed
TABLE 7.3. Bioavailability in dogs of ivermectin from

tablet formulation at three dose levels.
Dose Measurement Peak plasma
administered technique concentration
Radiotracer counting 3.3 nglml
Fluorogenic 19.1 ng/ml
derivatizationa
100 ILg/kg Direct photometricb 40 ng/ml
a From Kojima Yamamoto, and Nakanishi 1987.
b From Pivnichny Shim, and Zimmerman 1983.
at 3 to 4 hours; this is additional evidence of the dose-independent

pharmacokinetics obtained from this tablet formulation for dogs. A recent
study has also demonstrated the bioequivalence of ivermectin when the
total dose is administered to dogs either (a) in tablets, each containing 46
micrograms of the drug, or (b) in twice as many tablets, each of which is
formulated to contain 23 micrograms of ivermectin (Kojima et al. 1987).
Swine
One marked example of the effect of route of administration on bioavail-
ability can be found in a bioequivalence study in swine treated with
ivermectin by subcutaneous injection and by oral administration. Two
groups of 4 pigs were each given ivermectin at a dose of 200 JLg/kg
parenterally in a propylene glycol-glycerol formal (60: 40 v/v) vehicle or
orally in propylene glycol solution. The drug concentration in these
dosing solutions was 10 mg/ml ivermectin. Plasma samples were collected
at appropriate time intervals and analyzed using the fluorogenic-
derivatization technique described previously (Tolan et al. 1980). The
results are shown in Figure 7.7.
Mter either subcutaneous or oral administration of ivermectin to swine,
the plasma concentration of the drug reached a peak and then declined
exponentially with time. It is significant that the terminal portions
resulting from these 2 dosing regimens are nearly parallel, suggesting that
ivermectin is eliminated rapidly in swine following either route of
absorption. The concentration of ivermectin in the plasma, however,
attains its peak, tp , faster following the oral dose than the parenteral: that
these peaks appeared at 0.5 days and 2 days, respectively, demonstrates a
higher absorption rate via oral administration than subcutaneous. These
results in swine are in accord with the bioavailability obtained with
subcutaneous injections in cattle and sheep. The slower absorption
associated with the parenteral route has been attributed to precipitation of
the drug at the injection site (Lo et al. 1985). The area under the plasma
concentration-vs.-time curve (AUC) was larger after subcutaneous ad-
ministration, indicating that a greater fraction of the dose is absorbed than
after oral dosing. The bioavailability of ivermectin after oral administra-
tion, relative to that after subcutaneous administration, was estimated to
be 41%.
The feed route offers another opportunity to study the bioavailability of
ivermectin in swine following oral administration. Two groups of 6
Yorkshire barrows were each given a single oral dose of 400 JLg/kg body
weight ivermectin in premix formulations prepared for homogeneous
distribution in the swine feed. One premix formulation was composed of
0.22% ivermectin adsorbed on ground corn cobs. The second premix was
prepared to contain 0.6% ivermectin adsorbed on the same ground corn
cob substrate. Plasma samples were collected from the pigs during 14
7. Pharmacokinetics ofIvermectin in Animals and Humans 125
22
20
18
E
0,
c 16
c·
gc
0
14
Q)
0
12
c
8 10
til
E
t/)
til 8
a.
c
U 6
Q)
E
Q) 4
~
2
0
0 2 3 4 5 6 7 8
Time, (days)
FIGURE 7.7. Bioavailability of ivermectin in swine administered orally and

parenterally. Dose: 200 ~g/kg. Vehicles: Oral: propylene glycol; parenteral:
propylene glycol-glycerol formal mixed solvent (60: 40 (v/v). Each point is the
average of 4 animals.
days following the dosing, and these samples were analyzed for iver-
mectin. These bioavailability studies were also used to evaluate the
efficacy of ivermectin in feed formulations. Two studies were run, with
swine dosed at 400 I-tg/kg body weight in all the feed consumed: in both
studies, the drug reached its peak concentration in the plasma within 4 to
8 hours, in agreement with the results of oral administration shown in
Figure 7.7. These results are shown in Figure 7.8. Note that the drug is
rapidly adsorbed by the oral route, that it is eliminated rapidly from swine
(t1l2 ~ 0.5 day), and that good experimental agreement in the bioavail-
ability results was found between these 2 independent swine oral studies.
Horses
One additional clear example of the significant effect of formulation and
route of administration on the bioavailability of ivermectin has recently
been reported in a study in pregnant mares. A paste formulation and an
aqueous micellar formulation of this drug were administered (Asquith et
20~---r----'-----r---~-----r----~
15
c
o
-
:;:
2
c
CD
u 10
c
o
u
c
~
E
~
CD
..:i 5
Time (hours)
FIGURE 7.8. Bioavailability of ivermectin in swine following single oral adminis-

tration of adsorbate feed premixes.Drug (0.6% or 0.22%) adsorbed on ground corn
cobs. Dose: 400 ILg/kg body weight. Closed and open circles designate two
independent replicate bioavailability studies. The range of standard deviations
from these 10 data points was 0.4 ng/ml to 5.9 ng/ml ivermectin. (Reproduced,
with permission, from Lo et al. 1985.)
al. 1987). The dose for both treatments was 200 JLg/kg, and the liquid
micellar solution (1%) was administered by nasogastric intubation. A
striking difference in bioavailability was found; the solution is absorbed
much more rapidly than the paste product. Although the liquid formula-
tion attained its peak concentration within 4 to 5 hours, the paste product
took 15 hours to reach its maximum plasma concentration. Accordingly,
the peak plasma concentration obtained from the solution was higher than
that offered by the paste, and the bioavailability (AUC) was 20% higher
following the nasogastric oral liquid dose.
Humans
Disposition of ivermectin in humans was studied in 4 healthy volunteers
after an oral dose (no i. v. formulation exists) of 3H-ivermectin labeled in
the C(22) and C(23) positions. Concentrations of ivermectin in blood

peaked at around 4 hours postdose and slowly decreased afterward (see
Figure 7.9). Peak plasma concentrations of radioactive metabolites were
about twice those of parent drug and occurred somewhat later (ca. 7h).
The elimination phase of ivermectin does not consist of a simple
exponential, and examination of individual plasma-concentration profiles
reveals secondary peaks suggestive of enterohepatic recycling. The
half-life was estimated at about 12 hours. Ivermectin and its metabolites
seem to be excreted in bile with minimal (:5 1.0% of administered dose)
urinary excretion. Secretion of ivermectin in human milk has been
observed to a minor extent.
Urine, feces, and plasma of subjects given radioactive ivermectin were
analyzed to establish the chemical identities of metabolites. Positive
identification has been obtained for the presence of 3"-0-desmethyl-22,23-
dihydroavermectin B'a, and 22,23-dihydroavermectin B'a monosaccha-
ride in urine and feces, respectively. Plasma metabolites that are less
polar than the parent drug have been tentatively characterized as fatty
acid ester conjugates of the monosaccharides or aglycone of the drug
(S.-H.L. Chiu, personal communication).
Some of the metabolites are, systemically, longer lived than the parent
drug (see Figure 7.9). Monitoring the appearance and disappearance of
radioactively labeled metabolites in plasma gives an apparent half-life of
about 3 days.
The absolute bioavailability of ivermectin in humans is unknown.
Twelve healthy volunteers were given single doses of the commercially
available tablet, unformulated ivermectin in a capsule or a solution of
1------.
100 . - - - - - - - - - - { I--------------------~
---;\~ ....'-----......
E
-.!!!.c-
....... 10 Metabolites
Cl
c:
Ivermectin
o 10 20 40 60 80
Time (hours)
FIGURE 7.9. Mean concentrations ofivermectin and metabolites in human plasma
after administration ofa 14-mg dose of 3H-labeled drug (average from 4 subjects).
E
c,
c: 80
c::0
~ 60
'E
Q)
0
c: 40
8
c:
U
Q) 20
E
~
Q)
~
Time (hours)
FIGURE 7.10. Mean plasma concentrations of ivermectin in humans after adminis-

tration of 12-mg single doses to 12 healthy volunteers as either capsules, tablets,
or a hydroalcoholic solution.
ivermectin (0.6 mg/ml) in 40% aqueous ethanol. Mean plasma concen-

tration-vs-time curves are shown in Figure 7.10. The bioavailabilities of
the two solid formulations were similar, while the alcoholic solution was
substantially more bioavailable (mean AUC ratios of 1.73 and 1.96
relative to capsules and tablets, respectively). Formulation seems to have
no effect on the time at which maximum concentrations in plasma are
observed (ca. 4h). Such behavior is consistent with absorption of
60
E
-- CI
c:
50
.2
c:: 15mg
40
iU
-=c:
Q)
0 30
c:
0
0
c: 20
U
CD
E
~
CD 10
~
0
0 2 4 6 8 24 36 48 72
Time (hours)
FIGURE 7.11. Mean concentrations of ivermectin in human plasma after adminis-

tration of single doses of 6, 12, and 15 mg to 12 healthy volunteers.
ivermectin being limited by dissolution rate. These results put an upper

limit of some 50% to 60% on the bioavailability of ivermectin from
capsules and tablets (data from University of Liverpool and Merck and
Co., Inc.).
Investigation of the effect of dose on observed plasma concentrations in
12 healthy volunteers (see Figure 7.11) indicated that, in the therapeutic
range, the pharmacokinetics of the drug are concentration-independent
(Porras et al. 1987). Observed AUC and Cmax values increased signifi-
cantly when the dose was increased from 6 to 12 to 15 mg (single doses).
Ratios of dose-adjusted AUC estimates for the 12 versus 6 mg doses, the
15 versus 6 mg doses, and the 15 versus 12 mg doses were 0.81, 0.86, and
1.07, respectively. The corresponding Cmax ratios were 0.88, 0.92, and
1.05. In all cases the 95% confidence interval included unity. There was
no detectable effect of dose on the time taken to reach maximum plasma
concentration. These results are strongly supportive of dose propor-
tionality.
REFERENCES
Asquith RL, Lane TJ, Plue RE, Seward RL, Kivipelto J (1987) The bioavailability
of ivermectin in horses when administered in a liquid formulation by nasogastric
intubation versus in an oral paste. Equ. Vet. Sci. 7:28-30
J. Vet. Pharrn. & Therap. 7:1-16
Fink DW (1982) Some specific ftuorogenic reactions in pharmaceutical and
environmental applications. Trends in Anal. Chern. 1:254-258
Fink DW (in press) Ivermectin: analytical profile. In Florey K (ed) Analytical
Profiles of Drug Substances, Academic Press, New York
Hibbert H, Carter NM (1928) Studies on the reactions relating to carbohydrates
and polysaccharides. XVII. Structure of the isomeric methylidene glycerols.
J. Am. Chern. Soc. 50:3120-3127
Kojima K, Yamamoto K, Nakanishi Y (1987) Determination of 22,23-
dihydroavermectin B'a in dog plasma using solid-phase extraction and high-
performance liquid chromatography. J. Chrornatogr. 413:326-331
Kojima K, Yamamoto K, Katae H, Nakanishi Y (1987) Bioavailability of oral
ivermectin in dog [sic]. Nippon Juigaku Zasshi 49:899-900
Lo P-KA, Fink DW, Williams JB, Blodinger J (1985) Pharmacokinetic studies of
ivermectin: effects of formulation. Vet. Res. Cornrnun. 9:251-268
Lo P-KA, Williams JB (1983) Solubilization of ivermectin in water. U.S. Patent
4,389,397
Marriner SE, McKinnon I, Bogan JA (1987) The pharmacokinetics of ivermectin
after oral and subcutaneous administration to sheep and horses. J. Vet. Pharrn.
Therap. 10:175-179
Miwa GT, Walsh JS, VandenHeuvel WJA, Arison B, Sestokas E, Buhs R,
Rosegay A, Lu AYH, Walsh MAR, Walker RW, Taub R, Jacob TA (1982) The
metabolism of avermectins B'a, H 2B'a, and H 2B'b by liver microsomes. Drug
Metab. Dispos. 10:268-274
Pivnichny JV, Lawrence AA, Stong JD (1987) A robotic sample preparation
scheme for the high performance liquid chromatographic determination of

ivermectin in animal plasma. J. Chromatogr. Sci. 25:181-186
Pivnichny JV, Shim J-SK, Zimmerman LA (1983) Direct determination of
avermectins in plasma at nanogram levels by high-performance liquid chroma-
tography. J. Pharm. Sci. 72:1447-1450
Porras AG, Chiou R, Kukovetz W, Hall-Gregg M, Stubbs RJ, Mesinger M,
Beubler N, Jaeger A (1987) Dose proportionality ofthe anthelmintic ivermectin
in man. Abstract presented at the 2nd National Meeting of the AAPS, June
5-12, 1987, Boston, Mass.
Prichard RK, Steel JW, Lacey E, Hennessy, DR (1985) Pharmacokinetics of
ivermectin in sheep following intravenous, intra-abomasal or intraruminal
administration. J. Vet. Pharm. & Therap. 8:88-94
Schnitzerling HJ, Nolan J (1985) Normal phase chromatographic determination of
nanogram quantities of ivermectin in cattle blood or plasma. J. Assoc. Offic.
Anal. Chem. 68:36-40
Wilkinson PK, Pope DG, Baylis FP (1985) Pharmacokinetics of ivermectin
administered intravenously to cattle. J. Pharm. Sci. 74:1105-1107
CHAPTER 8
Metabolism and Tissue Residues

Shuet-Hing Lee Chiu and Anthony Y.H. Lu
I. Introduction
II. Tissue Residues in Cattle, Sheep, and Swine
III. Absorption, Distribution, and Excretion in Rats
IV. In Vitro Liver Metabolism of Ivermectin
V. In Vivo Liver Metabolism of Ivermectin
VI. In Vivo Fat Metabolism of Ivermectin
VII. Conclusion
I. Introduction
Ivermectin is used widely as an antiparasitic agent in food-producing
animals. As in the case of any such drug, the residual tissue concentration
of the therapeutic agent, or tissue residue, is a safety concern to the
meat-consuming public. To evaluate the toxic potential of the residual
tissue concentration of ivermectin and its metabolites, metabolism studies
have been carried out in target species (cattle, sheep, swine) using the
radiolabeled drug. Comparative metabolic studies were done in a labora-
tory animal, the rat, and in liver microsomes from various species.
Radiolabeled ivermectin was synthesized either by incorporating tri-
tium atoms in the C5 position by oxidation of the C5-alcohol to ketone
with manganese dioxide followed by reduction back to the alcohol with
sodium borotritide, or in the C22-23 positions by catalytic tritiation of
avermectin. In these synthetic reactions, the 2 components ofivermectin,
dihydro-B 1a (H 2B 1a) and dihydro-B 1b (H 2B 1b) are synthesized and purified
separately and then mixed in a ratio of 80: 20 with respect to H 2B 1a to
HzB 1b for dose preparation. Most of the studies were carried out with
C22-23-labeled drug (22,23-[3H]ivermectin). The tritium label has been
shown to be stable by volatility studies (Jacob et al. 1983).
This chapter addresses only those studies on ivermectin tissue residue
132 Shuet-Hing Lee Chiu and Anthony Y.H. Lu
and metabolism that employed the radiolabeled drug. For a review of

those studies that used unlabeled ivermectin, the reader is referred to
Chapter 9.
II. Tissue Residues in Cattle, Sheep, and Swine

Tissue residues are often assayed as total radioactivity in the tissue or
body fluids after an animal is dosed with a radioactive drug. The total
tissue residues include the free drug, its metabolites, and any tissue-
bound residues as they are generally classified (Weber 1980; Dorough
1980). The "free" and "bound" residues are differentiated by their
extractability by methods such as exhaustive extraction, denaturation, or
solubilization (Weber 1980). While toxicological evaluation of the free
drug residue is possible when the levels and structures of these residues
are known, evaluating the bound residue is extremely difficult (Lu et al.
1987).
Ivermectin residues in animal tissues are essentially free residues; the
total radioactivity in "edible" tissues (liver, kidney, muscle, and fat) of
cattle, sheep, swine, and rats is nearly quantitatively extractable with
organic solvents (Chiu et al. 1986, 1987). Thus, the bound residue is not a
concern with ivermectin.
Tissue-residue studies were carried out on animals dosed once at 0.3 or
0,4 mg/kg body weight either subcutaneously (cattle, swine), intraru-
minally (cattle, sheep), or orally (rats). Animals were slaughtered over a
period of 1 to 28 days after dosing (Campbell et al. 1983; Jacob et al.
1983). About 25 tissues and body fluids were collected from the target
species and assayed to examine the distribution of the radioactive
residue. Urine and feces were usually collected over the whole period of
the study and assayed for the recovery of the radioactive dose. Radioac-
tivity in tissues and feces was assayed by combustion of the aqueous
homogenates followed by scintillation counting. Radioactivity in plasma
and other fluids was analyzed directly by scintillation counting. Table 8.1
shows the tissue residue levels from 4 steers slaughtered 7, 14,21, and 28
days after receiving a single subcutaneous dose of 22,23-[3H]ivermectin at
0.3 mg/kg body weight. Among the 25 tissues and fluids analyzed, brain
tissue showed the lowest residue levels (4 ppb 7 days after dosing). Of the
edible tissues, liver and fat tissue levels were the highest, with depletion
half-lives of 4.8 and 7.6 days, respectively. Similar tissue-residue distribu-
tion patterns exist in the sheep, swine, and rats, although the residue-
depletion half-lives are much shorter in sheep and rats (1-2 days),
reflecting more rapid metabolism in these animals (Table 8.2). The
half-lives of the parent drug and total tissue residue are in close agreement
in liver tissue in all species (except swine), suggesting equally efficient
depletion of the parent drug and its metabolites. The parent drug depletes
relatively faster in fat tissue than the total residue (about twofold) in cattle
8. Metabolism and Tissue Residues 133
TABLE 8.1. Total radioactive residue levels

(Ppb) in tissues and fluids from cattle dosed
subcutaneously with 22,23-[3H]ivermectin at
0.3 mg/kg body weight. a
Days post
dose 7 14 21 28
Abomasum 44 17 10 1
Adrenals 29 6 7 2
Bile 273 54 22 1
Bone marrow 92 21 23 9
Brain 4 1 0 0
Cecum 33 9 3 0
Colon 44 II 9 0
Fat 270 83 69 29
Heart 41 8 3 0
Intestine 22 5 6 1
(small)
Kidney 68 6 7 2
Liver 782 55 68 II
Lung 66 12 4 1
Lymph gland 41 13 20 6
Muscle 23 2 4 0
Pancreas 83 16 9 I
Plasma 45 II 6 3
Rumen 34 to 10 2
Rumen fluid 7 I 0 0
Spleen 38 8 5 3
Thymus 64 21 9 I
Thyroid 58 16 8 6
Tongue 27 4 4 0
Injection site 70 28 33 39
a Tissues and fluids were collected from I steer per

time point. Radiolabeled ivermectin was at a spe-
cific activity of 0.1 mCi/mg. All samples except
plasma were assayed by combustion of the tissue
homogenate in water.
TABLE 8.2. Depletion half-lives of 22,23-eH]ivermectin total residue and

unchanged drug in liver and fat tissues of cattle, sheep, swine, and rat. a
tl/2 (day)
Liver Fat
Species Total residue H2B1a H2B1b Total residue H 2B1a H2B1b
Cattle 4.8 4.7 4.0 7.6 4.3 3.2
Sheep 1.2 1.2 1.0 1.8 1.1 0.9
Swine 5.2 3.8 3.2 5.1 3.8 4.0
Rat 0.95 0.90 - b
1.3 - b
- b
a Results are from studies in which all species except swine were dosed at 0.3 mg/kg body
weight, subcutaneously (steers), intraruminally (sheep), or orally (male rat). Swine were
dosed SC at 0.4 mg/kg body weight. Unchanged drug H2B1a and H 2B1b were analyzed by
RIDA.
b Not calculated due to insufficient data.
and sheep, probably because the fat tissues of these species eliminate
nonpolar metabolite conjugates less efficiently. This phenomenon will be
discussed at more length in the section on In Vivo Fat Metabolism
of Ivermectin. There is no difference between liver and fat residue
depletion in swine.
The distribution pattern of tissue residue remains unchanged when the
drug is dosed via different routes. Of the 3 routes-subcutaneous,
intraruminal, and percutaneous-in steers, the intraruminal dosing gives
the lowest residue levels in liver and fat. The depletion half-lives,
however, vary with the dosing method. Residue levels in the brain re-
main the lowest by all dosing routes (S.H.L. Chiu, M. Green, F.P. Baylis,
D. Eline, and T.A. Jacob, unpublished results).
Radiolabeled ivermectin is excreted mainly in feces in all 4 species
studied. Regardless of whether the drug was administered intraruminally
or subcutaneously to cattle, only 0.5% to 2.0% of the dose was eliminated
in urine; the remainder appeared in feces (Jacob et al. 1983). Recovery of
the radioactive dose 7 days after dosing varies between 62% to 83% in
cattle, sheep, and rats.
III. Absorption, Distribution, and Excretion of

Ivermectin in Rats
To establish the relevance of rodent toxicology studies to human safety
issues, comparative metabolism studies of ivermectin were also carried
out in the rat. Total radioactivity in rat plasma was compared following 5
dosing routes: gavage, subcutaneous, intramuscular, intraperiton~al, or
intravenous (Jacob et al. 1983). Orbital blood samples taken during the
first 24 hours showed that by 4 hours after dose, total radioactivity in
blood from all dosing routes converged to about 0.03 ppm. Furthermore,
depletion rates of blood radioactivity during the subsequent 8 to 24-hour
period were quite similar. Radioactive residue levels in "edible" tissues
increased proportionately with the dose, ranging from 0.06 mg/kg to
7.5 mg/kg (125-fold dose increase). At the dose used in food-producing
animals (0.3 mg/kg), the tissue-to-blood ratio is generally greater than 1.
At 1 day post dose, the ratio is highest in testicular fat (24 : 1), followed by
kidney (18: 1), liver (11: 1), and muscle (3: 1). By oral dosing, plasma
levels peaked between 4 to 8 hours post dose and depleted at a half-life of
about 1 day in both plasma and liver at all dosage levels. Intravenous
dosing resulted in a rapid distribution of radioactivity to lung, heart, liver,
and kidney within 2 hours (Figure 8.1). During the first 24 hours after
dosing, residues in these tissues deplete at half-lives of approximately 12
hours. Residue in the fat tissue, on the other hand, does not peak until 12
hours and depletes relatively slower, remaining as the highest residue
concentration among tissues.
2 '·'v'lung
• OL"...
"- Hearl
1 X
X
, ". o"::::-L "'v• -___
--ol.:;:-r.~.
X"_.('
0_
~.
",.....
--"'[01 __
,..... ~'7-.X·-·-.- __ x........ --_
..~ .............. -0
.6 /
/ Kidney "> . ~..
' ........ ~
Kidney
o llvi'f·...
..... ...... ..;:....... -- Hear I
'. /
.3
I ~
f-X~ .... ~o
E
-11
I
----....x Lung jt
0.
0.
o
..
ClJ
.1
~
I.'
"'C
II>
ClJ X
0::
.06
I '
' ......
....... _ ---....... Blood
' ...
.03 ..........................
"x- --x-···_·-·x·- __ -•• .x.,.........Br·oln ...
.....
X ."',
' . ...., .. ,
.01~--~--~----~--~--~~--~----u
o 4 8 12 16 20 24 28
Hours post dose

FIGURE 8.1. Depletion of total radioactive residues in tissues and blood of rats
dosed intravenously at 0.3 mg/kga.
a Ivermectin was dissolved in 0.5 ml rat plasma for i. v. dosing.
Absorption and distribution of the 2 components of ivermectin are

similar in rats. For all rat tissues analyzed, depletion of the major
component H2B1a showed half-lives of 1 to 1.5 days. The minor com-
ponent H 2B1b , however, appears to deplete slightly faster, with half-lives
of 0.4 to 0.8 day (Table 8.3). This phenomenon is also observed in cattle
but not particularly noticeable in sheep and swine.
IV. In Vitro Liver Metabolism of Ivermectin

The toxicological evaluation of free drug or metabolite residue in animal
tissues requires the elucidation of metabolites, structures, as well as
quantification of the parent drug and its metabolites. Because of the low
TABLE 8.3. Total radioactive residues in blood and tissues of male Sprague Dawley rats
dosed subcutaneously at 0.3 mg/kg with 22,23-[3H]H 2B ,a or 22,23-CH]H 2B ,b . a
Tissue residues (ppm) Depletion til'
Post-dose I day 2 days 4 days (day)
Tissues H,B'a H,B'b H,B,. H,B'b H,B,. H,B'b H,B,. H,B'b
Liver 0.320 0.229 0.130 0.082 0.043 0.010 1.06 0.66
Kidney 0.220 0.185 0.118 0.068 0.041 0.009 1.25 0.69
Fat pad 0.493 0.497 0.274 0.168 0.110 0.027 1.40 0.72
Muscle 0.109 0.067 0.044 0.023 0.017 0.004 1.16 0.75
Lung 0.166 0.108 0.064 0.041 0.027 0.004 1.19 0.63
Heart 0.126 0.105 0.073 0.044 0.027 0.005 1.36 0.68
Blood 0.019 0.013 0.009 0.004 0.003 0.000 1.14 0.42
Plasma 0.023 0.017 0.012 0.005 0.004 0.001 1.20 0.75
• Three rats (200 g each) were dosed at each time point with radiolabeled drug at specific activity of
0.1 mCi/mg. Blood was collected by heart puncture. Fat pad was removed from the testicular areas
of the animals.
tissue-residue levels in animals treated with ivermectin at normal slaugh-

ter time (14-28 days after dosing), it is extremely difficult to isolate
directly from tissue sufficient amounts of pure metabolites for structural
identification. In this regard, in vitro metabolism systems such as liver
homogenates or liver micro somes are superior because conditions can be
manipulated to maximize the formation of metabolites. The relatively
"clean" system (compared to the in vivo tissue matrix) also facilitates the
isolation and purification of metabolites (Lu et al. 1987). The in vitro
metabolism of ivermectin was studied with 22,23-eH]H 2B 'a and 22,23-
[3H]H 2B 'b by incubation with liver microsomes from rat, steer, and swine
in the presence of an NADPH-regenerating system (Miwa et al. 1982;
Chiu et al. 1984). Using these systems, rat and cattle liver microsomes
form 24-hydroxymethyl-H2B,a (V, Figure 8.2) and 24-hydroxymethyl-
H 2B 'b (VI) as major metabolites, I while swine microsomes give rise to
3"-O-desmethyl-H2B,a (IX) and 3"-O-desmethyl-H2B 'b (X) as major metab-
olites. Minor metabolites from cattle and rat in vitro systems include
monosaccharides of the major metabolites (24-0H-H2B'a-MS, VII; 24-
OH-H2B'b-MS, VIII). These compounds are used as references to
characterize in vivo metabolites from tissues.
I Nomenclatures 24-hydroxymethyl-H2B 1a and 24-hydroxymethyl-H 2B 1b were

used for metabolites V and VI in all previous publications and abbreviated as
24-0H-H2Bla and 24-0H-H2Blb, respectively. These abbreviated nomenclatures
are also used in this review. The correct nomenclatures are either 24a-hydroxy-
H 2B 1a and -H2Blb; or 24-desmethyl-24-hydroxymethyl-H2B and -H 2B ,b . ,a
OH
6
OCH,
HO
HOn
A· B·
HlC 0 H,c 0
R, Rz R,
HaB,. A CHIQi, H
II HaB,. A CH, H
m HaB ,. - Monosacchoride H CHICH, H
m Ha B,. - Monosotc.horide H CH, H
'lZ' 24 -OH-HIB,. A CHaCH, OH
1ZI 24-0H-HzB,. A CHl OH
1ZII 24-0H-HzBl.-Mono~horide H CHICH, OH
"illI 24-0H-HzB,.-Mono,oCChoride H CH, OH
IX 3--0- [)nmelhyl-HIB,. B CHaCHl H
X 3--0- Oumethyl-HaB,. B C~ H
FIGURE 8.2. Structures of ivermectin and derivatives.
v. In Vivo Liver Metabolism of Ivermectin

Despite the low concentration of ivermectin residues in tissues, they can
be extracted by organic solvents nearly quantitatively, purified, and
assayed chromatographically. Several methods have been reported for
quantifying the parent drug at low levels in biological samples. These are
analytical methods based on high-performance liquid chromatography
either by direct UV detection (Pivnichny, Shim, and Zimmerman 1983) or
by fluorescence detection of ivermectin derivatives (Tolan et at. 1980;
Tway, Wood, and Downing 1981; Stong 1987). For assay of the radioac-
tive parent compound in tissues, a microgram-scale reverse isotope
dilution assay (RIDA) method is especially valuable (Chiu et at. 1985).
This method compares the specific activity of parent-drug residue in

tissue before and after dilution of radioactivity with a known amount of
unlabeled ivermectin during isolation. Thus, the assay provides an
absolute amount of the parent drug in the tissue sample assayed,
independent of losses during extraction and purification. Results of the
analyses of unchanged parent drug in radioactive liver residue from cattle,
sheep, swine, and rat are shown in Table 8.4 (Chiu et al. 1987). The
unaltered drug (H2Bta and H 2B tb) accounts for at least 50% of the total
radioactive residue in tissues of cattle up to 14 days after dosing, in sheep
5 days after dosing, and in swine and rats, 7 and 3 days after dosing,
respectively. The rest of the residue consisted of more polar metabolites.
The in vivo liver metabolism of ivermectin by cattle, sheep, swine, and
rats correlates with that obtained by in vitro liver microsomes from the
same species. Three polar metabolites-24-0H-H2B ta , 24-0H-H2B ta-
MS, and 24-0H-H2B tb-account for 50%, 28%, and 39% of the total
metabolites in liver tissue of cattle, rat, and sheep, respectively. The rest
of the metabolites are complex mixtures of minor metabolites that are
present in much lower concentrations. Similarly, 3"-0-desmethyl-H2B ta
and -H2B tb , the major in vitro metabolites of swine liver microsomes,
TABLE 8.4. Total radioactive residue and unaltered

22,23- 3H-ivermectin in liver tissue of steers, sheep, swine,
and rats. a
% of total
residue
Species Days Total residue (Ppb) H 2B1a H 2B 1b
Steer 7 622 ± 223 56 7.7
14 104 ± 43 49 3.4
21 48 ± 17 36 3.9
28 31 ± 18 37 2.5
Sheep 1 212 ± 69 54 21
3 105 ± 44 51 27
5 23 ± 9.9 56 12
7 44 ± 31 ND ND
Swine 1 194 ± 110 37 17
7 110 ± 32 31 20
14 22 ± 13 20 6.5
28 2.7 ± 0.6 ND ND
Rat 1 152 66 5.8
3 47 56 9.0
a All animals except swine were dosed at 0.3 mg/kg body weight,
subcutaneously (steer), intraruminally (sheep), or by gavage (rat).
Swine were dosed subcutaneously at 0.4 mg/kg body weight. Three
animals were dosed at each point and tissues collected separately for
steer, sheep, and swine.Percentages of parent drug represent aver-
ages from RIDA of at least 2 animals at each point. Twenty-four rats
were dosed and tissues composited for each time point. N.D., not
determined.
account for 69% of total metabolites (or 45% oftotal residue) in composite
liver-tissue samples from 7 and 14 days post dose animals. Confirmation
of in vivo metabolite structures has been achieved by chromatographic
comparison with in vitro metabolites either directly (in at least 2 modes of
HPLC separation), or after chemical derivatization (such as acid hydroly-
sis or fluorescence derivatization) (Chiu et al. 1987). These rigorous
comparative tests are carried out for the identification of all major
metabolites and some of the minor metabolites from cattle, sheep, rat,
and swine.
VI. In Vivo Fat Metabolism of Ivermectin

The radioactive residue levels of ivermectin in fat tissues are either higher
than (rat, sheep, swine) or comparable to (cattle) those in the liver.
Similar to what one obtains in liver tissue, the parent drug constitutes at
least 50% of the total tissue residue during the early withdrawal periods,
i.e., 14 days post dose in steers, 3 days post dose in sheep, 7 days post
dose in swine, and 3 days post dose in rats. The percentage of parent
drug, however, decreases rapidly at later time points (Table 8.5). With the
TABLE 8.5. Total radioactive residue and unaltered

22,23- 3H-ivermectin in fat tissue of steers, sheep, swine,
and rats. a
% of total
residue
Species Oays Total residue (ppb) H 2B 1a H 2B1b
Steer 7 220 ± 58 61 5.9
14 88 ± 6 50 2.3
21 45 ± 21 18 0.9
28 33 ± 9 18 1.1
Sheep 1 245 ± 112 71 24
3 153 ± 62 51 9.3
5 63 ± 25 28 5.5
7 73 ± 35 19 3
Swine 1 384 ± 234 43 16
7 152 ± 43 50 13
14 28 ± 15 25 10
28 6.3 ± 3 NO NO
Rat 1 232 50 13
3 137 66 12
a All animals except swine were dosed at 0.3 mg/kg body weight,
subcutaneously (steer), intraruminally (sheep), or by gavage (rat).
Swine were dosed subcutaneously at 0.4 mg/kg body weight. Three
animals were dosed at each point and tissues collected separately for
steer, sheep, and swine. Percentages of parent drug represent
averages from RIOA of at least 2 animals at each point. Twenty-four
rats were dosed and tissues composited for each time point. NO,
not determined.
exception of swine, a group of metabolites less polar than the parent drug
is present in the fat tissue of these animals; they are particularly notable in
tissues of sheep and cattle, accounting for 60% to 70% of total residue in
the fat when these animals were withdrawn from dosing for a period of 5
or 28 days, respectively. Upon saponification or esterase treatment, these
nonpolar metabolites regenerate polar products that resemble chromat-
ographically the polar liver metabolites of ivermectin (e.g., 24-0H-
H2BIJ. Furthermore, the nonpolar metabolites give rise to products
identical to monosaccharide and aglycone of 24-0H-H2Bla when sub-
jected to p-toluenesulfonic acid-catalyzed methanolysis conditions (Fig-
ure 8.3). These results lead to the hypothesis that the polar ivermectin
metabolites produced in the liver are esterified as fatty acid esters and
stored in fat tissue as nonpolar entities. These nonpolar compounds can
be converted back to the polar metabolites chemically or enzymatically,
as shown in Figure 8.4 (Chiu et al., 1988). Two approaches were taken to
substantiate this hypothesis. First, the regeneration of polar metabolites
from the nonpolar metabolites was studied with cholesterol esterase, and
the structures of the hydrolysis products were confirmed as identical to
known liver metabolites (24-0H-H2Bla, -H2B 1b , -H2Bla-MS). Secondly,
samples of fatty acid (oleic acid) esters of 24-0H-H2Bla and -H2B 1b were
synthesized and directly compared with the nonpolar metabolites isolated
from a 2-kg fat sample from steers. These studies have established the
general structural class of the ivermectin nonpolar metabolites as being
the fatty acid conjugates of 24-hydroxymethyl derivatives of ivermectin.
The ivermectin metabolism pathway in fat tissue presents a rather
uncommon, but not unique, phenomenon in xenobiotic metabolism.
Although formation of less polar metabolites deviates from the most
common metabolic pathway-in which more polar metabolites are
formed for readiness of eliminations-several precedents have been
reported in recent years (Caldwell and Marsh 1983). Formation of fatty
acid esters of xenobiotics containing an alcoholic hydroxyl group, in
particular, have been reported for tetrahydrocannabinol (Leighty,
Fentiman, and Foltz 1976), pyridopyrimidine (Schmid et al. 1980), and
etofenamate (Dell et al. 1982). The metabolic pathway of ivermectin is an
interesting addition to these examples.
In contrast to cattle, sheep, and rat, swine do not follow the fat
metabolism pattern observed in the other three species. The fat metabo-
lite profile of swine resembles that of the liver tissue, i.e., 2 major
metabolites slightly more polar than the parent drug are present. These
are confirmed to be the 3"-0-desmethyl derivatives of H 2B 1a and H 2B 1b by
chromatographic comparison with authentic samples generated by swine
liver microsomes. The absence of ester conjugation of swine liver
metabolites in fat tissue can probably be attributed to the fact that there is
nO primary hydroxyl functional group in these metabolite structures. As a
result, these molecules are less favorable substrates for esterification
compared to 24-hydroxymethyl- H 2B 1a and -H2B1b produced in the livers
OH
polar metabolite
011
non-polar metabolite
in fat tissue
011
polar metabolite aglycone

pTSA = p-toluenesulfonic acid
R = fatty acid acyl
FIGURE 8.3. Scheme of hydrolysis reactions of dihydroavermectin-B1a non-polar
metabolites in fat tissue with p-toluenesulfonic acid (pTSA) or cholestrol esterase.
of the other 3 species. The similarity of liver and fat metabolite profiles in
swine explains the close resemblance of the residue depletion half-lives in
the two tissues of swine (Table 8.2).
VII. Conclusion
The absorption, excretion, distribution, and metabolism of tritium-labeled
ivermectin have been studied in target food-producing species (cattle,
sheep, swine) and a laboratory species (rat). The drug is absorbed rapidly
via subcutaneous, intraruminal, or oral dosing routes. Fecal excretion is
Steer Liver Microsomes Enzymatic Esterification
.---------------- --- .
(In vitro) (In vivo)
I I
I I Steer, Sheep, Rat
,
I I
I I
I I
I
,
I
I I
I
.
I
I
Polar Non-polar
Ivermectin Metabolites Metabolites
Steer (Liver) (Fat)
Sheep
Rat A
I
I
I
:
I
I I :
I I I
I I I
I I I
I IL ________________________ ~I
I
- _____________ - _ - ____ 1I I :
Rat Liver Microsomes Cholesterol Esterase Hydrolysis

(In vitro) (In vitro)
FIGURE 8.4. Relationship of ivermectin metabolites in the liver and fat tissues of
sheep, rat, and steer with in vitro metabolites by steer and rat liver microsomes.
the major pathway of drug elimination in all species studied. Only about
2% of dose is excreted via urine. Highest residual tissue concentrations of
radioactivity are present in the liver and fat tissues, while lowest levels
are observed in the brain. Metabolism of radiolabeled ivermectin in liver
and fat of cattle, sheep, and rat is similar. Some differences are noted in
swine. The parent drug is the major liver residue in all four species in the
early withdrawal time period. The major liver metabolite in cattle, sheep,
and rat is 24-hydroxymethyl-H2B 1a , which is further conjugated to fatty
acids as esters and deposited in the fat tissue. The major liver metabolites
in swine are 3"-O-desmethyl-H2B 1a and 3"-O-desmethyl-H2B 1b • These
drug-like compounds are also present in fat tissue as major metabolites. In
vivo metabolism of ivermectin has been found to correlate closely with
that by in vitro liver microsomes of the same species.
REFERENCES
Caldwell J, Marsh MV (1983) Interrelationships between xenobiotic metabolism

and lipid biosynthesis. Biochern. Pharrn. 32:1667-1672
Campbell we, Fisher MH, Stapley EO, Albers-Schonber G, Jacob TA (1983)
Ivermectin: A potent new antiparasitic agent. Science 221:923-928
ivermectin residue in animal tissues by high performance liquid chroma-
tography-reverse isotope dilution assay. J. Agric. Food Chern. 33:99-102
Chiu SHL, Carlin JR, Taub R, Sestokas E, Zweig J, VandenHeuval WJA, Jacob
TA (1988) Comparative metabolic disposition of ivermectin in fat tissues of

cattle, sheep, and rats. Drug Metab. & Dispos. 16:728-736
Chiu SHL, Sestokas E, Taub R, Buhs RP, Green M, Sestokas R, VandenHeuvel
WJA, Arison BH, Jacob TA (1986) Metabolic disposition of ivermectin in
tissues of steers, sheep, and rats. Drug Metab. & Dispos. 14:590-600
Chiu SHL, Sestokas E, Taub R, Smith JL, Arison B, Lu AYH (1984) The
metabolism of avermectin-H2Bla and -H2B lb by pig liver microsomes. Drug
Metab. & Dispos. 12:464-469
Chiu SHL, Taub R, Sestokas E, Lu AYH, Jacob TA (1987) Comparative in vivo
and in vitro metabolism of ivermectin in steers, sheep, swine, and rat. Drug
Metab. Rev. 18:289-302
Dell HD, Fiedler J, Kamp R, Gaug W, Kurz J, Weber B, Wuensche C (1982)
Etofenamate fatty acid esters, an example of a new route of drug metabolism.
Drug Metab. & Dispos. 10:55-60
Dorough HW (1980) Classification of radioactive pesticide residues in food-
producing animals. J. Environ. Path. Toxicol. 3:11-19
Jacob TA, Buhs RP, Carlin JR, Chiu SHL, Miwa GT, Rosegay A (1983) The
metabolism and tissue residue profiles of ivermectin. In Proceedings of the
MSD AGVET Symposium on Recent Developments in the Control of Animal
Parasites, XXII World Veterinary Congress, Perth, Australia
Leighty EG, Fentiman Jr AF, Foltz RL (1976) Long-chain metabolites of 6,9_ and
6,8-tetrahydrocannabinols identified as novel fatty acid conjugates. Res. Comm.
Chem. Pathol. Pharm. 14:13-28
Lu AYH, Wislocki PG, Chiu SHL, Miwa GT (1987) Tissue drug residues and their
toxicological significance. Drug Metab. Reviews 18:363-378
Rosegay A, Avermitilis S, Lu AYH, Walsh MAR, Walker RW, Taub R, Jacob
TA (1982) The metabolism of avermectins B la , H 2B la and H2B lb by liver
microsomes. Drug Metab. & Dispos. 10:268-274
Pivnichny JV, Shim J-SK, Zimmerman LA (1983) Direct determination of
avermectins in plasma at nanogram levels by high-performance liquid chroma-
tography. J. Pharm. Sci. 72:1447-1450
Schmid J, Prox A, Baeur E, Koss FW (1980) Abstract No. 434, 7th European
Workshop on Drug Metabolism, Zurich
Stong JD (1987) Determination ofivermectin by fluorescence derivatization. Anal.
Chem. 59:266-270
of avermectin in plasma at nanogram levels using high-performance liquid
Tway PC, Wood Jr JS, Downing GV (1981) Determination ofivermectin in steers
and sheep tissues using high-performance liquid chromatography with fluo-
rescence detection. J. Agric. Food Chem. 29:1059-1063
Weber NE (1980) Persistent residues: Interface with regulatory decisions.
J. Environ. Path. Toxicol. 3:35-43
CHAPTER 9
Chemical Assay for Ivermectin

in Edible Tissues
George V. Downing
For any animal drug a chemical assay is needed to demonstrate the

reduction of drug amounts in edible tissues to negligible levels at
appropriate times after the administration of the drug. In the case of
ivermectin this was a particular analytical challenge. Because ivermectin
is very potent, it is dosed at quite low levels, typically 0.2 mg/kg: hence
the residues, even immediately following dosing, are in the parts per
billion (ppb) level. Further, the no-effect level (NOEL) of 0.2 mg/kg in
animal safety studies results in negligible residues being defined as 10 to
20 ppb for many tissues. Since an acceptable analytical method must
provide reliable data at the negligible residue level, the actual analytical
method should have a limit of detection approximately tenfold lower than
the negligible level.
To develop a method of adequate reliability and sensitivity, it was felt
that only a chromatographic method would have the requisite specificity.
Ultraviolet detectors available at that time, however, were not sensitive
enough to detect the low levels of drug. Fortunately, Tolan and colleagues
(1980) descovered an aromatization reaction that resulted in a fluorescent
derivative with the desired sensitivity; estimated to be 0.2 ng/ml in
plasma. The isolation of the drug from tissue and the subsequent for-
mation of the fluorescent derivative by a modified procedure is the basis
of the analytical method that has been successfully applied for determing
residue levels in the edible tissues of many species (Tway, Wood, and
Downing 1981). The sensitivity of the method is estimated to be about
1 ng/gm.
The fluorescent derivative has been used by other investigators to
analyze ivermectin in plasma (Marriner, McKinnon, and Bogan 1987),
including the use of the more recently available solid adsorbents to
simplify the isolation (Kojima, Yamamoto, and Nakanishi 1987). A
modification of the method utilizing solid adsorbents was also developed
to determine invermectin levels in human milk as well as in plasma
(Chiou, Stubbs, and Bayne 1987). The milk assay has a sensitivity of
0.05 ppb.
9. Chemical Assay for Ivermectin in Edible Tissues 145
With the development of more sensitive ultraviolet detectors, methods

for the determination of ivermectin in plasma have been reported using
ultraviolet detection with a sensitivity of 2 ng/ml (Pivnichny, Shim, and
Zimmerman 1983). A method utilizing normal-phase chromatography and
ultraviolet detection with a sensitivity of 4 to 5 ng/ml (Schnitzerling and
Nolan 1985) is unique in that the dihydro avermectin B la and dihydro
avermectin Bib components cochromatograph, as contrasted to the
reverse-phase chromatographic methods, where the two components are
separated. Recently, a method has been published (Alvinerie et al. 1987)
for the determination in bovine milk using ultraviolet detection and a
sensitivity of 2 ng/ml.
Radioactive studies have demonstrated that the parent drug is an
adequate marker substance to monitor the depletion of residues in tissues
(see Chapter 8.). Even though the drug is metabolized, the parent
compound remains a major component of the total residue. In addition,
the major component of ivermectin-namely the dihydro avermectin B la
(H2BIJ-is depleted at the same rate as or more slowly than the minor
component, dihydro avermectin Bib; since H 2Bla is typically present at
concentrations 10 to 20 times higher than H 2Blb , it is used as the marker
compound in the assay.
In the method described here (and in detail in Appendix I), the tissue is
homogenized with acetone-water and the drug extracted into isooctane. (a
modification of the extraction procedure has also been developed to
measure levels in bovine milk, in which ammonium hydroxide and
ethanol are added to the milk, which is then extracted with ethyl
acetate/hexane.). The isooctane extract is evaporated and the residue
taken up in methanol and chilled to precipitate fatty material (fat is treated
somewhat differently). The extract is further purified by acetonitrile/
hexane cleanup to remove hexane-soluble impurities; water is then added
to move the drug into a fresh hexane phase. This hexane phase is evap-
orated and the purified extract taken up in methanol. An aliquot of the
methanol is evaporated and the residue derivatized by heating with
methyl imidazole/acetic anhydride/dimethyl formamide. The black reac-
tion mixture is cleaned up by adding chloroform and passing the solution
over a small silica gel column. The chloroform is evaporated and the
sample taken up in methanol for injection onto the HPLC column.
Detection is by fluorescence at an excitation wavelength of 364nm and
emission at 418nm. Tables 9-1 and 9-2 summarize date for recovering
ivermectin from tissues and plasma, using the HPLC/fluorescence deter-
minative residue method.
Confirmation that an observed residue is indeed the drug is an
important aspect of the chemical analysis of residues, both from an
analytical and regulatory standpoint. Often this is done by a mass
spectroscopic method, usually gas chromatography/mass spectroscopy.
Ivermectin is particularly unsuited to this type of assay, however; not
146 George V. Downing
TABLE 9.1. Summary of recoveries of

ivermectin from tissues by the
HPLClfluorescence method.
Added
Tissue ng/gm Percent Recovery
Injection 9.1 99, 88, 99
Site 14 93
18 94,94
36 92
Fat 9.1 110,77

14 86
18 83
27 85
45 91
Kidney 9.1 88,88

18 78
27 89
Liver 9.1 99
14 86,86
18 83
27 85,81
45 87
91 88
Muscle 9.1 99,88

18 100
27 96
Range = 77%-110%
Average = 90%
Coefficient of Variation = ± 7%
TABLE 9.2. Summary of recoveries of

ivermectin from plasma by the
HPLC/fluorescence method.
Added
(ng/ml) Percent Recovery
0.91 110, 110, 110
1.82 110, 110, 110
2.73 73,73
3.63 83,83
4.54 88,88
5.45 92,73
7.27 96, 83, 83, 96
9.09 77,88
14.54 83,89
Range = 73% 110%

Average = 91%
Coefficient of Variation = ± 13%
9. Chemical Assay for Ivermectin in Edible Tissues 147
only is it nonvolatile but it is also extremely poorly ionized in the mass

spectrometer. While heroic efforts using MS/MS (Tway et al. 1984)
indicated that the technique had the potential for an identification/
confirmation method, it did not seem to be a practical analytical ap-
proach. Instead, some unique chemical properties of ivermectin were
used to form 2 different derivatives; along with the highly specific
fluorescent detection, they serve to confirm the identity of the drug
residue. When ivermectin is treated with 1% sulfuric acid in methanol,
both of the sugar moieties are removed, resulting in the aglycone;
treatment with 1% sulfuric acid in isopropanol, however, removes only
one of the sugars, resulting in the formation of the monosaccharide. Both
of these compounds form fluorescent derivatives which have retention
times different from each other and from the parent drug. Thus, for
confirmation the original reaction must result in a fluorescent peak with
the proper retention time, and the subsequent acid-treatment products
must produce 2 new peaks, each with the appropriate retention times; the
original peak, of course, disappears because it is converted to the new
product.
The method described in detail in Appendix I has proven to be both
versatile and rugged; it has been applied to thousands of samples from
many species over the years. In addition to monitoring edible tissues, it
has been used to monitor plasma levels in rat, dog, mouse, and human in
connection with various studies. It has also been used to determine levels
in other tissues from other sources, including those from a snake that had
been overdosed with a mouse which had been dosed with ivermectin. The
depletion profiles that have been developed using data from this method
show considerable variation between species, and between methods of
administration. Some of these results are given in Table 9-3, where the
time to peak and the half-life of depletion values are given for a number of
studies. It should be noted that most injection routes show a prolonged
release pattern as compared to oral dosing.
TABLE 9.3. Depletion profiles of ivermectin in

various species.
Species Route Peak Plasma t1l2 Liver tl/2
Cattle SC 27 hrs 2.7 days 4.8
Oral 24-48 2.6 3.4
Horse 1M 72 4.2,7.1,2.1
Oral 6 3.5
Sheep Oral 6 0.7,2,5 1.5
Pig SC 48 4.6
Human OraI 1-4 10-12 hrs

148 George V. Downing
Acknowledgments
The development and application of the ivermectin tissue assay represent
the work of many people.
I would like to acknowledge the contributions of the following people:
Dr. Patricia C. Tway, Mr. James S. Wood, Jr., Dr. Teresa A. Wehner,
Ms. Krystyna T. Czeh, Ms. Eugenia H. Greff, Mr. William E. Tait, and
Mr. Richard J. Varsolona.
REFERENCES
Alvinerie M, Sutra JF, Galtier P, Toutain PL (1987) Determination of iver-

mectin in milk by high performance liquid chromatography. Ann. Rech. Vet.
18:269-274
Chiou R, Stubbs, RJ, Bayne WF (1987) The determination ofivermectin in human
plasma and milk by HPLC with Fluorescence detection. J. Chromatogr.
416:196-199
Kojima K, Yamamoto K, Nakanishi Y (1987) Determination of 22, 23-
dihydroavermectin B'a in dog plasma using solid-phase extraction and high-
performance liquid chromatography. J. Chromatogr. 413:326-331
Marriner SE, McKinnon I, Bogan JA (1987) The pharmokinetics of ivermectin
after oral and subcutaneous administration to sheep and horses. J. Vet. Pharm.
Therap. 10:175-179
Pivnichny JV, Shim JSK, Zimmerman LA (1983) Direct determination of aver-
mectins in plasma at nanogram levels by high performance liquid chromatogra-
phy. J. Pharm. Sci. 72:1447-1449
Schnitzerling HJ, Nolan J (1985) Normal phase liquid chromatographic determi-
nation of nanogram quantities of ivermectin in cattle blood or plasma. J. Assoc.
Office. Anal. Chem. 68, (1):36-40
Tolan JW, Escola P, Fink DW, Mrozik H, Zimmerman LA (1980) Determination
chromatography with fluorescence detection. J. Chromatogr. 190:367-376.
Tway PC, Downing GV, Slayback JRB, Rahn GS, Isensee RK (1984) Confir-
matory assay for ivermectin in cattle tissue using chemical ionization
mass spectrometry/mass spectrometry (MS/MS).Biomed. Mass Spectrom.
11 ,(4): 172-176
Tway PC, Wood JS, Downing GV (1981) Determination ofivermectin in cattle and
sheep tissues using high performance liquid chromatography with fluorescence
detection. J. Agric. Food Chem. 29:1059-1063
CHAPTER 10
Safety of Ivermectin
in Target Animals
J.D. Pulliam and J.M. Preston
I. Introduction
II. Safety in Horses
III. Safety in Cattle
IV. Safety in Sheep and Goats
V. Safety in Swine
VI. Safety in Dogs
I. Introduction
As part of the overall avermectin development program, target animal
safety studies were conducted in many countries by Merck personnel and
by independent investigators in universities and other research institu-
tions. In many instances anticipated commercial formulations were
studied, but during early development and when excessive dosing de-
manded, alternative or more concentrated formulations were used. The
various test formulations were compared for safety by bridging studies
involving assessment of tolerance and bioavailability. Control groups
were given vehicle, saline, or left untreated; many studies included both
treated and untreated controls.
II. Safety in Horses

Clinical signs of the acute toxic syndrome (toxicity) were first identified
when horses were given ivermectin intramuscularly at 12.0 mg/kg (60
times use level). All 4 treated horses showed depression, ataxia, mydria-
sis, lower lip droop, and depressed respiratory rate, and I was euthanized
after becoming recumbent. The other 3 recovered gradually over 2 weeks.
Some horses in the same trial given 3.0 and 6.0 mg/kg also developed
mydriasis (Washko, unpublished; Campbell and Benz 1984).
Biologically significant changes in hematologic or serum biochemical
values, with the possible exception of serum iron, have not been observed
150 I.D. Pulliam and I.M. Preston
in samples from ivermectin-treated horses. In the first of 2 experiments,

Herd and Kociba (1985) observed small-although statistically sig-
nificant-changes in 8 serum biochemical constituents, but changes in
these same measures were not seen when they repeated the study. They
concluded that these minor alterations were biologically insignificant and
probably related to feeding, handling, and husbandry. Asquith and
Kulwich (1981) also found no untoward hematologic or blood biochemical
changes following ivermectin treatment.
Following oral administration ofivermectin paste at 2.0 mg/kg (10 times
the use level) for 2 consecutive days, 5 of 11 horses exhibited transient
impairment of vision, depression, and ataxia; 2 of the animals were
dehydrated during 5 days after initial treatment (Swan, unpublished;
Egerton, Seward, and Robin 1983).
Ivermectin oral liquid was given as 3 repeated treatments of 0.6 and
1.0 mg/kg at 2-week intervals to foals aged between 3 weeks and 3
months. After the third dose, 2 of 6 foals given 0.6 mg/kg had transient
front limb lameness, which was considered unrelated to treatment, as no
such lameness occurred after the first or second administration of
0.6 mg/kg, nor was lameness or any toxic signs seen when ivermectin oral
liquid was given at 1.0 mg/kg (DiPietro et al. unpublished). Ivermectin in
liquid at 0.6 mg/kg and in paste at 1.0 mg/kg was given orally as 3
repeated treatments at 2-week intervals to foals aged 29 to 59 days with no
signs of toxicosis (Asquith, unpublished). Ivermectin in paste was also
given 3 times at 2-week intervals at 0.6, 1.2, or 1.8 mg/kg to horses in
another trial. The repeated doses of 3 or 6 times use level elicited no toxic
signs, but 9 times the use level was associated with a slow pupillary light
response and a decreased menace reflex after 2 of the 3 treatments
(DiPietro, unpublished).
Prior to marketing, ivermectin was given by intramuscular injection at
200 or 300 JLg/kg to 824 horses in 30 clinical field trials. Two horses had
transient swellings in reaction to the death of Onchocerca sp. microfilar-
iae in the skin, and 8 had injection site swellings that lasted less than 14
days and cleared without treatment. Similar field trials with 434 horses
aged from 6 weeks to 25 years were conducted at ranches, farms, and
stables to assess the field safety of the oral paste formulation. Except for 1
horse that had a transient allergic ventral subcutaneous edema, attributed
to activity against Onchocerca sp. microfilariae, no adverse effects were
seen.
Subcutaneous edema, particularly along the ventral midline, attributed
to death of Onchocerca cervicalis microfilariae has also been reported by
Herd and Donham (1983 a,b), but the swellings generally resolved within
3 to 4 days without additional therapy.
Ivermectin's effect on reproduction and breeding performance in mares
and stallions was also investigated. Fifty-eight mares were given oral
doses of 0.6 mg/kg during the first 2 weeks following breeding, and at
10. Safety of Ivermectin in Target Animals 151
2-week intervals thereafter, for a total of 6 doses during the period of

organogenesis (McKissick et al. 1987). Other groups of mares were
treated 6 or 7 times with 0.6 mg ivermectin/kg at 2-month intervals
beginning at about 90 days of pregnancy (Asquith et af. 1988; Godbee,
unpublished; Egerton, Seward, and Robin 1983). No anatomical or
functional teratogenic defects were discernible in the progeny of either
treatment group. Ivermectin administered once at 0.6 mg/kg to 12
stallions had no deleterious effect on libido, seminal characteristics,
spermatozoal morphology, testicular measurements or weights, concen-
trations of testosterone, or the capacity to produce spermatozoa (Squires,
unpublished; Egerton, Seward, and Robin 1983).
Due to septic lesions at the iqjection site and clostridial myositis, which
resulted in 2.4 injection site infections and 1.5 deaths reported per 100,000
horse doses sold, the parenteral formulation was withdrawn after 17
months in the U.S. market. The incidence of adverse reaction reports
associated with the use of the oral paste formulation, as reported to the
manufacturer and based on the sale of tens of millions of doses, has been
0.0003% for reported incidents and 0.0002% for deaths. None of these
reported reactions has been confirmed to be caused by ivermectin paste.
III. Safety in Cattle

The acute toxicity of ivermectin given subcutaneously to cattle was
investigated at doses of up to 8.0 mg/kg (Washko, unpublished; Leaning,
Roncalli, and Brokken 1983). Calves given 8.0 mg/kg developed ataxia
and became recumbent within 24 hours after dosing. They had general
motor depression, increased respiratory rates, muscular fasiculations,
mydriasis, and extensor rigidity of the limbs. One calf died 3 days after
treatment, 2 were euthanized, and the fourth recovered. Changes in blood
glucose, urea nitrogen, alkaline phosphatase, lactate dehydrogenase,
sodium, potassium, hematocrit, and hemoglobin from hemoconcentration
reflected the depressed physiologic state and lack of fluid intake in the
moribund cattle. Decreased serum iron concentrations within 1 day of
treatment were characteristic of the toxic syndrome, but levels returned
to normal in 7 to 10 days. No specific gross or histologic changes were
associated with ivermectin toxicosis.
Brem and Bulman (1986) examined the hematologic and biochemical
values of cattle 1 and 7 days after each of 3 treatments at 200 ILg/kg
subcutaneously at 30-day intervals. Little or no effect was noted except
for improvement of erythrocyte indices, which may have resulted from
control of parasites.
Oral and subcutaneous formulations of ivermectin at concentrations
equivalent to the marketed products, 4 mg/ml and 10 mg/ml, respectively,
or equivalent volume of vehicle were given at 4.0 mg/kg (20 times use
152 J.D. Pulliam and J.M. Preston
level). Affected cattle showed signs of central nervous system depression,

including listlessness and ataxia, within the first day following administra-
tion; the condition progressed to recumbency and in some instances
death. Control cattle given vehicle formulations, all of which included
propylene glycol, exhibited similar signs. The contributory effect of the
vehicle was further demonstrated by Brokken (unpublished) when he
gave single doses as high as 6.0 mg/kg (30 times use level) in a more
concentrated formulation (92 mg/ml) without ill effects (Leaning, Ron-
calli, and Brokken 1983).
Two trials using 208 Jersey and Friesian heifers were conducted to
determine the effect of ivermectin at 400 JLg/kg (twice use level) on the
developing bovine embryo and fetus when administered on 3 occasions
during organogenesis. Overall, treatments were administered at 12 time
periods between 7 and 56 days after insemination. No effects on preg-
nancy were detected that could be attributed to ivermectin treatment,
nor were any teratogenic effects observed in the calves that were subse-
quently delivered (Leaning, Roncalli, and Brokken 1983).
To demonstrate the safety oftreating cows during the second and third
trimesters of pregnancy, 21 cows were allocated to treatments by breed
and on the basis of breeding dates. Ivermectin was given at 400 JLg/kg
(twice use level) monthly, starting as early as the 90th day of pregnancy,
for a total of 6 administrations. No adverse effects were observed, and
normal calves were born, which indicated that there is no drug-related
risk in dosing pregnant cows (Leaning, Roncalli, and Brokken 1983).
A group of 10 bulls was evaluated for breeding performance, including
semen quality, before and after treatment at 400 JLg ivermectin/kg.
Treatment had no deleterious effects on semen concentration, motility,
volume, spermatozoal morphology, scrotal and testicular measurements,
testicle weights and histomorphology, the capacity to produce spermato-
zoa, libido, or breeding performance (Galloway, unpublished; Leaning,
Roncalli, and Brokken 1983).
Despite a distinct lack of problems during an extensive series of field
studies, a low incidence of adverse reactions to death of Hypoderma spp.
larvae following treatment with ivermectin was observed, primarily in
France. Ivermectin is highly effective against Hypoderma spp.larvae; the
death of migrating (first-instar) H. bovis after treatment may be followed
by hemorrhage into the spinal canal, which can result in paresis. Cattle
with first-instar H. lineatum in the esophageal tissues at the time of
treatment may experience an eosinophilic, edematous esophagitis; but
there have been only a few reports of this reaction. In France, most
reactions were reported during an indemnification period. The label
warning was revised, and few adverse reactions involving Hypoderma
spp. have since been reported. Hundreds of millions of doses of the
injectable cattle product have been sold as of this writing, with an overall
incidence of 0.0001% reported adverse reactions. The other reported
adverse reactions were incidental concurrent problems.
The tolerance of cattle to treatment at twice the recommended dose

was investigated in 53 field trials comprising 2367 cattle. In the iver-
mectin- and vehicle-treated cattle, most of the problems related to
transient injection site swelling. It was concluded that dosing cattle with
ivermectin at twice use level did not result in an increased incidence of
health problems compared with animals given vehicle only (Leaning,
Roncalli, and Brokken 1983).
In Australia, an injectable formulation of avermectin B, (abamectin) is
marketed for use in cattle. Although used widely as a pesticide, this is the
only abamectin formulation developed for commercial use in animals thus
far. Twenty-one trials were conducted with 477 treated and 444 control
cattle to evaluate tolerance. Dose levels ranged from 300 JLg/kg to
8.0 mg/kg. Safety was examined in 2 acute toxicity studies, 16 field trials
at twice use level, 2 breeding cow studies, and in bulls. The acute toxicity
trials demonstrated that early signs of toxicosis included depression and
ataxia, and that the maximum dose tolerated was about 1.0 mg/kg. In 1
of the field trials, mild signs of toxicity were seen in young calves at
727 JLg/kg. In toxicity trials at levels of 2.0 to 8.0 mg/kg or greater, the
animals showed more severe signs of toxicity, including ataxia, which
progressed to paresis, recumbency, decreased lip and tongue tone,
drooling, mydriasis, coma, and death. Abamectin at 300 JLg/kg had no
effect in the reproduction safety studies in cows and bulls. Following
reports of adverse reactions in calves aged 1 week to 3 to 4 months
(sometimes associated with overdosing), the label was revised to warn
against treating animals under 4 months of age. It has been speculated that
this increased susceptibility may have been due to increased permeability
of the blood-brain barrier to avermectins in young animals (Button et ai.
1988). Also in Australia, idiosyncratic toxic reactions were reported in a
herd of Murray Grey cattle, which were found to have higher levels of
abamectin in the central nervous system than would normally be expected
(Seaman et ai. 1987).
IV. Safety in Sheep and Goats

The acute toxicity syndrome to ivermectin has not been clearly identified
in sheep. Sheep given 8.0 mg/kg of ivermectin in propylene glycol were
ataxic within 3 hours; they staggered, leaned against the sides of the pen,
crossed their legs, were incoordinated, and fell. All were depressed, and 1
became recumbent with reflexes present but delayed. After 24 hours there
was marked improvement with only mild incoordination and depression;
by 3 days after dosing they appeared clinically normal. Two control sheep
given an equivalent volume of propylene glycol vehicle, however, devel-
oped comparable signs, and 1 died. No biologically significant or charac-
teristic changes were seen in a series of hematologic and biochemical
assays before and after treatment. No gross or histologic changes were
found upon necropsy at the termination of the trial (Washko, unpub-

lished; Hotson 1983). Sheep given 4.0 mg/kg ivermectin in propylene
glycol in this experiment (and in another in South Mrica) also had
depression and incoordination. Those in South Mrica also had hemoglo-
binuria, as did the equivalent vehicle-treated controls (Schroder, unpub-
lished; Hotson 1983). These reactions were attributed to the propylene
glycol vehicle and are not considered to be a characteristic of ivermectin
toxicity in sheep.
A micellar formulation was developed to reduce the incidence of
coughing that occurred following drenching with the propylene glycol
vehicle. Doses of ivermectin in this currently marketed formulation, up to
and including 4.0 mg/kg when given by stomach tube, produced no
adverse clinical reactions (Sutherland and Swan, unpublished; Hotson
1983).
A teratology study was conducted in New Zealand wherein 136 ewes
were given ivermectin at 400 ILg/kg, 138 were given vehicle, and 52 were
untreated controls. Groups of ewes were treated twice with a 21-day
interval, covering the period from Day 0 to 43 of pregnancy. Records
were kept of returns to service, numbers oflambs born, and any incidence
of abnormalities. All dead lambs were carefully examined, and a sample
of apparently normal lambs was necropsied for thorough examination. No
adverse effects resulted from the treatment of ewes in early pregnancy
(Hotson 1983). Breeding animal safety was further established in an
Australian study with 36 ewes given ivermectin at 400 ILg/kg at 14-day
intervals from approximately Day 51 of pregnancy until weaning of the
lambs at 56 days of age. No adverse effects were seen (Hotson 1983). The
injectable ivermectin formulation was given to ewes at 300 ILg/kg on 3
occasions at 7-day intervals during the first 3 weeks of pregnancy and to
other groups of pregnant ewes at 300 ILg/kg during weeks 7,8, 11, 12, 15,
19, and 20 of gestation. It was concluded that ivermectin administered
subcutaneously during the period of organogenesis or the last 14 weeks of
gestation had no adverse effect on the ewe or the fetus.
Ivermectin was administered orally 6 times at 400 ILg/kg at 21-day
intervals to 10 Merino rams. Ten other rams served as untreated controls.
A total of 29 semen samples were collected over a 122-day period. Semen
was evaluated for volume, density, color, motility, pH, percentage of live
sperm, and sperm morphology. One testis was removed from each ram 9
days after the last treatment (Day 114) and histological examinations
performed on testicular sections. No adverse effect on semen quality was
observed. There were no histological differences between the testicular
tissue of treated and control animals 9 days after the sixth treatment. It
was concluded that repeated treatment with ivermectin at 400 ILg/kg did
not impair the reproduction potential of rams (SchrOder et al. 1986). The
injectable ivermectin formulation was also given to rams at 300 ILg/kg
twice at a 7-day interval. Beginning a week before the first treatment and
at weekly intervals until Day 83, each ram was examined and weighed,
alid semen collected and testicular measurements taken. Semen assess-

ment included evaluation of volume, appearance, wave motion, progres-
sive motility, sperm concentration, percent dead, and cytological ex-
amination. The administration of ivermectin subcutaneously to rams at
300 ltg/kg twice with a 7-day interval had no adverse effect on breeding
performance (Harvey, unpublished).
Extensive field trials were conducted using both the propylene glycol
and aqueous micellar formulations at 400 ltg/kg (double use dose) in a
number of countries, with sheep of different breeds and ages under
various conditions of management, nutrition, and climate. More than
12,500 sheep were treated with ivermectin and, in many instances,
concomitant treatments such as flukicides or vaccines were given. The
only reaction was a low incidence of coughing immediately following
dosing. Ivermectin was confirmed to be safe under a variety of conditions
encountered in field use (Hotson 1983). In similar studies with the
injectable formulation, the only observation was a low incidence of
transient pain reaction at the time of injection. Following the sale of
hundreds of millions of doses of the oral sheep product, the incidence of
adverse reactions, as reported to the manufacturer, has been 0.00002%.
A syndrome of acute toxicity to ivermectin has not been clearly
identified in goats, either. Goats injected intramuscularly with ivermectin
at levels up to 2.0 mg/kg showed evidence of intense pain during and
immediately following injection but no signs of systemic toxicity. It
should be noted that this is not a recommended route of administration for
ivermectin in any species. Subcutaneous treatment is recommended for
cattle, sheep, and swine. Oral doses of 4.0 mg/kg were given to goats in 3
trials. In the first trial, the large volume (5.0 mllkg-40 times use level) of
micellar formulation was given by 30-ml plastic syringe into the goats'
mouths, resulting in gagging and most probably aspiration of the formula-
tion, since approximately half of both ivermectin- and vehicle-treated
groups died of pneumonia. In 2 subsequent trials, when the formulation
was given intraesophageally at 4.0 mg/kg (20 times use level), no signs of
toxicity were seen. It was concluded that the effects seen in the first trial
(4.0 mg/kg level) resulted from poor dosing technique, as both sub-
sequent, carefully controlled studies found no effects at 4.0 mg/kg in
goats. Njanja, Bell, and Wescott (1985) gave groups of goats 0.4, 0.8, and
1.6 mg/kg by subcutaneous injection without eliciting signs of toxicity
other than pain reactions that lasted 1 to 2 minutes.
V. Safety in Swine
The signs of acute toxicity to ivermectin in swine were observed in a
study where groups of swine were given 1, 10, 50, and 100 times the use
level of 300 ltg/kg. At 30.0 mg/kg, the pigs were lethargic and ataxic
within 24 hours following treatment. Bilateral mydriasis, intermittent
tremors, labored breathing, and recumbency occurred. Hematologic and

biochemical changes reflected the depressed physiological condition of
the animals, which were laterally recumbent and had decreased food and
water intake. These effects were noted in values for alkaline phosphatase,
cholesterol, triglycerides, glucose, phosphorus, potassium, creatinine,
direct bilirubin, and blood urea nitrogen. A posttreatment leukocytosis
and neutrophilia observed in some of the pigs given 30.0 mg ivermectin/kg
may have been an inflammatory response to the necrosis of connective
tissue, fat, and muscle observed at the injection sites following injection
of the large-volume dose. Serum iron values were decreased 8 to 14 days
following treatment; this transient response has also been associated with
the acute toxic syndrome in horses and cattle. Other than the inflamma-
tory reactions at multiple injection sites, no specific gross or histologic
changes were attributable to ivermectin. Pigs treated with ivermectin at
levels up to 15.0 mg/kg (50 times use level) did not exhibit signs oftoxicity
(Pulliam, unpublished; Campbell and Benz 1984).
A breeding-animal safety trial was conducted to determine the safety of
ivermectin for the developing embryo and fetus during the initial 30 days
of pregnancy when implantation and organogenesis occur. Forty-two
Yorkshire-cross sows were divided into 6 replicates of 7 treatments. All
sows in a replicate were individually mated to the same boar and
randomly allocated to treatments. Ivermectin, or an equivalent volume of
vehicle only, was administered by subcutaneous injection in the neck at a
dose level of 600 JLg/kg (twice use level) on 2 occasions (Days 1 and 18, 6
and 24, 12 and 30) during early pregnancy. An unmedicated control group
was included in each replicate. The first day after mating that a sow
refused to accept the boar was recorded as Day 1 of gestation. Pregnant
sows were observed for return to estrus by exposing them daily to a boar
for the first 45 days after mating. The sows were observed daily through
gestation, farrowing, and weaning of the pigs at 4 weeks of age. All pigs
that were dead at birth or that died prior to termination of the trial were
necropsied. At least 25% of each sex within each litter were randomly
selected for detailed anatomical dissection at about 35 days of age. The
clinical and histopathologic data collected during the trial did not reveal
significant differences (p > 0.05) among the litters from groups of sows
treated with vehicle or with ivermectin at 600 JLg/kg at various times
during the initial 30 days of pregnancy. Nor was there a significant
difference (p > 0.05) observed between the litters of unmedicated con-
trols and those of the vehicle- and ivermectin-treated groups (Brokken et
al. 1983).
To demonstrate the safety of treating pregnant sows during the latter
two thirds of pregnancy, ivermectin was given at 600 JLg/kg (twice use
level) at 28-day intervals. In this study, 21 pregnant Yorkshire-cross sows
were divided into 7 replicates of 3 sows each following individual hand
mating to the same boar. Commencing around 45 days of gestation, each
sow within a replicate received either no treatment, ivermectin, or an

equivalent volume of vehicle at 28-day intervals, so that 3 treatments
were given prior to farrowing. Pigs that were dead at birth and those that
died prior to termination of the trial were necropsied. A statistical
evaluation of the data did not detect significant differences (p > 0.05)
between treatment groups with regard to sow weight gain or the weight
and number of pigs farrowed. No correlation between the incidence of
death or abnormalities of the pigs and the treatments was detected
(Brokken et al. 1983).
The use of ivermectin in boars was studied in 10 Duroc-cross boars, 9 to
10 months old. During a 7- to 10-week acclimation period, the boars were
trained to mount a dummy sow and ejaculate. On Day 0, ivermectin was
given at 600 ILg/kg, and a comparable volume of vehicle was given to the
control boars. Semen was collected on Days -7, 0, 3, and 7 and then
weekly until Day 63 following treatment. The semen was evaluated for
volume, motility, progressive motility, abnormal sperm, sperm concen-
tration, and pH. No differences were detected between the ivermectin-
treated boars and the vehicle-treated controls that would indicate impair-
ment of breeding soundness following dosing with ivermectin (Brokken et
al. 1983).
Use trials with pigs of various breeds and age, including sows, gilts,
barrows, boars, and piglets were conducted in the United States and
Europe under field conditions. Ivermectin was given at 300 ILg/kg in the
U.S. trials and at 600 ILg/kg in the European trials. The trials were run
under various conditions of management and nutrition; in many in-
stances, concomitant treatments were given. No adverse reactions that
could be attributed to treatment were observed in any of the pigs during
the observation period of about 2 weeks (Brokken et al. 1983). Incidental
concurrent adverse reactions following the sale of tens of millions of
doses of the swine product have been reported to the manufacturer at a
rate of 0.00014%.
VI. Safety in Dogs

The acute toxic syndrome of ivermectin toxicosis in dogs was manifest as
mydriasis, depression, tremors, ataxia, stupor, emesis, drooling, and
coma, with an LDso for Beagles of approximately 80.0 mg/kg (Mac-
Donald, unpublished). The highest single oral dose without effect was
2.0 mg/kg; for continuous daily dosing for 14 weeks it was 0.5 mg/kg
(Norbury, unpublished). Single oral doses of 2.5 mg/kg produced mydria-
sis. Increasing doses of 5.0 mg/kg elicited mydriasis and tremors, while
severe signs-including ataxia-were seen at oral doses of 10.0 mg/kg.
One of 4 dogs given 40.0 mg/kg and 2 of 4 dogs given 80.0 mg/kg died
(MacDonald, unpublished).
In a 14-week study with daily oral dosing, dogs given 1.0 mg/kg or more
developed mydriasis. At 2.0 mg/kg, 4 of 8 dogs additionally lost a small
amount of weight and developed tremors, ataxia, anorexia, and dehydra-
tion. Euthanasia and necropsy of these 4 dogs revealed agonal gastro-
intestinal hemorrhage and/or congestion. No other treatment-related
gross or histological changes were found (Norbury, unpublished). Sub-
cutaneous injection of ivermectin at 4.7 mg/kg produced mydriasis and
salivation, while depression, ataxia, and deaths occurred at 9.4 mg/kg
(Seward, Brokken, and Plue 1986).
In assessing the physiologic effects of ivermectin, an intravenous dose
of 0.5 mg/kg had no significant effect on blood pressure or heart rate of
anesthetized dogs, nor did it modify blood pressure or heart rate
responses to autonomic drugs in a standard assay. An intragastric dose of
0.5 mg ivermectin/kg did not affect evoked or basal gastric secretions in
dogs with chronic gastric fistula (Clineschmidt, unpublished).
Safety in breeding bitches was evaluated in 2 studies. A dose of
0.5 mg/kg was administered to 14 pregnant bitches on Days 5, 15,25, and
35 of gestation and to 15 pregnant bitches on Days to, 20, 30, and 40 of
gestation. No teratogenic effects were observed in 48-day fetuses from
these bitches treated repeatedly with ivermectin during the first 40 days of
gestation. Ivermectin also had no effect on the reproductive status of the
pregnant bitches, as measured by the number of implants, resorptions,
and live and dead fetuses per pregnant female (Robertson, unpublished).
In the second study, bitches were given repeated oral doses of
ivermectin at 600 JLg/kg at least twice at monthly intervals before
breeding, and then on Days 10,25, and 45 of gestation, and again monthly
after whelping. No adverse effect on reproduction was observed in
treated dogs as compared to the controls (Gilman, unpublished).
The effect of ivermectin on spermatogenesis, fertility, and the re-
productive performance of 12 Beagle studs was evaluated. Ivermectin
was administered orally at 0.6 mg/kg at monthly intervals for 8 months
while controls were given water. Semen was collected every 3 days from
28 days before treatment until 83 days after (38 collections/dog). All studs
were then bred to 2 nontreated bitches each; litter size, birth weight, and
pup abnormalities and mortalities were evaluated. After all pups were
whelped, each stud was euthanized and necropsied, and the testes and
epididymides were examined microscopically. No adverse effect on
reproduction was observed in treated dogs as compared to the controls
(Daurio et al. 1987).
Following extra-label use of ivermectin products formulated and
marketed for horses (EQVALAN®), and for sheep, cattle, and swine
(IVOMEC®), it appeared that, within the Collie breed, some dogs were
unusually sensitive to the toxic effects of ivermectin (Campbell, Blair,
and Seward, 1983; Seward, Blair, Plue, and Brokken, 1983). In a
controlled study, 8 Collies affected with Collie eye anomaly (CEA) and 8
Collies clinically free of CEA were studied to determine whether iver-
mectin was more toxic to members of the Collie breed and if CEA was
related. Groups were given 50, 200, or 600 p,g ivermectin/kg orally. One
dog given 200 p,g/kg and 1 given 600 p,g/kg developed severe signs of
toxicity including ataxia, depression, tremors, recumbency, and mydria-
sis: 1 died and 1 was euthanized. Two other dogs in these groups had mild
transitory signs. Ivermectin tissue assays showed high concentrations of
ivermectin in the central nervous system of the 2 Collies that died. Collie
eye anomaly was not related to the response to ivermectin (Pulliam,
Seward, and Henry 1985). The range of Collie sensitivity to ivermectin
was further studied by Paul et al. (1987), who found that Collies are
sensitive to doses from 100 to 2500 p,g/kg, and that the length of time
between dosing and onset of clinical signs may indicate the eventual
severity of toxicosis and serve as an important prognosticator for the
clinician. Physostigmine, 1 mg b.i.d., was found to have limited benefit in
treating these dogs, and then only in the more severely depressed and/or
comatose stages of ivermectin toxicity. Its use as the sole therapeutic
agent or complete antidote for ivermectin toxicosis was not recommended
(Tranquilli et al. 1987).
The unapproved (extra-label or off-label) use of products designed for
large animals continues to result in iatrogenic toxic deaths in dogs
(Seward, Brokken, and Plue 1986). Many of these reactions appear to be
due to the accidental or erroneous administration of excessive amounts of
ivermectin (Preston 1983).
The only currently available formulation for the dog is a tablet to be
administered monthly at a dose rate of 6 p,g/kg for the prevention of
canine heartworm disease. This represents a wide margin of safety, not
only in normal dogs-where a single dose of 2.0 mg/kg and daily oral
dosing of 0.5 mg/kg for 14 weeks had no effect-but also in ivermectin-
sensitive Collies, which tolerate doses of 50 p,g/kg without effect (Pul-
liam, Seward, and Henry 1985; Paul et al. 1987).
In dogs with circulating microfilariae, vomiting, excess salivation,
blood in feces, soft stool/diarrhea, and depression have been observed
after ivermectin treatment (Schlotthauer et al. 1986). Although the
shocklike reaction in dogs infected with Dirofilaria immitis that develops
after administration of diethylcarbamazine did not occur following iver-
mectin treatment (Boreham and Atwell 1983), reactions following iver-
mectin treatment ofmicrofilaremic dogs may occur, and deaths have been
seen (Jackson, Seymour, and Beckett 1986).
The incidence of incidental concurrent adverse reactions, as reported
to the manufacturer and based on the sale of tens of millions of doses of
ivermectin tablets for use in dogs, is 0.001%. None of these reported
reactions has been confirmed to be caused by ivermectin.
REFERENCES
Asquith RL, Kivipelto J, Harvey JW, Bauer JE (1988) Comparative effects and
safety of ivermectin in pregnant mares. J. Equ. Vet. Sci. 8:32-35
Asquith RL, Kulwich R (1981) Safety and therapeutic activity of ivermectin as an
equine anthelmintic. J. Equ. Vet. Sci. 1: 18-20
Boreham PFL, Atwell RB (1983) Absence of shock-like reactions to ivermectin in
dogs infected with Dirofilaria immitis. J. Helminthol. 57:279-281
Brem JJ, Bulman GM (1986) An assay of biochemical parameters in cattle
medicated at therapeutic level with ivermectin. Vet. Argent. 3:365-373
Brokken ES, Barth D, Foster AG, Pulliam JD, Wallace DH (1983) Ivermectin: a
new broad-spectrum antiparasitic agent for swine. In Proceedings of the MSD
AGVET Symposium on Recent Developments in the Control of Animal Para-
sites, XXII World Veterinary Congress, Perth, Australia, pp 239-258
Button C, Barton R, Honey P, Rickford P (1988) Avermectin toxicity in calves
and an evaluation of picrotoxin as an antidote. Austral. Vet. J. 65:157-158
J. Vet. Pharm. & Ther. 7:1-16
Campbell WC, Blair LS, Seward RL (1983) Ivermectin vs. heartworm: the present
status. Proc. 1983 Heartworm Symp., Orlando, Fla., pp 146-149
Daurio CP, Gilman MR, Pulliam JD, Seward RL (1987) Reproductive evaluation
of male Beagles and the safety of ivermectin. Am. J. Vet. Res. 48: 1755-1760
Egerton JR, Seward RL, Robin B (1983) Ivermectin as an antiparasitic agent for
horses. In Proceedings of the MSD AGVET Symposium on Recent Develop-
ments in the Control of Animal Parasites, XXII World Veterinary Congress,
Perth, Australia, pp 49-55
Herd RP, Donham JC (1983a) Efficacy of ivermectin against Onchocerca cervica-
lis microfilarial dermatitis in horses. Am. J. Vet. Res. 44:1102-1105
Herd RP, Donham JC (1983b) Control of equine cutaneous nematodiasis by
ivermectin. In Proceedings of the MSD AGVET Symposium on Recent Devel-
opments in the Control of Animal Parasites, XXII World Veterinary Congress,
Herd RP, Kociba GJ (1985) Effect of ivermectin on equine blood constituents.
Equ. Vet. J. 17:142-144
Hotson IK (1983) The development of ivermectin as an antiparasitic agent in
sheep. In Proceedings of the MSD AGVET Symposium on Recent Develop-
ments in the Control of Animal Parasites, XXII World Veterinary Congress,
Jackson RF, Seymour WG, Beckett RS (1986) Lower dose of ivermectin as
a microfilaricide, 0.05 mg/kg. In Proc. 1986 Heartworm Symp., New Orleans,
pp 15-18
Leaning WHD, Roncalli RA, Brokken ES (1983) The efficacy and safety
evaluation of ivermectin: A new injectable antiparasitic agent for cattle. In
Proceedings of the MSD AGVET Symposium on Recent Developments in the
Control of Animal Parasites, XXII World Veterinary Congress, Perth, Austra-
lia, pp 25-41
McKissick GE, Sutherland IH, Foix J, Olson G (1987) The safety of ivermectin
administered orally to pregnant mares. Equ. Vet. Res. 7:357-367
Njanja JC, Bell JF, Wescott RB (1985) Apparent lack of toxicity in adult East
African goats on parenterally administered ivermectin. Bull. Anim. Health
Prod. Afr. 33:123-127
Paul AJ, Tranquilli WJ, Seward RL, Todd KS, DiPietro JA (1987) Clinical
observations in Collies given ivermectin orally Am. J. Vet. Res. 48:684-685
Preston JM (1983) Adverse reactions to unapproved applications (Letters to the
Editor) Vet. Rec. 112:286
Pulliam JD, Seward RL, Henry RT (1985) Investigating ivermectin toxicity in
Collies. Vet. Med. 80:33-40
Schlotthauer JC, Stromberg BE, Paul AJ, Todd KS, McCall JW, Dzimianski MT,
Blagburn BL, Hendrix CM (1986) Safety and acceptability ofivermectin in dogs
with naturally acquired patent infections of Dirofilaria immitis. Proc. 1986
Heartworm Symp., New Orleans, pp 29-32
SchrOder J, Swan GE, Barrick RA, Pulliam JD (1986) Effect ofivermectin on the
reproductive potential of breeding rams. J. So. Afr. Vet. Assoc. 57:211-213
Seaman JT, Eagleson JS, Carrigan MJ, Webb RF (1987) Avermectin B\ toxicity in
a herd of Murray Grey cattle. Austral. Vet. J. 64:284-285
Seward RL, Blair LS, Plue RE, Brokken ES (1983) The efficacy and safety of
ivermectin in dogs. In Proceedings of the MSD AGVET Symposium on Recent
Developments in the Control of Animal Parasites, XXII World Veterinary
Congress, Perth, Australia, pp 259-266
Seward RL, Brokken ES, Plue RE (1986) Ivermectin vs heartworms-A status
update. Proc. 1986 Heartworm Symp., New Orleans, pp 1-8
Tranquilli WJ, Paul AJ, Seward RL, Todd Jr. KS, DiPietro JA (1987) Response to
physostigmine administration in Collie dogs exhibiting ivermectin toxicosis.
J. Pharm. & Ther. 10:96-100
CHAPTER 11
Environmental Aspects of
Ivermectin Usage in Livestock:
General Considerations
B.A. Halley, R.J. Nessel, and A.Y.H. Lu
I. Introduction
II. Excretion of Ivermectin
III. Environmental Burden
IV. Soil Binding and Translocation
V. Stability in Soil and Photodegradation
VI. Toxicity to Soil Microorganisms
VII. Toxicity to Earthworms
VIII. Acute Toxicity to Freshwater Organisms
IX. Environmental Safety Assessment
X. Cattle Feedlot Environmental Safety Study
I. Introduction
A detailed analysis of ivermectin's effect on the environment was an
integral component of the overall program to develop ivermectin as an
antiparasitic drug for food-producing animals. The analytical studies were
designed to determine whether using ivermectin in animals would result in
any harmful or undesirable effects on the environment. These studies
measured ivermectin's physical properties and its mobility, distribution,
and stability in soil and water. Additional studies investigated the drug's
effect on a variety of environmentally important organisms. These
studies, combined with the clinical use pattern of ivermectin in cattle,
sheep, and swine, provided the means to assess ivermectin's environmen-
tal impact.
11. Environmental Aspects of Ivermectin Usage 163
II. Excretion of Ivermectin

When ivermectin is used in food-producing animals, the drug enters the
environment as a result of excretion. Radiolabeled drug was used to
determine the routes of excretion; the data in Table 11.1 show that
most of the radioactivity was excreted in the feces, regardless of animal
species or the route of administration (Chiu et al., in preparation). Less
than 2% of the radioactivity was found in the urine. (The rat data are
included in Table 11.1 for comparison.) Analysis by high-performance
liquid chromatography (HPLC) of acetone extracts of feces showed
that the major component in feces was the parent drug, and that the re-
mainder of the radioactivity consisted of various ivermectin metabolites
(Halley et al., 1989a). Based on both in vivo and in vitro metabolism
studies (Chiu et al. 1985; Chiu et al. 1986; Miwa et al. 1982); and a com-
parison of retention times on HPLC, the polar metabolites are be-
lieved to include the monosaccharides and aglycones of ivermectin,
as well as ivermectin's 24-hydroxymethyl derivatives and the corre-
sponding monosaccharides and aglycones. The druglike metabolites
would most likely be the 3"-O-desmethyl derivatives (Chiu et al. 1984;
Chiu et al. 1987).
III. Environmental Burden

The amount of ivermectin entering the environment can be predicted from
the use pattern of this product in food-producing animals. Accordingly,
the environmental burden is determined by the dose, use frequency,
excretion pattern, and disposal of the excreted wastes. The approved use
level of ivermectin is 0.2 mg/kg body weight for cattle and sheep, and
0.3 mg/kg body weight for swine. The animals may be kept on a pasture,
in a small independent feedlot, or in a large commercial feedlot. In
general, the animals receive only 1 drug treatment, but up to 3 or 4
treatments per year may be used in year-round parasite-control programs
(Halley et aI., 1989b).
Manure generated in the feedlot is spread and plowed into the field as
fertilizer. Since commercial feedlots produce the largest amounts of
manure for farm use, an assessment of this particular use pattern should
represent the uppermost environmental burden associated with the use of
ivermectin in farm animals. Table 11.2 shows that the calculated amount
of ivermectin in soil, based on a U.S. Environmental Protection Agency
Publication (U.S. EPA 1974), is in the parts-per-trillion to parts-per-
billion range (Halley et al., 1989b). These calculated levels do not take
into account the degradation of ivermectin during the period that the
animals are kept in the feedlot. The degradation half-life of ivermectin, in
TABLE 11.1. Excretion patterns and composition of radioactive residues in the feces of animals treated with 3H-labeled ivermectin.
% of dose % of radioactivity excreted
Dosage given Recommended excreted' in in the feces b as
Animal Dose route in study use level
species in study (mg/kg) (mg/kg) urine feces ivermectin polars druglike
Steer subcutaneous 0.3 0.2 1.5 62 39-45 54 5
Steer intraruminal 0.3 0.2 0.5 80
Sheep intraruminal 0.3 0.2 0.5 69 61-69 25 4
Swine subcutaneous 0.4 0.3 0.4 36 23-43 43 14-34
Rat oral 0.1 0.9 83 78 15 6
, Excreta were collected 3 days post dose from rats, 7 days post dose from steers, sheep, and swine.
b The composition of radioactive residues excreted in the feces was determined by HPLC profiles and reverse isotope dilution assay. Ranges represent
values from several determinations. Druglike component refers to the radioactive material eluted from the reverse-phase-HPLC slightly ahead of
ivermectin (Le., slightly more polar).
C Not determined.
soil or feces-soil mixtures, has been shown to be in the range of 91 to 217

days in the winter and 7 to 14 days in the summer (Halley et aI., 1989a).
Furthermore, the calculation assumes that all the drug-related material
found in the feces is the parent compound, although Table 11.1 clearly
shows that only about half the residue in feces is ivermectin. Other
studies have established that metabolites or degradation products are less
toxic than ivermectin (Halley et aI., 1989a). Thus, in the environment, the
ivermectin level present in the soil following the spreading and plowing of
manure should be considerably less than the concentrations shown in
Table 11.2.
TABLE 11.2. Environmental burden associated with the use of ivermectin in

various animal species.
Ivermectin
Animal in feces' Ivermectin in
species Assumptions (ppb) soil"
Feedlot cattle animal weight: 270 kg 19 0.2 ppb
dose: 54 mg
feedlot period: 130 d
total waste: 2860 kg
manure application: 27 metric tons/ha
plowing depth: 15 cm
Pastured cattle animal weight: 270 kg 19 0.016 mg/m2
dose: 54 mg
pasture period: 130 d
total waste: 2860 kg
animals/ha: 3
Pastured sheep animal weight: 36 kg 18 0.013 mg/m2
dose: 7.2 mg
total waste: 3% kg
animals/ha: 18
Feedlot swine animal weight: 25 kg feeder; 150 kg 18 0.2 ppb
breeder
dose: 7.5 mg; 45 mg
total waste: 588 kg; 1949 kg
manure application: 27 metric tons/ha
Water-washed animal weight: 25 kg feeder; 150 kg 18 0.04 ppb
swine waste breeder
dose: 7.5 mg; 45 mg
total waste: 588 kg; 1949 kg
spreading rate: 36,600 liter/ha
• Ivermectin concentrations expressed as total drug-equivalent without degradation during
the feedlot or pasture periods.
166 B.A. Halley, R.J. Nessel, and A.Y.H. Lu
IV. Soil Binding and Translocation

The distribution and movement of a chemical in the environment depends
on the compound's physical and chemical properties. Table 11.3 shows
that ivermectin is rather insoluble in water and has a high octanol-water
partition coefficient. These combined properties would indicate that
ivermectin should have a high affinity for soil organic matter and a limited
movement in the environmental compartments. Soil-binding studies with
3H-labeled ivermectin confirm that it is tightly bound to soils, as evi-
denced by the high partition coefficient (KD = 227 to 333) and high
organic-carbon binding constant (Koc = 12,600 to 15,700) (Halley et at.,
1989a).
A soil column leaching experiment was conducted to study the translo-
cation of ivermectin, its metabolites, and degradation products in soil. In
this experiment, steer feces containing radiolabeled ivermectin and
metabolites were mixed with four types of soil; each mixture was applied
to the top of a 30-cm soil column. Water was allowed to percolate through
the soil columns for 7 weeks, and the leachates were collected. The
amount of radioactivity recovered in the column leachates was 27% for
silt loam, 48% for clay loam, 10% for loam, and 43% for pasture sandy
loam (Halley et at., 1989a). HPLC analysis of the leachates showed no
detectable levels of ivermectin. At least half of the radioactivity remained
in the top 5 cm of the column for each soil type. Thus, only the
metabolites or degradants percolated through the columns, whereas
ivermectin remained tightly bound to the soils. These studies indicate that
ivermectin is rather immobile in soil and will not readily translocate into
ground water.
TABLE 11.3. Environmental fate of ivermectin.

Parameter Experimental data Possible consequences
Water solubility very insoluble-5 ppm in rather immobile in the
saturated solution environment
Octanol/water partition 1651 strong affinity to lipid, soil
coefficient and organic matter
Vapor pressure" less than 1.5 x 10-9 mm Hg virtually nonvolatile
UV absorption maxima at 237, 245, and subject to photodegradation
253 nm
Soil binding partition coefficient, very tight soil binding;
Ko = 227-333; organic run-off in surface water
carbon-binding constant, and leaching into ground
Koc = 12,600-15,700 water should be minimal
" Vapor pressure of avermectin, which differs from ivermectin only by having a double bond
at C-22 and C-23.
V. Stability in Soil and Photodegradation

Strong soil binding and lack of mobility could result in the accumulation
of ivermectin in soil. To determine the potential for accumulation, the
stability of ivermectin was studied under a variety of experimental
conditions. Table 11.4 shows that the degradation half-life of ivermectin
varies greatly, depending on the environmental conditions (Halley et at.,
1989a). In the laboratory, or outdoors during the winter months, iver-
mectin in feces/soil mixtures degraded slowly, with half-lives ranging
from 90 to 240 days. In contrast, ivermectin degraded very rapidly in the
summer, with half-lives of 7 to 14 days. Furthermore, the compound is
rapidly decomposed by sunlight: its half-life is only 3 hours when exposed
to sunlight as a thin, dry film on glass (Figure 11.1). These results indicate
that ivermectin will not accumulate in soil when administered to livestock
in the recommended manner.
VI. Toxicity to Soil Microorganisms

The toxicity of ivermectin to soil microbes was determined by studying its
effects on soil respiration and nitrification. Feces from treated steers were
added to pasture and forest soil so that the mixture contained 30 ppb of
ivermectin and metabolites. This concentration is about ISO-fold higher
than the field concentration shown in Table 11.2. Even at this exaggerated
level, ivermectin and its metabolites had no effect on soil respiration (as
measured by cumulative CO 2 levels) and soil nitrification (as measured by
the rates of conversion of ammonium ion to nitrite and nitrate). Soils
containing feces from control or treated steers gave virtually identical
results (Halley et ai., 1989a). In other studies, ivermectin had no
antifungal or antibacterial effects at concentrations as high as 2000 ppm
(see Chapter 2).
TABLE 11.4. Degradation half-lives of ivermectin under various

experimental conditions.
Condition
Laboratory, dark, room temperature, in soil/feces 93-240 days
mixtures
Outdoors, winter, in soil or in soil/feces mixtures 91-217 days
Outdoors, summer, in soil or in soil/feces mixtures 7-14 days
Outdoors, thin, dry film on glass, sunlight 3 hours
100
-----------------{]
o 8 16 24
Cumulative exposure, hours
FIGURE 11.1. Photodegradation of H 2B1a • The half-life of tritium-labeled H 2B1a ,
exposed to sunlight ( • ) as a thin, dry film on glass in July in Rahway, New
Jersey, was about 3 hours. Foil-covered sample ( 0 ) was subjected to the same
conditions.
VII. Toxicity to Earthworms

Ivermectin's toxicity to the earthworm, Eisenia foetida, was studied
under controlled laboratory conditions. The compound was added to soil
at concentrations of 12 to 200 ppm. Following 28 days of continuous
exposure, the LC so was extrapolated to be 315 ppm (Halley et al., 1989a).
The no-effect level was 12 ppm.
VIII. Acute Toxicity to Freshwater Organisms

Chlorella pyrenoidosa, a freshwater, unicellular, nonmotile chlorophyte,
was used to determine the toxicity of ivermectin toward algae. When
C. pyrenoidosa was exposed to ivermectin at concentrations of 1 to 10
ppm for 14 days, no effects on overall cell growth, mean specific growth
rate, or lag period were observed (Halley et al., 1989a). The only
observable effect was on the maximum standing crops (cells/mI). Thus, at
these relatively high concentrations, ivermectin had only a moderate
effect on the growth characteristics of this alga.
Freshwater organisms such as the water flea Daphnia magna, rainbow

trout Salmo gairdneri, and bluegill sunfish Lepomis macrochirus are
highly sensitive to ivermectin (Halley et al., 1989a). Table 11.5 shows that
the LCso value of ivermectin was approximately 0.025 ppb for D. magna,
3.0 ppb for rainbow trout, and 4.8 ppb for bluegill sunfish. Removal of the
oleandrose groups of ivermectin to form the monosaccharides and
aglycones reduced the compounds' toxicities to Daphnia. Leachates-
containing more polar products but no ivermectin-collected from either
soil columns or feces/soil columns were much less toxic to Daphnia.
Due to its tight soil binding, the small amount of ivermectin soluble in
water should be easily immobilized by suspended soil particles. Table
11.6 clearly shows that the binding of ivermectin to soil reduces the
mortality of the daphnids by as much as several orders of magnitude.
Thus, the toxicity of ivermectin to freshwater organisms in their natural
environment will be greatly reduced because of tight soil binding and
rapid photodegradation.
IX. Environmental Safety Assessment

The major route of introducing ivermectin into the soil is via the use of
feces from food-producing animals as fertilizer. The ivermectin levels in
soils listed in Table 11.2 are therefore valuable in assessing environmental
TABLE 11.5. Toxicity of ivermectin and metabolites to aquatic organisms.

Length Approximate
of LC 50' nontoxic level
Test species test (hrs) Test compound (ppb) (ppb)
Water fiea Daphnia 48 ivermectin 0.025 0.01
magna
H2B1• monosaccharide 0.4 0.1
H2B1a aglycone >17b >9
soil column leachates 0.48; 3.2
feces leachates -6.5
Rainbow trout Salmo 96 ivermectin 3.0 0.9
gairdneri
Bluegill sunfish Lepomis 96 ivermectin 4.8d
macrochirus
a Concentration lethal to 50% of exposed population.
b LC50 could not be accurately determined since the highest concentration of the aglycone
used was 17 ppb.
C The concentration of ivermectin, its metabolites, and degradation products in the soil
column or feces leachates limited the toxicity testing at only 1 or 2 concentrations. Not
enough data could be collected to accurately calculate the LC 50 values or the nontoxic
levels.
d Extrapolated value.
e Not determined. Toxicity observed at all concentrations in definitive test.
TABLE 11.6. Effect of soil on Daphnia magna toxicity.a

Concentration
%DMSO No effect causing 50%
Condition usedb level (ppt) death (ppt)
Ivermectin 0.5 10 15-30
Ivermectin plus
New Jersey garden soil
Expt. 1 0.5 >10 100
Expt. 2 0.1 >100 1000
Ivermectin plus
Iowa clay loam 0.1 >1000
a Dilute solutions ofivermectin were mixed with soil, 2.5 g per 100 ml of
solution, and stirred for 1 hour. The soil suspensions were centrifuged
and the mortality of daphnids was determined following 48 hours of
exposure.
b Original trials contained a final concentration of 0.5% DMSO (vehi-
cle). This was reduced to 0.1% in later trials.
C Because of the wide range of ivermectin concentrations needed in
experiments involving soil, not enough data points were available to

calculate the concentration causing 50% death.
safety. These levels are calculated from the ivermectin use pattern and
the rate of manure application. Based on the negligible effect at 30 ppb on
respiration and nitrification, and the no-effect level of 12 ppm on
earthworms, ivermectin would not adversely affect soil microbes and
other organisms. Phytotoxicity is also not expected, due to ivermectin's
lack of toxicity to freshwater algae over the concentration range of 1 to
10 ppm. Nor has phytotoxicity been observed with abamectin, the related
compound now being developed as a pesticide.
Aquatic organisms, such as water fleas and fish, are highly sensitive to
ivermectin toxicity. However, due to its low water solubility, high
octanol-water partition coefficient, and tight soil binding, ivermectin will
be immobile in soil and will not translocate or leach into ground water.
Based on the soil-binding data and the Freundlich equation, Halley and
colleagues (1989b) have calculated that the ivermectin concentration in
runoff and ground water should be below the no-effect level of 0.01 ppb
observed for D. magna (Table 11.5). Thus, in the environment, the
concentration of ivermectin in water should be sufficiently low to cause
no adverse effects on freshwater organisms. Other studies have shown
that ivermectin (or abamectin) does not appear to be very toxic to birds
(Kelso 1984), chickens (Mousa et al. 1986; Zeman 1987), ducks (see
Chapter 13), and mammals (Nessel, Jacob, and Robertson 1983). The
effect of ivermectin on manure-dwelling or -feeding insects is discussed in
Chapter 12.
x. Cattle Feedlot Environmental Safety Study

A cattle feedlot environmental fate study was also carried out to evaluate
ivermectin's potential ecological impact (Nessel et ai., 1989). Five cattle
were housed in each of 2 identical, unpaved, dirt feedlot pens at a density
of approximately one animal per 200 square feet. Animals in the treated
group were given a single dose (0.2 mg/kg) of ivermectin. Control animals
received no treatment. Surface runoff and percolated water collected
during a 28-day trial period contained no significant amounts « 0.01 ppb)
of ivermectin. The collected water also showed no ivermectin-related
toxicity to D. magna (no-effect level, 0.01 ppb). Furthermore, none of
the subsurface (8-30 cm depth) core soil samples collected after 28
days contained any detectable amounts of ivermectin (detection limit
= 1 ppb). This field trial supports our laboratory data and indicates that
ivermectin in feces is unlikely to cause adverse environmental effects in
surface or subsurface waters.
Our studies on environmental safety have been based on the currently
approved uses of ivermectin. As new programs or treatment methods are
developed, their potential environmental impact will need to be assessed.
Acknowledgements
We would like to thank Drs. Peter Wislocki, William VandenHeuvel and
S.H.L. Chiu for their constructive comments on this manuscript.
REFERENCES

ivermectin residue in animal tissues by high-performance liquid chroma-
tography-reverse isotope dilution assay. J. Agric. Food Chem. 33:99-102
Chiu SHL, Sestokas E, Taub R, Buhs RP, Green M, Sestokas R, VandenHeuvel
WJA, Arison BH, Jacob TA (1986) Metabolic disposition of ivermectin in
tissues of cattle, sheep, and rats. Drug Metab. & Dispos. 14:590-600
Chiu SHL, Sestokas E, Taub R, Smith JL, Arison B, Lu AYH (1984) The
metabolism of avermectin-H 2B 1a and -H2B1b by pig liver microsomes. Drug
Metab. & Dispos. 12:464-469
Chiu SHL, Taub R, Sestokas E, Lu AYH, Jacob TA (1987) Comparative in vivo
and in vitro metabolism of ivermectin in steers, sheep, swine, and rat. Drug
Metab. Rev. 18:289-302
Halley BA, Jacob TA, Lu AYH (1989a) The environmental impact of the use of
ivermectin: Environmental effects and fate. Chemosphere, in press.
Halley BA, Nessel RJ, Lu AYH, Roncalli RA (1989b) The environmental safety of
ivermectin: An overview. Chemosphere, in press.
Kelso PC (1984) Ivermectin cures scaly mite infestation. Vet. Med. 79:446
Rosegay A, Avermitilis S, Lu AYH, Walsh MAR, Walker RW, Taub R, Jacob
TA (1982) The metabolism of avermectins B1a , H2B 1a , and H 2B 1b by liver

microsomes. Drug Metab. & Dispos. 10:268-274
Mousa S, Grad N, Sokkar I, Raheem MA (1986) Investigations on the efficacy of
ivermectin for ectoparasites and nematodes in chickens. I. Testing for drug
safety to chickens. II. Anthelmintic efficacy of ivermectin for experimental
ascaridiasis in chickens. Assiut Vet. Med. J. 17(33):223, 225-233
Nessel RJ, Jacob TA, Robertson RT (1983) The human and environmental safety
aspects of ivermectin. In Leaning WHD, Siegmund OH, Fraser CM (eds)
Recent Developments in the Control of Animal Parasites, MSD-AGVET,
Rahway, NJ, pp 98-100
Nessel RJ, Wallace DH, Wehner T, Tait W, Gomez L (1989) Environmental fate
of ivermectin in a cattle feedlot. Chemosphere, in press.
US EPA (1974) Development document for effluent limitations guidelines and new
sourCf! performance standards for the feedlots-points source category. Wash-
ington, DC; EPA-4401l-74-004-a
Zeman P (1987) Systemic efficacy of ivermectin against Dermanyssus gallinae (de
Geer, 1778) in fowls. Vet. Parasitol. 23:141-146
CHAPTER 12
Use of I vermectin and Abamectin
in Livestock:
Effects on Cattle Dung Fauna
R.A. Roncalli
I. Role of Animal Droppings in the Ecosystem

II. Effect of Ivermectin on Larvae of Dung-Breeding Diptera
III. Effect of Ivermectin and Abamectin on Afro-Asian Dung Beetles
(Onthophagus gazella and Onthophagus binodis)
A. Onthophagus gazella
B. Onthophagus binodis
IV. Effect of Ivermectin on Other Dung Insects
V. Effect of Different Ivermectin Formulations on Degradation of Cattle
Fecal Pats
VI. Conclusions
I. Role of Animal Droppings in the Ecosystem

Animal droppings have played an important role in our ecosystem for
millions of years. In fact, they help preserve it by returning valuable
nutrients to the soil as they decompose.
Dung fauna encompass a range of inhabitants, including insects,
nematodes, bacteria, yeast, and fungi. The number of invertebrates and
micro-organisms involved in the decomposition of dung can be very high.
Putnam (1983), for instance, reported that there may be "as many as 1011
bacteria per ml of sheep dung and 106 fungal mycelia and 10 10 actino-
mycetals per mt." In cattle dung there may be as many as 2,000,000 yeast
cells per gram. It has been suggested that in 10 grams (dry weight) of
decaying cow dung, there can be as many as 500 nematodes, 100
collembolans, 15 mites,S enchytraids, and several hundred insect larvae
(chiefly beetle and dung-fly). The assemblage of inhabitants and their role
in the disintegration of dung varies, however, as may their cumulative
effects. This depends on many factors, including seasonality, micro-
174 R.A. Roncalli
climate, ecosystem, and artificial control of the dung community. Physi-

cal factors such as rain, snow, and frost (White 1960; Bastiman 1970),
trampling or disturbance by animals-especially birds-are of great
importance in the breakdown of dung.
The role played by the different members of the dung community in the
degradation process will vary according to the habitat. In fact, while
dung-burying beetles playa major role in the dispersal of dung in pasture
in some tropical areas, this does not appear to be the case in temperate
areas, where the number of dung-burying beetles is much lower. The
degradation of fecal material in temperate areas appears to be primarily a
decomposition process. Earthworms have also been reported as major
organisms responsible for dung degradation (White 1960).
Early in the 1960s, the Commonwealth Scientific and Industrial Re-
search Organization (CSIRO) in Australia and the United States Depart-
ment of Agriculture (USDA) encouraged the introduction of Afro-
Asian dung-burying beetles into each country's ecosystem. One of the
main purposes of this plan was to reduce the prevalence of dung-breeding
flies-such as Australia's bush fly and the hom fly-by the use of the
beetles to degrade the dung. Over the last 20 years, 41 dung beetle species
have been released in Australia; of these, 21 are known to have become
established (Ridsdill-Smith in press). Six species of dung beetles were
released in the United States (Fincher 1986), and most of these species are
now established. It appears that in both Australia (Doube 1986) and the
United States (Legner and Warkentin 1983) the results of the projects to
control buffalo and hom flies were disappointing. As a result, the official
USDA program was terminated in October 1987.
II. Effect of I vermectin on Larvae of

Dung-Breeding Diptera
A number of Muscidae, such as the hom fly, the face fly, and the stable
fly, are often associated with the dung of herbivorous mammals. The
hom fly, Haematobia irritans, and its subspecies, H. irritans exigua
(buffalo fly) and H. thirouxi potans, are biting species that annoy cattle
and cause considerable losses (i.e., reduction in average daily gain).
Drummond (1987) estimated that in the United States alone the hom fly
causes annual losses to the cattle industry of $679 million. Roberts (1952)
has postulated that cattle having to contend with large numbers of buffalo
flies-one of Australia's major pests-will be less productive (i.e.,
decreased weight gain and milk production). Moreover, Drummond
(1987) reported that the face fly (Musca autumnalis) generates annual
losses of $59 million to the U.S. cattle industry (feedlot and range cattle);
those losses appear to be the result of an increase in the incidence of
12. Effects on Cattle Dung Fauna 175
pinkeye, with a resulting effect on weaning and yearling weight. Accord-

ing to the same author the annual loss in feedlot cattle and dairy cows
caused by the stable fly (Stomoxys calcitrans) in the United States is $418
million.
During the past 9 years a number of trials have been conducted to
assess the effect of ivermectin and abamectin on larvae of cattle
dung-breeding Diptera. In laboratory tests, Schmidt and Kunz (1980)
determined the effect of ivermectin on stable flies (S. calcitrans) and hom
flies (H. irritans). Ivermectin prevented the flies' development from
larvae to adults. The effective dose of ivermectin was 0.048 ppm (LC50)
and 0.186 ppm (LC90) for stable flies, and 0.003 ppm (LC50) and 0.006 ppm
(LC90) for hom flies.
Meyer, Simco, and Lancaster (1980) reported that ivermectin given to
cattle subcutaneously at 200 ILg/kg had an effect on larvae of the face fly
(M. autumnalis) for 14 days after treatment. Miller and colleagues (1981)
showed that a single injection to cattle of 200 ILg ivermectin/kg controlled
larvae of hom flies (H. irrit9ns) for up to 4 weeks after treatment. When
ivermectin was given daily at a dose of 1 ILg/kg orally, it killed all hom fly
larvae in the manure. An oral dose level of 5 ILg/kg killed all of the face
flies, about 60% of the stable flies (S. calcitrans), and 90% of the house
flies (M. domestica) in the manure. Ivermectin, given by subcutaneous
injection at a dose of 5 ILg/kg/day, prevented development of hom flies in
the manure. On the other hand, injections of 10 ILg/kg daily did not
control stable fly larvae.
The results of a trial conducted by Schmidt (1983) at the USDA
Livestock Laboratory (Kerrville, TX) showed that intramuscular treat-
ment of cattle with ivermectin at a dose rate of 200 ILg/kg prevented
emergence of adult hom flies (H. irritans) from manure for up to 28 days
after treatment. Some H. irritans developed into adults from manure
collected at 35 days after treatment and then aged for 14 days. According
to Ridsdill-Smith (1988), a single subcutaneous injection of abamectin at a
dose rate of 200 ILg/kg killed bush fly (M. vetustissima) larvae for 2 weeks
after injection.
In a trial carried out in Australia (Burrows and Picton, unpublished
results), H. irritans exigua (buffalo fly) were cultured from feces collected
from 5 cattle injected subcutaneously with 200 ILg abamectin/kg body
weight. No flies emerged from the feces of the treated cattle for up to 21
days following treatment.
The results of 2 trials conducted in South Africa (D' Assonville and Soli,
unpublished results; D' Assonville, unpublished results) showed that
larval stages of the face fly M. xanthomelas, one of the vectors for
Para./ilaria bovicola, and H. thirouxi potans, the hom fly of central, east,
and southern Africa, were killed in feces of cattle treated with sustained-
release boluses delivering 40 ILg ivermectin/kg/day.
176 R.A. Roncalli
III. Effect of Ivermectin and Abamectin on

Afro-Asian Dung Beetles
(Onthophagus gazella and Onthophagus binodis)
A. ONTHOPHAGUS GAZELLA
A series of trials (see Tables 12.1 and 12.2) was conducted in Australia to
examine the effects of ivermectin and abamectin fecal residues on larval
and adult stages of Onthophagus gazella. Cattle were treated once,
subcutaneously, at 200 or 300 I-'g/kg body weight. The trials were carried
out in sequential order, to assess the effect of the drugs' residues in feces
at different time intervals following the placement of the beetles.
In these trials feces from treated and untreated animals were placed in
buckets, along with pairs of O. gazella, and observed for the beetles'
viability, production of brood balls, and emergence of new beetle
populations for up to 35 days after treatment. The feces were not exposed
to sunlight, eliminating any effects resulting from degradation of iver-
mectin and abamectin (which may occur under field conditions). The
results of these trials indicate that neither ivermectin nor abamectin at the
above doses had any effect on adult O. gazella. Both compounds did,
however, prevent larval development: for ivermectin, up to 21 days
following treatment, for abamectin, up to 28 days.
TABLE 12.1. Ivermectin vs dung beetles (Onthophagus gazella): ivermectin in

cattle trials (Australia).
Results
Dose
Investigator(s) Drug Route (I'g/kg) Adults Larvae
J. Picton (1980) ivermectin S.C." 300 No effect Failed to
unpublished develop
44-52 hours
after
treatment
J. Picton & ivermectin S.C. 300 No effect Failed to
R. W. Butler develop 1,3,
unpublished and 5 days
after
treatment
J. Picton & ivermectin S.C. 300 No effect Failed to
R. W. Butler develop 7 and
unpublished 14 days after
treatment
No effect at 21,
28 and 35
days after
treatment
"S.C. = subcutaneous.
TABLE 12.2. Ivermectin vs dung beetles (Onthophagus gazella): ivermectin and

abamectin in cattle trials (Australia).
Results
Dose
Investigator(s) Drug Route (lLg/kg) Adults Larvae
J. Picton & ivermectin S.C." 200 Not Failed to
R. O. determined develop 7 and
Burrows 10 days after
treatment
abamectin S.C. 200 Not Failed to
determined develop 7, 10
and 14 days
after
treatment
R.O.Burrows abamectin S.C. 200 No effect Failed to
& J. Picton develop 21
days after
treatment
Unaffected 28
days after
treatment
" S.C. = subcutaneous.
In Texas, O. gazella are active in the early fall. During September-

October 1980, Schmidt (1983) observed that the presence and burrows of
adult beetles were approximately the same in manure from untreated
cattle as in manure of cattle treated with ivermectin.
B. ONTHOPHAGUS BINODIS
Ridsdill-Smith (1988) conducted a trial in southwestern Australia to assess

the effect of abamectin, given to cattle subcutaneously at the prescribed
dose, on the development of Onthophagus binodis in manure. Dung was
collected 1, 2, 4, 8, and 11 weeks after treatment with abamectin and
transferred to plastic boxes. Pairs of O. binodis were placed in dung of
untreated cattle and abamectin-treated cattle. While the survival of adult
dung beetles was not influenced by the treatment, the production of brood
balls was decreased for 2 weeks in dung from cattle treated with
abamectin. The survival of immature dung beetles was affected for 4
weeks. In the same trial, levamisole given orally did not seem to influence
the beetles. Ridsdill-Smith suggested replacing the treatment with
abamectin with another anthelmintic from September to November,
which is the main breeding period of beetles in southwestern Australia.
The use of abamectin in other months should not influence dung beetle
breeding.
The abamectins are not the only compounds to affect the larval stages
of the dung beetle. A number of published reports have indicated that
178 R.A. Roncalli
various anthelmintics and insecticides (phenothiazine, dichlorvos, cou-

maphos, etc.) affect dung beetles, while thiabendazole and levamisole
have little effect. Also of interest is the fact that superphosphate, which is
used in many sites to improve pastures, decreased brood ball production
and survival of the dung beetle O. granulatus (Holm and Wallace 1987).
IV. Effect of I vermectin on Other Dung Insects

Schmidt (1983), in his trial conducted in Texas, also evaluated the effect
of ivermectin (200 ltg/kg intramuscularly) on native insects. Emergence
of some insects-such as sphaerocerids, sepsids, and 2 species of
Gymnodia-from feces of treated animals was severely reduced. In
addition, more parasitic wasps were present in the control manure. On the
other hand, about the same number of gnats and of staphylinids emerged
from both treated and untreated feces. Mites were plentiful in both types
of feces.
Wall and Strong (1987), in a trial conducted at Bristol University
(U.K.), collected dung from cattle treated with an experimental
sustained-release bolus delivering 40 ltg/kg/day as well as from untreated
cattle and observed the dung for 100 days. In this trial the dung from
treated animals contained "few, or in some cases no Coleoptera or
Diptera." Aphodius spp. was the most abundant (89.4%) ofthe Coleoptera
found in the untreated dung. The authors expressed concern about the
potential adverse impact ofivermectin usage on the condition of pastures.
In their study on the effect of treating cattle with a sustained-release
ivermectin device (above), Wall and Strong recorded the number of
earthworms appearing in dung pats, and underlying soil, at various
intervals after the initiation of treatment. The resulting data did not show
any clear-cut effect of ivermectin on the number or time of appearance of
earthworms.
V. Effect of Different Ivermectin Formulations on

Degradation of Cattle Fecal Pats
Schmidt (1983) observed the aging of manure from animals treated with an
intramuscular injection of ivermectin at a dose of 200 ltg/kg. Several pats
of manure collected from the pens of treated and untreated cattle were
molded in pans and placed on the surface of a stony-shaped area inside a
wire enclosure. At the end of the 29-day test, the consistency and form of
the manure from the treated and untreated groups appeared to be the
same. Schmidt stated that "since manure from treated cattle also
appeared to disintegrate at the same rate as manure of untreated animals,
on the basis of the results reported here, the manure of treated animals
should pose no greater threat of pollution than untreated manure in

grazing areas."
Wall and Strong (1987) reported that feces of untreated cattle were well
decomposed 40 days after deposition, and for the most part had
disappeared after 100 days. On the other hand, feces of cattle treated with
the ivermectin bolus releasing 40 ILg/kg/day were solid and intact after
100 days. Strong and Brown (1987) inferred that this was due to the fact
that Coleoptera were almost absent from treated feces of any age. In this
trial, feces were collected from housed cattle, mixed, and placed into a
circular mold of corrugated cardboard and transferred to a 5m x 10m
enclosure in a pasture.
McKeand, Bairden, and Ibarra-Silva (1988) reported on the effect of
ivermectin given as a pour-on formulation at a rate of 500 ILg/kg to calves
at 3, 8, and 13 weeks after turnout to pasture in western Scotland. One
week after the third treatment fecal pats were selected, labeled, and left in
situ. After 9 weeks there was no marked difference in the rates of
breakdown of the feces between the dung pats of the ivermectin-treated
calves and those of the untreated controls. Although the fauna present in
the cattle dung were not monitored, these authors suggest that if the fauna
involved in the decay of dung were adversely affected by the ivermectin,
these species would have rapidly repopulated the pats.
The results of the 3 trials described appear to indicate that there is a
difference with regard to ivermectin formulations and degradation of
cattle fecal pats. No difference was, in fact, detected between the
injectable and the pour-on formulations versus the untreated controls.
There was a difference, however, when an experimental formulation
(sustained-release bolus) was used. Two of the trials were conducted
under conditions which may not have been an accurate representation of
a true field situation. Future trials should be run under field conditions as
was done in the study by McKeand and colleagues (1988), i.e., evaluation
of dung pats left in situ. In fact, there are several factors to consider in
running these trials. Under natural circumstances, dung beetles (which
are mostly flying insects) regularly migrate to and from dung pats. Also,
disintegration of fecal pats will be accelerated by trampling by animals,
the weathering effects of wind and rain, or other environmental factors.
VI. Conclusions
Ivermectin and abamectin have an effect on some important dung fauna,
such as the larval stages of Diptera and of some dung beetles. The effect
on biting flies (i.e., the horn fly) could be considered beneficial, but its
practical impact under field conditions remains to be determined. Iver-
mectin and abamectin appear to affect dung beetles mainly in the larval
stages, and this could be considered harmful. The importance of these
180 R.A. Roncalli
findings to those situations in which beetles play a major role in dung

degradation is still under evaluation.
Although the population dynamics of coprophagous insects is not well
known and varies greatly from area to area, the use of ivermectin and
abamectin is not expected to pose a threat to dung beetle populations.
Localized use of the abamectins on particular herds of cattle at infrequent
intervals is unlikely to significantly affect overall beetle population sizes.
In fact, only a portion of a single generation of dung beetles will be
affected, as suppression of larval development will occur for a limited
period of time after treatment of cattle. Not all cattle will be treated in the
same herd or in neighboring herds at the same time, and feces from
untreated cattle will always be available to sustain beetle populations.
Another factor to consider is the effect of ivermectin on earthworms,
which play an important role in the degradation of dung in numerous
countries (e.g., the United Kingdom, New Zealand). The estimated LC so
of ivermectin for earthworms under controlled laboratory conditions was
300 mg/kg (ppm) of soil, a value which is much more than can be
encountered in the environment; the no-effect level of ivermectin was
12 ppm (Halley, Nessel, and Lu (1988). The final word on these issues will
be obtained only through a series of controlled field studies in different
areas of the world. A number of these trials have already been initiated,
and the results will provide valuable information on this area of research,
which is still not well understood.
REFERENCES
Bastiman B (1970) Problems of pasture contamination and herbage rejection under
intensive grazing. In 11th Progress Report of Experimental Husbandry and
Experimental Horticultural Stations, NAAS, pp 62-69
Doube BM (1986) Biological control of the buffalo fly in Australia: the potential of
the Southern Africa dung fauna. Misc. Pubs. Entomol. Soc. of Amer. (No.
61):16-34
Drummond RO (1987) Economic aspects of ectoparasites of cattle in North
America. In The Economic Impact of Parasitism in Cattle, MSD AGVET
Symposium, Montreal, Quebec, Canada, pp 9-24
Fincher GT (1986) Importation, colonization, and release of dung-burying scar-
abs. Misc. Pubs. Entomol. Soc. of Amer. (No 61):69-76
Halley BA, Nessel RJ, Lu AYH (1988) Environmental aspects of ivermectin
usage in livestock: general considerations. (Chapter 11)
Holm E, Wallace MMH (1987) The influence of superphosphate on the establish-
ment of introduced dung beetles in south-eastern Australia. J. Aust. Inst. Agric.
Sci. 5:202-203
Legner EF, Warkentin RW (1983) Questions concerning the dynamics of Ontho-
phagus gazella (Coleoptera:Scarabaeidae) with symbovine flies in the lower
Colorado desert of California. Proc. 51st Ann. Con! Calif. Mosquito & Vector
Contr. Assoc. pp 99-101
McKeand J, Bairden K, Ibarra-Silva AM (1988) The degradation of bovine fecal

pats containing ivermectin. Vet. Rec. 122:587-588
Meyer JA, Simco JS, Lancaster JL (1980) Control of face fly larval development
with ivermectin, MK-933. Southwest. Entomol. 5:207-209
Miller JA, Kunz SE, Oehler DD, Miller RW (1981) Larvicidal activity of Merck
MK-933, an avermectin, against the horn fly, stable fly, face fly and house fly.
J. Econ. Entomol. 74:608-611
Putnam RJ (1983) Carrion and dung: The decomposition of animal wastes. In The
Institute of Biology's Studies in Biology No. 156, Edward Arnold Publishers,
London, pp 1-62
Ridsdill-Smith TJ (in press) Some effects ofivermectin on non-target organisms in
cattle dung. Proc. 5th Austral. Conf. Grassland Invert. Ecol.
Ridsdill-Smith TJ (1988) Survival and reproduction of Musca vetustissima Walker
(Diptera:Muscidae) and a scarabaeine dung beetle in dung of cattle treated with
avermectin Bl. J. Aust. Entomol. Soc. 27:175-178
Roberts FHS (1952) Insects Affecting Livestock. Angus and Robertson, Sydney,
Australia, pp 117-124
Schmidt CD (1983) Activity of an avermectin against selected insects in aging
manure. Envir. Entomol. 12:455-457
Schmidt CD, Kunz SE (1980) Testing immature laboratory-reared stable flies and
horn flies for susceptibility to insecticides. J. Econ. Entomol. 73:702-703
Strong L, Brown TA (1987) Avermectins in insect control and biology: a review.
Bull. Entomol. Res. 77:357-389
Wall R, Strong L (1987) Environmental consequences of treating cattle with the
antiparasitic drug ivermectin. Nature 327:418-421
White E (1960) The distribution and subsequent disappearance of sheep dung on
Pennine moorland. J. Anim. Ecol. 29:243-250
CHAPTER 13
Abamectin Use in Crop Protection
P.G. Wislocki, L.S. Grosso, and R.A. Dybas
I. Introduction
II. Environmental Fate Of Abamectin
A. Photolysis
1. Water
2. Soil
3. Thin Film
B. Hydrolysis
C. Soil Metabolism
D. Solubility
E. Soil Binding
F. Plant Uptake
G. Bioconcentration in Fish
H. Fate in the Aquatic Environment
I. Fate in the Terrestrial Environment
J. Conclusions
III. Effects Of Abamectin On Non-Target Organisms
A. Aquatic Organisms
1. Invertebrates: Acute Toxicity
2. Fish: Acute Toxicity
3. Chronic Toxicity
4. Plants
5. Conclusions
B. Terrestrial Organisms
1. Soil micro-and macroorganisms
2. Bees
3. Birds
4. Mammals
5. Conclusions
IV. Risk Assessment
A. Aquatic organisms
B. Terrestrial organisms
V. Conclusions
13. Environmental Aspects of Use in Crop Protection 183
I. Introduction
The use of any chemical as a pesticide results in the exposure of the
environment to the chemical. The extent of this exposure depends on the
way the chemical is applied-pattern, rate, and frequency-as well as its
persistence in the environment. Humans can control all these parameters
save the last. A chemical's persistence is determined by its physico-
chemical properties: half-life in soil and water, photolytic stability, and
soil-binding ability. These same factors also affect the chemical's bio-
availability to living organisms.
In Chapter 20 Dr. Dybas discusses the agricultural uses of abamectin. It
will be used at a maximum rate of 0.025 lb active ingredient/acre
(28 g/hectare) 3 times/season, or at a maximum rate of 0.02 lb active
ingredient/acre (22 g/hectare) up to 10 times/season. Abamectin is
also being used by homeowners to control fire ants at a use rate of only
0.00011 lb active ingredient/acre (123 mg/hectare). The compound is
formulated in a bait used at a rate of more than a million particles/acre.
The bait particles are rapidly scavenged by foraging fire ants. This use will
not lead to significant exposure to the environment and therefore is not a
topic of discussion in this chapter.
The risk to nontarget organisms associated with the use of a pesticide
depends upon both its toxicity to nontarget organisms and their exposure
to the pesticide. As a pesticide, abamectin is very toxic to target pests. It
is also toxic to certain nontarget organisms; however, the pesticide's
properties and use patterns reduce the exposure of nontarget organisms.
In this regard abamectin has many positive attributes. It degrades rapidly
and its physiocochemical properties prevent it from bioconcentrating or
bioaccumulating in the environment. An analysis of abamectin's physio-
chemical properties and their effect on its environmental fate per se, as
well as the chemical's toxicity and risk to nontarget organisms, follows.
The nomenclature used in this chapter should be clarified. The product
that is being developed for agricultural use is designated abamectin. It
refers to a mixture of not less than 80% avermectin B la and not more than
20% avermectin Bib. (Figure 13.1). Certain studies presented here used
radioactively labeled avermectin Bla ; they are designated accordingly.
II. Environmental Fate of Abamectin

The studies on the environmental fate of a compound describe not only
how a compound is degraded in the environment but also the extent to
which it is available to biological organisms. This section addresses
abamectin's rate of degradation, its bioavailability, and its uptake into
biological systems.
184 P.G. Wislocki, L.S. Grosso, and R.A. Dybas
4' OCH 3
CH~ oJ-: H
O~:CH3 O. H
HO~ ~ "1314~
CH 30 12
CH3'
a-Component R = C 2 Hs > 80%

b-Component R = CH 3 < 20%
ABAMECTIN = AVERMECTIN B1
FIGURE 13.1. The structure of abamectin.
A. PHOTOLYSIS
Abamectin undergoes rapid photolysis under a number of different

conditions. Photolysis occurs when abamectin is exposed to light in
water, on soil particles, or as a thin film on either inert (glass) or biological
surfaces, such as leaves. The degradation occurs by oxidative and
photooxidative mechanisms.
1. Water
In a water photolysis experiment avermectin B1a was exposed to sunlight
and its degradation was followed by HPLC. Under these conditions
avermectin B 1a degraded with a half-life of less than 12 hours. Products
formed during the course of photolysis included a less polar degradate
identified as a photo isomer, the delta-8,9 Z isomer of avermectin B 1a , and
2 fractions more polar than avermectin B 1a, designated the moderately
polar and polar fractions. During the course of this degradation the
characteristic UV absorption at 245 nm is lost. (Absorption is due to the
presence of a diene moiety at carbons 8 through 11 in the molecule.) This
loss of absorbance also occurs when abamectin is photolyzed in organic
solvents (Mrozik et al. 1988). The terminal degradation products of water
photolysis were contained in the polar fraction. Its toxicity and that of the
other fractions are discussed later.
2. Soil
When soil containing avermectin B la was exposed to sunlight on glass
plates, it degraded to mUltiple products rapidly, with a half-life of 21
hours.
3. Thin Film
Abamectin is exceptionally susceptible to degradation when it is present
on a surface as a thin film. Such degradation occurs even in the absence of
light; however, it is greatly accelerated in the presence of light. Both
under laboratory light and in simulated sunlight, a thin film of abamectin
degrades with an initial half-life of 4 to 6 hours. Both avermectin Bla
and avermectin BIb degrade at the same rate (J. MacConnell and R.
Demchak, personal communication). The degradation products include
the delta-8,9 isomer of avermectin B la and a polar and moderately polar
fraction. HPLC profiles of the polar degradates generated on plants and
on glass demonstrated that these degradates were similar. The degradates
on glass contained a multitude of components, some of which had from 1
to 6 additional oxygens compared to the parent compound. Other
components had lost the features of the macrocyclic ring, including the
loss of UV absorbance at 245 nm; only the sugar portion of the molecule
was recognizable by NMR or mass spectroscopy. These data indicate that
extensive degradation occurs when abamectin is exposed to light.
B. HYDROLYSIS
The facility with which avermectin B la undergoes hydrolysis was deter-

mined at pH 5, 7, and 9 (the pH range normally found in surface and
ground water). Mter 28 days at 25°C, avermectin B la remained un-
changed, indicating that it was not readily hydrolyzed under these
conditions.
C. SOIL METABOLISM
In collaboration with Donald Bull (Bull et al. 1984), soil metabolism

studies were performed at concentrations of 0.1, 1.0, or 50 ppm in 3 soil
types: Lufkin fine sandy loam, Houston clay, and construction-grade
sand. Avermectin B la had half-lives ranging from 20 to 47 days. It
degraded to at least 13 products. The major product was identified by
NMR and mass spectroscopy as the 8a-hydroxy derivative of avermectin
Bla • This compound appears to be the result of microbial metabolism,
since sterilized soil did not metabolize abamectin. Under anaerobic
conditions the rate of metabolism decreased significantly.
D. SOLUBILITY
A major factor which determines the behavior of a chemical in the
environment is its solubility. Because abamectin is quite insoluble in
water, determining its actual solubility was very difficult. For this reason
the method of May and Wasik (1978) was used. By this dynamic method it
was determined that the solubility of avermectin B1a was 7.8 ppb in water.
E. SOIL BINDING
The bioavailability of a compound to biological systems in the environ-

ment is strongly influenced by its ability to bind to soil. A compound that
binds tightly and is not mobile will result in a much lower level of
exposure than one which does not bind tightly. The ability of abamectin to
bind to soil was examined by 3 different methods: soil thin-layer
chromatography (TLC) , batch equilibrium binding studies, and soil
column leaching studies. In all 3 systems it was determined that avermec-
tin B1a is an immobile pesticide. When its mobility was determined by soil
TLC using 6 different soils-including Lakeland sand, which contained
only 0.1% organic matter-averrnectin B 1a did not move from the origin;
when compared with standards, it was classified as an immobile pesticide.
To determine the binding constant of abamectin to soil, batch equilibrium
studies were performed. The results of these studies indicated that
avermectin B 1a binds very tightly to soil particles in relation to the organic
matter content of the soil. The binding constant, Koc, based on the
organic content of the soils, was approximately 4000 or greater, demon-
strating the immobile nature of abamectin. In a soil column leaching
study, averrnectin B1a was applied to 38-cm-high columns of soil. Water
equivalent to 22 to 23 inches of rainfall was applied to the columns over a
28-day period beginning immediately or after 29 days of aging. Approxi-
mately 80% to 99% of the radioactivity, which included avermectin B1a
and polar degradates, remained in the top 6 cm of the column. Some
radioactivity, including a small amount of avermectin B1a and polar
degradates, did pass through the column, apparently due to channeling
effects in the column. There was no significant difference between the
aged and unaged columns, indicating that any degradation products
formed during the aging period are not more mobile than the parent
compound. These studies further demonstrate the immobile nature of
abamectin and indicate its lack of bioavailability in the environment.
F. PLANT UPTAKE
The ability of plants to take up abamectin residue from the soil is a

measure of these residues bioavailability. This is a pragmatic issue
because crops are often rotated into fields where pesticides have been
used to treat previous crops. A rotational crop study (Moye et al. 1987)
was peJformed using 3 different soil types that had organic matter
contents from 0.14% to 38.3%. Although total applications of up to
0.36 lbs active ingredient per acre were used, there was no biocon-
centration of residues. Indeed, the highest average level of total radioac-
tive residues of avermectin B 1a found in any plant part was only 11.6 ppb.
This level is quite low compared to total radioactive residues in soil of
1070 ppb. Examination of the properties of the plant residues by
extraction indicated that 95% of the residues were not avermectin
B1a-related: they could not be extracted with acetone, which removes
avermectin B 1a and its degradates. Plant metabolism studies have shown
that radioactive fragments from avermectin B1a can be incorporated into
the natural products of plants. Therefore, the radioactive residue in the
plants is due to their uptake of small fragments of the avermectin B1a
molecule and the incorporation of 1- and 2-carbon fragments into natural
plant constituents.
G. BIOCONCENTRATION IN FISH
The fact that chlorinated pesticides concentrate and accumulate in fish

and organisms higher on the food chain led to a general concern about the
ability of pesticides to bioconcentrate and bioaccumulate in the environ-
ment. Not only are the possible ecological effects of concern, but so also
is the possible hazard to humans from consuming high levels of biocon-
centrated residues. Because abamectin is very insoluble in water and has
an octanol water partition coefficient of 9900, it seemed possible that it
could bioconcentrate. When its bioconcentration factor in bluegill sunfish
was determined, however, abamectin was shown not to bioconcentrate.
The study was peJformed under dynamic (flow-through) conditions by
ABC Laboratories (Columbia, MO). Bluegill sunfish were exposed to
avermectin B1a at a concentration of 0.099 ppb for 28 days. The residues
in the fish reached a steady-state concentration at 10 days. From days 28
to 42 the fish were kept in water without avermectin B1a for the
depuration phase of the experiment. The concentration of residues in
the whole fish was 6.8 ppb on day 28, but this decreased rapidly to only
0.32 ppb on day 42. Using the uptake and depuration rates, it was
calculated that abamectin has a steady-state bioconcentration factor of
only 52. Therefore, abamectin neither accumulates nor persists in fish.
This is probably due to the large size of the molecule. Avermectin B1a has
a molecular weight of 872. Its size prevents it from being a truly lipophilic
compound. Brooke, Dobbs, and Williams (1986) have reported this
phenomenon with other high molecular-weight compounds. These results
are consistent with animal metabolism studies that have shown that the
compound neither accumulates nor persists in rats or goats (Maynard et
al., unpublished data). Therefore, abamectin will neither bioconcentrate

in individual organisms nor bioaccumulate in the food chain.
H. FATE IN THE AQUATIC ENVIRONMENT
To assess abamectin's fate in the aquatic environment a simulated field

study was performed in collaboration with Wildlife International (Easton,
MD). Abamectin was applied by 1 of 2 methods to tanks of water 2 feet
deep containing a 3-inch layer of pond sediment. The water was either
sprayed directly with the diluted emulsifiable concentrate (EC) formula-
tion to simulate the drift route of exposure, or soil was sprayed directly
with the diluted formulation and added to the water immediately or 4 days
later to simulate runoff. Samples of water and sediment were taken at
various times for up to 52 days after treatment and analyzed for
abamectin.
The abamectin in tanks treated directly with spray (drift simulation)
was rapidly distributed between the sediment and the water. Between the
zero-time sampling and the 19-hour sampling, the levels of abamectin in
the water decreased 4- to IS-fold. After the initial distribution and rapid
degradation the level of abamectin decreased with a half-life of about 4
days. The half-life of abamectin in sediment was approximately 2 to 4
weeks. If one assumes an application rate of 0.025 lb active ingredient/
acre (the highest maximum application rate for any use) and that 5% of
the use rate impacts on nearby surface water, then the level of abamectin
in the water the day after the application was only 26 ppt. By 2 days the
level decreased to 8 ppt. This decreased to less than 1.25 ppt (the limit of
detection) by 15 days after treatment. The levels of abamectin in the water
over time are shown graphically in Figure 13.2.
In the simulated runoff study, soil was added to the water immediately
after treatment with abamectin or after 4 days of exposure to the
elements. Water and sediment from tanks treated with pre-aged soil did
not contain detectable levels of abamectin. Sediment and water from the
tanks to which soil was added immediately did contain abamectin
residues. Since abamectin binds very strongly to soil, one can assume a
runoff of 0.1 % of the applied abamectin. Also, in common agricultural
practice an area tenfold greater than the pond size is needed to sustain a
viable farm pond. Using these assumptions, the levels of residues in the
water were calculated to be in the range of 1 to 3 ppt for 15 days after the
addition of soil. The residues in the water dropped to 0.28 ppt by day 31
and to less than 0.03 ppt by day 51. Residues ofabamectin in the sediment
trended downward during the 51-day period and fell within the range of
100 to 300 ppt. The levels of abamectin in the water over time are shown
graphically in Figure 13.3.
This study also examined loss of abamectin on the soil surface. A layer
3.0
2.5
i='
0-
~
2.0
(.)
ABAMECTIN
z 1.5
0
(.)
(!) 1.0
0
....J
.5
0
0 1 2 4 8 15
TIME (DAYS)
FIGURE 13.2. Decrease in the concentration of abamectin (ppt) over time in the
water of tanks subjected to drift simulation.
of soil that had been sprayed with abamectin was analyzed at 0, 1,2, and 4
days. Abamectin degraded rapidly, with a half-life of 8 hours (Figure
13.4).
This simulated field study demonstrates that abamectin degrades
readily both on soil and in the aquatic environment.
2.0
1.6
1.2
i='
0- .8
0-
(.)
.4
z 0
0
(.) -.4
(!)
0 -.8
....J
-1.2
-1.6
-2.0
0124 8 15 31 52
TIME (DAYS)
FIGURE 13.3. Decrease in the concentration of abamectin (ppt) over time in the
water of tanks subjected to runoff simulation.
2.0
1.6
CO 1.2
a..
-z
a..
() .8
0 .4
()
C!J
0
9
-.4
-.8
o 2 4
TIME (DAYS)
FIGURE 13.4. Decrease in the concentration of abamectin (ppt) over time in soil
sprayed directly with abamectin.
I. FATE IN THE TERRESTRIAL ENVIRONMENT
Soil dissipation studies and crop residue trials were petformed to deter-
mine the residues of abamectin on soil and crops (Tway et aI., unpub-
lished data). Because these studies are actual field studies, they most
accurately depict the residue levels to which terrestrial organisms will
actually be exposed. In the soil dissipation studies the initial levels of
abamectin in the top 2 inches of soil were from 10 to 20 ppb. This is V. to ~
of the theoretical level of 42 ppb expected to be found at zero time when
0.025 lb active ingredient/acre is applied to soil. Minimal residues were
found at the 2-to-4-inch level, and no residues were found at a soil depth
of 4 to 6 inches. Residues depleted rapidly and did not accumulate, even
after 12 weekly applications.
Residue data on crops were determined to establish tolerances in raw
agricultural commodities and food and feed additives. These residue
values can also be used to estimate the exposure of terrestrial organisms
to abamectin. Residue data are available for 4 crops: cotton foliage, celery
plants, citrus fruits, and tomatoes. The cotton foliage trials were per-
formed in South Africa at application rates different from those in the
United States when calculated at the U.S. rate of 0.02 lb active in-
gredient/acre, the initial residue values were approximately 1 ppm.
Within 3 days the residue values were approximately 5% of the initial
values. In 20 residue trials on celery plants, the plants were treated at 10
weekly intervals at a rate of 0.02 lb active ingredient/acre. At zero time
after the last of 10 weekly applications the average residue value in the
whole celery plants was 170 ppb. By 7 days after the application the
residues had declined to 9.5 ppb, which is less than 6% of the initial
residue levels. These data demonstrate that even repeated applications of
abamectin do not result in accumulated residues-they deplete rapidly.
Residue results from cotton foliage and celery plants can be used as
surrogate data for grasses and leaves and leafy crops, respectively, in risk
assessments. Due to their small surface-to-mass ratios, citrus fruits and
tomatoes have lower initial residues of abamectin. The zero-time residue
levels in citrus fruits treated with 0.025 lb active ingredient/acre average
from 14 to 45 ppb. The residues declined to approximately 20% of initial
values by 7 days. The initial residue level of abamectin on tomatoes was
less than 17 ppb. By 3 days after the application the residue levels were
less than 5 ppb and usually less than 2 ppb, the limit of detection of the
analytical method. These data on citrus fruits and tomatoes can be used as
surrogate data for other fruits. Metabolism studies on citrus fruits (Iwata
et al. 1985; Maynard et al. 1989), cotton (Bull et al. 1984), and celery
(Moye et al., unpublished data) have also demonstrated the rapid
degradation of avermectin B'a on plants. These studies have shown that
the degradation products formed are the delta 8,9 isomer of avermectin
B'a, the moderately polar fraction, and the polar fraction. These fractions
are made up of multiple components. Because the polar fraction is the
major degradate, and because it is similar to the polar fraction generated
on glass, the glass-derived material was used for toxicology studies. The
results of these studies is discussed below. The delta 8,9 isomer is present
on plants at 10% to 20% of the total residue of it and its parent compound.
It also degrades at the same rate as its parent compound. Because the
delta 8,9 isomer is closely related to its parent, it was also tested for
toxicity in mice.
The soil dissipation and residue data indicate that, under field use
conditions, the residue levels of abamectin do not accumulate, even when
multiple applications are made; moreover, the residue levels deplete
rapidly, greatly limiting exposure.
J. CONCLUSIONS
The environmental fate data demonstrate that abamectin photodegrades

rapidly in the environment and is metabolized in the soil. Its solubility in
water is limited, and it binds to soil very tightly. It does not biocon-
centrate in fish and is not taken up from soil into plants. Both aquatic and
terrestrial studies confirm the rapid degradation of abamectin in the
environment, and its concomitant lack of accumulation and persistence.
III. Effects of Abamectin on Nontarget Organisms

Because it is impossible to target a chemical pesticide solely to the pest
causing crop damage, it is necessary to determine the effects of pesticide
use on nontarget organisms. The safety of pesticides to nontarget
organisms can be assessed by a comparison of the information derived

from a series of tests in various indicator organisms and actual residue
data. The toxicity data presented here are unpublished (Wislocki et at.).
A. AQUATIC ORGANISMS
Although abamectin will not be applied directly to surface water, some

exposure could occur either via drift during the course of application or
through runoff from treated fields. Therefore, it is important to assess
abamectin's toxicity to aquatic organisms that may inhabit waters adja-
cent to treated areas.
1. Invertebrates: Acute Toxicity

Invertebrates vary widely in their sensitivity to abamectin. (Studies were
performed by Springborn Life Sciences, Wareham, MA.) As seen in
Table 13.1, mysid shrimp are the most sensitive invertebrate. The 96-hour
LC50 for mysid shrimp, an estuarine organism, was 0.022 ppb. Pink
shrimp, however, had a 96-hour LC50 of 1.6 ppb, 2 orders of magnitude
higher than for the mysid shrimp. Daphnia, a freshwater invertebrate,
had a 48-hour LC50 of 340 ppt. Both blue crabs and Eastern oysters
were much less sensitive to the toxic effects of abamectin. They had
96-hour LC50 values of 153 and 430 ppb, respectively. The water
photolysis products of avermectin B1a and the major soil metabolite.
8a-hydroxyavermectin B1a were also tested for their acute toxicity to
Daphnia by ABC Laboratories (Columbia, MO). These degradation
products were much less toxic than the parent compound (Table 13.2).
The polar fraction-which is the terminal degradation product of the
photolysis of abamectin in water-had an LC50 of greater than 100 ppb,
more than 160 times that of avermectin B1a , which had an LC50 of
0.63 ppb. The moderately polar fraction (which is formed transiently) was
10 times less toxic than its parent compound. The delta-8,9 isomer and the
TABLE 13.1. Acute toxicity of abamectin to aquatic

invertebrates.
Invertebrate LC so (ppb)a
Daphnia (Daphnia magna) 0.34
Pink Shrimp (Panaeus duorarum) 1.6
Mysid shrimp (Mysidopsis bahia) 0.022
Eastern oysters (Crassostrea virginica)b 430
Blue crab (Callinectes sapidus) 153
a The LCso values are after 96 hours of exposure, except for

Daphnia, which represents 48 hours of exposure.
b Exposure occurred at the embryo-larval stage of the life
cycle
TABLE 13.2. Acute toxicity of avermectin B1a degradates to

Daphnia.
Degradatesa LCso at 4S hours (Ppb)
Polar >100
Moderately polar 6.S
delta S,9-isomer of avermectin B 1a 27.2
Sa-hydroxyavermectin B 1a 26
a All degradates were isolated from a water photolysis reaction, except

Sa-hydroxyavermectin B 1.. which was made synthetically.
8a-hydroxy derivative were 40 times less toxic than avermectin B1a•

Therefore, abamectin degrades to less toxic metabolites.
2. Fish: Acute Toxicity

The acute toxicity of abamectin has been determined in 5 different species
of fish (Table 13.3). Their response to abamectin is much more uniform
than that observed for invertebrates. The LC so for rainbow trout is
3.2 ppb, while that for bluegill sunfish is 9.6 ppb. (Studies were performed
by Springborn Life Sciences, Wareham, MA.) The sheepshead minnow,
an estuarine fish tested for toxicity, had an LC so of 15 ppb. The channel
catfish and carp had higher LC so values, of 24 and 42 ppb, respectively,
when determined by ABC Laboratories (Columbia, MO). There was only
a 1 order of magnitude difference among the fish species examined, while
there were 4 orders of magnitude difference among the invertebrate
species examined.
3. Chronic Toxicity
The results of the environmental fate studies indicate that chronic
exposure to abamectin should be minimal and without effects. Its rapid
photolytic degradation and strong binding to sediment preclude its being
bioavailable to aquatic organisms in a chronic manner. Nonetheless, 3
chronic exposure studies have been performed (Table 13.4). Full life-
cycle studies have been done with daphnia and mysid shrimp. (Studies
were performed by Springborn Life Sciences, Wareham, MA.) An early
TABLE 13.3. Acute toxicity of abamectin to fish.

Fish LC so at 96 hours of exposure (Ppb)
Rainbow trout 3.2
Bluegill sunfish 9.6
Sheepshead minnow 15
Channel catfish 24
Carp 42
TABLE 13.4. Chronic toxicity of abamectin to aquatic organisms.

Organism MATC (ppb)a ACRb
Daphnia (Daphnia magna), life cycle 0.03-0.09 6.5
Mysid (Mysidopsis bahia), life cycle 0.0035-0.0093 3.8
Rainbow trout, early life stage 0.52-0.95 4.6
a The maximum acceptable toxicant concentration (MATC) is expressed as
a value between the highest no-effect level (NOEL) and the lowest effect
level (LEL).
b The acute-to-chronic ratio (ACR) is obtained by dividing the LC so by the
MATC. The MATC for this calculation was determined by calculating the
geometric mean of the NOEL and LEL.
life-stage study was performed with rainbow trout by ABC Laboratories

(Columbia, MO). In all 3 cases abamectin had no chronic effects. The
difference between the LC so determined in each species and the no-effect
level (NOEL) observed in the chronic study was less than tenfold. The
NOELs observed were 30,3.5, and 520 ppt for Daphnia, mysid shrimp,
and rainbow trout, respectively. A common way of expressing the
no-effect level in a chronic study is to determine the maximum acceptable
toxicant concentration (MATC). This value is the geometric mean of the
NOEL and the lowest effect level (LEL). When the acute-to-chronic ratio
(ACR), the ratio of the LC so to the MATC, was determined, it was found
that the ACRs were in the range of 4 to 7. This indicates that abamectin
would not have chronic effects on these test organisms but could exhibit
acute effects.
4. Plants
Abamectin is essentially nonphytotoxic, permitting its use even on
sensitive ornamental plants. Its phytotoxicity to nontarget terrestrial
plants will be discussed in Chapter 20. However, 2 studies were per-
formed to assess its phytotoxicity to aquatic plants. (Studies were
performed by Springborn Life Sciences, Wareham, MA.) These studies
indicated that abamectin was relatively nontoxic to duckweed (Lemna
gibba) and to freshwater algae (Selenastrum capricornatum). The con-
centration of abamectin calculated to decrease frond production in
duckweed by 50% was 3.9 ppm. A concentration greater than 100 ppm
would be required to reduce the algae cell proliferation by 50%.
5. Conclusions
Aquatic organisms vary greatly in their sensitivity to abamectin. Aquatic
plants are only minimally affected by high concentrations of abamectin.
Certain invertebrates, however, are quite sensitive, with LC so values in
the ppt range, while other invertebrates are much less sensitive. Fish are
fairly uniform in their response to abamectin and are 2 to 3 orders of
magnitude less sensitive than mysid shrimp. Degradation products of

abamectin are 1 to 3 orders of magnitude less toxic than abamectin itself.
B. TERRESTRIAL ORGANISMS
Despite the diversity of terrestrial organisms, only a limited number can

be tested for the effects of pesticides. Certain organisms are chosen
because of their importance to the environment. These include nitrogen-
fixing bacteria, earthworms, bees, and birds. The effects of pesticides on
mammalian species are determined during the assessment of the toxicity
of the pesticide to humans when rodents and other mammalian species are
used for extensive testing. The results of these tests on mammalian
species are presented in Chapter 6.
1. Soil Micro- and Macroorganisms

The effect of abamectin on the bacteria, which convert ammonium ion to
nitrate, was studied in humic sandy soil and loam soil by TNO (The
Netherlands). Forty and 400 ppb abamectin had no effect on either the
decrease in the level of ammonium ion or on the increase of nitrate. At a
use rate of 0.025 lb active ingredient/acre, the maximum theoretical
concentration of abamectin in the top inch of soil immediately following
application would be approximately 85 ppb within the range of the
experimental concentrations. Therefore, abamectin should not affect
nitrogen-fixing bacteria.
Eisenia foetida was used to study the effects of abamectin on earth-
worms (Biospherics, MD). The 28-day LC50 was determined to be
38 ppm, indicating that the use of abamectin should not have an effect on
earthworms.
2. Bees
The toxicity of abamectin to bees was assessed in laboratory studies by
applying it directly to the bees, by putting it in their food, and by exposing
the bees to foliage that had been treated at various times prior to
harvesting (Atkins, personal communications). Abamectin was found to
be quite toxic to the bees. In 2 studies it had contact LC50 values of 0.002
and 0.017 JLg/bee at 24 and 48 hours, respectively. When abamectin was
fed to bees, it had an LD50 of 0.009 JLg/bee. However, citrus or alfalfa
foliage that had been treated with abamectin 24 to 48 hours earlier was not
toxic. Abamectin rapidly loses its toxicity to bees upon exposure to the
environment.
3. Birds
The bobwhite quail and mallard duck are the species normally chosen for
avian toxicity tests. In Table 13.5 the results of a number of studies with
TABLE 13.5. Toxicity of abamectin to birds.

Reproductive
effects,
NOEL
Bird LDso (mg/kg)' LCso (ppm)b (ppm)C
Mallard duck 383 12
Bobwhite quail 2000 3102
• Single oral dose.

b Birds were fed abamectin in the diet for 5 days and then fed regular diet
for 3 days.
CBirds were fed abamectin for 18 weeks from prior to mating through egg
laying.
d Not determined.
these two species are presented. The studies were performed by Wildlife
International (Easton, MD). Abamectin was not very toxic to the
bobwhite quail. The LD50 was greater than 2 g/kg. When quail were fed
abamectin in their diet for 5 days, followed by 3 days on untreated feed,
the LC50 was 3102 ppm. Mallard ducks were somewhat more sensitive:
the LC50 after 5 days of dietary exposure was determined to be 383 ppm.
An avian reproduction study was performed in the mallard duck. When
ducks were fed 3,6, or 12 ppm abamectin in their diet for 18 weeks, there
was no statistically significant effect on any parameter of reproduction.
These dietary levels are greater than those which would occur in the
environment.
4. Mammals
Mammalian toxicity data (discussed in detail in Chapter 6) indicate that
abamectin has an LD50 of approximately 10.0 mg/kg in mice. Studies
were also performed on the two degradation products, the delta 8,9
isomer of avermectin B 1a and the polar fraction. The delta 8,9 isomer is at
least eightfold less acutely toxic than its parent compound, but it
possesses the same level of subchronic toxicity. For this reason residue
data now include the level of this isomer. The polar fraction isolated from
the degradation of abamectin on glass is more than 500 times less acutely
toxic than abamectin and is greater than 20 times less toxic in the most
sensitive subchronic test.
5. Conclusions
Abamectin does not affect nitrogen-fixing bacteria, even at levels higher
than those which would be present in the environment. At high concen-
trations abamectin can have a toxic effect on earthworms and birds.
However, these effects occur at concentrations which are higher than
those found in the environment. The toxicity of abamectin to bees is
mitigated by its rapid degradation, such that residues of abamectin

present on foliage 24 to 48 hours after treatment are not toxic. Abamec-
tin's polar degradation fraction-the major degradation fraction-is
much less toxic than abamectin itself.
IV. Risk Assessment

In order to assess the risk of a pesticide to the environment one must
compare the exposure levels that cause a toxic effect with the exposure
levels expected from using the compound. Environmental fate data yield
an estimate of the environmental concentration, while toxicity studies
indicate the concentration at which toxic effects can occur.
A. AQUATIC ORGANISMS
The rapid photolytic degradation of abamectin in water to much less toxic

degradation products and its strong binding to soil indicate that levels of
abamectin in water should be low. This was shown to be the case in the
simulated field study, which should give the worst-case scenario for drift
and runoff. To simulate drift, the water was sprayed directly with the
diluted formulation. Under these conditions, some mixing occurred: this
would decrease the amount of surface degradation. Also, both the run-
off and drift tanks were covered with tarpaulins to prevent cross-
contamination; since the treatments were applied in the afternoon, the
tanks were not uncovered until the next morning, permitting mixing to
occur prior to environmental exposure. Water analyses also included
abamectin bound to particles suspended in the water. Therefore, the
residue values obtained represented a worst-case situation-and even
then they were very low.
The levels of abamectin in water resulting from runoff were in the range
of only I to 3 ppt during the I5-day period after treatment, and lower
thereafter. This level is below the lowest NOEL determined, which was
in the my sid shrimp chronic life-cycle study (Table 13.4). The mysid
shrimp is the most sensitive organism to the toxic effects of abamectin.
The levels of abamectin resulting from drift are initially high, assuming
that 5% of the applied pesticide impacted on the surface water. This level
of abamectin is greater than would be expected, given the rapid photolysis
abamectin would undergo as it drifted to the water and settled on the
surface rather than being sprayed directly into the surface. Initial residue
levels of 266 ppt are below the LC50 levels for even Daphnia, the
freshwater organisms most sensitive to abamectin. These levels are also
below the NOEL for rainbow trout determined in the early life-stage
study. By 2 days the residues have decreased to 8 ppt, below the LC50 for
mysid shrimp and below the NOEL for Daphnia. At 2 days the concentra-
tion of abamectin in water is at levels that may affect shrimp (8 ppt vs an
198 P.G. Wisiocki, L.S. Grosso, and R.A. Dybas
MATC of approximately 5 ppt). However, these are exaggerated residue

levels derived from the tank study. A stagnant 2-foot-deep body of water
is not representative of an estuary, which undergoes mixing from the
influx of fresh water and the action of tides. Therefore, the use of
abamectin should not have an acute or chronic effect on aquatic or-
ganisms.
B. TERRESTRIAL ORGANISMS
As indicated previously, the use of abamectin results in soil residue levels

of 10 to 20 ppb in the top 2 inches of soil immediately after application.
This level is not sufficient to have an effect on earthworms (LD 5o of
28 ppm) or nitrogen-fixing bacteria (NOEL greater than 400 ppb). In fact,
abamectin levels as high as 2000 ppm do not have antifungal or antibac-
terial activity (see Chapter 2). Abamectin foliar residues aged for 24 to 48
hours are not toxic to bees. Moreover, abamectin is not significantly toxic
to nontarget beneficial insects (see Chapter 20). All these facts suggest
that abamectin is a good candidate for use in integrated pest-management
programs.
The concern over the impact of pesticides on terrestrial organisms
centers on their effects on birds and mammals. It is also assumed that
protecting these animals protects other animals, such as amphibians and
reptiles. Abamectin is acutely toxic to birds. However, birds are much
less sensitive to abamectin than are other species. The lowest LC50
observed, 383 ppm, was in the mallard duck (Table 13.5). The NOEL for
reproductive effects in the mallard duck was 12 ppm. Therefore, if residue
levels of abamectin on avian feedstuffs are below 12 ppm, no adverse
acute or chronic effects on birds would be expected. One method used to
determine the levels of residues of pesticides on foodstuffs for terrestrial
animals is to use the data accumulated by Hoerger and Kenaga (1972) and
Kenaga (1973). Those data-from residue trials and tolerances for a
number of pesticides-were used to predict levels of residues in various
foodstuffs (Table 13.6). Based on a use rate of 0.025 lb active ingredient/
acre, theoretical residue levels greater than 12 ppm would not be
expected. Therefore, the use of abamectin should not pose a risk to avian
species.
By comparing the theoretical levels from Kenaga with those derived
from the actual use of the compound, it can be seen that abamectin does
not give the high initial levels predicted by Kenaga's calculations and it
depletes very rapidly. This is due to its rapid photodegradation on
surfaces (initial half-life of 4 to 6 hours). In the 20 celery trials performed
in the United States, in which repeated applications were made, the
average residue level was only 0.17 ppm, which is V. that proposed by
Kenaga. Again, the residues deplete rapidly, to less than 0.01 ppm by 7
days. Mammalian species vary greatly in their sensitivity to abamectin.
TABLE 13.6. The theoretical maximum and typical

concentrations of abamectin expected on terrestrial food items
immediately after application of abamectin at 0.025 lb active
ingredient (ai)/acre (citrus) and 0.02 lb ai/acre (cotton,
vegetables) (Kenaga 1973).
0.025 lb ai/acre 0.02 lb ai/acre
Maximum Typical Maximum Typical
Food items (ppm) (ppm) (ppm) (ppm)
Short grass 6.0 3.1 4.8 2.5
Long grass 2.7 2.3 2.2 1.8
Leafy crops, leaves 3.0 0.9 2.5 0.7
Forage, insects 1.5 0.8 1.2 0.7
Seed pods 0.3 0.12 0.2 0.1
Fruit 0.15 0.04 0.1 0.03
The lowest acute toxicity value, obtained in the weanling rat, was an LD50
of 1.5 mg/kg. If one assumes that a wild mammal will consume 20% of its
body weight in food/day, it would have to ingest residue levels of
7.5 ppm, 44 times higher than the average found on celery, to reach the
LD50 level. Several other factors should be mentioned. The exposure
levels assume that all mammalian foodstuffs have the same residue level
as the target crop. In actuality, this is not the case. Animals would also
consume foodstuffs outside the direct application area, where residue
levels would be lower. Also, laboratory animals were dosed by gavage,
which has been shown to yield a twofold greater toxicity to abamectin
than in feed administration. Thus an additional twofold safety factor
exists for wild mammalian species.
The lowest subchronic toxicity value was obtained in the materno-
toxicity-teratogenicity study in the CF-l mouse. The NOEL was
0.05 mg/kg by gavage and 0.1 mg/kg in the diet. In this study mUltiple
doses of abamectin were given. Using the dietary NOEL, a mammal
which ingests 20% of its body weight in food/day would have to consume
greater than 0.50 ppm of abamectin in the diet for several days. The low
initial residue levels-which are 3 times lower than this amount-and the
rapid degradation eliminate the possiblility of continuous exposure: no
maternotoxic or teratogenic effects should occur.
Residues in fruit are much lower than those in celery, and, therefore, no
effects from fruit consumption should occur. Insects are a major food
source for certain mammals. Given the low residue levels on plants and
the rapid degradation of abamectin, insects will not have high residues of
abamectin. Insects will not bioconcentrate residues of abamectin. A study
by Bull (1986) demonstrated that the half-life of abamectin in insects is
less than a half a day. Insects should not contain abamectin residues at
levels sufficient to be toxic to those animals consuming them.
V. Conclusions
Abamectin neither persists nor accumulates in the environment. It rapidly
photolyzes in water, on soil, and as a thin film on both inert and biological
surfaces. Soil metabolizes abamectin at a moderate rate. Its strong
binding to soil prevents it from entering the aquatic environment, reduces
its concentration in water, and limits its bioavailability. These character-
istics reduce the environmental exposure to abamectin to a level below
that which would cause effects on aquatic or terrestrial organisms.
Therefore, when label directions are followed, abamectin can be used in
an environmentally safe manner.
REFERENCES
Brooke DN, Dobbs AJ, Williams N (1986) Octanol water partition coefficients
(P): measurement, estimation and interpretation, particularly for chemicals with
P lOS. Ecotox. & Envir. Safety 11:251-260
Bull DL (1986) Toxicity and pharmacodynamics of avermectin in the tobacco
budworm, corn earworm and fall armyworm (Noctuidae Lepidoptera). J. Agric.
Food Chem. 34:74-78
JM, VandenHeuval WJA (1984) Fate of avermectin Bla in soil and plants.
J. Agric. Food Chem. 32:94-102
Hoerger FD, Kenaga EE (1972) Pesticide residues in plants. Correlation of
representative data as a basis for estimation of their magnitUde in the environ-
ment. In F. Coulston and F. Korte (eds) Environmental Quality, Vol 1,
Academic Press, New York, pp 9-28
Iwata I, MacConnell JG, Flor JE, Putter I, Dinhoff TM (1985) Residues of
avermectin B 1a on and in citrus fruits and foliage. J. Agric. Food Chem.
33:467-471
Kenaga EE (1973) Factors to be considered in the evaluation of the toxicity of
pesticides to birds in their environment. In F. Coulston and F. Korte (eds)
Environmental Quality and Safety, Vol 2, Academic Press, New York,
pp 161-181
May WE, Wasik SP (1978) Determination of the aqueous solubility of the
polynuclear aromatic hydrocarbons by a coupled liquid chromatography tech-
nique. Anal. Chem. 50:175-182
Maynard MS, Iwata Y, Wislocki PG, Ku CC, Jacob TA, (1989) The fate of
avermectin B 1a on citrus fruits: 1. Distribution and magnitude ofthe 14C residue
and avermectin B 1a on citrus fruits from a field study. J. Agric. Food Chem.
37:178-183
Moye HA, Malagodi MH, Yoh J, Leibee GL, Ku CC, Wislocki PG (1987)
Residues of avermectin B 1a in rotational crops and soils following soil treatment
with [14C] avermectin B 1a. J. Agric. Food Chem. 35:859-864
Mrozik H, Eskola P, Reynolds GF, Arison BH, Smith GM, Fisher MH (1988)
Photoisomers of avermectins. J. Org. Chem. 53:1820-1823
CHAPTER 14
Worker Safety Aspects of

L.S. Grosso, R.A. Dybas, and S.F. Rickard
I. Introduction
II. Worker Exposure to Abamectin During Airblast Treatment of Citrus
Groves
III. Worker Exposure to Abamectin during Reentry
I. Introduction
Abamectin is the avermectin selected for agricultural use. Its various
formulations, and the uses that have been developed, are described in
Chapter 20.
Although abamectin is applied to ornamental plants, crops, and trees at
extremely low rates, its high oral mammalian toxicity (Chapter 6) initially
raised concern regarding the safety of workers mixing, loading, and
applying the product, as well as those harvesting treated crops. An
established quantitative exposure model developed for registered pesti-
cides indicated a significant correlation between dermal exposure and
application rate (Reinert and Severn 1985). The exposure model was
based on data generated from exposure monitoring of workers in orchards
during ground application of pesticides using high-pressure or airblast
equipment. A linear regression analysis of the data failed to intercept the
origin, thus predicting substantial exposure to such products used at low
rates of application. The reported values are normalized to 3000 cm2 of
exposed skin, the amount for an applicator wearing long pants, a
short-sleeved open-necked shirt, and no gloves or facial protection. The
authors argue that the correlation (r=0.7) permits an estimate of exposure
during airblast application of any pesticide with a certain level of
confidence. It should be pointed out that the model does not consider the
bioavailability of individual products (such as dermal penetration), and
thereby again overestimates exposure.
The studies on which the exposure model was based were conducted
with pesticide products used at conventional rates of application (0.25 to
202 L.S. Grosso, R.A. Dybas, and S.F. Rickard
8.0 lbs active ingredient/acre). Obviously, the model greatly overes-

timates exposures for lower application rates. To demonstrate actual
exposure data for abamectin when applied with similar equipment, a
worker-exposure study was conducted using conventional airblast equip-
ment in typical California citrus groves, under typical working conditions.
Airblast equipment was chosen because this application method has been
reported to result in the highest levels of exposure. Moreover, the results
could be used as a "worst-case" approximation for other application
methods.
The safety to agricultural workers reentering and harvesting crops
treated with abamectin was also a concern during the early phases of
product development. It is widely accepted that reentry into pesticide-
treated sites can present a significant hazard to workers/harvesters. This
has been especially true with the acutely toxic organophosphates and
carbamates. Worker exposure is mainly by dermal contact with residues
sorbed to particulate matter. Measurement of dislodgeable residues is
therefore a means of estimating worker exposure to a pesticide during
routine harvesting practices (Gunther et al. 1977).
Workers can be exposed to foliar pesticide residues via dermal contact
and inhalation. Most evidence from field studies now indicates that
greater than 99% of total exposure to pesticides occurs by dermal contact
(hands and forearms) with residues bound to particulate matter on foliage
(Zweig et al. 1984). Transfer of pesticide residues to humans during
reentry activities is governed more by the physical nature of the residues
than by their chemical properties. Inhalation exposure to vapor phase and
particulate r~sidue also occurs, but it is minor-l% of the dermal
exposure. Inhalation exposure of workers to abamectin residues is
considered to be negligible due to the extremely low vapor pressure of the
compound «1.5x 10-9 torr), coupled with its low application rate.
Pesticide residues sorbed to particulate matter adhering to foliar or fruit
surfaces are measured as dislodgeable foliar residues (DFR) by tech-
niques developed by Iwata and colleagues (1977). Zweig and colleagues
(1984) have summarized the results of 35 field studies on human dermal
exposure and dislodgeable foliar residues (DFR) from harvesting opera-
tions on a number of crops with a variety of pesticides. Table 14.1 lists the
ratios of dermal exposure to dislodgeable foliar residues obtained.
Considering the variety of pesticides and crops studied, the calculated
ratios are very similar, and thus there appears to be a good correlation
between dermal exposure and dislodgeable pesticide residues. Therefore,
Zweig and colleagues (1984) have proposed that it should be possible to
calculate daily dermal exposure of workers/harvesters, as a first approxi-
mation, by multiplying the dislodgeable foliar residue concentration
(DFR: p,g/cm2 ) obtained experimentally by the empirical factor 5000. The
resulting Zweig-Popendorf Factor is 5000 cm2/hour.
To utilize the Zweig-Popendorf Factor, dislodgeable foliar residue
14. Worker Safety Aspects of Abamectin Use in Crop Protection 203
TABLE 14.1. Ratios of dermal exposure to dislodgeable foliar residues.

Pesticide Crop Activity Rxl0- 3
Carbaryl (day l) Strawberries Harvesting 4.34
Carbaryl (day 2) Strawberries Harvesting 2.82
Carbaryl (day 3) Strawberries Harvesting 6.17
Captan Strawberries Harvesting 6.17
Benomyl Strawberries Harvesting 7.19
Benomyl Apples Thinning 0.5
Chlorobenzilate Citrus Harvesting 5.338
OP Compounds Citrus Harvesting 5.1b
OP Compounds Peaches Harvesting 1.9"
Empirical Factor = 5 x 103
a Corrected for one side of the leaf surface.
b Geometric mean from 14 observations.
C Geometric mean from 9 observations.
(From Zweig et al., (1984) J. Agric. Food Chern. 32:1232.)
studies with abamectin were conducted on the labor-intensive crop,

ornamental chrysanthemums. Abamectin residues on chrysanthemum
were obtained from leaf-punch samples from greenhouse-grown cultivars.
To prove the correlation factor as proposed by Zweig and colleagues
(1984), an actual worker reentry study was conducted using greenhouse
grown chrysanthemums in Holland under conditions and practices typical
of that country.
The data developed for the tasks of mixing/loading and application of
abamectin to citrus trees, as well as for harvesting citrus and ornamental
crops, can be used in conjunction with the toxicology data as presented in
Chapter 6 to calculate the margins of safety (MOS) of the product to the
workers.
II. Worker Exposure to Abamectin during Airblast

Treatment of Citrus Groves
Abamectin is formulated as a 0.15 lb/gallon emulsifiable concentrate
(0.15 EC) for use on a variety of crops. Because exposure during airblast
application is generally considered to be a "worst case," typical condi-
tions of mixing/loading and application of the product in citrus groves
were tested. The study involved a total of 4 applications: 3 consisted of 3
tankloads and 1 of 4 tankloads, for a total of 13 monitorings during 4 work
periods of 4 hours each. Rates of exposure were calculated for workers
wearing long-sleeved chambray shirts, long denim pants, and shoes. Test
204 L.S. Grosso, R.A. Dybas, and S.P. Rickard
conditions also included the wearing of gloves, and the effectiveness of

minimal hygiene. Measured residues were corrected for recoveries from
field-fortified (spiked) controls. For residue values less than the sensitiv-
ity of the detection method, 50% of the method's sensitivity was used
as the magnitude of the residue. For example, with a sensitivity of
0.5 ILg/cm2 , nondetectable residues were reported as 0.25 ILg/cm2 for the
purposes of quantitation.
With the exception of exposure of the hands, dermal exposure was
measured with multilayer dosimeters affixed to full-body protective suits.
The dosimeters were placed at conventional locations to represent body
areas for which surface areas have been estimated: back of neck, front
and "V" of neck, shoulders/upper arms, forearms and wrists, thighs, and
ankles/lower legs. Three types of dosimeters were used for each worker.
These were:
• Conventional clothing I: 100% lightweight cotton shirt material on the
upper body; 65/35% polyester/cotton pants material on the lower body.
• Conventional clothing II: 65/35% lightweight polyester/cotton cham-
bray shirt material on the upper body; 100% cotton work-weight denim
material on the lower body.
• Protective clothing I: 100% cotton coverall material on upper and lower
body dosimeters.
A total of 42 dosimeters were tested on each person. Exposure of the
face was estimated by swabbing 4 areas of 20 cm2 each: forehead, left
cheek, right cheek, and throat. Because the workers wore goggles and
respirators, the swabs were taken from areas not covered by protective
gear. Exposure of hands was measured from total residue in hand rinses.
Exposure by inhalation was measured by sampling the air in the worker's
breathing zone with an MSA Fixt-Flo personal air pump, and charcoal
tube traps. For each type of measurement, samples spiked in the field
were used as concurrent recovery and stability controls.
There were no detectable residues in charcoal air-monitoring tubes;
therefore, inhalation exposures were not calculated. Using the industrial
hygiene convention, a lognormal distribution was chosen to represent the
data. This convention is based on the understanding that the conditions
conducive to (but not all necessary for) the occurrence of lognormal
distributions are found in occupational environmental data (Leidel,
Busch, and Lynch 1977).
Table 14.2 presents the geometric means and geometric standard
deviations calculated from the adjusted measured values. For risk-
assessment purposes, unprotected whole-body exposure was compared
to dermal exposure calculated from residues when wearing conventional
clothing II (chambray shirt and denim pants) and gloves over washed
hands.
As shown in Table 14.2, the exposure/minute was greater for mixer/
TABLE 14.2. Worker exposure to abamectin.

Adjusted
Unprotected whole Adjusted ng ai/g
body exposure ng ai/minute ai/handled
Mixer/loader Geometric mean 24,915 14,lll
GSDa 4.74 4.60
Sprayer Geometric mean 7547 7989
GSD 4.88 4.32
Wearing long pants

long sleeve shirt,
impermeable gloves,
and washing hands
Mixer/loader Geometric mean 604 342
GSD 1.51 1.29
Sprayer Geometric mean 414 439
GSD 1.79 1.60
a As an estimator of the variance of the data, the geometric mean is multiplied and divided
by the geometric standard deviation (GSD).
loaders than for sprayers. This is consistent with previously published

field studies (Reinert and Severn 1985). When gloves were not worn, the
exposurelg ai handled for mixers was also greater than for sprayers. Also,
most of the reduction in exposure was due to the wearing of gloves.
However, when gloves were worn, the exposurelg ai handled was
essentially similar for mixers and sprayers.
Table 14.2 also shows that wearing denim long pants and chambray
long-sleeved shirt can reduce total bodily exposure irrespective of facial
washings by more than 94% compared to unprotected whole-body
exposure (regardless of rate function); this is consistent with popular
assumptions of 80% to 100% protection by clothing (Reinert et al. 1987).
Also, most of the reduction in exposure was due to wearing gloves.
However, almost all of the exposure calculated for clothed skin resulted
from the fact that more than 50% of the totals were contributed by using
50% of the method's sensitivity in the calculations for these residues on
dosimeters that were below the sensitivity of the method (5ng/cm 2 ).
Actual exposure would be considerably lower than calculated exposure.
Cloth was highly effective in trapping abamectin residues. This was not
unexpected, as abamectin is known to bind to cellulosic and other
substrates very well.
To conclude, the actual rates of exposure for workers in this study
would be less than 50% of the calculated rates; the estimates are
extremely conservative, as 50% of the limit of sensitivity of the analytical
method was added in each case where there were no detectable residues.
Due to the low expected exposure to abamectin, workers were monitored
for a 4-hour period (1/2 workday) while applying 3 tankloads of spray
(1 replicate included 4 tankloads) rather than single tankloads (as is the

convention for these studies).
Calculations of margins of safety (MOS) to the exposure conditions
described may be computed utilizing the toxicology data from Chapter 6
as follows:
• Lowest no-effect level of abamectin technical based on matemotoxicity
in the CF-l mouse-0.05 mg/kg/day by oral gavage.
• Monkey dermal penetration is less than 1% using 3H-abamectin
technical; 3H-abamectin in the 0.15 EC undiluted; or 3H-abamectin
0.15 EC diluted in water
The CF-l mouse is extremely sensitive to abamectin and provides a
very conservative model for predicting the safety of abamectin to
humans. Data discussed in Chapter 6 demonstrate that primates (rhesus
monkeys) are far less sensitive to abamectin than the CF-l muose or any
of the other mammalian species tested. Also, human data available on
ivermectin, the human and animal health drug analog of abamectin,
further support the safety of abamectin in humans. In fact, the clini-
cal dosages of ivermectin in animals and humans range from 0.15 to
0.3 mg/kg, and even the lowest of these is lethal to the CF-l mouse.
The dermal penetration of abamectin is extremely low: less than 1%
when applied either as the technical material or in the 0.15 EC formula-
tion, diluted or undiluted. The following analysis compares the MOS for
workers who are unprotected with those who use typical protective
clothing and personal hygiene during airblast application of abamectin at
0.025 lb ai/acre to citrus. The exposure data for each of these test
conditions are presented in 2 different ways. The traditional expression of
pesticide exposure to applicators is an amount of active ingredient per
unit of time. The first set of MOS calculations, therefore, presents the
geometric mean exposures in ng/minute. These values are then adjusted
for the length of time the particular task is performed during a typical
workday; i.e., 1 hour for mixers/loaders and 8 hours for sprayers.
The second set of MOS calculations presents the geometric mean
exposures in ng of active ingredient per gram of active ingredient handled.
These values are then adjusted for the total amount of active ingredient
handled by each worker type during a typical workday in which a
maximum of 20 acres of citrus would be treated by airblast sprayer.
Therefore, mixers/loaders and sprayers would handle the following
amount of abamectin active ingredient per day
20 acres/day x 11.34 grams ai/acre = 226.8 grams ai/day
(0.025 lb ai/acre = 11.34 g ai/acre)
Margins of safety for the methods of exposure expression are calcu-
lated as follows:
Exposure expressed as active ingredient per time

Exposure = ng/min x 10- 6 mg/ng x 60 min/hr x 1/60 kg
Exposure = mg ai/kg body weight hour
Factoring for dermal penetration (dfp)-l%
Exposure = mg/kg/hr x 100- 1
MOS = NOEL (mg/kg) -:- Exposure (mg/kg/hr) x 11100
Exposure for mixer/loader = mg/kg/hr
Exposure for sprayers = mg/kg/Shr
Exposure expressed as active ingredient per gram active
ingredient handled
Exposure = ng/g ai x 10-6 mg/ng x 0.5 lbs ai/day x 454 gr/lb
x 1160 kg
Factoring for dermal penetration (dfp)-I%
Exposure = mg/kg/day x 11100
MOS for mixers/loaders/applicators = NOEL (mg/kg/day) -:- Exposure
(mg/kg/day) x 11100
Table 14.3 compares the Margins of Safety based on time of exposure

and amount of active ingredient handled between mixers/loaders and
applicators for whole body unprotected and when wearing conventional
clothing (chambray shirt and denim pants) and using personal hygiene.
Margins of Safety (MOS) have been presented for agricultural workers
who handle the product during mixing, loading, and application functions.
Actual human-exposure values have been measured for these tasks under
typical conditions reflective of product labels approved by regulatory
agencies in various countries. The exposure data developed for abamectin
TABLE 14.3. Abamectin margins of safety (MOS) based on maternotoxicity for

pesticide applicators-mixers/loaders and sprayers.
MOS based on time of exposure/per day·
Mixers/loaders Sprayers
Unprotected whole body exposure 200 83
Protected whole body exposure 8333 1562
MOS based on amount ai handled per dayb

Mixers/Loaders Sprayers
Unprotected whole body exposure 94 166
Protected whole body exposure 3846 3030
• Mixers/Loaders handle pesticides for a maximum of I hour per day to prepare sufficient
material for sprayers to work for a maximum of 8 hours.
b The quantity of active ingredient handled is based on a maximum of 20 acres treated
per day.
clearly show that the existing model available for applicators grossly
overestimates exposure. Based on the available safety data on the
avermectins in the most sensitive model (CF-l mouse), in addition to
human data and exposure data, we believe that there are adequate
margins of safety for workers who mix/load or spray abamectin using
airblast equipment with or without the use of conventional clothing.
III. Worker Exposure to Abamectin during Reentry

To estimate the safety of abamectin to workers harvesting ornamental
plants by utilizing the Zweig-Popendorf Correlation, a dislodgeable foliar
residue study was conducted in a commercial chrysanthemum propa-
gation greenhouse at DeLier, Netherlands. Abamectin was applied in
3 weekly applications of 27 g ai/hectare (0.024 Ib ai/acre) as 1.5 I
formulation of the 1.8EC (18 g/I EC) in 1500 I of water per hectare.
Surface extracts from the abamectin-treated chrysanthemum foliage
were prepared for assay by the method of Gunther and colleagues (1977)
to determine dislodgeable foliar residues to which workers might be
exposed while harvesting cuttings from the treated plants. Of the various
production tasks, harvesting of chrysanthemum propagation cuttings is
believed to represent the most labor-intensive and "worst case" for
worker exposure to pesticide residues in greenhouse operations. Because
DFR data are commonly used to estimate exposure for workers entering
treated crop areas to perform cultural tasks, they are used in making a risk
assessment to determine margins of safety (MOS) for the worker (Adams
1984). In this procedure, the empirically derived Zweig-Popendorf Factor
of 5000 cm2/hour (based on one surface of the leaf, as were these data) is
applied to the DFR data to estimate exposure. The Zweig-Popendorf
Factor provides a very conservative estimate of worker safety, since
exposure values are derived from harvesters wearing cotton gloves
(which are known to significantly overestimate residues on the hands).
Leaf-punch samples for residue analysis were collected directly into
separate glass jars from each of 2 untreated and 4 treated plots before and
at 2, 4, 8, 16, and 24 to 30 hours, and at 3 and 7 days after the third
application. Immediately after collection, surface residues of abamectin
were removed from the leaf discs, and the aqueous solutions were frozen
until analyzed by an HPLC-fiuorescence method for residues of abamec-
tin (avermectin B la and avermectin BIb) and its corresponding d8,
9-isomer. Residues were reported on a leaf surface area and used to
predict dermal exposure and margins of safety for workers. Extra leaf
punches from the control plots were collected to be used as control and
field fortification samples; these samples measured the abamectin stability
and recovery through sample preparation, storage, and assay.
The results of the study are summarized in Table 14.4. The dissipa-
TABLE 14.4. Dislodgeable foliar residues of abamectin and its 8,9-isomer on

chrysanthemum (average of 4 replicates).
Total residues (ng/em2) with
Sampling time standard deviation deviation
Before 3rd Spray 1.30 ± 0.52
2-hr. post spray 22.04 ± 1.74
4-hr. post spray 19.85 ± 1.68
8-hr. post spray 21.50 ± 3.95
16-hr. post spray 21.92 ± 2.33
24-30 hr. post spray 11.59±1.06
3 days 5.16 ± 0.62
7 days 2.16 ± 0.89
tion curve for both residue species is similar (Figure 14.1). The "prior
to last application" samples for all 4 treated plots had average aver-
mectin Bla/ a8,9-isomer residues of 1.2 ng/cm2 and 0.14 ng/cm2 avermec-
tin BIb. The residues at 2 hours following the last application aver-
aged 19.2 ng/cm 2 for the combined residues of Bla/ a8,9 isomer and
2.9 ng/cm2 avermectin BIb. Abamectin is unstable in the presence of
sunlight, but these residues did not decay through Hour 16, as the treated
chrysanthemums were protected from light in the late afternoon by
drawing close the greenhouse thermal shades. Between Hours 16 and 30,
after the thermal shades were opened, the residues decayed, with average
residues at 30 hours of 10.1 ng/cm2 avermectin B la and 1.5 ng/cm2
avermectin BIb. At 7 days after the last application, residues averaged 1.9
ng/cm2 avermectin B la and 0.23 ng/cm2 avermectin BIb, essentially
identical to the residues found at 7 days following the second of 3
applications. No buildup or accumulation of residues was observed
following multiple applications in greenhouse conditions. No difference
was found in the recoveries from leaf disc samples spiked in the field and
those spiked in the laboratory. Application of the Zweig-Popendorf
Factor to the DFR data from this study results in the following estimate of
worker exposure to abamectin residues. The average for total DFR
residues of abamectin and its a8,9-isomer was 22 ng/cm2 at 16 hours after
the third of 3 weekly sprays at the recommended rate of 27 g abamectin/
hectare, immediately prior to the reentry phase.
DFR = 22 ng/cm2 or 0.022 p,g/cm2

Zweig-Popendorf Factor = 5,000 cm2/hr
Assume 8-hour workday and 60-kg person
Exposure = 0.022 p,g/cm2 X 5000 cm2/hr x 8 hrs
= 880 p,g/person/day
OR
Exposure = 880p,g/person/day = 14.7 p,g/kg/day for a 60 kg person
25
23
•••
21
18
•
17
II)
:;jll
....'tI
III
II)
II: 8
2 3 4 5 6 7
TIllE (Days)
FIGURE 14.1. Abamectin 8,9 isomer dislodgeable residue dissipation.
The margins of safety (MaS) for exposure of harvesters to abamectin

dislodgeable foliar residues are calculated as a ratio of exposure to the
lowest NOEL for toxicological effects. For abamectin, the lowest NOEL
is 0.05 mg/kg/day for matemotoxicity in CF-I mice. The dermal penetra-
tion factor as developed in monkeys is less than 1%.
NOEL = 50 ILg/kg/day
Exposure = 14.7 ILg/kg/day
MOS = NOEL/exposure x 100 (dpf)
= 50 x 100 = 340
14.7
MOS = 340 for maternotoxicity
It was believed that this worker exposure of 14.7 ILg/kg/day from foliar
residues of abamectin estimated via the Zweig-Popendorf Factor was an
overestimate of exposure, resulting in a very conservative MOS.
To confirm the accuracy ofthe Zweig-PopendorfFactor with abamectin
under the same greenhouse conditions, an actual worker reentry study
was conducted in the Netherlands. This study measures the dermal
exposure of workers to foliar residues of abamectin during harvest of
propagation cuttings from greenhouse-grown chrysanthemums that had
been treated with abamectin (0.15 EC) at its maximum recommended use
rate. Residues from coveralls, hats, gloves, and hand rinses of harvesters
were used to estimate exposure, which was then used in a risk assessment
to determine margins of safety for workers involved in the culture and
production of greenhouse ornamentals.
The study was conducted in the same commercial chrysanthemum
propagation greenhouse at DeLier, Netherlands as that used for the DFR
study. Each of 4 workers was dressed in a disposable Tyvek spun-olefin
coverall and baseball hat with prepared dosimeters attached to the front
and the back of the hat, the chest, back, upper and lower arms, and upper
and lower legs. The workers wore cotton gloves. Each worker was
assigned to a separate treated plot for harvesting. At the end of each of
four I-hour work periods, the dosimeters and gloves were removed into
separate numbered bags. Mter the gloves were removed, each worker's
hands were rinsed 3 times. Control and fortified (spiked) samples from the
field, along with the experimental samples, were immediately frozen
following their collection for subsequent analysis using an HPLC-
fluorescence method which determines abamectin (avermectin B la and
avermectin Bib) and the ~8,9-isomer.
Abamectin residues were found only on the gloves, lower-leg do-
simeters, and in the hand-rinse solutions; in 1 case low-level residues
were found on the chest dosimeter. The data are presented in Table 14.5.
Average exposure was 70,863 ng/hr with a coefficient of variation (CV)
of 5.4%. Approximately 97% of the total exposure was on the hands, 3%
on the clothing, and less than 1% on the face and clothed body. Because
50% of the limit of quantitation was used when no residues were
measured on a given dosimeter, calculated exposures of the face and neck
(except in 1 case) and the clothed body were calculational artifacts that
were insignificant in estimating exposure. Body protection by long pants
TABLE 14.5. Individual worker exposure data.

Worker Total Face &
body no. (ng/hr) Hands neck Clothes Body
.101 76,546 74,184 80 2054 227
.102 70,069 66,088 132 3284 566
.103 68,299 66,212 80 1780 227
.104 68,537 66,488 80 1742 227
Average: 70,863 68,243 93 2215 312
S.D. 3869 3964 26 726 169
and long-sleeved shirt was minimal since 97% of exposure occurred on

the hands. When body exposure occurred, it was usually only the lower
leg dosimeters that had been in contact with treated foliage.
Residue concentration in cotton gloves is believed to grossly overes-
timate exposure from actual residues present and available for dermal
penetration of the workers' hands (Reinert et al. 1987). This is supported
by previously reported studies in which cotton gloves were clearly shown
to represent a "worst case" for estimating potential exposure of the
hands to dislodgeable residues. Because of its almost irreversible binding
to cellulosic and other fibers in the gloves, abamectin would not be
dislodged from gloves or penetrate through to the hand. This was shown
by finding 99.7% of hand exposure in the glove extracts and less than
0.3% of the total residue in the solvent hand rinse. Average residues in
gloves from other studies reported in the literature are usually 3 to 10
times those in solvent rinses.
From the exposure data presented above, MOS may be determined for
the greenhouse harvester of abamectin-treated chrysanthemums. The
following calculations assume an 8-hour workday, 1% dermal penetration
for abamectin, and a 60-kg person:
Exposure = 70.86 JLg/worker/hour = 567 JLg/worker/8-hr workday
= 9.45 JLg/kg/day dermal exposure for a 6O-kg person
= 9.45 JLg/kg/day x 1% dermal penetration
= 0.0945 JLg/kg/day systemic exposure
NOEL = Matemotoxicity in CF-l mouse 0.05 mg/kg/day
MOS NOEL
Systemic exposure
50 JLg/kg/day
0.0945 JLg/kg/day
= 529
The actual exposure value of 9.45 JLg/kg/day developed in this study
agrees well with the calculated value of 14.7 JLg/kg/day based on
measured foliar dislodgeable residues and using the Zweig-Popendorf
Factor. A similar DFR study was also conducted by Merck Sharp &
TABLE 14.6. Harvester exposure.

Worker MOS
Study site Basis Exposure Maternotoxicity
Netherlands Actual worker 9.45/Lg/kg/day 529
exposure
Netherlands Predicted exposure from DFR 14.7/Lg/kg/day 340
California Predicted exposure from DFR 10.7/Lg/kg/day 467
Dohme Research Laboratories in California (Jenkins et al. 1987). In this

study AVID 0.15 EC was applied to greenhouse-grown chrysanthemums
twice, 5 days apart, at 28 g abamectin per hectare (0.025Ib ai/acre) in 2338
I of water. At 2 hours after the second spray, total DFRs of abamectin and
its 8.9 isomer were present at 16 ng/cm 2, which compares favorably to the
22 ng/cm2 found in the Holland DFR study.
Using the California data and the Zweig-Popendorf Factor, daily
harvester exposure is estimated to be 10.67 JLg/kg/day (0.016 JLg/kg/day
x 5000 cm2 x 8 hours = 640 JLg/day = 10.67 JLg/kg/day). The MOS may
then be calculated accordingly [MOS = NOEL / exposure = 50 JLg/kg/
day/l0.67JLg/kg/day x 100 (dermal penetration)] as 467.
The worker-exposure data from the 3 greenhouse studies are summa-
rized in Table 14.6.
These MOS determinations in Table 14.6 are remarkably consistent,
whether calculated from actual worker-exposure data or from worker-
exposure estimates derived from measured foliar residues. The MOS
from the DFR studies are slightly more conservative than the MOS from
the worker reentry actual exposure study, which represented "worst-
case" exposure data. The MOS calculated from the worker reentry study
was based upon direct measurement of abamectin residues on harvesters
of cuttings from greenhouse chrysanthemums treated with the maximum
recommended rate of abamectin, and used in combination with the lowest
NOEL for toxic effects in the most sensitive species of test animal, the
CF-l mouse. The data show the model used for predicting harvester
exposure to pesticide dislodgeable foliar residues is reasonably accurate,
and further prove this work task may be performed safely with abamectin.
REFERENCES
Adams, JD (1984) Pesticide Assessment Guidelines, Subdivision K. Exposure:

Reentry Protection. US EPA Office of Pesticide Programs.
Gunther FA, Iwata Y, Carmen GE, Smith CA (1977) The citrus reentry problem:
research on its causes and effects and approaches to its minimization. Resid.
Rev. 67:1-139.
Iwata Y, Spear RC, Knaak JB, Foster RJ (1977) Worker reentry into pesticide-
treated crops. Procedure for the determination of dislodgeable pesticide resi-
dues on foliage. Bull. Envir. Contam. Toxicol. 18(6):649-655.
Jenkins n, Rosenthal HS, Mollet J, Brown RD, Norton J, Dybas RA (1987)

Dislodgeable residues of avermectin BI • and its delta 8,9-isomer following
application of AVID 0.15 EC to greenhouse grown chrysanthemums. ACS
Services. 193rd National Meeting, AGROCHEMICALS No.3.
Leidel NA, Busch KA, Lynch JR (1977) Occupational Exposure Sampling
Strategy Manual. National Institute for Occupational Safety and Health p 131.
Reinert JC, Nielsen AP, Davis JE, Hickey KD, Nigg HN, Ware GW, Lunchick C
(1987) Pesticide Assessment Guidelines, Subdivision U. Applicator Exposure
Monitoring, US EPA Office of Pesticide Programs, p 163.
Reinert JC, Severn DJ (1985) Dermal exposure to pesticides, the Environmental
Protection Agency's viewpoint in Honeycutt. In ACS Symposium Series 273,
American Chemical Society, Washington, DC, pp 257-268.
Zweig G, Gao R, Witt J, Popendorf W, Bogen K (1984) Dermal exposure to
carbaryl by strawberry harvesters. J. Agric. Food Chern. 32:1232-1276.
CHAPTER 15
Use of Ivermectin in Cattle, Sheep,

Goats, and Swine
G.W. Benz, R.A. Roncalli, and S.l. Gross
I. Introduction
II. Use in Cattle
A. Injectable Formulation
B. Oral Formulations
C. Topical Formulation
D. Sustained-Release Bolus
III. Use In Sheep
A. Oral Formulation
B. Injectable Formulation
IV. Use In Goats
A. Oral Formulation
B. Injectable Formulation
V. Use In Swine
A. Injectable Formulation
References
I. Introduction
This chapter presents a summary of the efficacy data that form the basis
for the clinical use of ivermectin in cattle, sheep, goats, and swine. At
least 2 adequate and well-controlled trials investigated each ivermectin
formulation's effect on each parasite species. Most of the trials were
conducted by MSDRL investigators, though investigators at other labora-
tories also contributed. The efficacy percentages presented here are those
upon which the regulatory approvals were based. Animal health applica-
tions of ivermectin have been reviewed (Campbell et al. 1983; Campbell
and Benz 1984; Benz 1985; Campbell 1985; Bennett 1986; Barragry 1987).
Clinical usages have also been summarized concerning cattle (Leaning,
Roncalli, and Brokken 1983), sheep (Hotson 1983), and swine (Brokken et
al. 1983). Recent reports that expand the basis of the label claims are
included.
216 G.W. Benz, R.A. Roncalli, and S.J. Gross
I. Use in Cattle
A. INJECTABLE FORMULATION
IVOMEC Injection is a sterile solution containing 1.0% w/v ivermectin.

The product is indicated for the treatment and control of gastrointestinal
nematodes (including hypobiotic Ostertagia fourth-stage larvae), lung-
worms, and certain other nematodes; grubs; sucking lice; psoroptic and
sarcoptic mange mites; sand tampans; and larvae of the Old World
screwworm fly. It is to be administered subcutaneously in front of or
behind the shoulder at 1.0 ml/SO kg body weight.
Endoparasite Activity
The recommended dose of 200 /Lg ivermectin/kg was selected upon
completion of numerous dose-titration trials. Adult Cooperia oncophora,
C. punctata, Trichostrongylus colubriformis, and Nematodirus helvetia-
nus each required a dose between 100 /Lg/kg and 200 /Lg/kg; fourth-
stage larvae of Haemonchus placei, C. oncophora, N. helvetianus, and
T. colubriformis required a dose of 200 /Lg/kg.
Trial data used for registration of label efficacy claims are summarized
in Table IS.I. In work performed subsequent to the original registrations,
hemorrhagic dermal lesions caused by Parafilaria bovicola were observed
to require at least 70 days for regression (Swan et al. 1983). Efficacy of
more than 99% was found for Thelazia spp. adults (Carmichael, SolI, and
Scherer 1985).
Infections with Ostertagia and Cooperia acquired up to at least 7 days
after an ivermectin injection, as well as with Dictyocaulus viviparus
acquired up to at least 14 days later, are controlled (Barth 1983; Armour et
al. 1985), presumably because the drug persists following injection. Based
on this persistent effect, control programs utilizing ivermectin injections
at 3, 8, and 13 weeks after turnout onto spring pasture have been effective
in controlling gastrointestinal parasites as well as lungworms (Ryan et al.
1986; Taylor, Mallon, and Kenny 1985; Jacobs, Fox, and Ryan 1987). In
such programs, cattle on pastures contaminated with D. viviparus infec-
tive larvae have been found to be resistant to lungworm disease in the
next grazing season, probably because infective larvae ingested between
treatments induced immunity; gastrointestinal nematodes in the second
season were not apparent but inhibited Ostertagia larvae did accumulate
(Armour et al. 1988).
Ectoparasite Activity
The dose rate confirmed as optimal against nematode parasites was also
found to be highly effective against several economically important
ectoparasites.
15. Use of Ivermectin in Cattle, Sheep, Goats, and Swine 217
TABLE 15.1. Efficacy in registration trials ofivermectin given subcutaneously at

200 p,g/kg against adult and fourth-stage nematode parasites of cattle.
% Reduction
Parasite Adults Larvae"

Gastrointestinal nematodes
Haemonchus placei 98 96
Mecistocirrus digitatus 100 NOb
Ostertagia ostertagi >99 98
O. ostertagi hypobiotic larvae >99
O.lyrata 99 NOb
Trichostrongylus axei >99 98
T. colubriformis 93 97
Cooperia spp. 97 97
C.oncophora 95 99
C. punctata 98 95
C. pectinata >99 NOb
Nematodirus helvetianus 84 NOb
N. spathiger 99 NOb
Strongyloides papillosus >99 NOb
Toxocara vitulorum 100 NOb
Bunostomum phlebotomum >99 >99
Oesophagostomum radiatum 98 >99
Lungworms
Dictyocaulus viviparus 100 >99
D. viviparus hypobiotic larvae 100"
Other nematodes
Parafilaria bovicola
Thelazia spp. >99
" Oeveloping fourth-stage larvae.
b No data are available.
C Includes fourth- and/or fifth-stage larvae.
d Subcutaneous lesions require 70 days for regression.
Grubs
The common cattle grubs in the Northern Hemisphere, Hypoderma bovis
and H. lineatum, are highly susceptible to ivermectin. All 3 larval stages
are killed following treatment. Efficacy in confirmation trials was 100% in
most instances, although a few third instars (larvae), which were near to
emergence from their host at the time of treatment, were expelled even
though they had been grossly affected and were probably dead. Killing
Hypoderma first instars during their normal migratory route may cause
adverse host-parasite reactions-such as eosinophilic esophagitis leading
to bloat (H. linea tum) or hemorrhage, (H. bovis) which affects the spinal
cord and most often leads to hindquarter paralysis. These reactions can
be avoided by not treating cattle with ivermectin during this migratory
season (as determined by local epidemiologic patterns).
The tropical cattle grub affecting cattle in Latin America, Dermatobia
hominis, is also controlled effectively (Roncalli 1984b).
Lice
Lice that ingest blood or body fluids from their host (i.e., sucking lice),
including Haematopinus eurysternus, Linognathus vituli, and Soleno-
potes capillatus, are highly susceptible to therapeutic treatment. Iver-
mectin given subcutaneously at 200 ILg/kg kills adult sucking lice as they
feed upon blood or its constituents; the drug persists sufficiently long so
that newly emerged larvae as well as nymphs are also killed as they begin
to feed. In the registration trials, viable lice were recovered from cattle
1 week after treatment but none thereafter (up to 8 weeks). Lice surviving
on ivermectin-treated cattle during the first week after treatment should
be considered infective to nonmedicated cattle. To avoid reinfestation,
ivermectin-medicated cattle should not be mingled with those untreated
during the balance of the winter season, when these parasites are still
active on their host.
Biting lice (Damalinia bovis) are not eliminated consistently from
cattle given a therapeutic injection of ivermectin. This may be related
to their superficial feeding pattern. A claim stated as "aid in the
control" of D. bovis is registered in Europe and some Southern
Hemisphere countries. In the United States, treatment of infested
cattle was ineffectual; whether there are susceptibility differences
among D. bovis populations from various geographic localities
has not been determined.
Mites
Among mange mites, Sarcoptes scabiei var. bovis is the most sensitive to
ivermectin. This species burrows more deeply into the dermis than do
other species and evidently consumes blood or body fluids directly. In
registration trials, live sarcoptic mites from medicated animals were
recovered at 1 week following injection with ivermectin, but not there-
after (up to 8 weeks).
Psoroptes ovis does not burrow into the skin as deeply as Sarcoptes.
Nonetheless, no live psoroptic mites were recovered during confirmation
trials after 2 weeks following treatment although, in subsequent trials, a
few P. ovis have been found on cattle treated 20 days previously (Wright
and Guillot 1984). An isolation period of at least 14 days following
treatment is adequate for use in control programs (Strickland and Gerrish
1987). Very few live mites can be recovered subsequent to treatment; the
persistence of ivermectin following injection presumably accounts for the
few recovered mites not being able to reestablish viable populations
(Guillot, Wright, and Oehler 1986).
Chorioptes bovis, living on the superficial dermal surface, is the least
responsive of the cattle mites studied; only an "aid in control" claim is
registered in certain countries.
Ticks
Ticks also are affected by ivermectin. Boophilus microplus and B.
decoloratus, which are single-host species, are more sensitive than
multihost ticks. Ingestion of blood by adult females is interfered with;
eventually they falloff, usually only partially engorged, and few eggs are
produced. The efficacy of therapeutic doses against Boophilus spp. is
sufficient for the registration of "aid in control" claims in several
countries. Boophilus strains resistant to organophosphates, DDT, synthe-
tic pyrethroids, and amidine were sensitive to ivermectin (Roncalli et al.
1984).
Effects against most multihost ticks are too transient for meaningful
efficacy claims; however, a claim has been registered for the sand tampan
(Ornithodoros savignyi) present in South Mrica (SolI et al. 1984).
Screwworms
Among fly larvae (screwworms) that develop in the wounds of cattle, a
therapeutic injection of ivermectin effectively controls Old World screw-
worm fly (Chrysomya bezziana) larvae for 14 days following treatment.
Established infestations are less effectively controlled and may require
additional measures. Screwworms of Cochliomyia hominivorax are not
controlled by such ivermectin treatment.
ORAL FORMULATIONS
IVOMEC oral solution for cattle contains 0.4% w/v ivermectin and
should be administered at 2.5 m1l50 kg body weight using standard dosing
equipment. IVOMEC cattle paste contains 0.153% w/v ivermectin and is
administered orally by depressing the handle of the Medigun applicator
completely to deliver 23 mg ivermectin per 113.5 kg (250 lb) body weight.
These formulations are indicated for the treatment and control of gastro-
intestinal nematodes, lungworms, grubs, and sucking lice.
Efficacy data used for registration of nematode efficacy claims for both
oral formulations are listed in Table 15.2.
Complete efficacy is lacking against mange mites following oral adminis-
tration of ivermectin to cattle, presumably because less ivermectin is
absorbed systemically and it persists for a shorter duration compared to
the subcutaneous route.
The common cattle grubs in the Northern Hemisphere, Hypoderma
220 G.W. Benz, R.A. Roncalli, and S.l. Gross
TABLE 15.2. Efficacy in registration trials of ivermectin

given orally at 200 jLg/kg against nematode parasites of
cattle.
% Reduction
Parasite Adults Larvae a
Haemonchus placei >99 100
O.lyrata >99 >99
Trichostrongylus axei >99 >99
T. colubriformis 100 >99
Cooperia oncophora >99 100
C. punctata >99 >99
Nematodirus helvetianus 97 98
Strongyloides papillosus >99 ND b
Bunostomum phlebotomum >99 100
Oesophagostomum radiatum 100 100
Trichuris ovis >99 NDb
Lungworms
a Developing fourth-stage larvae.
bovis and H. lineatum, were killed with 100% efficiency in development

trials and the same precautions observed with the injectable formulation
regarding potential adverse host-parasite reactions during the autumn
migratory season should be observed. Infestations of sucking lice (Hae-
matopinus eurysternus and Linognathus vituiz) present on cattle in the
development program were eliminated after 1 week following oral
treatment.
TOPICAL FORMULATION
IVOMEC Pour On for cattle contains 0.5% w/v ivermectin and is applied
along the topline in a narrow strip extending from the withers to the
tailhead. The volume of application is 1 mlllO kg body weight. This
formulation is for the treatment and control of gastrointestinal nematodes,
lungworms, grubs, sucking and biting lice, chorioptic and sarcoptic
mange mites, and as an "aid in control" of ticks (Boophilus microplus).
The recommended dose rate of 500 j.tg/kg was selected following a series
of dose-titration trials. Dose-confirmation data against gastrointestinal
nematodes and lungworms (Hotson et al. 1985; Alva-Valdes et al. 1986;
Yazwinski et al. 1986) are summarized in Table 15.3. The formulation has

applied topically at 500 jLg/kg against nematode parasites
of cattle.
% Reduction
Haemonchus placei >99 93
Trichostrongylus axei 96 99
T. colubriformis 80 84
Cooperia spp. 95 96
C.oncophora 96 ND b
C. punctata 100 NDb
Strongyloides papillosus 97 NDb
Oesophagostomum radiatum >99 100
O. venulosum 100 NDb
Trichuris spp. 87 NDb
Lungworms
" Developing fourth-stage larvae.
sufficient persistence of activity to control Ostertagia spp. infections

acquired up to 14 days after treatment and Dictyocaulus viviparus
acquired up to 28 days after treatment.
Ivermectin applied topically to cattle effectively controls several ecto-
parasites (Barth et al. 1986; Alva-Valdes et al. 1986; Hotson et al. 1985).
Infestations of Chorioptes bovis as well as Sarcoptes scabiei var. bovis
are controlled. However, the formulation does not completely eliminate
Psoroptes ovis and, in such cases, the use of IVOMEC Injection should be
considered. Sucking lice (Haematopinus eurysternus, Linognathus vituli,
and Solenopotes capillatus) as well as biting lice (Damalinia bovis) are
effectively controlled. Hom flies (Haematobia irritans) are controlled for
up to 35 days following product application. Northern Hemisphere grubs
(larvae of Hypoderma bovis and H. linea tum) and tropical grubs (Der-
matobia hominis) are controlled. The topical formulation "aids in the
control" of Boophilus microplus (Cramer et al. 1985).
SUSTAINED-RELEASE BOLUS
Development of a sustained-release bolus is in progress. Initial trials

utilizing an osmotic pump to deliver ivermectin at steady-state rate (Pope
et al. 1985) over 28 days demonstrated that prophylaxis against newly
ingested larvae (Egerton, Suhayda, and Eary 1986), as well as therapy

against established infections could be accomplished with a dose of
40 JLg/kg/day (Baggott, Batty, and Ross 1986). The same dose prevented
the engorgement by adult female multihost ticks (Soil, Carmichael, and
Gross 1987).
Prototype boluses (Eckenhoff et al. 1987; Wilkinson et al. 1987) for
commercial use (SolI, Carmichael, and Harvey 1988) are envisaged to
deliver ivermectin at steady-state rate over 120 days. There are 2 boluses
in development that are identical in all respects except that I bolus will
delivery 8 mg ivermectin daily for cattle expected to weigh 200 kg after
the end of the delivery period, whereas the other will deliver 12 mg daily
for cattle to attain 300 kg. Each bolus will be delivered orally into the
distal pharynx or anterior esophagus so as to preclude damage during
administration. After 120 days, the bolus is expected to promptly cease
delivery.
Use in Sheep
ORAL FORMULATION
IVOMEC or ORAMEC Liquid for sheep is a 0.08% w/v solution of

ivermectin to be given orally to sheep at 2.5 mlll0 kg body weight using
standard drenching equipment. Dose rates and equipment should be
checked before drenching commences to ensure proper volumes are
delivered. The oral formulation is for the treatment and control of
gastrointestinal nematodes, lungworms, nasal bots, and itch mites.
The recommended dose rate of 200 JLg/kg for sheep was selected
following a series of dose-titration trials. Dose-confirmation data against
gastrointestinal nematodes and lungworms, used for registration, are
summarized in Table 15.4. Strains of Haemonchus contortus, Ostertagia
circumcincta, and Trichostrongylus colubriformis tolerant to benzim-
idazoles, levamisole, and morantel (and an H. contortus strain tolerant to
rafoxanide) were fully susceptible to ivermectin. Haemonchus resistance
to ivermectin has been reported following repeated treatments in labora-
tory (Egerton, Suhayda, and Eary 1988) and, at short intervals, field
situations (Carmichael et al. 1987; van Wyk and Malan 1988; van Wyk et
al. 1987; Echevarria and Trindade, 1989). Trichostrongylus colubriformis
resistance under laboratory conditions also has been reported (Giordano,
Tritschler, and Coles 1987).
Efficacy also was demonstrated against all three instars (larvae) of nasal
bots, Oestrus ovis (Roncalli 1984a), and against itch mites, Psorergates

given orally at 200 p.g/kg against nematode parasites of
sheep.
% Reduction
Haemonchus contortus 99 98
H. contortus hypobiotic larvae lOO
H. placei >99 NDb
Ostertagia circumcincta >99 99
O. circumcincta 99
Trichostrongylus axei 99 99
T. colubriformis >99 >99
T. vitrinus >99 96
Cooperia curticei >99 93
C. oncophora 97 96
Gaigeria pachyscelis >99 99
Nematodirus battus >99 >99
N. filicollis 99 98
N. spathiger 98 99
Strongyloides papillosus >99 >99
Chabertia ovina >99 97
Trichuris ovis 97 94
Oesophagostomum columbianum lOO 97
O. venulosum lOO 97
Lungworms
Dictyocaulus filaria >99 96

ovis (Hotson 1983). Ivermectin delivered orally to sheep does not control
other mange infestations.
INJECTABLE FORMULATION
IVOMEC Injection, the same 1.0% w/v solution used in cattle, is also
employed for sheep. The product is administered subcutaneously at
0.5 ml/25 kg body weight, using care to ensure that the skin has been
penetrated before attempting delivery of the dose. It is indicated for the
treatment and control of gastrointestinal nematodes, lungworms, nasal
bots, mange mites, and itch mites.
The dose of 200 p,g ivermectin/kg utilized orally in sheep was selected for
this formulation based on other usages. Registration trials subsequently
confirmed high efficacy against gastrointestinal nematodes and lung-
worms (Table 15.5). Efficacy against Protostrongylus rufescens has also
been demonstrated (Dakkak et al. 1986).

given subcutaneously at 200 ILg/kg against nematode
parasites of sheep.
% Reduction
Haemonchus contortus 99 >99
Ostertagia circumcincta >99 >99
O. trifurcata 100 98
Trichostrongylus axei 100 NOb
T. colubriformis 98 >99
T. vitrinus 100 93
Nematodirus filicollis 92 93
N. spathiger NOb 95
Cooperia curticei 99 >99
Oesophagostomum >99 >99
columbianum
O. venulosum 100 NOb
Chabertia ovina 100 98
Trichuris ovis 87 NOb
Strongyloides papillosus NOb >99
Gaigeria pachyscelis 100 >99
Lungworms
Dictyocaulus filaria >99 >99
Protostrongylus rufescens 100 NOb
a Developing fourth-stage larvae.

Efficacy was confirmed in registration trials against the parasitic larval
stages of the nasal bot, Oestrus ovis.
Infestations caused by mange mites, Sarcoptes scabiei, and itch mites,
Psorergates ovis, also are controlled by ivermectin given subcutaneously.
In contrast to the oral formulation, which does not control Psoroptes ovis,
sheep given the injectable formulation do experience marked reductions
in Psoroptes numbers and clinical signs. A second injection 7 days later
may be required to eliminate all the living mites (Roncalli and Sutherland
1986). These reductions in psoroptic mites on sheep are presumably
related to the longer persistence of ivermectin following subcutaneous
administration, compared to oral administration, as well as to the feeding
patterns of Psoroptes on sheep (Kirkwood 1985).
Use in Goats
ORAL FORMULATION
IVOMEC or ORAMEC Liquid for goats is the same 0.08% w/v solution of
ivermectin as used in sheep, and is to be given at 2.5 mlll0 kg body
weight. Standard drenching equipment can be utilized provided proper

volumes are delivered. The product is indicated for the treatment and
control of gastrointestinal nematodes and lungworms.
The same dose rate of 200 JLg/kg, demonstrated to be efficacious in
sheep, is used in goats. Data from dose-titration (Swan and Gross 1985)
and dose-confirmation trials used for registration of efficacy claims are
summarized in Table 15.6. Ectoparasites affecting goats were not suffi-
ciently responsive to be included on the list of efficacy claims.
In goats, there is insufficient efficacy against nematodes to justify use of

the injectable formulation (lVOMEC Injection) otherwise recommended
in cattle and sheep.
Use in Swine
IVOMEC Injection containing 1.0% w/v ivermectin, the same formulation

recommended for use in cattle and sheep, also is utilized in swine for the
treatment and control of gastrointestinal roundworms, lungworms, kid-
ney worms, lice, and mites. A dilute formulation containing 0.27% w/v
ivermectin is recommended for young pigs.
The recommended dose of the 1.0% formulation is 1 m1l33 kg body
weight (300 JLg ivermectin/kg) adminstered subcutaneously. In young
pigs, especially those below 16 kg for which less than 0.5 ml 1.0%
IVOMEC Injection is indicated, dosing accuracy is important; the use of a

given orally at 200 JLg/kg against nematode parasites of
goats.
% Reduction
Gastrointestinal Nematodes
Haemonchus contortus >99 >99
Ostertagia circumcincta >99 tOO
Trichostrongylus colubriformis >99 >99
Nematodirus spathiger tOO tOO
Chabertia ouina tOO ND b
Strongyloides papillosus >99 tOO
Oesophagostomum tOO tOO
columbianum
Lungworm
Dictyocaulus filaria tOO NDb

226 G.W. Benz, R.A. Roncaili, and S.J. Gross
syringe that can deliver accurately as little as 0.1 ml is recommended.

Alternatively, young pigs should be given 1 ml of the 0.27% formulation
per 9 kg (20 lb) body weight.
Efficacy data used for the registration of nematode efficacy claims are
summarized in Table 15.7. Kidney worms, Stephanurus dentatus, are
controlled (Plue et al. 1986). Further, ivermectin given to sows 7 to 14
days before farrowing effectively controls transmission of Strongyloides
ransomi larvae to newborn pigs.
Infestations of lice (Haematopinus suis) and mange mites (Sarcoptes
scabiei var. suis) were effectively controlled during registration trials. In
mange control programs, all animals in a herd must be treated and a
consistent program of retreatment adhered to (Hogg 1984). Sows prefera-
bly should be treated prior to breeding and 7 to 14 days prior to farrowing
(Courtney, Ingalls, and Stitzlein 1983). Boars should be treated at least
twice annually, although frequency of and need for treatments depend on
exposure. All feeder pigs received for finishing should be treated before
placement in clean quarters; pigs exposed to contaminated soil, premises,
or other pigs may need retreatment if reinfestation occurs.
IVOMEC Injection in swine has a persistent drug level sufficient to
control sarcoptic mite infestations throughout the life cycle, from egg to
adult. Because the ivermectin effect is not immediate, however, care
must be taken to prevent reinfestation from exposure to untreated animals
or contaminated facilities. Generally, pigs should not be moved to clean

given subcutaneously at 300 #£g/kg against nematode
parasites of swine.
% Reduction
Ascaris suum 100 >99
Hyostrongylus rubidus 98 99
Oesophagostomum spp. 96 95
Strongyloides ransomi 99 NOb
Trichuris suis 80 NOb
Lungworms
Metastrongylus spp. 100 NOb
Kidney Worms
Stephanurus dentatus >99 >99
" Oeveloping fourth-stage larvae.
quarters or exposed to uninfested pigs for approximately 1 week after

treatment. Sows should be treated at least 1 week before farrowing to
minimize transfer of mites to newborn pigs. Louse eggs are unaffected by
IVOMEC Injection and may require up to 3 weeks to hatch; louse
infestations developing from hatching eggs may require retreatment.
REFERENCES
Alva-Valdes R, Wallace DH, Holste JE, Egerton JR, Cox JL, Wooden JW,
Barrick RA (1986) Efficacy of ivermectin in a topical formulation against
induced gastrointestinal and pulmonary nematode infections, and naturally
acquired grubs and lice in cattle. Am. J. Vet. Res. 47:2389-2392.
Armour J, Bairden K, Batty AF, Davison CC, Ross D (1985) Persistent
anthelmintic activity of ivermectin in cattle. Vet. Rec. 116:151-153.
Armour J, Bairden K, Ryan WG (1988) Immunity of ivermectin treated cattle to
challenge from helminth parasites in the following season. Vet. Rec. 122:223-
225.
Baggott DG, Batty AF, Ross DB (1986) The control of mature nematode
infections in cattle by sustained delivery ofivermectin. Proceedings of the 14th
World Congress on Diseases of Cattle, Vol. 1, Dublin, Ireland, pp 160-165.
Barragry TB (1987) A review ofthe pharmacology and clinical uses ofivermectin.
Can. Vet. J. 28:512-517.
Barth D (1983) Persistent anthelmintic effect of ivermectin in cattle. Vet. Rec.
113:300.
Barth D, Batty AF, Robin B, Preston JM (1986) Efficacy of a topical formulation
of ivermectin against cattle ectoparasites. Proceedings 14th World Congress of
Diseases of Cattle, Vol 1, Dublin, Ireland, pp 157-159.
Barth D, Preston JM (1985) Efficacy against somatic Strongyloides ransomi
larvae. Vet. Rec. 116:366-367.
Bennett DG (1986) Clinical pharmacology ofivermectin. J. Am. Vet. Med. Assoc.
189:100-104.
Benz GW (1985) Animal health applications of ivermectin. Southwest. Entomol.
(Suppl.) 7:43-50.
Brokken ES, Barth D, Foster AG, Pulliam JD, Wallace DH (1983) Ivermectin, a
new broad-spectrum antiparasitic agent for swine. In Proceedings of the MSD
AGVET Symposium "Recent Developments in the Control of Animal Para-
sites," in association with the XXII World Veterinary Congress, Perth,
Australia, pp 239-258.
Campbell WC (1985) Ivermectin: An update. Parasitol. Today 1:10-16.
Campbell WC, Benz GW (1984) Ivermectin: A review of efficacy and safety. J.
Vet. Pharm. Therap. 7:1-16.
Campbell WC, Fisher MH, Stapley EO, Albers-Schonberg G, Jacobs TA (1983)
Ivermectin: A potent new antiparasitic agent. Science 221:823-828.
Carmichael IH, Soll MD, Scherer H (1985) The use ofivermectin in the control of
bovine thelaziosis. 11th Conference World Association for the Advancement of
Veterinary Parasitology, Rio de Janeiro, Abstract 106.
Carmichael IH, Visser R, Schneider D, Soll MD (1987) Haemonchus contortus
resistance to ivermectin (Letter to Editor). J. S. Afr. Vet. Assoc. 58:93.
Courtney CH, Ingalls WL, Stitzlein SL (1983) Ivermectin for the control of swine
scabies: Relative values of prefarrowing treatment of sows and weaning

treatment of pigs. Am. J. Vet. Res. 44:1220-1223.
Cramer LG, Carvalho LAF, Bridl AA, Amaral NK, Barrick RA (1985) Topically
applied ivermectin: Efficacy against Boophilus microplus in cattle. 11th Confer-
ence World Association for Advancement of Veterinary Parasitology, Rio de
Janeiro, Abstract 126.
Dakkak A, Robin B, Kachani M (1986) Efficacy of ivermectin in the ewe. Rev.
Med. Vet. (Toulouse) 137:781-787 (in French).
Echevarria FAM, Trindade GNP (1989) Anthelmintic resistance by Haemonchus
contortus to ivermectin in Brazil: A preliminary report. Vet. Rec. 124: 147-148.
Eckenhoff B, Cortese R, Wright JC, Wilkinson PK, Zingerman JR, Pope DG
(1987) Osmotically-driven ruminal delivery system for long term rate controlled
delivery of ivermectin. Pharm. Res. (Suppl.) 4:46.
Egerton JR, Suhayda D, Eary CH (1986) Prophylaxis of nematode infections in
cattle with an indwelling rumino-reticular ivermectin sustained release bolus.
Vet. Parasit. 22:67-75.
Egerton JR, Suhayda D, Eary CH (1988) Laboratory selection of Haemonchus
contortus for resistance to ivermectin. J. Parasit. 74:614-617.
Giordano DJ, Tritschler JP, Coles GC (1987) Selection of ivermectin resistant
Trichostrongylus colubriformis in lambs. Proceedings 12th Conference World
Association for Advancement of Veterinary Parasitology, Montreal, Abstract
No.8A-5.
Guillot FS, Wright FC, Oehler D (1986) Concentration of ivermectin in bovine
serum and its effect on the fecundity ofpsoroptic mange mites. Am. J. Vet. Res.
47:525-527.
Hogg A (1984) Eradication of sarcoptic mange in swine with ivermectin. Proceed-
ings 8th International Pig Veterinary Society Congress, Ghent, Belgium, p 206.
Hotson IK (1983) The development of ivermectin as an antiparasitic agent in
sheep. In Proceedings of the MSD AGVET Symposium "Recent Developments
in the Control of Animal Parasites," in association with the XXII World
Veterinary Congress, Perth, Australia, pp 42-55.
Hotson IK, Bliss WJ, Cox JL, Roncalli RA, Sutherland IH (1985) Efficacy of
topically administered ivermectin against cattle parasites. lith Conference
World Association for Advancement of Veterinary Parasitology, Rio de Ja-
neirio, Abstract 112.
Jacobs DE, Fox MT, Ryan WG (1987) Early season parasitic gastroenteritis in
calves and its prevention with ivermectin. Vet. Rec. 120:29-31.
Kirkwood AC (1985) Some observations on the biology and control of the sheep
scab mite Psoroptes ovis (Herring) in Britain. Vet. Parasitol. 18:269-279.
Leaning WHD, Roncalli RA, Brokken ES (1983) The efficacy and safety
evaluation of ivermectin: A new injectable antiparasitic agent for cattle. In
Proceedings of the MSD AGVET Symposium "Recent Developments in the
Control of Animal Parasites," in association with the XXII World Veterinary
Congress, Perth, Australia, pp 25-41.
Plue RE, Brokken ES, Ericsson GF, Foster AG, Wallace DH (1986) A review of
ivermectin efficacy against adult and immature Stephanurus dentatus. Proceed-
ings 9th International Pig Veterinary Society Congress, p 375.
Pope DG, Wilkinson PK, Egerton JR, Conroy J (1985) Oral controlled-release
delivery of ivermectin in cattle via an osmotic pump. J. Pharm. Sci. 74:1108-
1110.
Roncalli RA (1984a) Efficacy of ivermectin against Oestrus ovis in sheep. Vet.

Med. Small Ani. Clinic. 79:1095-1097.
Roncalli RA (1984b) The biology and the control of Dermatobia hominis, the
tropical warble-fly of Latin America. Prevent. Vet. Med. 2:569-578.
Roncalli RA, Hotson IK, Benitez-Usher C, Bridi AA (1984c) Efficacy of iver-
mectin against Boophilus spp. in cattle. Proceedings of 29th Annual Meeting of
American Association of Veterinary Parasitologists, New Orleans, Abstract 12.
Roncalli RA, Sutherland IH (1986) The use of ivermectin injectable against
Psoroptes ovis (Acarina: Psoroptidae) in sheep. Proceedings 31st Annual
Meeting American Association of Veterinary Parasitologists, No. 10.
Ryan WG, Armour J, Bairden K, Fox MT, Jacobs DE (1986) Early season use of
ivermectin to control parasitic gastroenteritis and bronchitis in calves. Proceed-
ings of the 14th World Congress of Diseases of Cattle, Dublin, Ireland, Vol. 1,
pp 185-190.
Soll MD, Carmichael IH, Gross SJ (1987) Control of induced infestations of adult
Amblyomma hebraeum with sustained release ivermectin. Onderstepoort J.
Vet. Res. 54:17-20.
Soll MD, Carmichael IH, Harvey RG (1988) Prophylactic efficacy of sustained-
release ivermectin against induced nematode infestations in cattle. J. S. AIr. Vet.
Assoc. 59:9-11.
Soll MD, Swan GE, Schroder J, Bezuidenhout JD (1984) Efficacy of ivermectin
against the sand tampan (Ornithodoros savignyi). Vet. Rec. 114:70.
Strickland RK, Gerrish RR (1987) Infestivity of Psoroptes ovis on ivermectin-
treated cattle. Am. J. Vet. Res. 48:342-344.
Swan GE, Gross SJ (1985) Efficacy ofivermectin against induced gastrointestinal
nematode infections in goats. Vet. Rec. 117:147-149.
Swan GE, Soll MD, Carmichael IH, Schroder J (1983) Efficacy of ivermectin
against Parafilaria bovicola. Vet. Rec. 113:260.
Taylor SM, Mallon TR, Kenny J (1985) Comparison of early season suppressive
anthelmintic prophylactic methods for parasitic gastroenteritis and bronchitis in
calves. Vet. Rec. 117:521-524.
van Wyk JA, Malan FS (1988) Resistance of field strains of Haemonchus
contortus to ivermectin, closantel, rafoxanide and the benzimidazoles in South
Africa. Vet. Rec. 123:226-228.
van Wyk JA, Malan FS, Gerber HM, Alva RMR (1987) Two field strains of
Haemonchus contortus resistant to rafoxanide. Onderstepoort 1. Vet. Res.
54: 143-146.
Wilkinson PK, Baggott DG, Zingerman JR, Pope DG, Eckenhoff B (1987) In-vivo
performance of a new osmotically driven, long term rate controlled delivery
system for ivermectin in cattle. Pharm. Res. (Supplement) 4:46.
Wright FC, Guillot FS (1984) Effect of ivermectin in heifers on mortality and egg
production of Psoroptes ovis. Am. 1. Vet. Res. 45:2132-2134.
Yazwinski TA, Wood N, Holtzen HM, Presson BL, Kilgore RL (1986) Measuring
the anthelmintic efficacy of a new formulation of ivermectin. Vet. Med. Small
Ani. Clinic. 81:1175-1177.
CHAPTER 16
Use of Abamectin in Cattle

G.W. Benz and J.L. Cox
I. Introduction
II. Efficacy
A. Endoparasite Activity
B. Ectoparasite Activity
I. Introduction
This chapter presents a summary of the efficacy data that form the basis
for the clinical use of abamectin (avermectin B t ) in cattle. The product is
formulated in the same vehicle as used for IVOMEC (ivermectin)
Injection. At the present time, AVOMEC (abamectin) Injection is regis-
tered only in Australia.
At least 2 adequate and well-controlled trials investigated the effect on
each parasite species. Most of the trials were conducted by MSDRL
researchers, though investigators at other laboratories also contributed.
The efficacy percentages presented here are those upon which regulatory
approval was based. Other efficacy data obtained with abamectin but
formulated in a different vehicle, and not relied upon for registration,
have been presented (Benz and Ernst 1979; Egerton et al. 1979; Wescott
et al. 1980).
II. Efficacy
AVOMEC Injection is a sterile solution containing 1.0% w/v abamectin.
The product is indicated for the treatment and control of gastrointestinal
nematodes (including inhibited immature Ostertagia ostertagl), lung-
worms (Dictyocaulus v;v;parus), and sucking lice (Linognathus vitum
(Scott et al. 1985). It is to be administered subcutaneously in front of or
behind the shoulder at 1 mil 50 kg body weight.
16. Use of Abamectin in Cattle 231
A. ENDOPARASITE ACTIVITY
The recommended dose of 200 JLg abamectin/kg was selected upon

completion of dose-titration trials involving nematodes primarily, but also
taking into account data against sucking lice (Linognathus vitulz) and
tropical cattle ticks (Boophilus microplus).
Trial data based on subcutaneous injection of abamectin and used for
registration oflabel-efficacy claims are summarized in Table 16.1. Claims
registered in Australia include the following:
Haemonchus placei (adults and fourth-stage larvae)

Ostertagia ostertagi (adults, fourth-stage larvae, and hypobiotic larvae)
Trichostrongylus axei (adults)
Cooperia spp. (adults and fourth-stage larvae)
Chabertia ovina (adults)
Oesophagostomum radiatum (adults and fourth-stage larvae)
Dictyocaulus viviparus (adults and fourth-stage larvae)
Several of the trials involved comparisons of the efficacy of abamectin

and ivermectin (Table 16.2). Ivermectin and abamectin differ only by
reduction of one double bond, each product utilizes the same vehicle, and
each is administered subcutaneously at the same dose rate. Based on
comparable efficacies, the following additional nematode efficacy claims
based on ivermectin data have been registered in Australia:
TABLE 16.1. Efficacy in registration trials of abamectin

given subcutaneously at 200 ILg/kg against adult and
fourth-stage nematode parasites of cattle.
% Reduction
Parasite Adults Immatures'

Haemonchus placei 100 100
Ostertagia ostertagia >99 >99
O. ostertagi hypobiotic >99
larvae
Trichostrongylus axei >99 NDb
Cooperia spp. >99 100
Chabertia ouina 100 NDb
Oesophagostomum >99 100
radiatum
Lungworms
Dictyocaulus uiuiparus 100 100
• Developing fourth-stage larvae.
232 G.W. Benz and J.L. Cox
TABLE 16.2. Comparative efficacy in registration trials of abamectin and

ivermectin, each given subcutaneously at 200 ILg/kg, against adult and
fourth-stage nematode parasites. of cattle.
Abamectin Ivermectin
% Reduction % Reduction
Parasite Adults Immatures· Adults Immatures
Haemonchus placei 100 100 >99 100
Ostertagia ostertagi 100 >99 >99 >99
O. ostertagi hypobiotic >99 >99
larvae
Trichostrongylus axei 100 NOb >99 NOb
Cooperia spp. >99 100 >99 >99
Chabertia ovina 100 NOb 100 NOb
Oesophagostomum 100 100 100 >99
radiatum
• Oeveloping fourth-stage larvae.
Ostertagia lyrata (adults and fourth-stage larvae)

Trichostrongylus axei (fourth-stage larvae)
Trichostrongylus colubriformis (adults and fourth-stage larvae)
Strongyloides papillosus (adults)
Bunostomum phlebotomum (adults and fourth-stage larvae)
AVOMEC Injection effectively controls infections with Haemonchus
placei, Ostertagia spp., Cooperia spp., and Oesophagostomum radiatum
acquired up to at least 7 days after treatment, and Dictyocaulus viviparus
acquired up to at least 14 days after treatment. Data to substantiate these
claims were obtained using ivermectin (Bremner et al. 1983; Barth 1983;
Armour et al. 1985) as well as abamectin.
Since registration, the uses of abamectin for nematode control have
been reported (Tahir, Holroyd, and Copeman 1986; De Chaneet et al.
1988). The high efficacy of abamectin against oxfendazole-resistant
Trichostrongylus axei in cattle also has been reported (Eagleson and
Bowie 1986).
B. ECTOPARASITE ACTIVITY
The abamectin dose rate confirmed as optimal against nematode parasites

was also found to be highly effective against sucking lice, Linognathus
vituli. Among treated cattle, no live lice were found on Day 14 after
treatment until observations were terminated on Day 56.
Abamectin is sufficiently efficacious against tropical cattle ticks, Bo-
ophilus microplus, to justify an "aid-in-control" label claim. A single
subcutaneous injection at 200 JLg/kg significantly reduces the numbers of
16. Use of Abamectin in Cattle 233
engorged female ticks dropped from treated cattle for at least 21 days, and
the reproductive potential among surviving female ticks is reduced by
interference with egg production.
REFERENCES
Armour J, Bairden K, Batty AF, Davison CC, Ross D (1985) Persistent anthel-
mintic activity ofivermectin in cattle. Vet. Rec. 116:151-153.
Barth D (1983) Persistent anthelmintic effect of ivermectin in cattle. Vet. Rec.
113:300.
Benz GW and Ernst JV (1979) Anthelmintic activities of the B.a fraction of
avermectin against gastrointestinal nematodes in calves. Am. J. Vet. Res.
40: 1187-1188.
Bremner KC, Berrie DA, Hotson IK (1983) Persistence of the anthelmintic
activity of ivermectin in calves. Vet. Rec. 113:569.
De Chaneet GC, Casey R, Dixon FF, Besier RB, Mitchell RK (1988) Effect of
avermectin B. and benzimidazole anthelmintics on worm egg output of treated
cattle. Austral. Vet. J. 65:85-86.
Eagleson JS, Bowie JY (1986) Oxfendazole resistance in Trichostrongylus axei in
cattle in Australia. Vet. Rec. 119:604.
Egerton JR, Ostlind DA, Blair LS, Eary CH, Suhayda D, Cifelli S, Riek RF,
Campbell WC (1979) Avermectins, a new family of potent anthelmintic agents:
efficacy of the B.a component. Antimic. Agents Chemother. 15:372-378.
Scott PG, Burrows RO, Hotson IK, Cox JL (1985) Avermectin B. as an
anti-parasitic agent for cattle. 11th Conference World Association for Ad-
vancement of Veterinary Parasitology, Rio de Janeiro, Abstract 83.
Tahir MS, Holroyd RG, Copeman DB (1986) Treatment of beef calves with
ivermectin and avermectin B. in dry tropical Australia. 6th International
Congress of Parasitology, Brisbane, Australia, Abstract 651.
Wescott RB, Farrell CJ, Gallina AM, Foreyt WJ (1980) Efficacy of avermectin
B.a for treatment of experimentally induced nematode infections in cattle. Am.
J. Vet. Res. 41:1326-1328.
CHAPTER 17
Use of Ivermectin in Horses

W.C. Campbell, W.H.D. Leaning, and R.L. Seward
I. Products
A. Eqvalan paste™, Zimecterin paste™
B. Eqvalan liquid ™
c. Eqvalan injectable ™
II. Antiparasitic efficacy
A. General
B. Parascaris equorum
C. Strongylus vulgaris, S. edentatus, S. equinus
D. Small strongyles
E. Dictyocaulus arnfieldi (Lungworm)
F. Draschia sp. and Habronema sp.
G. Onchocerca cervicalis
H. Other helminths
I. Gastrophilus spp. (bots)
J. Sarcoptes scabei
I. Products
A. EQvALAN PASTE TM; ZIMECTERIN PASTE ™
This is a paste, for oral use, containing 1.87% w/w ivermectin. In the
United States, it is available in a prefilled syringe, calibrated so that each
increment represents the amount of drug needed for 250 lb (114 kg) of
body weight at the rate of 200 ltg/kg. Each syringe contains enough paste
to treat one mature horse. The syringe sold in international markets is
calibrated to deliver the amount of drug needed for 100 kg (220 lb) of body
weight at 200 ltg/kg.
B. EQv ALAN LIQUID ™
This product is a clear, ready-to-use liquid for professional administration

by stomach tube (nasogastric intubation) or as an oral drench. It contains
17. Use ofIvermectin in Horses 235
1% ivermectin and various excipients. The recommended dosage is

200 ILg/kg, and each ml contains enough ivermectin to treat 110 lb (50 kg)
of body weight.
c. EQVALAN INJECTABLE™
Eqvalan injectable ™ is a micellar formulation containing 20 mg of

ivermectin per milliliter of sterile aqueous solution (2.0% w/v) intended
for intramuscular injection. Manufacture and distribution of this product
were suspended in 1984. Adverse reactions have included some that were
severe or even fatal. In most instances contamination of the injection site
with bacteria (Clostridium sp.) appears to have been involved, while in
others the reaction appears to have been of the anaphylactoid type and
was associated with inadvertent intravenous injection or the presence of
polysorbate 80 in the micellar formulation.
II. Antiparasitic Efficacy

A. GENERAL
Most of the early studies on the efficacy of ivermectin in horses involved

intramuscular injection of the drug. Because the injectable product is no
longer commercially available its efficacy will not be discussed in detail
here, and readers are referred to reviews by Campbell and Benz (1983);
Egerton, Seward, and Robin (1984); Leaning (1984); Madigan (1984);
Weiss (1984); Slocombe and colleagues (1984); Klei and colleagues (1984);
and Campbell (1985).
The efficacy of parenterally and orally administered ivermectin in
horses is almost identical. For a host species such as the horse-where
widely different experimental protocols are used and where many species
of parasites are involved-the compilation of data is cumbersome; and for
practical purposes it may be more useful to tabulate the species or
"claims" for which approval has been granted by major governmental
regulatory agencies. This approach has the advantage not only of
simplicity but of subsuming the elements of statistical probability and
host safety that are integral parts of the registration process. Table 17.1
lists the "approved claims" for the oral use of ivermectin at 200 ILg/kg in
horses. The following summary of efficacy against individual species is
limited to orally administered drug, except where data obtained using the
injectable formulation are needed to introduce or supplement d~ta
obtained with the paste.
B. PARASCARIS EQUORUM
When given orally in a paste formulation, ivermectiil at 200 ILg/kg totally

eliminated the passage of P. equorum eggs in naturally infected horses
236 W.C. Campbell, W.H.D. Leaning, and R.L. Seward
TABLE 17.1. Horse parasites for which ivermectin oral

paste has been approved at a dosage of 0.2 mg/kg. 1
Large strongyles
Strongylus vulgaris, adult and arterial larval stage
S. edentatus, adult and tissue stage
S. equinus, adult
Triodontophorus spp., adult
Small strongyles
Cyathostomum spp., adult and L4 stage
Cylicocyclus spp., adult and L4 stage
Cylicostephanus spp., adult and L4 stage
Cylicodontophorus spp., adult and L4 stage
Gyalocephalus sp., adult and immature 2
Other nematodes
Parascaris equorum, adult, L3 and L4 stage
Oxyuris equi, adult and L4 stage
Trichostrongylus axei, adult
Habronema muscae, adult and cutaneous L3 stage
Draschia spp., cutaneous L3 stage 2
Onchocerca sp., microfilaria
Dictyocaulus arnfieldi, adult and L4 stage
Bots
Gastrophilus spp., oral and gastric stages
I Unless otherwise specified, approval refers to the Food and Drug
Administration of the United States of America.

2 Approved in country or countries other than United States.
(Corba et al. 1986), illustrating that the oral paste, like the injectable
solution, is active against adult P. equorum. An effect on the passage of
eggs was also observed by Puccini and colleagues (1984). In another
study, fecal ascarid egg counts in 14 naturally infected horses were
negative following treatment with ivermectin oral paste and remained at
zero for 10 to 12 weeks following treatment, while egg counts in the 6
control horses were substantially unchanged (E.T. Lyons, unpublished
data). Efficacy against adult ascarids was demonstrated more directly and
definitively in controlled studies in which ascarid burdens were deter-
mined at necropsy (French et al. 1988b; DiPietro et al. 1987). In both
studies ivermectin paste at 200 jLg/kg was 100% effective against the adult
worms.
Ivermectin is also active against the immature stages of P. equorum. In
two studies in naturally infected ponies, larval burdens were determined
at necropsy following treatment with the paste at 200 jLg/kg. When
necropsy was performed 2 weeks after treatment, the reduction in lung
larval burden was 100% (French et al. 1988b) and the reduction in
intestinal larval burden was also 100% (French et al. 1988b; DiPietro et al.
1987). When necropsy was delayed until 5 weeks after treatment, larvae
were recovered from the intestines; and, since treatment is highly active
against the early migrating stage, this probably reflected reinfection,
17. Use of Ivermectin in Horses 237
despite effort taken in one of the trials to prevent it. In one of the trials,
treatment was not associated with clear-cut improvement in clinical
condition but was associated with improved weight gain (French et al.
1988b).
To define the efficacy ofivermectin against immature P. equorum more
precisely, 2 trials were conducted on experimentally induced infections.
Groups of 6 horses were treated with ivermectin paste or liquid at
200 JLg/kg, or were maintained as untreated or vehicle-untreated controls
(DiPietro et al., in press; French et al., in press). Treatment was
administered on Day 11 of infection, when the majority of the parasites
would be in the migratory L3 larval stage. (Necropsy of indicator horses
at that time revealed larvae in the lungs but not in the small intestine).
At necropsy, 14 days after treatment, the efficacy of ivermectin against
P. equorum was determined to be 100% in both trials and with both
formulations (P < 0.001). It is not possible to conclude from the data
whether all of the larvae were killed in the lungs or on early entry into the
small intestine.
To determine efficacy against the later (L4) stage of P. equorum, 4 trials
were conducted with experimentally induced infections (A. Batty, unpub-
lished data; J.L. Duncan, unpublished data; T.R. Bello 1985; and
DiPietro et al., in press). Treatment was given on Day 28 or 29 of
infection. At necropsy on Day 42 or 43 of infection (14 days after
treatment), the immature ascarids in the small intestine were counted,
revealing reductions of 62%,95%, greater than 99%, and greater than 99%
in the paste-treated groups in the 4 trials, respectively. In 2 of the trials,
ivermectin liquid was also tested, and reductions of greater than 99%
were recorded in both (DiPietro et at., in press; J.L. Duncan, unpub-
lished data). Using an appropriate statistical method (Friedman's test) to
allow for heterogeneity of variances among treatments, the overall
reduction was calculated as 98.4% (P > 0.001). In the trial by DiPietro and
colleagues (in press), gross examination oflung and liver revealed similar
lesions in the treated and untreated foals, but the ascarids recovered from
the treated animals were significantly smaller than those from the
controls. Thus treatment evidently affected the parasites adversely,
perhaps by interfering with feeding, but did not kill them.
Thus the paste and liquid ivermectin formulations have been shown to
be highly active against both the early (L3) tissue phase and against the
later intestinal (L4) phase. The significance and differences of opinion
associated with this efficacy have been discussed by Boraski (1987).
c. STRONGYLUS VULGARIS, S. EDENTATUS, S. EQUINUS
The oral paste formulation, used at 200 JLg/kg, was 100% active against
adult S. vulgaris, as was the injectable formulation which was tested in
the same experiment (Klei et al. 1984).
The migratory fourth stage larvae of S. vulgaris are of special im-
portance because (1) they cause lesions of certain arteries (especially the
mesenteric), often giving rise to serious clinical signs; and (2) they are
particularly hard to eliminate by chemotherapeutic means. Extensive
studies by Slocombe and his colleagues (Slocombe et al. 1982; Slocombe
and McGraw 1984; Slocombe et al. 1984) showed that the injectable
formulation of ivermectin is highly active against the migratory fourth
stage larvae. He further showed that efficacy against late fourth stage
larvae may not be demonstrable if necropsy is performed only 2 weeks
after treatment; and he cautioned that treatment appeared to cause some
larvae to become situated more deeply in the wall of the ileocolic artery,
prompting a concern about the eventual fate of such larvae and the impact
of their putative slow disintegration within the arterial wall-especially
after repeated cycles of infection and treatment. Studies by Klei and his
colleagues (Klei et al. 1984a; Klei et al. 1984b) confirmed that the interval
between treatment and necropsy is critical for demonstrating the efficacy
of ivermectin against the migratory larvae of S. vulgaris. In their
experience, the oral paste (and the injectable formulation) at 200 #Lg/kg
was 99% effective against 8-week-old larvae, as revealed by necropsy 5
weeks after treatment. Klei also commented on the potential impact of
dead or dying larvae on the arterial walls. He noted,
Killing S. vulgaris larvae with ivermectin not only promoted a remarkable
resolution of arterial lesions, but also prevented their progression into the lesions
observed in the nontreated controls. Although ivermectin clearly kills intravascu-
lar S. vulgaris larvae, nematode bodies persist for some time following treatment,
as indicated by the numbers of dead larvae recovered. It is interesting to note that
the presence of the inactive or paralyzed worms alone does not provoke the
arterial lesions associated with S. vulgaris migrations. Observations of lesions in
livers and lungs in control animals confirm and extend previous observations that
S. vulgaris migrations are not restricted to the predilection sites in the mesenteric
vasculature. The severity and occurrence of lesions in extravascular sites were
markedly reduced in the ivermectin-treated foals, indicating that resolution of
these lesions also occurs within the 5-week period following treatment.
Repeated cycles of infection and ivermectin therapy have not resulted in
clinical disease or unique lesions (T.R. Klei, unpublished data). It is not
known whether ivermectin is active against the earliest parasitic (L3)
stage of S. vulgaris, but there is evidence that the drug (at least when
given intramuscularly) is effective prophylactically. A foal treated 2
weeks before inoculation with larvae and necropsied 4 weeks after
inoculation had no evidence of arterial infection, whereas the arteries of
control foals showed clear signs of the infection (Slocombe et al. 1984).
Apparently the ivermectin persisted in sufficient concentration to affect
the third or fourth stage larvae at some time during the 4-week period
following inoculation with infective larvae.
Ivermectin was tested at O.2#Lg/kg against adult S. edentatus in horses.
In 5 trials, side-by-side comparison of the paste and injectable formula-
tions gave reductions in worm burden of greater than 99% and 100%,
respectively (T.R. Klei, D.H. Wallace, T.A. Yazwinski, and J.L. Dun-
can, unpublished data). Against the L4 migratory stage of this species, an
efficacy of 100% was recorded for both formulations (J.L. Duncan,
unpublished data).
In 4 trials, the paste and injectable formulations were 100% effective in
removing adult S. equinus at a dosage of 0.2 ILg/kg (T.R. Klei, T.A.
Yazwinski, and D.H. Wallace, unpublished data).
D. SMALL STRONGYLES
The efficacy of ivermectin paste against strongyles in general, as deter-

mined by fecal examination, was reported by Asquith, Plue, and Seward
(1983). Baker and colleagues (1984) gave ivermectin at 200 ILg/kg in the
paste formulation (or the injectable formulation) to horses in which
benzimidazole treatment had already been shown to have little effect on
the output of strongyle eggs. The egg counts were reduced by 100% at 14
days after treatment, and remained reduced by 90% or more for at least 2
months. Similarly, Burrows, Thomson, and Lindsey (1985) used fecal
examination to demonstrate the efficacy of the paste formulation against
benzimidazole-resistant small strongyles and showed that egg counts
remained zero for at least 3 weeks. French and colleagues (1985) reported
that treatment with the paste formulation at 200 ILg/kg resulted in egg
counts "at or near zero for 8 to 10 weeks." These data have been
supplemented by data from controlled trials (based on necropsy and
recovery of parasites from treated and untreated horses) in which the:
efficacy of the paste formulation was compared directly with that of the'
injectable product. Against adult Cyathostomum spp., both formulations
were more than 99% effective (D.H. Wallace and J.L. Duncan, unpub-
lished data). Against adult Poteriostomum the injectable product was 63%
effective while the paste was 100% effective (T.A. Yazwinski, unpub-
lished data). Against unidentified species of small strongyle, immature
and mature, both formulations were more than 99% effective (D.H.
Wallace and J.L. Duncan, unpublished data). Thus the activity of the
paste formulation at least equals the activity of the injectable product in
these trials and in the trials previously reported (for references, see
Campbell and Benz, 1984).
E. DICTYOCAULUS ARNFIELDI (LUNGWORM)
Ivermectin paste was administered orally (200 ILg/kg) to ponies experi-

mentally infected with D. arnfieldi (Britt and Preston 1985). On the basis
of fecal examination and necropsy, the treatment was judged 100%
effective against the adult and immature stages (including inhibited
forms). Lyons, Drudge, and Tolliver (1985) also reported evidence of
efficacy against D. arnfieldi in donkeys.
F. DRASCHIA SP. AND HABRONEMA SP.
The cutaneous lesions ("summer sores") caused by third stage larvae of

Draschia sp. and Habronema sp. have been notoriously difficult to cure
by means of chemotherapy, but Herd and Donham (1984) showed that 1
or 2 intramuscular doses of ivermectin at 200 JLg/kg were highly effective.
A lindane ointment was used to prevent reinfection, and under these
circumstances "the typical summer sore was replaced by healthy pink
granulation tissue at 7 days and this healed after 1 to 5 weeks." Biopsy
was used to demonstrate the absence of larvae after treatment. One dose
was sufficient in 86% of the 31 horses treated, while the remaining
horses"":""which apparently became reinfected despite precautions-
required a second dose. In a subsequent trial with the oral paste
formulation, the results were difficult to assess quantitatively-because
of spontaneous cures in some of the untreated control animals-but did
support the finding of efficacy against cutaneous Habronema sp. in horses
(V.E. Reuter-Dallman, unpublished).
G. ONCHOCERCA CERVICALIS
Ivermectin is not known to be active against the adult stage of any species
of Onchocerca in any host species, and its lack of efficacy in the case
of Onchocerca spp. in the horse has been documented (Lyons, Drudge,
and Tolliver 1988). However, its activity against the skin-dwelling micro-
filariae of the genus Onchocerca has been known since Egerton and
colleagues (1981) demonstrated that a single subcutaneous injection at
200 JLg/kg was highly effective against the microfilariae of o. cervicalis in
horses. Subsequent studies have shown that intramuscular administration
of the same dosage is highly effective; and Herd and Donham (1983) have
reviewed the parasitological and clinical significance of the finding.
The efficacy of the paste formulation against the microfilariae of O.
cervicalis has also been reported (French et al. 1988a; Foil et al., 1985;
and Pollitt et al. 1986). In these studies, treatment resulted in both the
disappearance of microfilariae from the skin and the resolution of the
associated lesions. In the experience of Pollitt and colleagues (1986), "a
single oral dose of ivermectin caused all microfilariae of Onchocerca spp.
to disappear from the skin and all the horses showed marked clinical
improvement of their dermatitis 1 to 2 weeks after treatment. By 4 to 6
weeks it was impossible to determine the original sites of the lesions."
These authors went on to point out that ivermectin did not reduce lesions
attributed to hypersensitivity to midge bites, and therefore it could be
used in horses to differentiate between that form of dermatitis and
onchocercal dermatitis. The presence of o. cervicalis microfilariae in the
skin is by no means always associated with dermatitis, but where
dermatitis occurs, it can be treated successfully with ivermectin.
H. OTHER HELMINTHS
Ivermectin at 200 p.g/kg, given orally in the paste formulation, was
effective against Strongyloides westeri in foals (Ryan and Best 1985;
Schlichting and Stoye 1985). This observation is in accord with earlier
studies using the injectable formulation (see reviews cited previously).
Ponies with naturally acquired infections of Setaria equina were treated
intramuscularly with ivermectin. Dosages of 200 to 500 p.g/kg gave
moderate efficacy (80%-88%) (Klei, Torbert, and Ochoa 1980). Iver-
mectin was given to a pony with severe jaw disease attributed to invasion
of the tissue by Micronema deletrix, but the disease was not alleviated
(Keg et al. 1984). A pony continued to pass Anoplocephala spp. eggs in
the feces after treatment with ivermectin paste (Torbert, Kramer, and
Klei 1982), as would be expected in view of the lack of efficacy of this
drug against tapeworms in general.
I. GASTROPHILUS SPP. (BOTS)
The efficacy of parenterally administered ivermectin against the parasitic

larvae (stomach bots) of Gastrophilus intestinalis and G. nasalis was
reported by Egerton and colleagues (1981), who showed that sub-
cutaneous dosages as low as 20 p.g/kg were highly effective. The
effectiveness of the broad-spectrum intramuscular dosage of 200 p.g/kg
has been demonstrated by several groups (see reviews cited previously).
The oral paste is as effective as the injectable formulation, as illustrated
by a trial in which a single dose of the paste at 200 p.g/kg removed 99% of
G. intestinalis larvae and 100% of G. nasalis larvae (Torbert, Kramer, and
Klei 1982) and a trial in which the same dosage removed 100% of second
and third stage larvae of G. intestinalis (Britt and Preston 1985).
J. SARCOPTES SCABIEI
Ivermectin was highly effective against Sarcoptes scabiei (of goat origin)
in experimentally infected donkeys, Equus asinus (Abu-Samra et al.
1985). Similarly, systemic ivermectin treatment in conjunction with a
topical acaricidal dressing was reported curative against sarcoptic mange
(perhaps of fox origin) in horses (Christensson et al. 1984).
REFERENCES
Abu-Samra MT, Ali BH, Musa BE, Ibrahim KEE (1985) Experimental infection
of the domestic donkey (Equus asinus asinus) with a goat strain of Sarcoptes
scabiei, and treatment with ivermectin. Acta Tropica 42:217-224.
Asquith RL, Plue RE, Seward RL (1983) Field performance of the equine
parasiticide, ivermectin, in an oral paste. Equ. Vet. J. 3:90-91.
Baker NF, Miller IE, Madigan JE, Fulton R, Seward RL (1984) Anthelmintic
action of ivermectin, oxibendazole, and pyrantel pamoare against thiaben-
dazole-resistant small strongyles of horses. Equ. Pract. 6:8-19.
Bello TR (1985) Proc. 30th Ann. Mtg., Am. Assoc. Vet. Parasit., University of
North Carolina.
Boraski EA (1987) Ivermectin, efficacy against ascarids-new information. Calif.
Vet. 41:16-17.
Britt DP, Preston JM (1985) Efficacy of ivermectin against Dictyocaulus arnfieldi
in ponies. Vet. Rec. 116:343-345.
Burrows RO, Thomson BM, Lindsey MJ (1985) Efficacy of ivermectin against
nematodes of horses including small strongyles resistant to benzimidazole.
Austral. Vet. J. 61:343-344.
Campbell WC (1985) Ivermectin: An update. Parasit. Today 1:10-16.
Campbell WC, Benz GW (1983) Ivermectin: A review of efficacy and safety.
J. Vet. Pharm. & Therap. 7:1-16.
Christensson D, Lindstedt E, Wierup M, Aropsenius J (1984) Sarcopotic mange in
horses in Sweden. Svensk Vet. 36:15-17 (in Swedish).
Corba J, Andrasko H, Stoffa P, Holakovsky P (1986) Efficacy of Eqvalan™ and
Panacur™ paste against gastrointestinal nematodes of horses. Veterinarstvi
36:79-80 (in Slovak).
DiPietro JA, Lock TF, Todd KS, Davis JL (in press) Evaluation of ivermectin for
larvacidal effect in experimentally induced Parascaris equorum infections in
pony foals. Am. J. Vet. Res.
DiPietro JA, Lock TF, Todd KS, Reuter VE (1987) Evaluation ofivermectin paste
in the treatment of ponies for Parascaris equorum infections. J. Am. Vet. Med.
Assoc. 190:1181-1183.
Egerton JR, Brokken ES, Suhayda D, Eary CH, Wooden JW, Kilgore RL (1981)
The antiparasitic activlty of ivermectin in horses. Vet. Parasit. 8:83-88.
Egerton JR, Seward RL, Robin B (1984) Ivermectin as an antiparasitic agent for
horses. Proc. MDS AGVET Symposium: Recent Developments in the Control
of Animal Parasites, XXII World Veterinary Congress, Perth, Australia, Aug.
25-26, 1983, pp. 49-55.
Foil LD, Klei TR, Miller RI, Foil CS, French DD (1985) Pathogenesis of equine
ventral midline dermatitis and its association with Onchocerca cervicalis and
Haematobia irritans. Abst. 196, 11th Conf. World Assoc. Adv. Vet. Parasitol.,
Rio de Janeiro, Brazil, Aug. 5-9, 1985.
French DD, Klei TR, Chapman MR, Torbert BJ (1985) Field evaluation of
ivermectin against benzimidazole resistant strongyles in Louisiana horses.
Abst. 197, 11th Conf. World Assoc. Adv. Vet. Parasitol.
French DD, Klei TR, Foil CS, Miller RI, Foil LD, Chapman MR, McClure JJ
(1988a) Efficacy of ivermectin in paste and injectable formulations against
microfilariae of Onchocerca cervicalis and resolution of associated dermatitis in
horses. Am. J. Vet. Res. 49:1550-1554.
French DD, Klei TR, Taylor HW, Chapman MR, Wright FR (1988) Efficacy of
ivermectin in the oral paste formulation against naturally acquired adult and
larval stages of Parascaris equorum in pony foals. Am. J. Vet. Res. 49:1000-
1003.
French DD, Klei TR, Chapman MR and Taylor HW (in press) Efficacy of
ivermectin in the oral paste and oral drench formulations against migrating
larvae of Parascaris equorum. Am. J. Vet. Res.
Herd RP, Donham JC (1983) Efficacy ofivermectin against Onchocerca cervicalis

microfilarial dermatitis in horses. Am. J. Vet. Res. 44:1102-1105.
Herd RP, Donham JC (1984) Control of equine cutaneous nematodiasis by
ivermectin. Proc. MSD AGVET Symposium: Recent Developments in the
Control of Animal Parasites, XXII World Vet Congr., Perth, Australia, Aug.
25-26, 1983, pp 286-295.
Klei TR, Torbert BJ, Chapman MR and Turk MAM (1984a) Efficacy ofivermectin
in injectable and oral paste formulations against eight-week-old Strongylus
vulgaris larvae in ponies. Am. J. Vet. Res. 45:183-185.
Klei TR, Torbert BJ, Onchoa R (1980) Efficacy of ivermectin (22,23-dihydro-
avermectin B\) against adult Setaria equina and microfilariae of Onchocerca
cervicalis in ponies. J. Parasit. 66:859-861.
Klei TR, Turk MAM, Torbert BJ, Chapman MR (1984a) Ivermectin in parenteral
and oral paste formulations: efficacy against gastrointestinal parasites, migrat-
ing larvae of Strongylus vulgaris, and post-treatment responses of infected
ponies. Proc. MSD AGVET Symposium: Recent Developments in the Control
of Animal Parasites, XXII World Vet., Congr., Perth, Australia, Aug. 25-26,
1983, pp 271-285.
Leaning WHD (1984) The efficacy and safety evaluation of ivermectin as a
parenteral and oral antiparasitic agent in horses. In Milne FJ (ed) Proceedings of
the Twenty-Ninth Annual Convention of the American Association of Equine
Practitioners, Las Vegas, Nevada, Dec. 1983.
Lyons ET, Drudge JH, Tolliver SC (1985) Ivermectin: treating for naturally
occurring infections of lungworms and stomach worms in equids. Vet. Med.
80:58-64.
Lyons ET, Drudge JH, Tolliver SC (1988) Verification of ineffectual activity of
ivermectin against adult Onchocerca spp. in the ligamentum nuchae of horses.
Am. J. Vet. Res. 49:983-985.
Madigan JE (1984) Ivermectin use in the horse, reviewed. Calif. Vet. 38:29-
32,35.
Pollitt CC, Holdsworth PA, Kelly WR, Meacham CS, Sheahan B (1986) Treat-
ment of equine onchocerciasis with ivermectin paste. Austral. Vet. J. 63:152-
156.
Puccini V, Tassi P, Lo Storto E, Troiano P (1984) Field evaluation ofthe efficacy
of ivermectin paste against intestinal helminths in horses. Obiettivi e Documenti
Vet. 5:53-55 (in Italian).
Ryan WG, Best PJ (1985) Efficacy of ivermectin paste against Strongyloides
westeri in foals. Vet. Rec. 117: 169-170.
Schlichting CK, Stoye, M (1985) Effect of ivermectin on Strongyloides westeri in
foals. Prakt. Tierarzt. 66:128-130 (in German).
Slocombe JOD, McCraw BM, Pennock PW and Vasey J (1982) Effectiveness of
ivermectin against later 4th-stage Strongylus vulgaris in ponies. Am. J. Vet.
Res. 43: 1525-1529.
Slocombe JOD, McCraw BM, Pennock PW, Vasey J (1984) Ivermectin in the
control of Strongylus vulgaris in the equine. Proc. MSD AGVET Symposium:
Recent Developments in the Control of Animal Parasites, XXII World Vet.
Congr., Perth, Australia, Aug 25-26, 1983, pp 265-270.
Slocombe JOD and McGraw BM (1984) Evaluation of ivermectin against later
4th-stage Strongylus vulgaris in ponies at two and five weeks after treatment.
Canadian J. Compo Med. 48:343-348.
Torbert BJ, Kramer BS, Klei TR (1982) Efficacy of injectable and oral paste
formulations of ivermectin against gastrointestinal parasites in ponies. Am. J.
Vet. Res. 43:1451-1453.
Weiss J (1984) Value of ivermectin in treating certain parasitoses of horses. Rev.
Med. Vet. 135:425-433 (in French).
CHAPTER 18
Use of Ivermectin in Dogs

and Cats
W.C. Campbell
I. DOGS
A. Products
1. Heartgard 30
2. Heartgard 30 chewable
B. Antiparasitic efficacy
1. Endoparasites
a. Dirofilaria immitis (heartworm)
b. Dipetalonema reconditum
c. Toxocara canis
d. Toxascaris leonina
e. Ancylostoma caninum (hookworm)
f. Uncinaria stenocephala (hookworm)
g. Strongyloides stercoralis
h. Filaroides osleri
i. Capillaria spp.
j. Trichuris vulpis (whipworm)
2. Ectoparasites
a. Sarcoptes scabei
b. Demodex canis
c. Otodectes cynotis
d. Ctenocephalides cati
II. CATS
A. Products
B. Antiparasitic efficacy
1. Endoparasites
a. Aelurostrongylus abstrusus
b. Toxocara cati
c. Ancylostoma spp.
d. Other helminths
2. Ectoparasites
a. Nodoedres cat;
b. Otodectes cynotis
c. Ctenocephalides cati
246 W.C. Campbell
I. Dogs
A. PRODUCTS
1. Heartgard 30™
This is a product for the prevention of heartworm disease in dogs. It is for
use by, or on the order of, a licensed veterinarian, and consists of
color-coded packages of white, hard tablets containing ivermectin in the
amount of 65 ILg (blue code, for dogs weighing up to 25Ib), 136 ILg (green
code, for dogs weighing 26 to 50 lb), or 272 ILg (brown code, for dogs
weighing 51 to 100 lb). When administered according to this schedule, the
tablets provide a minimum ivermectin dosage of 6 ILg/kg.
The product is to be administered at monthly intervals, beginning
within one month after the first expected exposure to mosquitoes and
ending within a month after the last potential exposure. The indication for
use in heartworm prevention is based on efficacy studies summarized
below.
The product is not recommended for dogs under 6 weeks of age.
Certain dogs, particularly Collies, are exceptionally susceptible to iver-
mectin toxicity (see Chapter 10). In such dogs, adverse reactions have
been reported when ivermectin was given under experimental conditions
at 200 ILg/kg but not when given at 50 ILg/kg.
2. Heartgard 30 Chewable ™
The chewable formulation consists of tablets that are designed to be
readily accepted by dogs when proffered to them, thus obviating the need
to insert the medication into the dog's mouth. In other respects the
product's use is similar to that of Heartgard 30™ .
B. ANTIPARASITIC EFFICACY
1. Endoparasites
a. Dirofilaria immitis (heartworm):
Immatures. Ivermectin prevents the maturation of the immature (pre-

cardiac) stages of heartworm, thus preventing heartworm disease. The
studies that have demonstrated this efficacy are listed in Table 18.1 and
have been reviewed in detail elsewhere (Campbell 1987).
When treatment (50 ILg/kg) was given 1 day after inoculation of dogs
with infective larvae, heartworm subsequently failed to appear in the
heart (Blair, Jacob, and Ewanciw 1982). Because the third stage larva is
believed to molt about 2 days after inoculation, the success of treatment
was probably due to efficacy against the third stage larva. It is possible,
however, that the quantity of drug remaining in the dog's tissues after the
worms had molted was sufficient to affect the fourth stage directly.
18. Use of Ivermectin in Dogs and Cats 247
TABLE 18.1. Prophylactic efficacy of ivermectin against D. immitis when given

to experimentally infected dogs at various intervals after inoculation.
Time of Dosage (t-tg/kg) Efficacy' Investigators
Treatment
1-2 days 50 po x 1 100% Blair et al. 1982
1 month 3.0 po x 12 Poor McCall et al. 1986
3.0 po x 1 None Paul et al. 1986a
1.0 po x 1 Poor McCall et al. 1986
1.0 po x 1 Poor Paul et al. 1986a
2.0 po x 1 100% McCall et al. 1986
3.0 po x 12 100% McCall et al. 1986
3.3 po x 1 High Paul et al. 1986a
3.0 po x 1 100% McCall et al. 1981
12.5 po x 1 100% McCall et al. 1981
50poxl 100% Blair and Campbell 1980
50 po x 1 100% McCall et al. 1981
200 po x 1 100% McCall et al. 1981
200 sc x 1 100% McCall et al. 1981
400 po x 1 100% McCall et al. 1981
1.5 months 2.0 po x 1 Poor Paul et al. 1986a
2.0 po x 1 Variable McCall et al. 1986
3.3 po x 1 High 3 Paul et al. 1986b
6.0 po x 1 100% Paul et al. 1986b
12 po x 1 High4 Paul et al. 1986b
25 po x 1 100% Paul et al. 1986b
50 po x 1 100% Paul et al. 1986b
2 months 3.3 po x 1 Variable McCall et al. 1986
6.0 po x 1 100% McCall et al. 1986
12poxl High McCall et al. 1986
25 po x 1 100% McCall et al. 1986
50 po x 1 100% McCall et al. 1986
50 po xl 100% Blair and Campbell. 1980
200 po x 5 100% Egerton et al. 1980
3 months 50 po x 1 Variable Blair and Campbell 1980
50 po x 1 Variable McCall et al. 1981
Abbreviations: po. orally; sc, subcutaneously.
, 100%, no worms found in any treated animal; high, 95% to 100% reduction in all treated
animals; variable, 100% reduction in some treated animals and <95% reduction in others;
poor, <95% reduction in all treated animals.
2 Dogs held for 1 year after treatment instead of usual 1/2 year.
3 Two male worms in 1 of 8 dogs.
4 One male worm in 1 of 8 dogs.
When treatment is given 1 month after inoculation of dogs with

infective larvae, the subsequent absence of worms in the heart can be
attributed to efficacy against the fourth stage larvae (references in Table
IS.1). This stage is extraordinarily susceptible to the action of
ivermectin-perhaps more so than any stage of any other helminth.
Single oral doses as low as 1 JLg/kg have been highly active, but
248 W.C. Campbell
administration of such a dosage has not given uniform results, and in

practice of dosage of 6 ILg/kg is used to ensure complete prophylaxis in all
cases.
A marked falling-off in efficacy is observed when treatment is delayed
until 3 months after inoculation (Table 18.1), and this probably reflects the
increasing proportion of the worm population that has molted to the fifth
stage and become refractory to the drug.
Repeated cycles of infection and ivermectin treatment have been used
to demonstrate the immunogenicity of immature D. immitis (Blair and
Campbell 1981; Blair, Jacob, and Ewanciw 1982).
Adults. In contrast to the exceptional susceptibility of the immature
stages in the body tissues of infected dogs is the apparently total
refractoriness of the adult worms in the lumen of the heart and associated
blood vessels. Even dosages approximating the maximum tolerated
dosage for dogs have not resulted in the death of the adult worms. The
adult worms are undoubtedly exposed to the drug because the drug is
demonstrably present in the blood plasma of treated animals in general.
The drug may not reach the inside of the worms, however, and may not
affect them adversely even if it does. Ivermectin does not kill or paralyze
adult heartworm in vitro (Anantaphruti et al. 1982); and it has also been
shown that, under in vitro conditions, radiolabeled ivermectin does in fact
get widely distributed in the tissues of the worms but does not affect them
adversely in any visible way (R. E. Howells, unpublished data). Despite
these indications of the insusceptibility of adult D. immitis to ivermectin,
certain observations suggest that the drug may have a suppressive effect
on the reproduction of the worm following in vivo treatment (Anantaph-
ruti et al. 1982; Knight et al. 1986; Lok et al. 1988). The nature of this
effect is unclear. While Anantaphruti and colleagues observed a scarcity
of advanced microfilariae in the anterior part of the uteri, Lok and
colleagues observed an accumulation of fully developed microfilariae and
a decrease in the early embryonic stages. The latter authors suggested
that treatment of the host dog with ivermectin might block an early stage
in the embryogenesis of Dirofilaria within that host, without preventing
the more advanced embryos from continuing their development into
microfilariae. Alternatively, or in addition, the accumulation of fully
developed microfilariae within the uteri might be due to blockade of the
mechanism by which microfilariae are expelled from the vulva (Lok et al.
1988).
Microfilariae. The microfilariae of D. immitis are highly susceptible to
ivermectin, and the use of ivermectin as a microfilaricide has been
discussed in detail elsewhere (Jackson 1987). Many authors have re-
ported the suppression of peripheral microfilaremia after the treatment of
dogs with ivermectin at various dosages (Table 18.2). This could reflect
paralysis and sequestration of microfilariae, but evidence of killing-at
TABLE 18.2. Dosages of ivermectin found

capable of reducing or eliminating
microfilariae of D. immitis in dogs. Degree
and duration of effect depend on various
factors; see text.
Dosage Investigators
(,.t.g/kg)
Various Jackson and Seymour 1981
2.0 po x 31 Schlotthauer et al. 1986
3.lpoXl 2 Blair et al. 1983
10 po x 31 Schlotthauer et al. 1986
12.5 po x 1 Blair et al. 1983
12.5 po x 1 Swartz 1985
50 po x 12 Blair et al. 1983
50 po xl Boreham et al. 1983
50 po x 12 Jackson 1984
50 po x 12 Jackson et al. 1986
200 po x 12 Blair et al. 1983
200 po x 1 Anantaphruti et al. 1982
200 sc x 1 Plue et al. 1983
200 sc x 1 Simpson and Jackson 1985
250 po x 12 Knight et al. 1986
400 po x 31 Schlotthauer et al. 1986
500scxl McManus and Pulliam 1984
Abbreviations: po, orally; sc, subcutaneously.
1 At intervals of approximately 1 month.
2 Preceded by adulticide treatment in at least some
groups.
least at higher dosages-has been found by examination of various

visceral tissues at biopsy or necropsy of treated dogs (McManus and
Pulliam 1984; Simpson and Jackson 1985).
Following the administration of a single dose of ivermectin at 250 or
500 J-tg/kg per os (p.o.), suppression of microfilaremia was barely
detectable at 4 hours, significant at 8 hours, and very marked at 18 hours
(Knight et al. 1986; McManus and Pulliam 1984). The time required for
suppression following lower dosages has not been determined precisely,
but 90% of dogs receiving 50 J-tg/kg p.o. became microfilaria-negative
within 3 weeks, and the remainder became negative within 4, or (in a few
cases) more weeks (Jackson 1987). Of the 121 dogs so treated, 96%
became negative after a single dose, and 100% were negative after
receiving mUltiple doses.
If microfilaricidal treatment is not preceded or followed by adulticidal
treatment, the suppression of microfilaremia will be temporary. Treat-
ment may reduce the reproductive capacity of the adults (Anantaphruti et
al. 1982), but they will again shed microfilariae if given sufficient time.
Although ivermectin at 50 J-tg/kg is used routinely as a microfilaricide
250 W.e. Campbell
by some veterinarians, this use has not been approved by government

regulatory agencies. Such approval has not been sought by the manufac-
turer of the drug, and no recommendation or claim is made by the
manufacturer with respect to microfilaricidal use.
As in the case of adult D. immitis, the microfilariae are not killed by
ivermectin in vitro (Devaney and Howells 1984) although preliminary data
suggest that ivermectin can be lethal to microfilariae when host leuko-
cytes are present in the medium (E. C. McManus, unpublished data).
Further work is needed to determine why both stages are virtually
unaffected in vitro, while one is susceptible in the blood-vascular system
of the host and the other is not.
b. Dipetalonema reconditum
Lindemann and McCall (1983) gave ivermectin orally at 250 ltg/kg to
three Beagle dogs experimentally infected with D. reconditum. Treatment
was followed by a reduction of more than 99% in the number
of circulating microfilariae. When the dogs were necropsied, no adult
D. reconditum were found in the abdominal cavity or connective tissues,
but the authors cautioned that the absence of adult worms in this
experiment is not necessarily a reflection of drug efficacy.
c. Toxocara canis
Treatment of patent naturally acquired infections of T. canis in dogs
reduced the worm burden by 91% when the ivermectin dosage was
200 ltg/kg subcutaneously (s.c.) and 97% when the dosage was 400 ltg/kg
s.c. (Anderson and Roberson 1982).
The immature stages of T. canis are important in two quite different
contexts-the intestinal and the extraintestinal. Against the intestinal
fourth stage larvae (in dogs treated 2 weeks after inoculation with
infective eggs) ivermectin was 97% effective at 200 ltg/kg s.c. and 98%
effective at 400 ltg/kg s.c. (Anderson and Roberson 1982). Dosages of 50
and 100 ltg/kg s.c. gave no significant reduction in worm burden.
Yazwinski, Tilley, and Greenway (1982) also treated dogs 2 weeks after
inoculation with infective eggs of T. canis, but they reported 100%
efficacy at dosages of 100 ltg/kg s.c. and even 50 ltg/kg s.c. This
discrepancy may be due to the difference in intensity of infection, the
mean control worm burden being 232 in the former trial and only 5.4 in the
latter.
The importance of the extraintestinal (L2) larvae of T. canis lies in their
role in transuterine and transmammary transmission from bitch to pup;
and it has been notably difficult to block this transmission by chemo-
therapy. Using experimentally infected mice as a model, Abo-Shenada
(1982) found that L2 larvae were affected by ivermectin at high dosage
(2000 ltg/kg) given p.o. or s.c. on Days 2 to 7 post inoculation. In mice so
treated the larvae failed to migrate from the host liver to the brain and
muscles; but when treatment was given after the larvae had completed
their migration (Days 8 to 13) even that very high dosage failed to reduce
their number. Studies in dogs, however, suggest that it is possible to
affect T. canis in the tissues in such a way that pups born to treated
bitches are essentially free of infection. Bitches inoculated with T. canis
eggs were treated with ivermectin around the 42nd day of gestation (when
the larvae are reactivated from the tissues) and the pups subsequently
born to those bitches were necropsied either at birth or at weaning (Shoop
et al. 1988). Treatment regimens that resulted in greater than 99%
reduction in worm burden and in the passage of eggs in the feces were:
1000 ~g/kg s.c. on Days 20 and 42 of gestation; 500 ~g/kg s.c. on Days 38,
41,44, and 47 of gestation; or 1000 ~g/kg s.c. on Day 20 followed by 500
~/kg on Days 42, 47, and 53 of gestation.
d. Toxascaris leonina
T. leonina appears to be less sensitive to ivermectin than T. canis, but
very little comparative information is available. Using experimentally
infected dogs, Anderson and Roberson (1982) found that dosages up to
400 ~g/kg s.c. did not give satisfactory activity against the adult worms.
Yazwinski, Tilley, and Greenway (1982) treated dogs 4 weeks after
inoculation with T. leonina eggs (when the worms would have been in the
L4 stage). They recorded 100% efficacy with a dosage of 200 ~g/kg s.c.,
but the mean number of worms recovered from control dogs was small.
c. Ancylostoma caninum (hookworm)

The canine hookworms are among the most susceptible of parasites to the
action of the avermectins. In a controlled trial against adult A. caninum in
dogs, abamectin was 98% effective at 5 ~g/kg, and avermectin B2a was
100% effective at 3 ~g/kg (Blair and Campbell 1978). Ivermectin, too, was
found to be highly active against hookworms (Table 18.3). Yazwinski,
Tilley, and Greenway (1982) reported high activity at 50 ~g/kg, sub-
cutaneously, but did not test lower dosages. Egerton, Eary, and Suhayda
(1985) reported that a single oral or subcutaneous dose at 24 ~g/kg was
more than 99% effective against both the L4 and adult stages.
In one study, ivermectin at 10 ~g/kg p.o. not only resulted in the
expulsion of all of the hookworms but affected the eggs within the female
worms (Wang et al., in press). Eggs were teased out of the uterine tissue
of female worms passed in the feces of treated dogs (and from female
worms recovered from untreated dogs at necropsy) and cultured under
conditions suitable for hatching and development of infective larvae.
Such larvae were recovered from cultures of eggs obtained from un-
treated dogs but not from dogs treated with ivermectin at dosages varying
from 10 to 100 ~g/kg. Similarly, when ivermectin was added directly to
252 W.C. Campbell
TABLE 18.3. Dosages of ivermectin reported to be at least 95% effective in

reducing the numbers of various nematode parasites in dogs. All treatments
refer to single dose; p.o. = oral, s.c. = subcutaneous.
Dosage
Parasite (ILg/kg) Reference
Toxocara canis
L4 stage 50 s.c. Yazwinski et al. 1982
L4 stage 200 s.c. Anderson and Roberson 1982
adult 400 s.c. Anderson and Roberson 1982
Toxascaris leonina
immature 200 s.c. Yazwinski et al. 1982
Ancylostoma caninum
L4 stage 24 p.o. Egerton et al. 1985
L4 stage 50 s.c. Anderson and Roberson 1982
L4 stage 12 s.c. Egerton et al. 1985
L4 stage 50 s.c. Yazwinski et al. 1982
adult 12 p.o. Egerton et al. 1985
adult 12 s.c. Egerton et al. 1985
adult 10 p.o. Wang et al. in press
Ancylostoma braziliense
Uncinaria stenocephala
L4 stage 24 p.o. Egerton et al. 1985
L4 stage 12 s.c. Egerton et al. 1985
adult 24 p.o. Egerton et al. 1985
adult 24 s.c. Egerton et al. 1985
Trichuris vulpis
immature 10 s.c. Yazwinski et al. 1982
Capillaria aerophila
See text.
Strongyloides stercoralis
See text.
cultures of A. caninum eggs, development was suppressed by 95% at

10 ILg/ml and by 100% at 100 ILg/ml (Wang et al., in press). These same
authors showed that ivermectin was lethal to infective (L3) larvae of
A. caninum in vitro at 0.025 ILg/ml, and immobilized or killed adult
A. caninum in vitro at concentrations ranging from 0.7 to 7.0 ILg/ml-the
severity of the effect being proportional to the concentration of drug in the
medium.
f. Uncinaria stenocephala (hookworm)
Egerton, Eary, and Suhayda (1985), using similar techniques for each
species, found U. stenocephala to be only slightly less susceptible than
A. caninum to the therapeutic effect of ivermectin. Dosages that gave at
least 95% reduction in the number of U. stenocephala are listed in Table
18.3. Against the adult stage, treatment at 12 ILg/kg was 80% effective
when given orally but 90% effective when given subcutaneously (Egerton,
Eary, and Suhayda 1985). When given by either route, a dosage of
24 JLg/kg was more than 96% effective against the adults and against L4
larvae (Egerton, Eary, and Suhayda 1985).
g. Strongyloides stercoralis spp.
In a preliminary probe, ivermectin was given as a single oral dose at
800 JLg/kg to 1 of 2 dogs that had severe (corticosteroid-enhanced)
strongyloidiasis. The untreated dog, when necropsied 6 days after treat-
ment, yielded 56,556 adult intestinal worms and 976,185 intestinal larvae.
The treated dog yielded no intestinal adults and only 2 larvae (L.M.
Aikens and G.A. Schad, personal communication 1988). Negligible
numbers of worms or larvae were recovered from extraintestinal sites in
either dog.
h. Filaroides osleri
There seems to be no published evidence of the efficacy of ivermectin
against F. osleri, but Clayton (1983) mentioned en passant that a single
injection at 400 JLg/kg had shown "promising results in a limited number
of experimental F. osleri infections."
i. Capillaria spp.
A single dog with nasal capillariasis (C. aerophilia) was given ivermectin
orally at 200 JLg/kg. The nasal discharge stopped within 7 days and the
feces became free of eggs within 14 days. There was no relapse for at least
8 months (Evinger, Kazacos, and Cantwell 1985). A single dog with
urinary capillariasis (c. plica) was injected subcutaneously with iver-
mectin at 200 JLg/kg (after temporary alleviation of clinical signs had been
achieved by fenbendazole therapy). The ivermectin treatment was fol-
lowed by apparent clinical and parasitological cure (passage of eggs in
urine ceased within 7 days) (Kirkpatrick and Nelson 1987).
j. Trichuris vulpis (whipworm)
When tested against immature (4-week-old) T. vulpis in experimentally
infected dogs, ivermectin was 100% effective at 100 JLg/kg s.c. (Yaz-
winski, Tilley, and Greenway 1982), but the control dogs had a mean of
only 6.0 worms. Against mature T. vulpis, treatment was more than 99%
effective at 100 JLg/kg s.c., and more than 95% effective at 200 JLg/kg p.o.
(J. R. Egerton, unpublished data).
2. Ectoparasites
a. Sarcoptes scabiei
Dogs with natural infestations of Sarcoptes scabiei were treated sub-
cutaneously with ivermectin at various dosages. At 200 JLg/kg or higher,
treatment resulted in complete cure by Day 7 after treatment. Dosages as
254 W.C. Campbell
low as 50 p,g/kg were also effective but took 2 to 3 weeks to achieve their
full effect (Yazwinski et al. 1981). The efficacy of ivermectin was
confirmed in a clinical trial involving almost 300 dogs. Despite confine-
ment of the dogs "in a heavily contaminated environment which was
conducive to repeated reinfestation during treatment," 2 doses of the
drug at 200 p,g/kg s.c., given 14 days apart, provided marked clinical
improvement and complete parasitological cure 2 weeks after the second
injection (Scheidt et al. 1984). Additional evidence of efficacy was
provided by Gravino, de Caprariis, and Agresti (1985), who used
400 p,g/kg s.c., and gave 3 treatments at 8-day intervals.
b. Demodex canis
Despite anecdotal reports of ivermectin's efficacy in demodectic mange,
there is little or no scientific evidence to support it. In one clinical study,
ivermectin was administered subcutaneously to 4 dogs with generalized
amitraz-resistant demodecosis. The dosage was 400 p,g/kg and was
repeated weekly for a total of 8 treatments. Despite this rather intensive
treatment, and despite the apparent clinical and parasitological cure of 1
dog, no reduction in mite number could be demonstrated in the remaining
3 dogs (Scott and Walton 1985). All 4 dogs, however, showed some
clinical improvement, and the authors suggested that this might be a
reflection of decreased mite activity in the skin. Perhaps the dosage used
in that study was not high enough, because Gravino, de Caprariis, and
Giglio (1985) reported a favorable response in an (uncontrolled?) study in
which 20 dogs were given 600 p,g/kg s.c. at weekly intervals for a total of 5
treatments. On the other hand, Belot, Parent, and Pangui (1984) reported
that demodectic mange responded to 2 doses of ivermection at 400 p,g/kg
s.c. given 15 days apart. These 3 trials were conducted in the United
States, Italy, and France, respectively, so there may be geographical
strains with different degrees of susceptibility. Additional controlled
studies are needed to clarify the matter.
c. Otodectes cynotis
The efficacy of ivermectin versus O. cynotis was demonstrated by
Yazwinski and colleagues (1981) in a controlled clinical trial using natural
infections. A single subcutaneous injection at 200 p,g/kg gave complete
cure in 14 days, while a single treatment at 400 p,g/kg resulted in cure in 7
days. Similarly, Molina (1986) found that a single injection at 200 p,g/kg
gave complete cure in 14 days, and Chauve (1984) also found that dosage
to be fully effective.
d. Ctenocephalides cati
Despite anecdotal reports of efficacy, neither abamectin nor ivermectin
has been shown to have significant activity against fleas on dogs. In
a controlled study, ivermectin was given orally, at 50 p,g/kg/day or
500 JLg/kg/week, to dogs exposed repeatedly to C. cati. The treatment

neither reduced the number of fleas to a detectable degree, nor prevented
reinfestation (Blair, Egerton, and Ewanciw 1984).
II. Cats
A. PRODUCTS
Ivermectin has not been approved by regulatory agencies for use in cats in
any form, nor has any application for such use been submitted by the
manufacturer.
B. ANTIPARASITIC EFFICACY
1. Endoparasites
a. Aelurostrongylus abstrusus
Blagbum and colleagues (1987) gave ivermectin orally at various dosages
to cats with natural infection of A. abstrusus, but there was little if any
evidence of efficacy. Kirkpatrick and Megella (1987) tested ivermectin
against the same parasite in a single naturally infected cat. After an
injection of ivermectin at 200 JLg/kg, larvae continued to be detectable in
the feces. After a second injection at 400 JLg/kg larvae could no longer be
found in the feces, but no improvement in the pulmonary disease could be
demonstrated radiographically, and the findings in this single animal are
of doubtful significance.
b. Toxocara cati
In a single treated cat, the passage of T. cati eggs in the feces ceased after
treatment of the cat with ivermectin at 200 JLg/kg s.c. (Kirkpatrick and
Megella 1987). The limited evidence of efficacy provided by this single
instance is supported by observations made by Blagbum and colleagues
(1987) in several cats with very light natural T. cati infection. In those
cats, oral treatment at 100 or 300 JLg/kg appeared to be highly effective,
but additional studies would clearly be needed to determine the efficacy of
ivermectin against this common ascarid.
c. Ancylostoma spp.
Hookworms in cats, like those in dogs, appear to be very sensitive
to ivermectin. Using cats with very light natural infections, Blagbum
and colleagues (1987) obtained evidence that dosages ranging from 10 to
300 JLg/kg p.o. and s.c. were active against A. braziliense; and in one cat
infected with A. tubaeforme, a dosage of 10 JLg/kg p.o. appeared to be
effective.
256 W.C. Campbell
d. Other Helminths
It is not known whether ivermectin is effective against Capillaria sp. in
cats. Ivermectin is not known to be active against any species of flukes or
tapeworms, and the incidental observations of Blagburn and colleagues
(1987) on several tapeworm species in cats are in accord with this general
finding.
2. Ectoparasites
a. Nodoedres cati
According to Bigler, Waber, and Pfister (1984), a single s.c. injection of
ivermectin at 1000 ILg/kg was effective in controlling an outbreak of
notoedric mange. This was a clinical report, not the record of a laboratory
trial, but all 17 treated cats were judged completely cured. Suarez and
Bertero (1986) used the same high dosage successfully against N. cati and
reported that all skin scrapings were negative for mites 7 and 30 days after
treatment.
b. Otodectes cynotis
Chauve (1984) reported that ivermectin at 200 ILg/kg s.c. was effective
against naturally acquired o. cynotis infection in cats. Chauve and
Reynaud (1984) extended this observation to experimentally induced
infestations, and showed that mites disappeared 14 days after the cats had
received a single injection at 666 ILg/kg. Further clinical reports of the
effectiveness of the 200 ILg/kg s.c. dosage, and of a dosage of 400 ILg/kg
s.c., were provided by Church (1985) and Franc, Dorchies, and Soubey-
roux (1985), respectively.
c. Ctenocephalides cati
In a clinical note, Bigler, Waber, and Pfister (1984) reported that adult
populations of fleas (presumably C. cati) were "decimated" by an
injection of ivermectin at the very high dosage of 1000 ILg/kg. However,
there appears to be no evidence, obtained under controlled conditions, of
the efficacy of ivermectin against fleas on cats.
REFERENCES
Abo-Shenada MN (1982) The effect of some anthelmintics on the second stage

larvae of Toxocara canis. Parasitol. 85:xiii (in Soc. Proc.).
Anantaphruti M, Kino H, Terada M, Ishii AI, Sano M (1982) Studies on
chemotherapy of parasitic helminths (XIII): Efficacy of ivermectin on the
circulating microfilariae and ~embryonic development in the female worm of
Dirfilaria immitis. Jap. J. Parasitol. 31:517-530.
Anderson DL, Roberson EL (1982) Activity of ivermectin against canine intesti-
nal helminths. Am. J. Vet. Res. 43:1681-1683.
Belot J, Parent R, Pangui JL (1984) Canine demodectic mange: Clinical observa-

tions in a treatment trial with ivermectin. Point Vet. 16:618-620.
Bigler B, Waber S, Pfister K (1984) Successful treatment of Notoedres cati
infestation with ivermectin. Schweiz. Arch. Tierheilkd. 126:365-367 (in
German).
Blagbum BL, Hendrix CM, Lindsay DS, Vaughan JL (1987) Anthelmintic
efficacy of ivermectin in naturally parasitized cats. Am. J. Vet. Res. 48:670-
672.
Blair LS, Campbell WC (1978) Efficacy of avermectins against Ancylostoma
caninum in dogs. J. Helminthol. 52:305-307.
Blair LS, Campbell WC (1979) Efficacy of avermectin Bla against microfilariae of
Dirofilaria immitis. Am. J. Vet. Res. 40:1031-1032.
Blair LS, Campbell WC (1981) Immunization of ferrets against Dirofilaria immitis
by means of chemically abbreviated infections. Para. Immunol. 3:143-147.
Blair LS, Egerton JR, Ewanciw EV (1984) Ctenocephalides felis: failure of
ivermectin given per os in miglyol vehicle to control experimental flea
infestations of dogs in a double blind controlled trial. Proc. Amer. Assoc. Vet.
Parasitol. July 15-17, New Orleans, LA.
Blair LS, Jacob L, Ewanciw DV (1982) Immunization of dogs against Dirofilaria
immitis by means of chemically abbreviated infections. Mole. Biochem. Par-
asit., Supplement, p. 678.
Blair LS, Malatesta PF, Ewanciw DV (1983) Dose-response study of ivermectin
against Dirofilaria immitis microfilariae in dogs with naturally acquired infec-
tions. Am. J. Vet. Res. 44:475-477.
Blair LS, Williams E, Ewanciw DV (1982) Efficacy of ivermectin against
third-stage Dirofilaria immitis larvae in ferrets and dogs. Res. Vet. Sci.
33:386-387.
Boreham PFL, Atwell RB (1983) Absence of shock-like reactions to ivermection
in dogs infected with Dirofilaria immitis. J Helminthol 57:279-281.
Campbell WC (1987) Ivermectin and heartworm. Sem. Vet. Med. & Surg. (Sm.
Anim.) 2:48-55.
Chauve C (1984) Prospects for a parenteral treatment for otodectic mange in dogs
and cats. Sci. Vet. Med. Compo 86:41-43.
Chauve C, Reynaud M-C (1984) Parenteral treatment of otodectic mange in cats:
efficacy ofivermectin. Sci. Vet. Med. Compo 86:189-192 (in French).
Church D (1985) Treat and control infestations of ear mites. Vet. Med. 54:4.
Clayton HM (1983) The management and treatment of respiratory nematode
infections in small animals. Vet. Ann., 23rd issue, pp 254-259.
Devaney E, Howells RE (1984) The microfilaricidal activity ofivermectin in vitro
and in vivo. Trop. Med. Parasit. 35:47-49.
Egerton JR, Eary CH, Suhayda D (1985) Dose-titration studies of ivermectin
against experimental Ancylostoma caninum and Uncinaria stenocephala infec-
tions. Am. J. Vet. Res. 46:1057-1059.
Evinger JV, Kazacos KR, Cantwell HD (1985) Ivermectin for treatment of nasal
capillariasis in a dog. J. Am. Vet. Med. Assoc. 186:174-175.
Franc M, Dorchies PH, Soubeyroux H (1985) Studies on the treatment of
otoacariasis in cats with ivermectin. Rev. Med. Vet. (France) 136:685-686 (in
French).
Gravino AE, deCaprariis D, Agresti A (1985) Impiego della 22,23-diidroaver-
258 W.C. Campbell
mectina nel trattamento dello rogna sarcoptica del cane. Acta Med. Vet.
31:43-47.
Gravino AE, deCaprariis D, Giglio M (1985) Impiego della 22,23-diidroaver-
mectina nel trattamento della Demodicosi del cane. Acta Med. Vet. 31:185-189.
Jackson RF (1987) Microfilaricides. Sem. in Vet. Med. & Surg. (Sm. Anim.)
2:44-47.
Jackson RF (1984) Ivermectin again. Am Heartworm Soc Bull 10:9.
Jackson RF, Seymour GW (1981) Efficacy of avermectins against microfilariae of
Dirofilaria immitis. In Otto GF (ed) Proceedings of the Heartworm Symposium
1980. Vet. Med. Pub. Co., Edwardsville, KS, pp 131-136.
Jackson RF, Seymour WG, Beckett R (1986) Routine use of 0.05 mg/kg
ivermectin as a microfilaricide, in Otto GF (ed): Proceedings of the Heartworm
Symposium, 1986. Washington, DC, American Heartworm Society, pp 37-39.
Kirkpatrick CE, Megella C (1987) Use of ivermectin in treatment of Aeluro-
strongylus abstrusus and Toxocara cati infections in a cat. J. Am. Vet. Med.
Assoc. 190:1309-1310.
Kirkpatrick CE, Nelson GR (1987) Ivermectin treatment of urinary capillariasis in
a dog. J. Am. Vet. Med. Assoc. 191:701-702.
Knight DH, Campbell WC, Weiner DJ, Washabau RJ (1986) Microfilaricidal
efficacy of ivermectin in adulticide treated and untreated heartworm infected
dogs. In Otto GF (ed) Proceedings of the Heartworm Symposium 1986.
American Heartworm Society, Washington, D.C., pp 19-27.
Lindemann BA, McCall JW (1983) Microfilaricidal activity ofivermection against
Dipetalonema reconditum. J. Vet. Pharm. & Therap. 6:75-76.
Lok JB, Harpaz T, Knight DH (1988) Abnormal patterns of embryogenesis in
Dirofilaria immitis treated with ivermectin. J. Helminthol. 62:175-180.
McCall JW, Lindemann BE, Porter CA (1981) Prophylactic activity of avermec-
tins against Dirofilaria immitis, in Otto GF (ed): Proceedings of the Heartworm
Symposium, 1980. Edwardsville, KS, Veterinary Medicine, pp 126-130.
McCall JW, Dziminanski MT, Plue RE (1986) Ivermectin in heartworm pro-
phylaxis: Studies with experimentally induced and naturally acquired infec-
tions. In Otto GF (ed) Proceedings of the Heartworm Symposium 1986.
McManus EC, Pulliam JD (1984) Histopathologic features of canine heartworm
microfilarial infection after treatment with ivermectin. Am. J. Vet. Res.
45:91-97.
Molina CG (1986) The activity of ivermectin against Octodectes cynotis in
naturally infested dogs. Vet. Mex. 17:39-40.
Paul AJ, Todd KS, Sundberg JP (1986a) Efficacy of ivermectin against Dirofilaria
immitis larvae in dogs 30 and 45 days after induced infection. Am. J. Vet. Res.
47:883-884.
Paul AJ, McCall JW, Todd KS (1986b) Efficacy of ivermectin against Dirofilaria
immitis larvae in dogs forty-five days after induced infection. In Otto GF (ed)
Proceedings of the Heartworm Symposium 1986. American Heartworm Soci-
ety, Washington, D.C., pp 33-35.
Plue RE, Seward RL, Acre KE, Cave JS, Schlotthauer JC, Stromberg BE (1983)
Clearance of Dirofilaria immitis, microfilariae in dogs using 200 mcg/kg
ivermectin subcutaneously, in Otto GF (ed): Proceedings of the Heartworm
Symposium, 1983, Edwardsville, KS, Veterinary Medicine, pp 153-160.
18. Use ofIvermectin in Dogs and Cats 259
Scheidt VJ, Medleau L, Seward RL, Schwartzman RM (1984) An evaluation of

ivermectin in the treatment of sarcoptic mange in dogs. Am. J. Vet. Res.
45: 1201-1202.
Schlotthauer JC, Stromberg BE, Paul AJ (1986) Safety and acceptability of
ivermectin in dogs with naturally acquired patent infection of Dirofilaria
immitis. In Otto GF (ed) Proceedings of the Heartworm Symposium 1986.
Scott DW, Walton DK (1985) Experiences with the use of amitraz and ivermectin
for the treatment of generalized demodicosis in dogs. Am. Anim. Hosp. Assoc.
21:535-541.
Shoop WL, Egerton JR, Eary CH, Suhayda D (1988) Control of Toxocara canis
transmission from bitch to offspring with ivermectin. Program and Abstracts,
Am. Soc. Parasitol., Winston-Salem, NC, pp 59-60.
Simpson CF, Jackson RF (1985) Lesions in the liver and kidney of Dirofilaria
immitis-infected dogs following treatment with ivermectin. Zeitschrift fur
Parasitenkunde 71:97-105.
Swartz EO (1985) Treating for microfilaremia caused by heartworms. Vet Med
80:59.
Suarez CJM, Bertero SI (1986) Tratamiento de la sarna notoedrica del gato con
ivermectina. Vet. Argent. 3:340-344.
Wang C, Huang XX, Zhang YQ, Yen QY, Wen Y (in press) Efficacy of ivermectin
in hookworms as examined in Ancylostoma caninum infections. J. Parasit.
Yazwinski TA, Pote L, Tilley W, Rodriguez C, Greenway T (1981) Efficacy of
ivermectin against Sarcoptes scabiei and Otodectes cynotis infestations of
dogs. Vet. Med. Small Anim. Clinic. 76:1751.
Yazwinski TA, Tilley W, Greenway T (1982) Efficacy of ivermectin in the
treatment of artificially induced canine, mixed, gastrointestinal helminthiasis.
Vet. Med. Small ANim. Clinic. 77:225-226.
CHAPTER 19
Use of Ivermectin in Laboratory

and Exotic Mammals and in
Birds, Fish, and Reptiles
M.D. SoIl
I. Introduction
II. Laboratory Animals
A. Mice
1. Endoparasites
a. Aspiculuris tetraptera
b. Nematospiroides dubius
c. Strongyloides spp.
d. Syphacill obvelata
e. Toxocara canis
f. Trichinella spp.
g. Trichuris muris
2. Microfilariae
3. Ectoparasites
a. Mites
b. Triatomid bugs
c. Cuterebra fontinella
B. Rats
a. Angiostrongylus cantonensis
b. Trichinella spiralis
c. Syphacia muris
C. Guinea Pigs
1. Ticks
2. Tsetse flies
3. Mange mites
D. Hamsters
E. Gerbils
F. Rabbits
1. Mange mites
2. Tsetse flies
3. Ticks
G. Ferrets and Jirds
19. Use of Ivermectin in Mammals, Birds, Fish, and Reptiles 261
III. Wild, Zoo and Exotic Pet Animals

A. Rodents
B. Hedgehogs
C. Carnivores
1. Foxes
a. Endoparasites
h. Ectoparasites
2. Wolves
3. Cape Hunting Dog
D. Ungulates
1. Camels
a. Endoparasites
h. Ectoparasites
2. Alpacas and Llamas
3. Reindeer
4. Deer
a. Endoparasites
b. Ectoparasites
5. Bighorn sheep
6. Impala
7. Other Antelope
8. Bison
9. Buffalo
10. Wild boar
11. Other species
E. Elephants
F. Nonhuman primates
1. Mite infections
2. Nematode infections
IV. Birds
A. Nematodes
a. Ascaridia
h. Capillaria
c. Pelecitus
d. Oxyspirura
e. Syngamus
B. Mites
a. Dermanyssus and Ornithonyssus
h. Cnemidocoptes spp.
c. Sternostoma tracheolum
d. Cytodites nudus
e. Pterolichidae
V. Fish
A. Copepods
B. Nematodes
262 M.D. SoIl
VI. Reptiles
A. Snakes
B. Chelonians
VII. Conclusion
I. Introduction
The first observations on the antiparasitic properties of the avermectins
were made in laboratory animal screens. Ivermectin is potent (allowing
for administration of very small doses); it can be administered either
orally or parenterally; and it is effective against both arthropod and
nemathelmintic endo- and ectoparasites. These attributes have led to the
extensive use of the compound for treating parasitic infections in a variety
of laboratory, wild, zoo, and exotic mammals, as well as in reptiles, birds,
and fish. Apart from domestic animals, ivermection has been approved by
various regulatory authorities for use in camels and reindeer only. The
treatment of any other species with the product is considered to be
"extralabel" usage.
II. Laboratory Animals

A. MICE
Ivermectin's efficacy has been tested in mice against naturally acquired

parasite infections as well as in a number of mouse-parasite model
systems.
1. Endoparasites
a. Aspiculuris tetraptera
Ivermectin added to the drinking water for 24 hours to provide a dose of
1.0 to 1.6 mg/kg body weight was fully effective in removing naturally
acquired infections of this parasite from laboratory mice (Hasslinger and
Wiethe 1987).
b. Nematospiroides dubius
Oral treatment at 5 mg/kg on Days 3 or 6 post infection was reported to
remove all worms present (Sayles and Jacobson 1983); Rajasekariah and
colleagues (1986) reported that a much lower dose of 0.3 mg/kg com-
pletely eliminated adult worms from mice.
c. Strongyloides spp.
Grove (1983) reported that, following ivermectin treatment of mice with
01 mg on Days 0 and 1 (migration phase) and 3 and 4 (intestinal phase)
after infection with S. ratti, the treated animals subsequently passed no
larvae in the feces. Moreover, he found that a single dose of .05 mg 6 days
after infection eradicated intestinal adult worms. Rajasekariah and col-
leagues (1986) reported that 2 treatments at 0.3 mg/kg are required to
eliminate adult parasites from mice.
Grove (1983) reported that S. stercora lis responds similarly to iver-
mectin: daily doses of .01 mg of ivermectin on Days 0, 1, 2, and 3 after
infection greatly reduced parasite numbers in the muscles, while a single
dose of 0.1 mg given 1 hour or 6 days after infection completely eradicated
larvae from the muscles.
d. Syphacia obvelata
Ostlind, Nartowicz, and Mickle (1985) reported that feeding a diet
containing 0.0005% ivermectin for 6 days to mice naturally infested with
pinworms was more than 99% effective against both adult and immature
worms; moreover, a single oral treatment at a dose rate of 2 mg/kg was
94% to 100% effective against adult and 86% to 97% effective against
immature S. obvelata.
e. Toxocara canis
Treating mice harboring immature T. canis with ivermectin at 2 mg/kg for
2 to 7 days after infection (Abo-Shenada and Herbert 1984), or 0.4 mg/kg
on Days 15 to 28 after infection (Carillo and Barriga 1987) alters the
distribution of the larvae: Larvae are retained in the liver and lungs, and
fewer migrate to the brain. None of these dose rates completely eradica-
ted the parasite from the tissues, however.
f. Trichinella spp.
Treating mice with avermectin at a dose rate of 3 mg/kg reduced the
number of intestinal T. spira lis 80% to 90% (Campbell, Blair, and Lotti
1979). The number of parasites was not reduced significantly, however,
following ivermectin treatment (5-10 mg/kg) of mice harboring the
migrating and muscle stages of a number of Trichinella spp. (Sanmartin-
Duran et al. 1986). Ivermectin was effective against intestinal forms of T.
nelsoni and T. nativa in this study, reducing their number by 91% to 96%
(relative to controls), following treatment at 2 and 5 mg/kg on Days 1 and
6 post infection.
g. Trichuris muris
Although some degree of efficacy was apparent after a repeated dose of
ivermectin at 10 mg/kg, mice treated for T. muris infestation were not
completely cured (Rajasekariah et al. 1986).
2. Microfilariae
Ivermectin's microfilaricidal activity has been demonstrated against
immature stages of a number of filarial parasites in mice.
264 M.D. SoU
Single doses of ivermectin as low as 0.2 mg/kg destroyed 100% of

Onchocerca lienalis microfilariae within 5 days after treatment. The same
dose, administered at various times prior to microfilariae infection,
reduced parasite recoveries by 92% for infection 4 days after treatment
(Bianco et al. 1986).
Devaney an Howells (1984) reported that Brugia pahangi microfilare-
mia was reduced by 87% 24 hours after treatment of mice orally at
5 mg/kg; the peritoneal microfilariae of this species were unaffected.
O. lienalis larvae in the cutaneous and subcutaneous tissues, and Dirofi-
laria immitis microfilariae in the blood of mice were also killed at this
dose.
Zahner and colleagues (1987) reported that subcutaneous treatment of
Dipetalonema viteae-infected Mastomys natalensis with ivermectin at
0.1 mg/kg once or 0.05 mg/kg on 5 consecutive days reduced microfilare-
mia 100% over a 42-day trial period. Higher doses (0.2 mg/kg once or
5x 1.5 mg/kg) were required to remove microfilariae of Litomosoides
carinii. Lower doses removed circulating microfilariae of both species for
shorter periods of time. Activity against adult D. viteae was demonstrated
following 5 consecutive daily treatments at 0.2 mg/kg and higher;
ivermectin did not affect the numbers of adult L. carinni at these dose
rates.
3. Ectoparasites
The efficacy of ivermectin has been evaluated against some parasitic
arthropod species in the mouse.
a. Mites
Silverman, Blatt, and Lerro (1983) reported that feeding mice a commer-
cial diet containing 2 mg of ivermectin/kg of feed for two 6-day periods,
1 week apart, effectively eliminated Myobia musculi mites. Efficacy of
the product against murine mites was confirmed by Wing, Courtney,
and Young (1985), who demonstrated that 2 subcutaneous injections at
0.2 mg/kg, given 1 week apart, effectively eliminated Mycoptes musculi-
nus and Myobia musculi from laboratory mice. Single treatments reduced
infections only temporarily.
Baumans et al. (1988) reported that spraying mice with a .01% solution
of ivermectin was effective in eliminating symptoms of mite infection for
12 weeks after treatment.
b. Triatomid bugs
Treating mice with ivermectin at 0.2 mg/kg was reported to have caused
high mortality and a reduction in egg-laying potential of Rhodnius prolixus
and Hemiptera triatominae feeding on these animals (Azambuja et al.
1985).
19. Use ofIvermectin in Mammals, Birds, Fish, and Reptiles 265
c. Cuterebra fontinella
The C. fontinellalmouse model was used to demonstrate the systemic
insecticidal efficacy of the avermectins. Ostlind, Cifelli, and Lang (1979)
reported an avermectin mixture to be effective at dose levels of 0.15
mg/kg and higher. Drummond (1980) subsequently confirmed this activity
using ivermectin in mice with induced C. fontinella infections.
B. RATS
Rats have been used as a model system to assess ivermectin's basic

toxicology, as well as for extensive studies on the drug's mode of action,
pharmacology, and pharmacokinetics (Calcott and Fatig 1984; Chiu et al.
1986; Olsen and Snowman 1985; Pong and Wang 1980; Pong, Wang, and
Fritz 1980; Williams and Yarbrough 1979).
Antiparasitic evaluation has been limited to a few endoparasite-rat
models, although efficacy has been demonstrated against natural pinworm
infestations.
a. Angiostrongylus cantonensis
Although treatment of 6-week-old infections of A. cantonensis at a dose
of 0.1 mg/kg resulted in a lower fecal egg and larval count in rats
(indicating an effect on the fecundity of the parasite), ivermectin did not
have a demonstrable vermicidal effect at this level. Oral treatment at the
higher dose of 1 mg/kg, 3 days after induced infection, did result in a
significant reduction in the number of worms recovered (Ishii et al. 1983).
b. Trichinella spiralis
Although treatment at 0.3 mg/kg was only 84% effective against migrating
larvae and 83% effective against encysted forms, increasing the dose rate
to 3 mg/kg was reported to increase efficacy to 94% and 99% respectively,
against these 2 stages of T. spiralis (Rapic, Dzakula, and Matic-Piantanida
1982). Daily treatment at 0.6 mg/kg for 3 days was found to be only 54%
effective against intestinal T. spira lis (Alcaino, Gorman, and Imbert
1984).
c. Syphacia muris
Treatment of rats with ivermectin at 0.2 mg/kg/day for 5 consecutive days
was 99% effective against natural infections of this parasite (Battles et al.
1987). There was no difference in body weight between treated and
control animals, and no toxic signs were observed after treatment.
c. GUINEA PIGS
1. Ticks
In screening tests, ivermectin has been shown to be completely effective
against the lone star tick (Amblyomma americanum) when adult ticks are
266 M.D. SolI
fed on treated guinea pigs (Hunt 1982). Efficacy has also been demon-
strated against larval, nymphal, and adult Rhipicephalus sanguine us-as
well as A. americanum-when these ticks were fed on guinea pigs treated
at dose rates of up to 2.0 mg/kg (Wilkins et al. 1981).
2. Tsetse Flies
Mortality of tsetse flies (Glossina palpalis) occurred among flies fed
exclusively on guinea pigs treated with ivermectin at dose rates greater
than 2.0 mg/kg (Distelmans, D'Haeseleer, and Mortelmans 1983).
3. Mange Mites
Ivermectin has been used successfully to treat mange caused by the mite
Trixacarus caviae in a guinea pig colony. The compound was adminis-
tered at 0.2 mg/kg subcutaneously on 3 occasions with 7-day intervals
(Harvey 1987).
D. HAMSTERS
Ivermectin's efficacy against the human hookworm Necator americanus

was evaluated against a strain of the parasite adapted to laboratory
hamsters. Doses of 30 mg/kg once, or 10 mg/kg twice, were required for
complete clearance of immature N. americanus; lower doses of 15 mg/kg
once, or 7.5 mg/kg twice, completely eliminated infestation in hamsters
carrying adult parasites (Rajasekariah et al. 1986).
E. GERBILS
The avermectin's efficacy against Trichostrongylus colubriformis was

assessed in gerbils early in the development phase ofthese compounds. A
mixture of avermectin B 1a and B2a was found to be 100% effective
against 6-day-old T. colubriformis when administered at 0.03 mg/kg
(Ostlind and Cifelli 1981).
F. RABBITS
The resolution of lesions on the ears of rabbits caused by psoroptic mite

infestation gave the first indication of the potential activity of this class of
compounds against arthropods (Wilkins et al. 1980). Numerous studies
have subsequently been conducted to demonstrate ivermectin's efficacy
against natural and induced mange mite infections in rabbits. The rabbit
has also been used as a model to assess the systemic effect of ivermectin
on tsetse flies and ticks feeding on treated animals.
1. Mange Mites
Although subcutaneous injection of ivermectin at 0.2 mg/kg is reported to
be effective for treating rabbits naturally infected with psoroptic and
sarcoptic mange mites (Ashmawy and Fahmy 1987; Macchioni, Marcon-

cini, and Sbrana 1983); Mousa et al. 1986a), some investigators have
found that a single treatment at 0.2 or 0.4 mg/kg does not always eliminate
psoroptic mites (Wright and Riner 1985; Prosl and Kanout 1985). Wright
and Riner (1985) suggest that the susceptibility of 2 psoroptic mite
species, Psoroptes ovis and P. cuniculi, to ivermectin treatment in rabbits
may be different. Romero and Valentini (1984) report that 2 treatments at
0.25 mg/kg, 7 days apart, were effective in eradicating psoroptic and
sarcoptic mite infection; Prosl and Kanout (1985) recommend 2 treat-
ments at 0.4 mg/kg, 4 to 6 days apart, for eradication of P. cuniculi
infection. Curtis, Housley, and Brooks (1988) report that 2 ivermectin
treatments at 0.4 mg/kg with an 18-day interval are safe and effective for
treating rabbits with ear mite infections but suggest that subsequent
injections may be necessary in rabbitries with poor sanitation. Two
subcutaneous ivermectin treatments 3 weeks apart are also reported to be
effective against Notoedres cati infection in rabbits (Manurang, Partou-
torno, and Stevenson 1985).
2. Tsetse Flies
The reproductive capacity of Glossina palpalis fed on rabbits treated with
ivermectin at 2.0 mg/kg has been shown to be significantly reduced
(Abbeele, D'Haeseleer, and Goossens 1986).
3. Ticks
In an assessment of the suitability of the rabbit as a model for testing
systemic ascaricides, ivermectin has been shown to be effective against
induced infestations of Ornithodoros moubata for up to 7 days after
treatment at dose rates of 0.1 to 0.4 mg/kg (Frossard 1981).
G. FERRETS AND JIRDS
Because the ferret (Mustela putorius) is susceptible to infection with the

canine heartworm Dirofilaria immitis, this species was initially used to
assess the effect of ivermectin on the precardiac stages of heartworm.
Ferrets, artificially infected with D. immitis 2 to 42 days prior to
treatment, were given ivermectin at dose rates between 0.0125 and
2.0 mg/kg (Blair and Campbell 1980; Blair, Williams, and Ewanciw 1982).
The results of these studies suggest that the minimum effective dose for
suppression of maturation of D. immitis larvae in ferrets lies between
.0125 and 0.05 mg/kg, which is higher than the dose required for this
purpose in dogs.
In contrast to the exquisite sensitivity of the precardiac sL.tges of D.
immitis to ivermectin, this compound has not been shown to be effective
in preventing the development of adult Brugia pahangi in the jird
(Campbell 1982; Denham 1982).
268 M.D. Soil
III. Wild, Zoo, and Exotic Pet Animals

Although ivermectin is a useful antiparasitic agent for the treatment of a
variety of parasitic diseases encountered in wild, zoo, and exotic pet
animals, its overall usage in these species is relatively small and does not
warrant full-scale development projects to assess all aspects of safety and
efficacy for the numerous species represented here.
The injectable formulation is registered for use in camels in Algeria,
Jordan, Morocco, and Sudan; it has also been approved under an IR4
(minor species) development program for use in reindeer in the United
States. Apart from these applications, none of the commercially available
formulations is registered for use in other nondomesticated species;
considerable extra-label usage of these products does occur, however.
This section reviews the reported use of ivermectin in these species.
A. RODENTS
In addition to the treatment of natural and induced parasitic infections of a

number of rodent species under laboratory conditions, squirrels have
been successfully treated with ivermectin for mite infestation (Notoedres
douglassO. Subcutaneous treatment at a dose rate of 0.5 mg/kg resulted in
almost complete resolution of skin lesions within 5 days; no mites were
recovered from subsequent multiple skin scrapings (Evans 1984).
B. HEDGEHOGS
Ivermectin is reported to be effective against Crenosoma striatum

(95.9%) and Capillaria spp. (100%) when administered at dose rate of
3 mg/kg to hedgehogs (Barutzki, Laubmeier, and Forstner 1987).
C. CARNIVORES
1. Foxes
a. Endoparasites
Blagbum and colleagues (1986) report a study in which wild foxes were
trapped and an assessment of helminth infection made by fecal flotation.
Animals were then divided into groups and treated with a number of
anthelmintics, including ivermectin at 0.4 mg/kg. Of the 9 foxes positive
for Ancylostoma spp. eggs prior to ivermectin treatment, none was
positive by 3 weeks after treatment.
In a study involving more than 500 female arctic and silver foxes, those
treated with ivermectin subcutaneously at 0.2 mg/kg had better concep-
tion and parturition rates than untreated animals (Kopczewski et al.
1987). The numbers of cubs with Toxocara canis eggs in the feces at 30,
60, and 75 days of age was approximately 50% lower for those cubs born
to ivermectin-treated vixens than for those born to controls.
b. Ectoparasites
The efficacy of ivermectin has been assessed against 2 mite species-
Sarcoptes scabiei var canis and Cheyletiella blakei-in foxes.
Sarcoptic mange mites were eradicated from foxes treated with iver-
mectin at dose rates of 0.2 or 0.4 mg/kg once or 0.2 mg twice with a
5-week interval (Berge and Smeds 1984). These authors recommend a
treatment regimen involving an initial dose of 0.4 mg/kg, followed by a
second dose of 0.2 mg/kg, 2 to 3 weeks later, to control this parasite in
foxes.
In a study involving polar foxes, Malczewski and colleagues (1984)
reported that 6 months after 2 subcutaneous ivermectin treatments at
dose rates up to 0.5 mg/kg, 6 weeks apart, no C. blakei mites were
recovered from treated animals; these foxes were also free from clinical
signs of infestation after treatment.
2. Wolves
Treatment of captive wolves with ivermectin by intramuscular injection,
the direct oral route, or in treated meat was found to kill the adult and
nymphal stages of the biting louse, Trichodectes canis (Taylor and
Spraker 1983). Free-ranging packs were subsequently successfully
treated with injections and treated meat. Offspring of previously infested,
treated adults were also reported to be free of lice.
3. Cape Hunting Dog

Following treatment of adult heartworm (Dirofilaria immitis) with thiace-
tarsamide sodium, a single oral dose of ivermectin at 0.2 mg/kg was used
as a microfilaricide in 2 Cape hunting dogs (Loomis and Lee 1984). These
authors warn that heartworm should be considered when importing
canines to or from an enzootic heartworm area; they suggest that
ivermectin at 0.005 mg/kg once a month be considered for prophylactic
treatment.
D. UNGULATES
Because wild ungulates are physiologically similar to the domestic species

for which the ivermectin products are registered and because they harbor
parasites similar to those found in cattle, sheep, pigs, and horses, many
ungulate species are treated with ivermectin.
1. Camels
Ivermectin given at 0.2 mg/kg subcutaneously effectively treats and
controls gastrointestinal nematodes and sarcoptic mange mites in camels
(Camelus dromedarius) (Ibrahim et al. 1981; Tager-Kagan and Robin
1986; Jones 1987). Studies have demonstrated a wide safety margin, but
270 M.D. So11
specific trials have not been done to demonstrate safety in breeding

camels. Camels should not be treated within 28 days of slaughter for
human consumption.
a. Endoparasites
Efficacy against endoparasites was established in 3 controlled studies
involving 32 camels (Robin, Koenig, and Anstey, in press). Treatment at
the recommended dose rate of 0.2 mg/kg provided effective control of the
following gastrointestinal nematodes:*
Haemonchus spp. (adults) H. contortus
H. lonistipes
Ostertagia spp. (adults)
Trichostrongylus spp. (adults) T. axei
T. colubriformis
T. probolurus
T. vitrinus
Camelostrongylus mentulatus (adults and L4)
Impalaia tuberculata (adults)
Nematodirus spathiger (adults)
Chabertia ovina (adults)
Oesophagostomum spp. (adults) O. columbianum
o. venulosum
In these studies, efficacy was also demonstrated against Cooperia spp.,
although the control animals had only small numbers of these parasites.
Boyce and colleagues (1984) report that for 4 trials in which ivermectin
was administered to camels at 0.2 mg/kg the number of trichostrongylid
eggs (Haemonchus, Trichostrongylus, and Cooperia) was effectively
reduced in the feces. Although subcutaneous treatment had no effect on
the number of Trichuris spp. eggs in the feces up to 10 weeks after
treatment, oral treatment reduced Trichuris egg counts by more than 85%.
Robin, Koenig, and Anstey (in press) reported variable efficacy against
Trichuris globulosa following subcutaneous treatment at 0.2 mg/kg.
b. Ectoparasites
A single injection of ivermectin at 0.2 mg/kg subcutaneously markedly
reduces the numbers of Sarcoptes scabiei var cameli and frequently leads
to resolution of clinical signs of scab. Two injections may be necessary to
provide complete control. The efficacy of 2 treatments with a 14- to
16-day interval was confirmed by Hashim and Wasfi (1986), and Opfer-
man (1985), who report the absence of live mites on animals 3 weeks to
12 months after treatment.
* Data on file, MSDRL

Variable efficacy has been demonstrated against the oestrid larvae of

Cephalopina titillator in the nasal passages of camels treated at 0.2 mg/kg
(Tager-Kagan and Robin 1986; Robin, Koenig, and Anstey 1988).
2. Alpacas and Llamas

Ivermectin, injected subcutaneously at 0.2 mg/kg, has been used to
successfully treat sarcoptic mange in alpacas (Alva and Guerrero 1986).
In llamas, this dose was effective against lice infestation (Microthoracius
spp.) (Fowler 1986) and nasal myaisis caused by the oestrid larvae of
Cephenemya sp. (Fowler and Murphy 1985).
Administration of ivermectin at 0.2 mg/kg once a month for 6 months
was used to treat a herd of llamas prophylactically for Parelaphostrongy-
lus tenuis infection (Krogdahl, Thilsted, and Olsen 1987). These authors
recommended that prophylactic treatment at 2 two-monthly intervals
should be considered if llamas are grazed on pastures inhabited by deer
(Odocoileus virginianus).
3. Reindeer
Ivermectin, administered subcutaneously at 0.2 mg/kg, is approved for
treating and controlling warbles (Oedemagena tarandi) in reindeer (Rang-
ifer tarandus) in the United States. Under an IR4 (minor species)
development project conducted in Alaska, a high degree of efficacy was
established in extensive studies (Dieterich 1985). Safety was demon-
strated and no adverse reactions were elicited following treatment of
animals at 5 and 10 times the recommended dose. The withdrawal period
is 56 days.
Ivermectin's efficacy in reindeer is supported by a field study con-
ducted in Sweden (Nordkvist et al. 1984). Grazing reindeer treated with
ivermectin at 0.2 mg/kg were returned to pasture for 150 days during the
winter. Efficacy, determined by slaughtering and processing 5 treated and
5 control animals at the end of this time, was reported to be at or near
100% against Oedemagena tarandi, Cephenemyia trompe, Dictyocaulus
viviparus, and Elaphostrongylus rangiferi. The high level of efficacy,
despite the long treatment/slaughter interval, is due to the reindeers'
decreased exposure tei infective larvae in the winter months. The average
weight loss ofthe treated group over the winter was significantly less than
that of the controls.
4. Deer
Ivermectin's efficacy has been evaluated against various endo- and
ectoparasites in red deer (Cervus elaphus), white-tailed deer (Odocoileus
virginianus), and mule deer (0. hemionus).
272 M.D. SolI
a. Endoparasites
Dictyocaulus viviparus. The high level of efficacy of ivermectin against
lungworm in domestic animals has also been demonstrated in red deer.
Treatment at 0.1 mg/kg subcutaneously 1 to 4 weeks after induced
infections was 100% effective against this parasite (Mackintosh and
Mason 1985). Oral treatment (0.2 mg/kg) of red deer on pasture reduced
fecal larval counts to low levels by 20 days after each of 4 treatments
(Bowie, Mackintosh, and Mason 1987). Mean larval output and pro-
portion of deer shedding D. viviparus larvae at 27 and 33 days after
treatment were significantly lower for ivermectin-treated animals than for
those treated with oxfendazole.
Parelaphostrongylus. In white-tailed deer, ivermectin has been shown to
be effective against certain immature stages of Parelaphostrongylus
tenuis (the meningeal nematode), but not against the adults. Kocan (1985)
reports that treatment at 0.1 or 0.2 mg/kg subcutaneously is effective
against larvae penetrating the abomasum. Treatment on Days 10 and 30
after infection was ineffective. This could be due to the apparent inability
of the avermectins to cross the blood-brain barrier, and the fact that the
larvae of this parasite are generally located in the spinal cord by Day 10
after infection. Fecal larval output was reduced completely by 10 days
after treatment of deer with patent infections, but larvae reappeared in the
feces by Day 28. These findings, together with the absence of eggs and
larvae in the lungs of animals slaughtered 12 days after treatment, indicate
activity against first stage larvae in the lungs and possibly an effect on egg
production (Kocan 1985).
Samuel and Gray (1988) report that ivermectin treatment (0.2 or 0.4
mg/kg) of white-tailed deer artificially infected with the muscle nematode
Parelaphostrongylus andersoni, eliminated first stage larvae from the
feces for up to six weeks. This was attributed to suppressed larval
production by adult females and/or destruction of larvae in the lungs.
Oral ivermectin treatment had little effect on the adults of another
metastrongyloid nematode Elaphostrongylus cervi in red deer (Watson
1986).
b. Ectoparasites
Lice. Treatment of mule and white-tailed deer with ivermectin at
0.2 mg/kg intramuscularly was reported to reduce the number of Linog-
nathus africanus by 90% within 7 days of treatment. A second treatment
21 days later further reduced louse numbers but did not eradicate them
(Foreyt, Rice, and Kim 1986). The interval between treatments in this
case was probably too long relative to the life cycle of the parasite to
completely eliminate the infection.
Fly Strike. Treatment of a single red deer stag suffering from fly strike
with 40 mg of ivermectin subcutaneously (total dose) reportedly aborted
the strike within 3 days (Fletcher 1984).
Warbles. Ivermectin treatment has been reported to be effective for

treatment of warbles in farmed deer (Rafferty 1982). Petrov and col-
leagues (1986) report the use of a "briquette containing salt and a
larvacide such as ivermectin" for treatment and prophylaxis of warbles in
wild deer in the hunting areas of Bulgaria.
5. Bighorn Sheep
Desert bighorn sheep (Ovis canadensis), naturally infested with Psorop-
tes ovis, were successfully treated for scabies with ivermectin adminis-
tered intramuscularly at 0.5 or 1.0 mg/kg-either as a single or repeated
dose with a 14-day interval (Meleney, Wright, and Guillot 1980; Kinzer et
al. 1983). Ivermectin was also effective against this parasite in free-
roaming sheep treated by administering 35 mg of the drug loaded in frozen
gelatin "bullets" and fired from a modified rifle (Meleney, Wright, and
Guillot 1980).
Activity against lungworm has also been demonstrated. Under labora-
tory conditions, Miller and colleagues (1987) report that a single sub-
cutaneous injection (0.2 mg/kg) eliminated first stage Protostrongylus
larvae from the feces of bighorn sheep examined 4 weeks after treatment.
Good efficacy was confirmed by monitoring fecal larval counts of
bighorns treated subcutaneously (0.5 mg/kg) at the time of translocation,
over an extended period of time (Miller and Hobbs 1988).
6. Impala
Horak and colleagues (1984) evaluated the efficacy of ivermectin against
the endo- and ectoparasites of free-living impala treated subcutaneously
at 0.2 mg/kg. Based on parasite recoveries, a high level of efficacy was
demonstrated against the nematode parasites Longistrongylus sabie,
Haemonchus krugeri, Trichostrongylus thomasi, T. colubriformis, Im-
palia tuberculata, Gaigeria pachyscelis, and Oesophagostomum colum-
bianum. Activity was variable against Cooperia hungi, Cooperoides
hamiltoni, and Strongyloides spp.
Of the 4 tick species present on these animals, only Boophilus
decoloratus was markedly affected by treatment. The efficacy of iver-
mectin at this dose was high against the sucking lice Linognathus
aepycerus and L. nevilli but poor against the biting lice Damalinia
aepycerus and D. elongata.
This spectrum of activity is similar to that observed against gastroin-
testinal nematodes and lice on goats treated at 0.2 mg/kg subcutaneously
(unpublished).
7. Other Antelope
Based on fecal egg counts, ivermectin at 0.2 mg/kg is reported to be
effective against gastrointestinal nematodes of the genera Nematodirus,
Trichuris, Trichostrongylus, Ostertagia, Cooperia, Haemonchus, and
274 M.D. Soil
Chabertia, as well as sarcoptic mange mites. Studies were conducted on

8 unspecified antelope species, under zoo conditions, (Frolka and Rot-
inska 1985).
8. Bison
Studies have been conducted to provide data on the use of ivermectin at
0.2 mg/kg subcutaneously for controlling and treating hypodermosis
(Hypoderma bovis) in the American buffalo (Bison bison), under the
auspices of the IR4 (minor species) system (Schillhom, Sikarskie, and
Braselton 1987). These studies demonstrated safety at doses at 0.2 mg/kg
and 1.0 mg/kg. Efficacy against Hypoderma bovis was evaluated under
field conditions, by treating animals at 0.2 mg/kg in the fall. The mean
number of grub lesions in treated animals the following spring was zero
compared to 11.3 for the control animals (Schillhom, Sikarskie, and
Braselton 1987). The efficacy of ivermectin against Hypoderma in the
bison is therefore comparable to that established in cattle for this parasite.
9. Buffalo
The efficacy of ivermectin against louse infection (Haematopinus tuber-
culatus) has been assessed in buffaloes (Lau and Singh 1985). Sub-
cutaneous treatment at 0.2 mg/kg was less effective (85%) than treatment
at 0.4 mg/kg (100%) on evaluation 7 to 14 days later. Efficacy waned to
45% to 50% by 33 days after treatment in this study. It was suggested that
a second treatment 7 to 14 days after the first may result in better efficacy
over a longer period of time.
10. Wild Boar

Kutzer (1986) reports that ivermectin (0.5 mg/kg administered as a single
oral dose) was effective against Sarcoptes suis, Ascaris suum, Meta-
strongylus spp, and Oesophagostomum spp in wild boar. Efficacy against
Trichuris and Capillaria was poorer. The drug was also administered by
mixing in feed. In the case of severe mange infestation, and where animals
are treated in game reserves and hunting grounds, retreatment is recom-
mended after 7 days (Kutzer 1986).
11. Other Species

Numerous anecdotal accounts exist for the use of ivermectin for treating
parasitic infections in a wide variety of other ungulate species. The low
dose volume required to obtain a high level of endo- and ectoparasitic
efficacy, even following intramuscular or oral administration, has resulted
in ivermectin's widespread use in projectile darts, or mixed with feed or
lick formulations, for treating parasitic conditions in a number of other
ungulates both in the wild and in zoos.
E. ELEPHANTS
Five Asian and 4 African elephants infested with lice (Haematomyzus

elephantis) were treated with 2 doses of ivermectin orally at levels
between .059 and .087 mg/kg, 5 to 6 weeks apart (Karesh and Robinson
1985). The authors selected the oral route because, previously, injection
had produced local inflammation at the site of administration. Although
no lice were found 7 days after the first treatment, some were present
prior to the second. No reinfestation occurred following the second
treatment.
Ivermectin treatment (0.2 mg/kg intramuscularly) was effective in
reducing fecal egg counts and improving the physical condition of
approximately 70 orphaned African elephant calves (P. R. Trembath,
personal communication 1988).
F. NONHUMAN PRIMATES
The efficacy of ivermectin for treating primates naturally infected with

arthropod and nematode parasites has been evaluated. Primates have also
been used in a study to evaluate the efficacy of the compound against
induced filarial infections.
1. Mite Infections
Joseph and colleagues (1984) reported that rhesus macaques with pul-
monary acariasis (Pneumonyssus) treated with ivermectin at 0.2 mg/kg
subcutaneously had only dead, fragmented mites in their lungs at ne-
cropsy, while control monkeys had numerous live mites. Histopathologi-
cally, inflammatory lesions were most severe in control animals and
decreased progressively with time in treated monkeys. Efficacy against
lung mites in this species has also been reported by Brack and Rietschel
(1986).
Efficacy has been demonstrated against Psorergates mites in stump-
tailed macaques (Bowman and Griffith 1987).
2. Nematode Infections
Feeding thiabendazole-medicated pellets for 1 week to rhesus monkeys
infected with Strongyloides fuelleborni did not eliminate the infection.
Brack and Rietschel (1986) subsequently treated the animals with iver-
mectin at 0.2 mg/kg subcutaneously for lung mite infections and found
this treatment to be effective against both the mites and the persisting
Strongyloides infection. No eggs were present in the feces 1 month after
treatment. Efficacy has been reported against natural Strongyloides spp.
infection in squirrel-monkeys (Battles, Greiner and Collins 1988).
An apparently healthy cymologous monkey (Macacafusicularis) which
died after routine inhalation anesthesia was found to be infected with the
276 M.D. SolI
filarioid nematode Edesonfifaria mafayensis (Kornegay et af. 1986). The

animal had been treated with ivermectin 17 days prior to surgery: the
authors postulated that treatment had killed many of the nematodes,
resulting in parasitic emboli which led to pulmonary infarction. The
authors advised that this potential complication should be considered
when monkeys with filariasis are treated prior to anesthesia. The apparent
effect on the adult parasite is surprising given the fact that other adult
filarid parasites are relatively refractory to ivermectin treatment (except
at high doses).
In an induced infection study, ivermectin given orally at 0.2 and 0.3
mg/kg was ineffective in preventing development of Brugia mafayi
infective larvae innoculated into leaf monkeys (Presby tis cristata) at the
time of treatment (Mak et af. 1987). This is in line with the finding of
Denham (982) and Campbell (982), who reported ivermectin to be
ineffective against the developing stages of Brugia pahangi in jirds.
IV. Birds
Ivermectin, administered orally, topicallY, and by injection, has been
used for the treatment of mite and nematode infestations in wild and
domesticated birds. Coles (987) considers the drug to work equally well
in these species when given orally or by injection and recommends oral
dosing.
In chickens, somnolence has been observed at doses of 5.4 mg/kg
(Zeman 1987). Listlessness and ataxia occurred at 16.2 mg/kg, but a dose
of 48.6 mg/kg was required to cause mortality. Mousa and colleagues
(1986b) reported that doses of 0.1 to 15.0 mg/kg of ivermectin were
nontoxic to chickens.
A. NEMATODES
a. Ascaridia
A single oral ivermectin treatment at 0.1, 0.2, or 0.4 mg/kg was found to
be ineffective against the tissue phase of Ascaridia galli in chickens
infested 12 days before treatment. These dosages were, however, 96% to
100% effective against adult worms present 40 days after infection (Mousa
et af. 1986b). These authors therefore recommended a repeated treatment
with a 30-day interval for controlling this parasite. The inclusion of
ivermectin in poultry feed at a level of 4 ppm was 84% effective against a
challenge infection of A. galli (R. L. Kilgore, unpublished data).
Successful treatment of Ascaridia cofumbae infection in pigeons required
2 oral doses 0.5 mg) 3 weeks apart (Schepkens, Duchatel, and Vinde-
vogel 1985).
Okaema (1988) reports that subcutaneous injection of guinea fowl
(Numidia meleagris) at 0.14 mg/kg was effective in controlling natural

infections of Ascaridia galli, Heterakis gallinarum, and Sublura suctoria.
b. Capillaria
A single ivermectin treatment, administered orally (1.5 mg, total dose) or
by intramuscular injection (0.3 mg, total dose) was effective in removing
Capillaria spp. from naturally infected pigeons (Schepkens, Duchatel,
and Vindevogel 1985).
Intramuscular treatment (0.4 mg/kg) of 28 raptors (7 genera) reduced
the numbers of Capillaria spp. eggs in the feces by 96%, 8 weeks after
treatment (Kollias, Griener, and Heard 1987). These authors recom-
mended a repeat treatment 3 to 4 weeks after the first for controlling this
parasite.
c. Pelecitus
Oral treatment with ivermectin at 0.2 mg/kg did not result in complete
resolution of lesions attributable to subcutaneous filariasis (Pelecitus sp.)
in a yellow-collared macaw (Allen et al. 1985). Although a needle aspirate
revealed dead microfilariae 14 days after treatment, surgery was required
to remove adult worms.
d. Oxyspirura
Artificial infections of Oxyspirura sp. in the eyes of chickens were treated
successfully with a single dose (.05 mg) ofivermectin instilled into the eye
(Thomas-Baker 1986). A dose of .01 mg/bird was not effective against this
parasite when administered orally or intramuscularly. Natural infections
in the eyes of wood partridges were also removed by intraocular
administration of .05 mg.
e. Syngamus trachea
Seagulls infected with S. trachea have been successfully treated with oral
ivermectin (Rowley 1987).
B. MITES
a. Dermanyssus and Ornithonyssus

Although a noticeable effect was seen on mites (Dermanyssus gallina e)
on fowls treated intraabdominally with ivermectin at 0.6 mg/kg, dose
rates of 1.8 to 5.4 mg/kg were required to obtain adequate efficacy
(Zeman 1987).
Neither ivermectin treatment at 0.5 mg/kg subcutaneously nor the
inclusion of the compound in the feed at levels of 0.4, 2.0, or 4.0 ppm for
28 days was effective against northern fowl mites (Ornithonyssus syl-
viarum) on hens (R. L. Kilgore, unpublished data). These findings are
278 M.D. SoH
supported by those of DeVaney (1985), who reported that low levels of

ivermectin were not effective against this mite.
b. Cnemidocoptes spp.
Ivermectin has been used by intramuscular injection (0.2 mg/kg) and
topical application to treat "scaly-leg" in cage birds ranging in size from
budgies to macaws (Kelso 1984; Tassi 1984; Hogan et al. 1984; Ryan
1986). Heavily infested birds may require a second injection (Kelso 1984;
Hogan et al. 1984) or 2 oral treatments at 0.2 mg/kg 10 to 16 days apart
(Ryan 1987). Topical application of a 1% solution to crusts on the beak,
eyelids, legs, and pericloacal areas of affected birds was reportedly
effective in all but the most severe cases (Tassi 1984). Kummerfeld and
Nolte (1987) reported effective spot treatment of budgerigars for Cne-
midocoptes pilae infestation at levels of 0.4 to 20.0 mg of ivermectin/kg.
c. Sternostoma tracheolum
Dead mites were recovered from the lungs and air sacs of canaries
injected intramuscularly with .02 to .06 mg of ivermectin per bird
(Brownell 1984). In this study, clinical recovery occurred 10 days after
treatment, and the birds began singing 4 days later. Although no mortality
occurred in birds treated at these high levels, immobility was reported.
d. Cytodites nudus
Subcutaneous injection of golden pheasants infested with this air-sac mite
was effective only at high doses (50 mg/kg) (Grimm and Centurier 1986).
e. Pterolichidae
Three doses of ivermectin (0.2 mg/kg orally) at 4-week intervals were
effective for the treatment of quill mite infestation in an adult ostrich
(Hoover, Lochner, and Mullins 1988).
v. Fish
Ivermectin has been used to control parasitic copepods on fish; the drug
has shown some efficacy against nematodes. Dosage levels in fish are
critical: the toxic level is close to the therapeutic level in some species.
Although no mortalities or toxic effects were reported following oral
ivermectin treatment of Atlantic salmon or rainbow trout at 0.2 mg/kg,
the margin of safety is narrow, with mortalities occurring at twice this
dose (Palmer et al. 1987). Mortality of mottled sculpin, 5.5 cm in length,
was reported to have occurred after injection of more than .05 ml of a
.05% ivermectin solution (Heckman 1985).
A. COPEPODS
Compared to untreated controls, numbers of parasitic copepods (Lep-

eopthirus salmonis and Caligus elongatus) on Atlantic salmon and the
numbers of fish with skin lesions were reduced by Day 60 following
administration of ivermectin in feed pellets at intervals of 14 to 21 days
(Palmer et al. 1987). These copepods were also reduced in number
following a single oral dose of ivermectin at 0.2 mg/kg. In this study,
ivermectin was mainly active against adult cope pods and, although
reinfestation with juveniles occurred, they did not survive to the adult
stage.
Lernea spp. on goldfish given an intramuscular injection of a .00008%
solution ofivermectin, administered at a dose rate of 0.1 ml/50 g, began to
disintegrate 4 days after treatment. After 6 days, the fish were clear of
parasites. Ulcers at the point of attachment were healed by Day 10
(Hyland and Adams 1987).
B. NEMATODES
Ivermectin administered intramuscularly or in the water was moderately

effective against Rhabdochona sp. in mottled sculpin (Heckman 1985).
Although subcutaneous treatment at dose rates of 0.1 to 10.0 mg/kg was
effective against Anguillicola crassus-a nematode found in the air
bladder of eels-all ivermectin-treated eels died within 2 weeks (Tara-
chewski, Renner, and Mehlhorn 1988).
IV. Reptiles
Although adverse reactions were not recorded in snakes given ivermectin
at 0.2 mg/kg (Lawrence 1984; Stanchi and Grisolia 1986), and the corn
snake (Elapha guttata) tolerated doses of up to 1.0 mg/kg without
apparent adverse clinical signs, chelonians were found to be sensitive to
the effects of the compound at much lower levels (Teare and Bush 1983).
Red-footed tortoises were killed by treatment at 0.4 mg/kg; paralysis
occurred with doses as low as 0.05 mg/kg when it was repeated within
72 hours. Doses which induced minimal signs of toxicosis ranged from
0.025 mg/kg in leopard tortoises to 0.1 mg/kg in box tortoises. A dose
of 0.05 mg/kg was considered safe in red-footed tortoises if treatment
intervals were not less than 7 days (Teare and Bush 1983).
A. SNAKES
Ivermectin treatment at 0.2 mg/kg subcutaneously was effective in the

treatment of snakes for mite infestation and pythons for tick infestation
(Lawrence 1984). This dose was reported to be 83% effective against
280 M.D. SolI
Ophinyssus sp. mites by Stanchi and Grisolia (1986). These authors

recommended that treated snakes should be placed in parasite-free
housing after treatment.
B. CHELONIANS
Repeated treatment at .05 mg/kg weekly for 2 to 6 weeks reduced

shedding of eggs and larvae (Proatracti sp. and Kolicephalus sp.) in the
feces of red-footed tortoises (Teare and Bush 1983).
Conclusion
Ivermectin is efficacious against a wide-range of endo- and ectoparasites,
in a wide variety of hosts, and is formulated in a variety of dosage forms.
Although mammals tolerate ivermectin very well, it must be empha-
sized that the safety of the drug for nondomestic species-with the
exception of camels and reindeer-has not been proven in controlled
studies. Use in nonmammalian species is not without risk, especially for
reptiles and fish, where the therapeutic dose may be close to the toxic
level.
Similarly, because the compound's efficacy against the wide array of
parasites infecting laboratory, zoo, wild, and exotic animals has not been
proven in controlled studies, clinical response to "extra-label" usage of
the product should be monitored closely.
REFERENCES
Abbeele J van den, D'Haeseleer F, Goossens M (1986) Efficacy of ivermectin on

the reproductive biology of Glossina palpalis. Annales de la Societe Beige de
Medicine Tropicale 66:167-172.
Abo-Shenada MN, Herbert IV (1984) Anthelmintic effect of levamisole, iver-
mectin, albendazole and fenbendazole in larval Toxocara canis infection in
mice. Res. Vet. Sci. 36:87-91.
Alcaino H, Gorman T, Imbert R (1984) Treatment of experimental trichinelliasis
in the rat with febantel and ivermectin. Parasitologie al Dia 8:79-83.
Allen JL, Kollias GV, Griener EC, Boyce W (1985) Subcutaneous filariasis
(Pelecitus sp.) in a yellow collared macaw. Avian Dis. 29:891-894.
Alva J., Guerrero C (1986) Use ofivermectin against sarcoptic mange in alpacas.
IVITA, Boletin de Divulgacion No. 21:23-27.
Ashmawy K, Fahmy MM (1987) Acaricidal efficiency of Ivomec against psoroptic
and sarcoptic mites infesting rabbits. J. Egypt. Vet. Med. Assoc. 47:315-319.
Azambuja PD, Lima Gomes JEP, Lopes F, Garcia ES (1985) Efficacy of
ivermectin against the bloodsucking insects Rhodnius prolixus and Hemiptera
triatominae. Memoirs o/the Institute Oswaldo Cruz 80:439-442.
Barutzki D, Laubmeier E, Forstner MJ (1987) Endoparasite infestation of
hedgehogs with a contribution to therapy. Tierarztl. Prax. 15:325-331.
Battles AH, Adams SW, Courtney CH, Mladinich CRT (1987) Efficacy of
ivermectin against natural infection of Syphacia muris in rats. Lab. Ani. Sci.
37:791-792.
Battles AH, Greiner EC, Collins BR (1988) Efficacy of ivermectin against natural
infection of Strongyloides spp. in squirrel-monkeys. Lab. Ani. Sci. 38:474-485.
Baumans V, Havenaar R, Van Herck H, Rooymans TP (1988) The effectiveness
ofIvomec and Neguvon in the control of murine mites. Lab. Anim. 22:243-245.
Berge GN, Smeds E (1984) Efficacy of ivermectin against S. scabiei infestations of
foxes. Nord. Vet. Med. 36:156-161.
Bianco AE, Nwachukwa MA, Townson S, Doenhoff MJ, Muller RL (1986)
Evaluation of drugs against Onchocerca microfilariae in an inbred mouse
model. Trop. Med. Parasit. 37:39-45.
Blagburn BL, Swango LJ, Hendrix CM, Lindsay DS (1986) Comparative effica-
cies of ivermectin, febantel, fenbendazole and mebendazole against helminth
parasites of gray foxes. 1. Am. Vet. Med. Assoc. 189:1084-1085.
Blair LS, Campbell WC (1980) Suppression of maturation of Dirofilaria immitis in
Mustela putorius by a single dose of ivermectin. 1. Parasit. 66:691-692.
Blair LS, Williams E, Ewanciw DV (1982) Efficacy of ivermectin against third
stage Dirofilaria immitis larvae in ferrets and dogs. Res. Vet. Sci. 33:386-387.
Bowie, JY, Mackintosh CG, Mason PC (1987) Effect of ivermectin and oxfen-
dazole on shedding of larvae of the lungworm Dictyocaulus viviparus by Deer.
New Zeal. Vet. 1. 35:8-10.
Bowman TA, Griffith JW (1987) Comparison of treatments for Psorergates mites
in stumptailed macaques. Lab. Ani. Sci. 37;100-102.
Boyce W, Kollias G, Courtney CH, Allen J, Chalmers E (1984) Efficacy of
ivermectin against gastrointestinal nematodes in dromedary camels. 1. Am.
Vet. Med. Assoc. 185:1307-1308.
Brack M, Rietschel W (1986) Ivermectin for the control of Strongyloides
fuelleborni in rhesus monkeys. Kleintierpraxis 31:29.
Brownell JR (1984) Sternostoma tracheolum-continuing studies. Proc. 33rd
Western Poultry Disease Con!, Davis, CA, pp 91-93.
Calc ott PH, Fatig RO (1984) Avermectin modulation of gamma-aminobutyric acid
binding to membranes of rat brain, brine shrimp and a fungus. J. Antibiot.
37:797-801.
Campbell WC (1982) Efficacy of the avermectins against filarial parasites. Vet.
Res. Commun. 5:251-262.
Campbell WC, Blair LS, Lotti VJ (1979) Efficacy of avermectins against Trichi-
nella spiralis in mice. 1. Helminthol. 53:254-256.
Carrillo M, Barriga 0 (1987) Anthelmintic effect oflevamisole hydrochloride and
ivermectin on tissue toxocariasis of mice. Am. 1. Vet. Res. 48:281-283.
Chiu SHL, Sestokas E, Taub R, Buhs RP, Green M, Sestokas R, VandenHeuval
WJA, Arison BH, Jacobs TA (1986) Metabolic disposition of ivermectin in
tissues of cattle, sheep and rats. Drug Metab. & Dispos. 14:590-600.
Coles BH (1987) Ivermectin for birds. Vet. Rec. 120:604.
Curtis SK, Housley R, Brooks DL (1988) The use of ivermectin in rabbits for the
treatment of ear mites. Proc. 125th Mtg. AVMA, Portland, Oregon, Abstr. 113.
Denham DA (1982) The effects of some avermectins on the growth of Brugia
pahangi. Meth. & Find. Exper. CUn. Pharmacol. 4:347-350.
Devaney E, Howells RE (1984) The Microfilaricidal activity ofivermectin in vitro
and in vivo. Trop. Med. Parasit. 35:47-49.
282 M.D. SoIl
DeVaney JA (1985) Progress on control of northern fowl mites on caged laying

hens. Vet. Parasit. 18:289-295.
Dieterich RA (1985) An overview of Alaskan reindeer diseases and current control
methods. In Fennessy PF, Drew KR (eds) Biology ofDeer Production, Proc. of
Intern. Conf., Dunedin, NZ, pp 97-100.
Distelmans W, D'Haeseleer D, Mortelmans J (1983) Efficacy of systemic adminis-
tration of ivermectin against tsetse flies. Annales de la Societe BeIge de
Medecine Tropicale 63:119-120.
Drummond RO (1980) Mouse-Cuterebra animal systemic insecticide tests. Proc.
Entomol. Soc. Am. 5:230-231.
Evans RH (1984) Iverrnectin treatment of notoedric mange in two fox squirrels.
J. Am. Vet. Med. Assoc. 185:1437-1438.
Fletcher J (1984) Iverrnectin in fly strike. Vet. Rec. 115:475.
Foreyt WJ, Rice DH, Kim KC (1986) Pediculosis of mule deer and white-tailed
deer fawns in captivity. J. Am. Vet. Med. Assoc. 189:1172-1173.
Fowler ME (1986) Lice in llamas. AvialExotic Prac. 3:22-25.
Fowler ME, Murphy J (1985) Cephenemya sp. infestation in the llama. Calif. Vet.
39:10-12.
Frolka J, Rotinska J (1985) Possible use of Ivomec and Eqvalan in the treat-
ment of parasitic diseases of animals in zoological gardens. Veterinarstvi 35:
415-416.
Frossard TL (1981) Studies on the suitability of rabbits as a model for testing
systemic acaricides under laboratory conditions. Thesis, Munich University,
39 pp.
Grimm F, Centurier C (1986) Cytodites nudus-an air sac mite in golden
pheasants. Praktische Tierarzt 67:198-199.
Grove DI (1983) The effects of 22,23 dihydroavermectin BJ on Strongyloides ratti
and S. stercoralis infections in mice. Ann. Trop. Med. Parasit. 77:405-410.
Harvey RG (1987) Use of ivermectin for guinea pig mange. Vet. Rec. 120:351.
Hashim DH, Wasfi IA (1986) Ivermectin treatment of camels naturally infected
with sarcoptic mange. World Ani. Rev. 57:26-29.
Hasslinger MA, Wiethe T (1987) Oxyurids of small laboratory animals and their
control with iverrnectin. Tierarzt. Prax. 15:93-97.
Heckmann RA (1985) The efficacy of iverrnectin against nematodes of fish.
Proceed. of the 60th Annual Mtg. of the Am. Soc. for Parasit. Abstract 24.
Hogan HL, Joseph B, Henrickson R, Lowenstine L (1984) Efficacy and safety of
iverrnectin for scaly leg mite infestation in parakeets. Proc. of Annual Mtg. of
the Am. Assoc. of Zoo Veter. Abstract 156.
Hoover JP, Lochner FK, Mullins SB (1988) Quill mites in an osrich with rhinitis,
sinusitis and air saculitis. Compan. Anim. Pract. 2:23-26.
Horak IG, Boomker J, Kingsley SA, De Vos V (1984) The efficacy of iverrnectin
against helminth and arthropod parasites of impalla. J. S. Afr. Vet. Assoc.
54:251-253.
Hunt LM (1982) Guinea pig-Amblyomma americanum animal systemic insecticide
test. Proc. Entomol. Soc. Am. 7:253.
Hyland KPC, Adams SJR (1987) Iverrnectin for use in fish. Vet. Rec. 120:539.
Ibraham MS, Mohamed AR, EI-Balkemy FA, Omran H, EI-Mekkauri MF (1981)
Studies on the relation between the effect of IvomecT as a parasitic control and
the general health condition of camels. Res. Bull. No. 375, Oct. 1981, Zagazig
Univ., Egypt.
Ishii AI, Terada M, Kwo H, Hayashi M, Sano M (1983) Studies on chemotherapy

of parasitic helminths; effects of avermectin B)a on Angiostrongylus cantonen-
sis in rats. IntI. J. Parasit. 13:491-498.
Jones PGH (1987) Recent advances in parasite control as an aid to increasing
livestock health and productivity. J. Egypt. Vet. Med. Assoc. 47:347-364.
Joseph BE, Wilson DW, Hendrickson RV, Robinson PT, Benirschke K (1984)
Treatment of pulmonary acariasis in rhesus macaques with ivermectin. Lab.
Ani. Ad. 34:360-364.
Karesh WB, Robinson PT (1985) Ivermectin treatment of lice infestations in two
elephant species. J. Am. Vet. Med. Assoc. 187:1235-1236.
Kelso PC (1984) Ivermectin cures scaly mite infestation. Vet. Med. 79:446.
Kinzer HG, Meleney WP, Lange RE, Houghton WE (1983) Preliminary evalu-
ation of ivermectin for control of Psoroptes ovis in desert bighorn sheep.
J. Wildlife Dis. 19:52-54.
Kocan AA (1985) The use of ivermectin in the treatment and prevention of
Paralephostrongylus tenuis in white-tailed deer. J. Wildlife Dis. 21:454-455.
Kollias GV, Griener EC, Heard D (1987) The efficacy of ivermectin and
praziquantel against gastrointestinal nematodes and trematodes in raptors.
Proc. of 1st Conf. in Zool. and Avian Med., Florida, pp 408-409.
Kopczewski A, Malczewski A, Wroblewska M, Zdunkiewicz T (1987) Effect of
ivermectin on the fertility and health offoxes. Med. Weterynaryjna 43: 102-104.
Kornegay RW, Giddens WE, Morton WR, Knitter GH (1986) Verminous vasculi-
tis pneumonia and infarction in a cymologous monkey after treatment with
ivermectin. Lab. Ani. Ad. 36:45-47.
Krogdahl DW, Thilsted JP, Olsen SK (1987) Ataxia and hypermetria caused by
Parelaphostrongylus tenuis infection in llamas. J. Am. Vet. Med. Assoc.
190: 191-193.
Kummerfe1d N, Nolte CS (1987) Treatment of mite infestation in budgerigars and
finches using ivermectin spot-on methods. Kleintierpraxis 32:293-296.
Kutzer E (1986) Zur behandlung der Sarcoptesraude bei wild- und haus-
schweinen mit ivermectin. Dtsch. tiearztl. Wschr. 93:426-429.
Lau HD, Singh NP (1985) Efficacy of ivermectin in the control of the louse
Haematopinus turberculatus in buffaloes. Proc. of 11th Con! of the WAAVP,
Abstract 47.
Lawrence U (1984) Ivermectin as an ectoparasiticide in snakes. Vet. Rec.
115:441-442.
Loomis MR, Lee CD (1984) Canine heartworm infection in Mrican Cape Hunting
Dogs. Annual Mtg. of the Am. Assoc. of Zoo Veter. p. 137.
Macchioni G, Marconcini A, Sbrana L (1983) Sarcoptic mange in the domestic
rabbit and guinea pig. Annali della Facolta di Medicina Veterinaria de Pisa
36:287-294.
Mackintosh CG, Mason PC (1985) Anthelmintics and lungworm in deer. In
Fennesy PF, Drew KR (eds) Biology of Deer Production, Proc. of Conf.,
Dunedin, N.Z. pp 131-133.
Mak JW, Lam PL, Rain An, Suresh K (1987) Chemoprophylactic studies with
ivermectin against subperiodic Brugia malayi infection in the leaf monkey.
J. Helminthol. 61:311-314.
Malczewski A, Kopczewski A, Malczewski M, Zielinski J (1984) Mange due to
Cheyletiella blakei in polar foxes in Poland. Zentralblatt fur Bakteriologie
Mikrobiologie und Hygiene 258:412-413.
284 M.D. SolI
Manurang J, Partoutomo S, Stevenson P (1985) Treatment of rabbit ear mange

caused by Notoedres cati with ivermectin and neguvon. Panyakit Hewan
17:308-311.
Meleney WP, Wright FC, Guillot FS (1980) Identification and control ofpsoroptic
scabies in bighorn sheep. Proc. 84th Annual Mtg. of the U.S. Ani. Health
Assoc., Louisville, Kentucky, pp 403-407.
Miller MW, Hobbs NT (1988) Ivermectin for treating lungworms in mountain
sheep: an appropriate technology with a specific utility. Wildl. Soc. Bull.
16:236-238.
Miller MW, Hobbs NT, Rutherford WH, Miller LLW (1987) Efficacy of injectable
ivermectin for treating lungworm infections in mountain sheep. Wildl. Soc.
Bull. 15:260-263.
Mousa S, Gad N, Sokkar I, Raheem MA (1986a) Efficacy of a single injectable
dose of ivermectin for psoroptic and sarcoptic mange in rabbits. Assiut Vet.
Med. J. 17:237-239.
Mousa S, Gad N, Sokkar I, Raheem MA (1986b) Investigations on the efficacy of
ivermectin for ectoparasites and nematodes in chickens. Assiut Vet. Med.
J. 17:225-233.
Nordkvist M, Christens son D, Rehbinder C (1984) A deworming field trial with
ivermectin in reindeer. Rangifer 4:10-15.
Okaeme AN (1988) Ivermectin in the control of helminthiasis in guinea fowl. Vet.
Q. 10:70-71.
Olsen RW, Snowman AM (1985) Avermectin B1a modulation of gamma amino-
butyric acid benzodiazapine receptor binding in mammalian brain. J. Neuro-
chern. 44:1074-1082.
Opferman RR (1985) Treatment of sarcoptic mange in a dromedary camel. J. Am.
Vet. Med. Assoc. 187:1240-1241.
Ostlind DA, Cifelli S (1981) Efficacy of thiabendazole, levamisole and the major
natural avermectins against Trichostrongylus colubriformis in the gerbil. Res.
Vet. Sci. 31:255-256.
Ostlind DA, Cifelli S, Lang R (1979) Insecticidal activity of the antiparasitic
avermectins. Vet. Rec. 105;168.
Ostlind DA, Nartowicz MA, Mickle WG (1985) Efficacy of ivermectin against
Syphacia obvelata in mice. J. Helminthol. 59:257-261.
Palmer R, Rodger H, Drinan E, Dwyer C, Smith PR (1987) Preliminary trials on
the efficacy of ivermectin against parasitic copepods of Atlantic salmon. Bull.
Europ. Assoc. Fish Pathol. 7:47-54.
Petrov D, Nedelchev N, Meshkov S, Rusev I, Yuzer P, Paliiski E (1986) A study
of hypodermatosis in wild ruminants. Vet. Sbirka 84:26-28.
Pong SS, Wang CC (1980) The specificity of high affinity binding of avermectin
B1a to mammalian brain. Neuropharm. 19:311-317.
Pong SS, Wang CC, Fritz LC (1980) Studies on the mechanism of action of
avermectin B1a. J. Neurochem. 34:351-358.
Prosl H, Kanout AG (1985) Treatment of ear mange in rabbits with ivermectin.
Berliner und Munchener Tierarztliche Wochenschrift 98:45-47.
Rafferty GC (1982) Red deer endoparasites. Vet. Rec. 111:565.
Rajasekariah GR, Deb BN, Dhage KR, Bose S (1986) Response of laboratory-
adapted human hookworm and other nematodes to ivermectin. Ann. Trop.
Med. Parasit. 80:615-621.
Rapic D, Dzakula N, Matic-Piantanida D (1982) Anthelmintic activity of iver-

mectin against Trichinella spiralis larvae in rats. Vet. Arh. 52:131-139.
Robin B, Koenig K, Anstey MD (in press) Efficacy ofivermectin against internal
parasites in dromedaries (Camelus dromedarius). Revue Scientifique et Tech-
nique.
Romero JR, Valentini HG (1984) Efficacy ofavermectin in the treatment of mange
in rabbits. Vet. Argent. 1:871-874.
Rowley J (1987) Syngamus trachea in three seagulls. Compan. Anim. Pract. 1:46.
Ryan T (1986) Cnemidocoptic mite infestation in cage birds. Mod. Vet. Prac.
67:525-526.
Ryan T (1987) Cnemidocoptic mite infestation in cage birds. Cline. Ins. 2:305-306.
Samuel WM, Gray JB (1988) Efficacy of ivermectin against Parelaphostrongylus
andersoni in white-tailed deer. J. Wildl. Dis. 24:491-495.
Sanmartin-Duran ML, Sernandez L, Yebra MJ, Santamarina MT, Ubeira FM
(1986) Anthelmintic effect ofivermectin on the endogenous cycle of Trichinella.
Revista Iberica de Parasitologia 46:189-193.
Sayles PC, Jacobson RH (1983) Effects of various anthelmintics on larval stages
of Nematospiroides dubius. J. Parasit. 69:1079-1083.
Schepkens E, Duchatel JP, Vindevogel H (1985) Treatment of Ascarida columbae
and Capillaria spp. infestations in pigeons with ivermectin. Ann. Med. Vet.
129:475-485.
Schillhorn van Veen TW, Sikarskie JG, Braselton WE (1987). Safety, efficacy and
tissue residues of ivermectin used to control hypodermosis in Bison. Proc.
Con! of Res. Workers in Ani. Dis., Chicago, IL, Abstraact 205.
Silverman J, Blatt H, Lerro A (1983) Effect of ivermectin against Myobia musculi.
Lab. Ani. Sci. 33:487.
Stanchi NO, Grisolia CX (1986) Use ofivermectin in the treatment of snake mite
infection. Vet. Argent. 3:26.
Tager-Kagan P, Robin B (1986) Results ofa trial ofivermectin against parasites of
dromedaries in Niger. Rev. Elev. Med. Vet. Pays Tropic. 39:333-340.
Taraschewski H, Renner C, Mehlhorn H (1988) Effects of levamisole hydrochlo-
ride, metrifonate, fenbendazole, mebendazole and ivermectin on Anguillicola
crassus pathogenic in the air bladder of eels. Parasit. Res. 74:281-289.
Tassi P (1984) Topical treatment for scaly mites. Vet. Med. Small Ani. Clinic.
79:1447.
Taylor WP, Spraker TH (1983) Management of a biting louse infestation in a free
ranging wolf population. Proc. Annual Mtg. Am. Assoc. of Zoo Veterin.
pp 40-41.
Teare JA, Bush M (1983) Toxicity and efficacy ofivermectin in chelonians. J. Am.
Vet. Med. Assoc. 183:1195-1197.
Thomas-Baker B (1986) Ivermectin as a treatment for ocular nematodiasis in
birds. Proc. Annual Mtg. of Am. Assoc. Zoo Veterin. pp 99-100.
Watson TG (1986) Efficacy of drenching deer and wapiti with particular reference
to Elaphostrongylus cervi and Dictyocaulus viviparus. Proc. N .Z. Vet. Assoc.
Deer Branch Course No.3, Rotorua: 170-182.
Wilkins CA, Conroy J, Ho P, O'Shanny WJ, Capizzi T (1981) The effect of
ivermectin on the live mass, period of attachment and percent control of ticks.
In Whitehead GB, Gibson JD (eds) Tick Biology and Control, Rhodes Univer-
sity, Grahamstown, SA, pp 137-142.
286 M.D. SoIl
Wilkins CA, Conroy JA, Ho P, O'Shanny WJ, Malatesta PF, Egerton JR (1980)
Treatment of psoroptic mange with avermectins. Am. J. Vet. Res. 41:2112-
2113.
Williams M, Yarbrough GG (1979) Enhancement of in vivo binding of diazepam
by a novel anthelmintic agent, avermectin B)a. Eur. J. Pharmacol. 56:273-276.
Wing SR, Courtney CH, Young MD (1985) Effect ofivermectin on murine mites.
J. Am. Vet. Med. Assoc. 187:1191-1192.
Wright FC, Riner JC (1985) Comparative efficacy of injection routes and doses of
ivermectin against Psoroptes in rabbits. Am. J. Vet. Res. 46:752-754.
Zahner H, Sanger I, Lammler G, Muller HA (1987) Effect of ivermectin on
Dipetalonema uiteae and Litomosoides carinii infections of Mastomys natalen-
sis. Trop. Med. Parasit. 38:117-122.
Zeman P (1987) Systemic efficacy of ivermectin against Dermanyssus gallinae in
fowls. Vet. Pathol. 23:141-146.
CHAPTER 20

Richard A. Dybas
I. Introduction
II. Products
A. AVID®, AGRI-MEK®, VERTIMEC®
B. AFFIRM®
III. Applications
A. Red Imported Fire Ant (Solenopsis invicta)
B. Citrus
1. Citrus Rust Mite (Phyliocoptruta oleivora)
2. Broad Mite (Polyphagotarsonemus latus)
3. Citrus Red Mite (Panonychus dtn)
4. Citrus Thrips (Sdrtothrips dtn)
5. Beneficial Organisms
C. Ornamentals
1. Twospotted Spider Mite (Tetranychus urticae)
2. Leafminer (Liriomyza trifoliz)
3. Thrips Species
D. Cotton
1. Spider Mites (Tetranychus urticae, T. turkestani, T. pacijicus,
T. dnnabarinus)
2. Lepidoptera
3. Cotton Aphid (Aphis gossypil)
E. Vegetables (celery, tomato, pepper)
1. Leafminers (Liriomyza trifolii; L. sativae)
2. Twospotted Spider Mite and Tomato Russet Mite (T. urticae
and Aculops lycopersicz)
3. Armyworms (Spodoptera exigua and S. eridania)
4. Tomato Pinworm (Keiferia lycopersicelia)
5. Beneficial Arthropods
F. Pears
1. Pear Psylla (Psylia pyricola)
2. Twospotted Spider Mite, European Red Mite, Pear Rust Mite
(T. urticae, Panonychus ulmi, Epitrimerus pyn)
G. Miscellaneous Crops
1. Strawberry
2. Deciduous Tree Nuts
288 Richard A. Dybas
I. Introduction
Initial indications that the avermectins possessed activity against arthro-
pods were obtained in studies with avermectin B'a against the confused
flour beetle (Tribolium confusum) and the sheep blowfly larva (Lucilia
cuprina). Evaluation of the 8 major natural avermectins and several
hundred semisynthetic avermectin analogs and derivatives showed that
avermectin B" the major component of the fermentation of Streptomyces
avermitilis, was extremely toxic to acarines and to a wide range of
insects, induding members of the orders Diptera, Homoptera, Coleop-
tera, and Lepidoptera (Putter et al. 1981). Avermectin B, (MK-936), with
the common name abamectin, was selected for use against arthropod
pests of economic importance in horticulture and agriculture.
II. Products
A. AVID®, AGRI-MEK®, VERTIMEC®
Abamectin for use in foliar spray applications is formulated as a 1.8% w/v
emulsifiable concentrate containing 0.15 lb abamectin per U.S. gallon.
The oral LD50 of this product in rats is 650 mg/kg body weight, while the
dermal LD50 in rabbits is greater than 2000 mg/kg body weight. Rabbits
exhibit slight-to-moderate eye irritation and very slight-to-slight skin
irritation. Skin sensitization was not observed in guinea pigs with the
emulsifiable concentrate formulation.
The product is being developed for use on a number of horticultural and
agronomic crops to control phytophagous mites and insects. It is recom-
mended at rates in the range of 0.005 to 0.0251b abamectin/acre (5.4 to 27
grams ai/hectare), in volumes of water ranging from 5 to 1000 gallons per
acre, depending on the crop to be treated and application equipment. For
certain crop applications, e.g., for control of mites and insects on citrus,
pear, and deciduous tree nuts, the product is recommended in combina-
tion with a 0.25% narrow-range paraffinic oil in dilute sprays or a
minimum of 1 gallon oil per acre (9.5 liters/ha) in concentrate sprays.
The maximum recommended application rate for citrus, pears, and
deciduous tree nuts is 0.025 lb active ingredient (ai)/acre (20 fl oz
AGRI-MEK®/acre). The maximum recommended rate for ornamental
plants, cotton, vegetables, and strawberries is 0.02 lb ai/acre (16 fl oz
AGRIMEK®/acre).
A maximum of 0.075 lb ai/acre per growing season will be allowed on
citrus, pears, and tree nuts, and 0.20 lb ai/acre on vegetable crops.
With the exception of mild symptoms of phytotoxicity to certain ferns,
and slight spotting of carnation foliage, crops tolerate the product well
when the recommended application directions are followed. This applies
to abamectin sprayed in water or, where recommended, in combination
with paraffinic spray oils (Green et al. 1985a).
20. Abamectin Use in Crop Protection 289
B. AFFIRM®
Abamectin for use in controlling the red imported fire ant (Soienopsis
invicla) is formulated as a bait containing 0.011% w/w abamectin in
soybean oil on pregelled defatted corn grits. The oral LDso in rats and
dermal LDso in rabbits is greater than 2000 mg/kg, and the product is
nonirritating to rabbit skin.
AFFIRM® is intended for use at a rate of 1.0 Ib of bait per acre for
broadcast application through ground or aerial equipment, or 5 to 7
tablespoons of bait for treatment of individual fire ant mounds. The
product is recommended for nonagricultural lands such as lawns, turf,
golf courses, and parks. The U.S. EPA-approved label restricts bait
application to land other than pastures, range lands and crop land.
III. Applications
A. RED IMPORTED FIRE ANT (SOLENOPSIS INVICTA)
The red imported fire ant is an important pest in the southern United
States. The worker ants inflict painful stings to humans, they feed directly
on crops, and the ant mounds cause damage to farm machinery. Any
successful control strategy requires eliminating the worker ants and/or
the physogastric queen, who is responsible for the reproductive capacity
of the colony.
Studies conducted in the USDA's Gainesville, FL, laboratory demon-
strated that abamectin exhibits a rate-related delayed toxicity to red
imported fire ant (RIFA) workers. This delayed effect is an important
criterion for efficacy of a bait toxicant against RIFA, since the worker
ants must gather the bait toxicant and distribute it throughout the colony
before the effects of treatment are observed.
In laboratory studies, abamectin at a concentration of 0.1 % killed more
than 75% of RIFA workers after 10 days exposure. Delayed toxicity to
worker ants was observed; less than 10% kill occurred 1 day after
treatment at that concentration (Lofgren and Williams 1982; Williams
1983). When entire colonies were treated in the laboratory, 0.0025%
abamectin in soybean oil caused loss of all brood production and sterility
of the RIFA queen. Abamectin effects at this concentration appeared
irreversible. Lower abamectin concentrations-for example, 0.00025% in
soybean oil-caused a reduction in brood production; however, the
effects were transient, with a return to normal egg production in 16
weeks.
Histological examination of the ovaries of sterile queens showed
several tissue and cell abnormalities. The damage was characterized by
hypertrophy of the squamous epithelium that sheaths the ovarioles, and
pycnosis of the nurse-cell nuclei. According to the authors, the lack of
eggs containing yolk, the very small number of eggs produced, and the
small size of the eggs all point to an effect on the neurosecretory cells
(Glancey, Lofgren, and Williams 1982).
Abamectin applied at lib of bait per acre (50 mg ai/acre) resulted in
virtually complete elimination in worker brood (> 95%) and a significant
reduction in colony ant populations (> 85%) at 6 to 12 weeks post
treatment. Consistent with the irreversible effects on RIFA queens noted
in the laboratory, the USDA researchers found no worker brood in
treated colonies 18 weeks after application (Lofgren and Williams 1982;
Williams 1985).
Treating individual fire ant mounds with abamectin bait provided high
colony mortality, in the range of 55% to 82%, 12 to 21 weeks after
treatment with 1,4, and 7 tablespoons bait per mound. The 7-tablespoon
rate provided the fastest kill, although by 21 weeks there was no
difference in colony mortality between the 4- and 7-tablespoon rates. The
7-tablespoon rate of AFFIRM® is recommended for use by homeowners,
since more rapid kill of worker ants and elimination of fire ant colonies is
expected in this market (Greenblatt et al. 1986). Apperson, Powell, and
Browne (1985) and Lemke and Kissam (1987) also reported that AF-
FIRM® provided greater than 80% reduction in colonies with brood and
colony mortality following treatment of individual mounds.
Preliminary evidence from field trials suggests that fire ants do not
rapidly recolonize abamectin-treated areas. The gradule elimination of
worker populations and the presence of live sterile queens in treated
colonies may prevent adoption and establishment of newly mated phy-
sogastric queens in abamectin plots (Greenblatt et al. 1986). Therefore,
the slow action of abamectin appears to offer the advantage of providing
high mortality to treated colonies while delaying or preventing reinfes-
tation of treated areas.
B. CITRUS
1. Citrus Rust Mite (Phyllocoptruta oleivora)

The eriophyid citrus rust mite is an important pest in the United States,
Brazil, and other humid citrus-growing areas, causing extensive fruit
russeting, reduced fruit size and juice quality, and decreased internal
solids. When tested under laboratory conditions, abamectin was highly
toxic by contact to P. oleivora on leaf discs, with an LC90 of 0.02 ppm
(McCoy, Bullock, and Dybas 1982). Mortality was slow, especially at
low doses, requiring 48 to 72 hours to achieve maximum kill. No toxicity
to P. oleivora eggs was observed up to 25 ppm, the highest concentration
tested.
Preliminary field trials conducted on citrus indicated that abamectin
was effective in reducing initial P. oleivora populations to below eco-
nomic threshold levels, at application rates of 0.0125 to 0.05 lb ai/acre
(13.5-54.0 g ai/ha). Residual control of citrus rust mites was very short,
however, with reinfestation of treated fruit occurring 4 weeks post

treatment under some conditions.
Laboratory and field studies with tetranychid and eriophyid mites
showed that low, nonacaricidal concentrations of emulsified paraffinic
oil were effective in enhancing abamectin' s contact and residual toxi-
city against mite species (Dybas 1985). Subsequently, several authors
reported that applications of abamectin at 0.0125 to 0.025 lb ai/acre
(13.5-27.0 g ai/ha), in combination with a minimum of 1 gallon oil per
acre, resulted in enhanced activity and provided greater than 90%
reduction of initial rust mite populations. Furthermore, rust mite popula-
tions and fruit damage were controlled at these rates for up to 16 weeks
under the demanding conditions of Florida and Texas (McCoy, Bullock,
and Dybas 1982; Childers and Sorrell 1983; French 1984; Childers 1985).
These results clearly show that abamectin, when applied in combination
with a paraffinic oil on citrus, can control rust mite populations below
economic thresholds at one half to one quarter the rates effective with-
out oil.
Thus, abamectin combined with oil is recommended to control
P. oleivora infestations on all varieties of citrus, including lemons,
orange, grapefruit, and tangerines. Using this practice, essentially russet-
free fruit is obtained with abamectin treatments under virtually all
application conditions. There has been no evidence of phytotoxicity or
fruit injury from use of abamectin 0.15 EC formulation at recommended
rates in combination with paraffinic crop oil.
2. Broad Mite (Polyphagotarsonemus latus)

Abamectin is highly toxic to Polyphagotarsonemus latus adults on fruit
under laboratory conditions, with an LC90 of 0.05 ppm (Dybas and Green
1984). Brown reported that abamectin was not toxic to P. latus eggs
(unpublished); however, abamectin residues on citrus fruit treated at
0.1 ppm provided 100% mortality to emerging larvae. Lower concentra-
tions were less effective against the immatures, with 50% mortality
observed at 0.025 ppm.
Abamectin was effective against P. latus infestations in field trials at
rates of 0.0125 to 0.025 lb ai/acre in combination with 0.25% emulsified
paraffinic oil. Initial mite populations were reduced by greater than 90%,
with residual protection of fruit for 4 weeks or longer. Abamectin was
more effective against P. latus when applied in combination with oil,
consistent with the results obtained with P. oleivora (Brown, unpub-
lished).
3. Citrus Red Mite (Panonychus citrO

Citrus red mite feeding causes serious injury to citrus foliage, including
impaired photosynthesis and respiration and increased leaf drop. In
laboratory bioassays, abamectin had an LC90 of 0.24 ppm against adult

Panonychus citri (Dybas and Green 1984; Table 20.1).
Several researchers have reported that abamectin applied at rates of
0.0125 to 0.05 lb ai/acre reduced initial P. citri populations by 70% or
more within 7 to 14 days post treatment. In these trials, low doses of
paraffinic oil extended abamectin's residual mite control. However, field
trial results are inconsistent and have shown that residual control of
P. citri on foliage is highly variable, ranging from 2 to 6 weeks under
typical growing conditions (Elmer and Morse 1985; French 1986; Elmer
and Morse 1984; Brown and Jesser 1982; Brown and Jesser 1981; Childers
and Sorrell 1984; Hare 1986). Results of laboratory bioassays of field-
treated foliage have confirmed abamectin's short residual activity
on citrus. The LT50 for abamectin residues, applied to orange trees at
0.025 lb ai/acre plus oil, for citrus red mite was 3 days or less (Dybas,
unpublished). Because abamectin's effect on P. citri is inconsistent, the
manufacturer has decided not to seek a label claim for control of this
spider mite at this time.
Abamectin is unstable in the presence of sunlight. Several studies have
shown that when abamectin is exposed to light as a thin film on glass or on
foliage, it degrades rapidly, with a half-life (t1l2) of 4 to 6 hours
(MacConnell et ai., in press; Dybas et ai., in press). The limited nature of
abamectin's residual activity against P. citri in field trials is due to the
rapid decomposition and weak translaminar movement of the drug in
TABLE 20.1 Activity of abamectin against insect larvae and adult mites and
aphids.
Mite Species (Contact effect against adult mites) LC90 (ppm)A
Phyllocoptruta oleivora (Ashmead) (citrus rust mite) 0.02
Tetranychus urticae Koch (twospotted spider mite) 0.03
Panonychus ulmi (Koch) (European red mite) 0.04
Polyphagotarsonemus latus (Banks) (broad mite) 0.05
Tetranychus turkestani (Ugarov & Nikolski) (strawberry spider mite) 0.08
Tetranychus pacificus (McGregor) (Pacific spider mite) 0.16
Panonychus citri (McGregor) (citrus red mite) 0.24
Insect Species (Foliar residue bioassay)
Manduca sexta (L.) (tobacco hornworm) 0.02
Leptinotarsa decemlineata (Say) (Colorado potato beetle) 0.03
Heliothis virescens (F.) (tobacco budworm) 0.10
Epilachna varivestis Mulsant (Mexican bean beetle) 0.40
Trichiijjlusia ni (Huebner) (cabbage looper) 1.0
Heliothis zea (Boddie) (cotton bollworm) 1.5
Spodoptera eridania (Cramer) (southern armyworm) 6.0
Spodoptera jrugiperda (J .E. Smith) (fall armyworm) 25.0
A Mortality assessed 72-96 hours after exposure. Dybas, R.A., Green, A. St. J. (1984)
Avermectins: Their Chemistry and Pesticidal Activity. Proceedings, 1984 British Crop
Protection Conference-Pests and Diseases, Brighton, England 31:947-954.
citrus foliage (Wright et al. 1985a). In metabolism studies with tritium-

labeled product, Iwata and colleagues (1985) noted the short half-life of
abamectin on citrus fruit and foliage, while Bull and colleagues (1984)
demonstrated that abamectin applied to cotton foliage decomposed with a
half-life less than 1 day. Bull found 82% loss of abamectin by 2 days and
98% degradation by 8 days post application, indicating that the rate of
abamectin decomposition is rapid and unaffected by foliage type.
4. Citrus thrips (Scirtothrips citrO

In laboratory studies the LC95 of abamectin to first instar Scirtothrips citri
exposed to I-day-old foliar residues was 10.9 ppm (Grout and Morse
1986). Abamectin was less toxic to adult female S. citri than to immatures.
The LC90 of 2-hour foliar residues to 3 field colonies of adult female thrips
ranged from 40 to 192 ppm (Morse and Brawner 1986).
Experience to date indicates that abamectin is only moderately toxic to
S. citri. Several field trials conducted in California demonstrated that
single applications of abamectin at 0.01 to 0.02 lb ai/acre provided only
fair control of S. citri populations and fruit scarring on sensitive navel
oranges in the San Joaquin Valley (Morse and Brawner 1983; Morse,
Elmer, and Brawner 1984; Morse, Grout, and Brawner 1986; Morse and
Brawner 1987). More recent studies, however, have shown that sequen-
tial applications of abamectin at these rates, at petal fall and 7 days later,
may provide effective control of S. citri populations and maintain fruit
scarring and injury below economic levels for several weeks. Because of
the critical timing for abamectin citrus thrips sprays, additional experi-
ence is required under a variety of conditions to better define its utility
against this citrus pest.
5. Beneficial Organisms
Abamectin exhibits a relatively low order of intrinsic toxicity to the
predatory mite Euseius tularensis (LC90 9.5 ppm) compared to its activity
on citrus red mite (LC90 0.30 ppm) in slide-dip bioassays. These data
(Brown, unpublished) provide a safety factor of 31.7 based on the LC90
ratio for predator to prey and a safety factor of 54.6 at the LC50 level. In a
field trial by Brown (unpublished), populations of the predatory mite
E. tularensis rebounded rapidly following an application of abamectin at
0.025 lb ai/acre plus 1% crop oil. The numbers of predatory mites had
declined at 3 days following application but were reestablished to
control-plot levels by 6 days.
Morse and colleagues (1987) reported that field-aged foliar residues of
abamectin were relatively innocuous to selected beneficial arthropods of
citrus. One-day-old residues of abamectin, applied to lemon foliage at
0.0251b ai/acre (the highest rate recommended on the product's label) in
combination with 0.5% narrow-range oil, showed less than 10% mortality
to the hymenopterous scale parasite, Aphytis melinus, the mealybug

destroyer, Cryptolaemus montrouzieri, and the predator mite, Eusieus
stipulatus. These results are consistent with many earlier observations
that applying abamectin to citrus foliage at recommended rates is not
likely to disrupt natural beneficial populations and is compatible with
grower Integrated Pest Management (IPM) practices (Morse and Elliot
1985; Grafton-Cardwell and Hoy 1983; Hoy and Cave 1985; Lienk,
unpublished).
Under laboratory conditions on agar medium, abamectin at 1.5 ppm
reduced growth of Hirsutella thompsonii by 28% but had no effect on
Aschersonia aleyroidis conidial growth and germination at that concen-
tration. In field applications, however, abamectin has not had a negative
impact on growth nor disrupted the ability of these parasitic fungi to
control P. oleivora or the citrus whitefly, Dialeuroides citri, (McCoy,
Bullock, and Dybas 1982).
The cumulative results of studies on abamectin's effects on beneficials,
including predators, parasites, and entomophagous fungi, strongly sup-
port the utility of the product in established Integrated Pest Management
programs.
c. ORNAMENTALS
1. Twospotted Spider Mite (Tetranychus urticae)

Abamectin showed high intrinsic toxicity to Tetranychus urticae adults
under laboratory conditions in contact bioassays. The contact LC 90 to
adults on bean plants was 0.03 ppm (Table 20.1). Abamectin is toxic to all
stages of T. urticae except the egg stage.
Abamectin did not kill eggs at concentrations up to 4.5 ppm in
the laboratory; however, slight reduction in hatch of 3-day-old eggs
laid on bean leaves was observed. This effect was likely due to
abamectin's ability to penetrate the egg chorion at that age, resulting in
paralysis of the developing larvae and inhibition of eclosion (Green and
Dybas 1984).
Abamectin (AVID®) has displayed long residual activity in controlling
T. urticae populations on foliage in laboratory and greenhouse studies.
Because abamectin is nonovicidal at standard application rates, foliar
persistence is important in controlling larvae hatching from eggs present
at the time of application. In a laboratory bioassay on roses, abamectin
applied at 4.5 ppm provided 84% mortality to T. urticae adults 10 days
after treatment and 44% kill after 21 days (Green and Dybas 1984). Long
residual control of T. urticae was observed under commercial greenhouse
application conditions. Abamectin applied at 4.5 ppm in a full-cover spray
on roses maintained T. urticae populations below a threshold of 1 mite per
leaf for at least 28 days. Similarly, on chrysanthemum the same concen-
tration of abamectin held populations below 2 adults per leaf for over 42
days (Green and Dybas 1984; Green et al. 1985b). Several authors have
reported on abamectin's long residual control of mites on chrysanthe-

mum, carnation, and roses (Robb, Virzi, and Parrella 1985; Mollet et al.
1986).
The penetration of abamectin' s residues into the cuticle of plants has
been demonstrated in our laboratories (Presier, unpublished). This trans-
laminar movement provides a reservoir of abamectin within the meso-
phyll, accounting for the long residual activity to mites and insects feeding
on plant tissue. Green and colleagues (1985b) reported that abamectin
applied at 4.5 ppm showed translaminar activity in the control of
T. urticae on rose foliage in laboratory and greenhouse studies. Greater
than 80% mortality was observed at 7 days to T. urticae adults and
immatures feeding on the ventral side of leaves, following application of
abamectin to the dorsal surface. Similarly, Wright and colleagues (1985a)
reported control of T. urticae, but not aphids (Aphis fabae or Aphis
gossypiz), due to translaminar action by abamectin on chrysanthemum.
The LC50 for abamectin residues applied to either the dorsal or ventral
leaf surface was 30.1 to 33.4 ppm for T. urticae,' however, higher
concentrations, 450 ppm or greater, were required to achieve 50%
mortality of aphids. While aphids feed selectively on the phloem,
phytophagous mites such as T. urticae feed within the leaf parenchyma
and are likely to ingest a higher concentration of abamectin than the
aphids.
2. Leafminer (Liriomyza trifolii)
Abamectin is highly toxic to adult Liriomyza trifolii and early instar larvae
mining within leaf tissue. When applied to newly eclosed larvae on
chrysanthemum, abamectin provided 100% control at a 12 ppm concen-
tration. Greater than 90% control of third stage L. trifolii larvae was
obtained at that concentration; however, 48 ppm was required to provide
100% kill of third stage larvae and complete inhibition of pupation and
adult emergence (Parrella 1983).
Abamectin is not toxic to L. trifolii eggs. However, residues of
abamectin within leaf tissue, following application at 4.5 ppm, provided
100% control of newly hatched larvae (Green and Dybas 1984). Abamec-
tin applied on a weekly spray schedule, at concentrations in the range of
4.5 to 24 ppm to chrysanthemum, gerbera daisy, gypsophila, and other
flower varieties susceptible to attack by L. trifolii, provided over 85%
reduction in large mine formation and completed mines. Abamectin is not
repellent to Liriomyza flies, and in these trials it did not effectively reduce
feeding punctures and stippling by adult females (Parrella et al. 1988;
Green and Dybas 1984; Parrella 1983; Green et al. 1985a).
AVID® (abamectin) is recommended at a rate of 0.01 to 0.021b ai/acre
in full-cover foliar sprays for the control of Liriomyza trifolii on flowers
and foliage plants. Weekly applications of the product are recommended
during periods of continuous fly infestation to obtain high-quality,
leafminer-free crops.
3. Thrips Species
Several species of thrips attack ornamental foliage and floral crops

worldwide. Because of their secretive feeding habits and protected life
stages, these pests are normally difficult to control with nonsystemic,
nonfumigant pesticides.
Abamectin was only moderately toxic to adult western flower thrips
(Frankliniella occidentalis), the most important species affecting orna-
mentals. Broadbent and Kelly (1987a) reported that in a combined contact
and residual toxicity bioassay, abamectin at 40 ppm (the highest concen-
tration tested) produced only 51.9% mortality to adult F. occidentalis
confined on abamectin residues on petri dishes for 72 hours. Similarly,
only 45.6% mortality of adult thrips was observed following 72 hours
exposure to abamectin residues on chrysanthemum (Broadbent and Kelly
1987b). The activity against F. occidentalis at normal use rates is slow,
even when adult thrips are confined and forced to feed on treated foliage.
Oetting (1987) evaluated the toxicity of abamectin at 6 and 12 ppm as a
foliar residue on impatiens leaves to Ecinothrips americanus under
greenhouse conditions. When thrips were introduced onto foliar residues
within 24 hours after application, greater than 90% mortality was ob-
served in adult and second instar immatures exposed to 12 ppm abamectin
residues for 72 hours. However, the toxicity of abamectin foliar residues
decreased rapidly with time. Abamectin provided 60% or less mortality to
adult and immature stages of E. americanus introduced onto foliar
residues aged for 3 days or longer. No significant differences were
observed in the sensitivity of adult or immature thrips exposed to
abamectin foliar residues at any time period.
Oetting observed that egg hatch was slightly reduced when leaves were
sprayed at selected intervals before or after oviposition. The greatest
reduction in percent hatch (68%) was observed in eggs that hatched within
3 days following application of abamectin at 12 ppm. Although a few
thrips died during eclosion, most first instars emerged completely from
the egg and died after crawling on the leaf surface for a short time. The
reduction in the number of hatching thrips exposed to 12 ppm abamectin
residues was not sufficient for adequate control. In combination with the
short residual control of adults, these data suggest the need for mUltiple
applications of the product in commercial greenhouses to break the life
cycle of E. americanus or other thrips species.
Robb and Parrella (1986) evaluated abamectin for controlling F. occi-
dentalis in roses and carnations. The authors observed only moderate
activity, an average of 66% control of thrips populations, following 2
applications of abamectin at 0.003 lb ai/l00 gallons water, 5 days apart.
In a trial on dahlia against F. occidentalis, 3 applications of abamectin
at 0.02 lb ai/acre, 7 days apart, completely eliminated a low thrips
population. Following the same treatment regime against F. occidentalis
on carnations, however, abamectin provided only 50.4% control of a

heavy thrips population at 7 days after the third application (Taylor,
unpublished) .
Although abamectin appears to provide useful activity against difficult-
to-control thrips species, it has performed erratically in greenhouse trials
and has failed to provide consistent control following multiple applica-
tions. For these reasons, the manufacturer has decided not to label the
product for use against thrips species on ornamental crops until greater
experience is obtained.
D. COTTON
1. Spider Mites (Tetranychus urticae, T. turkestani,

T. pacijicus, T. cinnabarinus)
Several studies have shown that abamectin is highly toxic to tetranychid
spider mites under laboratory conditions, with LC90 values against adult
mites in the range of 0.02 to 0.16 ppm (Putter et al. 1981; Dybas and Green
1984; Preiser, unpublished).
In another study, Hilton (unpublished) compared the toxicity of
abamectin to a field isolate of the carmine spider mite (T. cinnabarinus)
from Uvalde, TX, and a laboratory colony of the twospotted spider mite
(T. urticae). Both populations were highly susceptible to abamectin, with
an LC so value of 0.0029 ppm for T. cinnabarinus compared to an LC so of
0.0087 for the laboratory colony of T. urticae. The T. cinnabarinus isolate
was highly resistant to monocroptophos, which is used extensively in the
Uvalde region. For monocroptophos, the LC so for the carmine mite was
11,036 ppm, compared to an LC so of 94.84 ppm for the twospotted spider
mite.
In other laboratory bioassays, abamectin was evaluated against 4 field
isolates of T. urticae from the San Joaquin Valley, CA. Abamectin was
equitoxic to all 4 strains, which had varying degrees of resistance to
dicofol and propargite, standard miticides used on cotton in the San
Joaquin Valley (Hilton and Dybas, in press). Therefore, results of
laboratory bioassays indicate that abamectin is effective against spider
mites, which are susceptible and resistant to standard cotton miticides.
Cross-resistance to abamectin has not been observed in spider mites in
limited studies. (Hoy and Cave 1985).
A number of authors have noted abamectin's ability to provide long
residual control of spider mites on cotton, despite its rapid photode-
composition following application. This persistence is due to abamectin's
translaminar action, which has been observed in both laboratory and field
bioassays with spider mites (Wright et al. 1985a; Dybas and Green 1984).
Bull and colleagues (1984) reported 82% and 98% decomposition of
abamectin on cotton after 2 and 8 days, with a half-life of less than 1 day.
Rapid penetration of abamectin into cotton foliage was observed within a
few hours of application, with 1.3% of the applied dose of tritiated

avermectin Bla absorbed into foliage by 8 days.
Results of laboratory bioassays of field-treated foliage have confirmed
abamectin's long residual activity on cotton. The LT50 for abamectin
residues, applied at 0.01 to 0.021b ai/acre, to the twospotted spider mite
was 21 to 28 days (Dunbar, Dybas, and Norton, in press). Wright and
colleagues (1985a) also reported that abamectin applied to cotton leaves at
30 ppm gave 100% kill of adult female T. urticae under glasshouse
conditions and outdoors for at least 28 days. At 3 ppm abamectin, greater
than 95% mortality was maintained for 14 days under glass and greater
than 85% control outdoors for 7 days.
The use of emulsified oil in combination with abamectin has been
shown to extend the toxicity of abamectin's foliar residues on various
crops under greenhouse and field conditions (Dybas 1985). Wright and
colleagues (1985b) reported that abamectin applied on cotton at 3 ppm in
combination with 0.25% paraffinic oil provided 77% or greater mortality
to spider mites for 21 days outdoors and 28 days under glasshouse
conditions.
Additionally, Mizell, Schifthauer, and Taylor (1986), reporting on the
complex interaction of host plant, light regime, and additives on the
stability of abamectin residues following spray application to a number of
crops, noted that the nonionic surfactant, Leaf Act 80A, and paraffinic oil
increased the length of abamectin's mite control on bean compared to
abamectin alone.
Abamectin has provided excellent control of mites in field trials on
cotton. Wynholds and Leigh (1986) reported 90% to 96% knockdown of
spider mite populations 13 to 18 days following abamectin applications at
0.008 to 0.0091b ai/acre, and residual control of 85% to 95% at 41 to 49
days after treatment. These researchers had previously reported 92% to
96% control of mite populations for 46 days following abamectin treat-
ment at 0.02 lb ai/acre (Wynholds and Leigh 1982). Royer and Edelson
(1987) showed that abamectin at 0.015 lb ai/acre reduced initial mite
populations by 98% at 7 days post treatment and provided 100% control of
populations after 21 days. The lower abamectin rate, 0.0075 lb ai/acre,
provided 89% to 93% control for the duration of the trial. No control of
the cotton aphid, Aphis gossypii, was observed with abamectin at these
application rates throughout the study. These results confirm laboratory
observations on abamectin's weak activity against aphid species on
cotton (Wright et al. 1985a).
Tuttle and Mullis (1984) evaluated the activity of abamectin against
carmine spider mites on cotton applied with and without paraffinic oil. At
0.01 and 0.021b ai/acre, abamectin without paraffinic oil gave 92% to 98%
control from 3 to 14 days post treatment. In combination with oil,
abamectin at O.Ollb ai/acre provided 98% reduction in mite populations 3
days after application; however, by 14 days, mite populations began to

rebound, and only 69% control was observed. The lower rate of abamec-
tin (0.005Ib ai/acre) in this trial was the least effective, providing 84% and
55% control at 3 days and 14 days post application, respectively.
Results obtained from several studies have shown that abamectin at
recommended rates of 0.01 to 0.02 lb ai/acre provides 80% or greater
reduction of initial spider mite populations by 7 to 10 days post applica-
tion and greater than 90% reduction by 14 days. Rates below 0.01 lb
ai/acre have been slower and less effective in reducing initial mite
populations and provide shorter residual control of spider mites (Tuttle
and Mullis 1984; Dybas, unpublished).
Field trials have shown that initial reduction of mite populations with
abamectin at 0.01 lb ai/acre at 7 days post treatment appears slow. This
occurs as a result of abamectin's lack of ovicidal action. Motile spider
mites are killed shortly after application; however, eggs present at the
time of treatment develop and hatch normally. An additional 3 to 4 days
are then required for the newly hatched larvae to accumulate a lethal dose
of abamectin.
Abamectin has demonstrated long residual spider mite control on
cotton and reduction in crop damage in season-long trials. Statistically
significant increases in seed cotton yield per acre over standard acaricide
treatments were observed with abamectin treatments of O.Ollb ai/acre in
large-scale field trials conducted in California under a U.S. Environmen-
tal Protection Agency (EPA)-approved experimental use permit on cotton
(Dunbar, Dybas, and Norton, in press).
2. Lepidoptera
Abamectin is highly selective in its toxicity to lepidopteran larvae, with
armyworm species being the least susceptible. Dybas and Green (1984;
Table 20.1) reported laboratory foliage bioassays LC90 values of 0.02, 1.5,
and 6.0 ppm for abamectin against tobacco hornworm (Manduca sexta),
corn earworm (Heliothis zea), and southern armyworm (Spodoptera
eridania) first instar larvae, respectively, indicating a 300-fold difference
in the sensitivity of southern armyworm compared to tobacco hornworm.
Similarly, Anderson and colleagues (1986) reported a 400-fold difference
in the toxicity of abamectin to Heliothis virescens larvae compared to
S. eridania at the LC 50 level in a foliage bioassay test. Substantial
differences in the susceptibility to abamectin among the noctuid species
H. virescens, H. zea, and S.frugiperda were also noted by Bull (1986). In
bioassays against lepidopterans, the first visible symptom of exposure of
the first instar larvae to abamectin is paralysis of the hind legs within 24
hours, followed by larval mortality over a 48- to 72-hour period (Wright,
Jenkins, and Villavaso, 1985).
Several authors have reported that abamectin is more toxic to lepidop-
teran larvae via ingestion than by contact with product residues

(Anderson et al. 1986; Bull 1986; Corbitt, Green, and Wright, in press;
Wright, Jenkins, and Villavaso. 1985). In a comparative evaluation,
Anderson and colleagues (1986) showed that abamectin was one fourth as
toxic as the pyrethroid fenvalerate to S. eridania larvae in the foliage
ingestion bioassay, whereas by topical application, abamectin was nearly
lOOO-fold less toxic to S. eridania than fenvalerate.
Abamectin was highly toxic to male and female pink bollworm (Pectin-
ophora gossypiella) moths under laboratory conditions. Applied topically
on either the dorsal thorax or the ventral surface of the last abdominal
segment of the moths, abamectin had an LC50 of 0.032 to 0.041 ug/insect
(Bariola and Lingren 1984). Abamectin prevented mating for up to 72
hours when males or females were treated with an LDlO or LD50 dose.
Additionally, oviposition was completely prevented in female moths
treated with an LDIO dose. Egg hatch was also significantly reduced by
abamectin (Bariola 1984).
In limited field trials conducted to date, abamectin has not provided
effective control of noctuid species on cotton (Merck data, unpublished;
Mink and Luttrell 1987). Abamectin's rapid photolytic breakdown follow-
ing application, its weak topical/contact activity against lepidopteran
larvae, the limited feeding habits of Heliothis species on cotton, and the
differential sensitivity among noctuide species are factors that have
severely limited the utility of abamectin in the control of Lepidoptera on
cotton.
3. Cotton Aphid (Aphis gossypii)
Abamectin was shown to be toxic to several aphid species under
laboratory conditions in contact bioassays, with LC 90 's in the range of 0.4
to 1.5 ppm (Putter et al. 1981; Dybas and Green 1984). Abamectin
demonstrated poor control of A. fabae and A. gossypii in foliar trans-
laminar bioassays, with concentrations as high as 450 ppm required to
produce 50% mortality in these species.
Aphids feed selectively on the phloem of leaf tissue, and this feeding
preference may account for the poor activity of abamectin in the control
of aphids compared to spider mites. Royer and Edelson (1987) confirmed
the weak activity of abamectin on aphids in a cotton field trial. This study
showed initial suppression of A. gossypii 7 days after treatment with
abamectin applied at 0.01 and 0.02 lb ai/acre; however, essentially no
control of cotton aphid populations was observed at 14 and 21 days post
treatment. Leigh and colleagues (1988) also showed poor control of a
mixed population of A. gossypii and A. craccivora following a single
application of abamectin in the San Joaquin Valley. The limited activity
observed against aphids was attributed to biological control, confirming
abamectin's safety to beneficial insects.
E. Vegetables (Celery, Tomato, Pepper)

1. Leafminers (Liriomyza trifolii and L. sativae)
Under laboratory conditions abamectin was highly toxic to early instar
L. trifolii larvae in foliage bioassays. Field-collected adult female
L. trifolii were allowed to oviposit on California blackeye peas for 4 hours
in the laboratory. To evaluate toxicity to the leafminer larvae within the
foliage, leaves were dipped in abamectin solutions after eggs had hatched
and the larvae developed to the second-third instar. The LC 95 for
L. trifolii was found to be 0.746 ppm. The toxicity of abamectin to L.
trifolii was independent of foliage type, with similar results observed on
celery, tomato, and chrysanthemum (Dybas, unpublished).
Schuster and Everett (1983) reported complete control of 1- and
3-day-old larvae in tomato foliage with abamectin applications at 6 and
12 ppm. At these concentrations, abamectin was nonovicidal to L. trifolii
eggs laid in leaf tissue; however, the treatment resulted in 90% or greater
kill of larvae during eclosion due to contact with residues in or on the
chorion or in the leaf mesophyll. In addition, abamectin residues of
12 ppm were moderately toxic to the adult flies. Mortality of adult L.
trifolii exposed in the laboratory to dried abamectin residues on filter
paper or tomato foliage was 24% to 38% after 48 hours exposure.
Although adult mortality was low, abamectin reduced egg laying and
stippling of foliage. Because these functions are initiated by the oviposi-
tor, paralysis of the ovipositor may have occurred as a result of exposure
to abamectin residues.
Bioassay of tomato foliage treated with abamectin at 0.01 Ib ai/acre
showed 40% to 60% mortality of L. trifolii males and females exposed to
fresh residues for 48 hours. However, little mortality was seen in adult
flies exposed to 24-hour field-aged residues, which is consistent with the
known rapid breakdown and dissipation of abamectin surface deposits
following application (Schuster and Taylor 1987). The field-treated foliar
residues were effective in reducing larval growth and survival in tomato
leaflets bioassayed 7 days after treatment. The length of residual toxicity
to L. trifolii larvae noted in this study supports earlier observations
regarding the trans laminar movement of abamectin in tomato foliage
and persistence of internal leaf residues following spray applications
(Schuster and Taylor 1987; Wright et al. 1985a).
Several authors have reported on the control of L. trifolii on tomato,
celery, and pepper in field trials with abamectin treatments at the
recommended field rates of 0.01 to 0.02 lb ai/acre (Trumble 1983a,b;
Carson, Trumble, and Nakakihara 1986; Carson, Nakakihara, and
Trumble 1986; Schuster and Everett 1987; Zehnder and Trumble 1984;
Schuster 1987; Royer, Edelson, and Cartwright, 1985; Chandler 1985).
Greater than 80% reduction in leaf mines and/or pupal emergence has
been observed in all trials. Results of limited studies have also shown
that, under certain conditions, the addition of the nonionic surfactant,
Leaf Act 80A, may improve leafminer control, presumably by enhancing
penetration of abamectin into the foliage and thereby extending the period
of residual protection (Merck data, unpublished).
2. Twospotted Spider Mite and Tomato Russet Mite
(T. urticae and Aculops lycopersici)
Laboratory studies have shown that abamectin is highly toxic to erio-
phyid and tetranychid mites (Dybas and Green 1984; McCoy, Bullock,
and Dybas 1982). The LC 90 of abamectin against T. urticae adults was
0.03 ppm. Royalty and Perring (1987) found that abamectin was very toxic
to A. lycopersici, with an LC 90 of 0.00959 ppm against adults. These
authors also noted that abamectin was much less toxic to the predatory
mite Homeopronematus anconai; thus, abamectin sprays should suffi-
ciently control tomato russet mite without reducing H. anconai popula-
tions.
Limited field studies have been conducted to date with abamectin for
control of mites on tomato with excellent results. Greater than 80%
control of mite populations was observed following abamectin applica-
tions at 0.01 to 0.021b ai/acre (Schuster 1984; Merck data, unpublished).
3. Armyworms (Spodoptera exigua and S. eridania)

Armyworm species are relatively insensitive to abamectin compared to
other insects (Table 20.1). The LC 90 for abamectin against southern
armyworm (Spodoptera eridania) was 6.0 ppm in a foliar bioassay (Dybas
and Green 1984), 1.96 ppm for beet armyworm (S. exigua) by diet
incorporation (Trumble et al. 1987; Moar and Trumble 1987), and greater
than 25 ppm for fall armyworm (S. Jrugiperda) in a foliar bioassay.
In field trials on celery and tomato, abamectin, at rates recommended
for leafminer control (0.01 to 0.02 lb ai/acre), was only marginally
effective in reducing Spodoptera infestations, providing 50% or less
control of larval populations and prevention of crop damage in season-
long programs (Carson, Trumble, and Nakakihara 1986; Schuster 1987;
Trumble 1983a; Trumble 1983b; Waddill 1982; Trumble 1985a; Carson,
Nakakihara, and Trumble 1986).
4. Tomato Pinworm (Keiferia lycopersicelia)

Under laboratory conditions, K. lycopersicelia larvae were found to be
highly susceptible to abamectin. The LC 95 for first instar larvae exposed
to fresh abamectin foliar residues was 0.031 ppm. However, abamectin
was shown to be considerably less toxic to pinworm larvae feeding
internally within the foliage. In this case, abamectin's LC 95 was 15.1 ppm
or approximately 500-fold higher than obtained for control of K. lycoper-
sicella larvae by contact with smface residues (Schuster and Taylor,

unpublished) .
Limited field studies conducted to date with abamectin on tomato at
rates of 0.01 to 0.02 lb ai/acre have provided good results against K.
lycopersicella. A 93% to 96% reduction of seasonal fruit damage from K.
lycopersicella was reported following weekly applications of abamectin at
0.01 to 0.02 lb ai/acre (Trumble 1983b). In another trial, 70% to 75%
reduction in seasonal damage was observed following 5 weekly appli-
cations of abamectin at 0.01 to 0.02 Ib ai/acre (Trumble 1985a), while in
yet another field trial, 62.5% to 69% reduction in fruit damage was re-
ported following 7 weekly applications (Carson, Trumble, and Nakaki-
hara 1986). Based on the experience obtained to date, abamectin is
recommended at a rate of 0.01 to 0.02 Ib ai/acre for control of K. lyco-
persicella on tomato.
5. Beneficial Arthropods
Trumble (1985a) reported that weekly treatments of celery with abamec-
tin at 0.01 lb ai/acre (its recommended rate of application for leafminer
control) had a minimal effect on both adult and larval stages of the
leafminer parasite complex. Abamectin applications did not adversely
impact seasonal percent parasitism, adult parasite mortality and survival,
or emergence of immature parasites of Diglyphus intermedius, D. begini,
Chrysonotomyia punctiventris, Chrysocharis parksi, C. ainsliei, and
Halticoptera circulus from treated foliage.
Abamectin at application rates of 0.01 to 0.02 lb ai/acre also did not
adversely affect the beneficial leafminer parasite complex on celery,
tomato, and pepper under field use conditions (Zehnder and Trumble
1984; Carson, Nakakihara, and Trumble 1986; Trumble 1985b; Chandler
1985). Parasite mortality from abamectin treatments were not statistically
different from the controls.
Results of several field trials support the fact that abamectin has a good
safety margin for beneficials and is useful for Integrated Pest Management
programs on celery, tomato, and peppers.
F. Pears
1. Pear psylla (Psylla pyricola)
Abamectin, at rates of 0.0125 to 0.025 lb ai/acre, has provided significant
reduction in Psylla pyricola nymphal populations in postbloom sprays on
pears. The addition of 0.25% paraffinic oil in dilute sprays, or a minimum
of 1 gallon oil per acre, extends abamectin's residual control of psylla
nymphs for 4 weeks or longer, and provides effective control of fruit
damage.
Burts (1983, 1984) reported that abamectin provided good control of
psylla populations throughout the season, and essentially no fruit russet-

ing at harvest due to pear psylla honeydew. Abamectin applied at 0.0125
and 0.025 lb ai/acre plus oil also provided good control of heavy psylla
populations, with minimal fruit damage (Burts 1985a,b). In this latter trial,
6% and 2% cull fruit were observed in the low- and high-rate abamectin
treatments, compared to 59% in the control plot. Riedl and Shearer (1987)
found excellent seasonal control of heavy pear psylla populations follow-
ing 3 applications of abamectin at 0.0125 and 0.025 lb ai/acre in combina-
tion with a paraffinic oil. At harvest, no fruit were downgraded in the
high-rate treatment, while only 2.9% fruit damage was observed at the
low abamectin rate compared to 21.8% in the control plot. Howitt and
Hays (1986) reported that, in a seasonal program, abamectin applied at
0.0231b ai/acre in combination with superior oil and with superior oil plus
the nonionic surfactant, Leaf Act 80A, provided effective control of
pyrethroid-resistant psylla populations. At harvest, the sooty fungus
rating on leaves was reduced by 96% and 99% in the abamectin oil and
oil! surfactant treatments.
Burts (1985a) reported that dormant-season sprays of abamectin at
0.01 lb ai/acre failed to reduce Psylla pyricola adult popUlations below
retreatment thresholds even after 2 applications. As a consequence,
abamectin is not recommended for prebloom control of overwintered
adult Psylla pyricola. Abamectin is most effective for control of psylla
nymphal populations after petal fall at the recommended rates of 0.0125 to
0.25 lb ai/acre in thorough coverage sprays.
2. Twospotted Spider Mite, European Red Mite, Pear Rust Mite
(Tetranychus urticae, Panonychus ulmi, Epitrimerus pyrO
The high susceptibility of tetranychid and eriophyid mites to abamectin
was discussed in earlier sections of this chapter. The results of several
field trials have shown that abamectin, applied at 0.0125 to 0.0251b ai/acre
in combination with 0.025% crop oil in dilute sprays, provides excellent
control of T. urticae, P. ulmi, and E. pyri populations.
Burts (1983, 1984, 1987) reported that abamectin at 0.025 lb ai/acre, in
combination with 0.25% oil, provided 89.5% to 97.9%, 95.8% to 100%,
and 83.5% to 93.9% control of European red mite, twospotted spider
mite, and pear rust mite popUlations in seasonal programs on pears. Only
slightly less control of mite populations was observed in these trials at the
0.0125 lb ai/acre rate: 84% to 90% for European red mite, 90.6% to 100%
for twospotted spider mite, and 73.3% to 88.9% for pear rust mite.
Cumulative results from field trials conducted to date indicate that
single applications of abamectin at recommended rates of 0.0125 to
0.025 lb ai/acre, in combination with paraffinic oil, reduces initial mite
populations by 85% or greater within 7 days after application, and
provides control of populations below retreatment levels for 4 weeks or
longer.
G. Miscellaneous Crops
1. Strawberry
Limited studies have shown that abamectin is effective against spider
mite infestations on strawberry and tree nuts. Carson and colleagues
(1987) reported that 7 biweekly applications of abamectin on strawberry
at 0.02 lb ai/acre in 50 to 100 gallons of water per acre were effective in
reducing T. urticae populations throughout the trial. Abamectin's activity
was not improved by the addition of 0.1 % oil in the spray tank. The
authors also reported that abamectin controlled a moderate density of
aphids under this spray regime.
Earlier, Trumble and Nakakihara (1984) had reported that abamectin
applied to strawberries at 0.04 lb ai/acre, alone and in combination with
0.1 % oil, reduced motile spider mite populations by an average of 92.7%
and 95.8%, respectively, over an 8-week period following a single
application. A second application of abamectin at the same rates, alone
and in combination with paraffinic oil, reduced mite levels but was less
effective, providing only 30.9% and 72.8% control over a 3-week period.
More recent studies have shown that abamectin controls spider mites
very effectively on strawberry following 2 applications, timed on a 7-day
spray schedule to kill hatching larvae (Dunbar, unpublished). In 3
replicated field trials, abamectin provided an average of 79% and 68%
control of T. vrticae over a 35-day period when applied twice at 0.02 and
O.Ollb ai/acre, compared to only 52% control of mite populations over the
same period from a single application at 0.01 lb ai/acre.
2. Deciduous Tree Nuts
Abamectin applied at 0.00625 lb ai/lOO gallons water in a high volume
dilute spray to almonds was shown to reduce initial popUlations of
Tetranychus pacificus by 85% at 6 days post treatment. Residual control
of spider mite populations was obtained for 48 days and was enhanced by
the combination of abamectin with a narrow-range spray oil (Sanderson
and Barnes 1983). The authors also reported that abamectin treatment did
not have a negative impact on beneficial species, including the predatory
mite, Metaseiulus occidentalis, and six-spotted thrips, Scolothrips sex-
maculatus. Payne, Dutcher, and Ellis (1985) reported that abamectin
applied at a rate of 0.02 lb ai/acre to pecans provided excellent control
(95%-100%) of the hickory shuckworm, pecan leaf scorch mite, and
pecan weevil. Abamectin, however, provided poor control of yellow
aphids and the black aphid in this field trial.
REFERENCES
Anderson TE, Babu JR, Dybas RA, Mehta H (1986) Avermectin Bl: ingestion and
contact toxicity against Spodoptera eridania and Heliothis virescens (Lepidop-
tera: Noctuidae) and potentiation by oil and piperonyl butoxide. J. Econ.

Entornol. 79: 197-201.
Apperson CS, Powell EE, Browne M (1985) Efficacy of individual mound
treatments of MK-936 and amdro against the red imported fire ant (Hymenop-
tera: Formicidae). J. Georgia Entornol. Soc. 19:508-516.
Bariola LA (1984) Pink Bollworms (Lepidoptera: Gelechiidae): Effects of low
concentrations of selected insecticides on mating and fecundity in the labora-
tory. J. Econ. Entornol. 77:1278-1282.
Bariola LA, Lingren PD (1984) Comparative toxicities of selected insecticides
against pink bollworm (Lepidoptera: Gelechiidae) moths. J. Econ. Entornol.
77:207-210.
Broadbent AB, Kelly RF (1987a) Residual toxicity of selected insecticides to
western flower thrips. Pestic. Res. Rep. E.C.P.U.A., 167.
Broadbent AB, Kelly RF (1987b) Combined contact and residual toxicity of
selected insecticides to western flower thrips. Pesticide Research Report
E.C.P.U.A., 168.
Brown RD, Jesser MJ (1981) Citrus red mite control in Riverside. Insectic. &
Acaric. Tests 6:41.
Brown RD, Jesser MJ (1982) Control of citrus red mite with carbamate,
pyrethroids, and avermectin B\. Insectic. Acaric. Tests 7:48-49.
Bull DL (1986) Toxicity and pharmacodynamics of avermectin in the tobacco
budworm, com earworm and fall armyworm (Noctuidae: Lepidoptera).
J. Agric. Food Chern. 34:74-78.
JM, VandenHeuvel WJA (1984) Fate of avermectin Bla in soil and plants.
J. Agric. Food Chern. 32:94-102.
Burts E (1983) Pear, pest control. Insectic. & Acaric. Tests 8:30.
Burts E (1984) Pear, pest control. Insectic. & Acaric. Tests 9:45-47.
Burts E (1985a) Pear, pest control. Insectic. & Acaric. Tests 10:36-37.
Burts EC (1985b) SN72129 and avermectin Bl, two new pesticides for control of
pear psylla, Psylla pyricola (Homoptera: Psyllidae). J. Econ. Entornol.
78:1327-1330.
Burts EC (1987) Pear, pest control. Insectic. & Acaric. Tests 12:60-61.
Carson WG, Nakakihara H, Trumble JT (1986) Impact of insecticides on
lepidopterous larvae and leafminers on tomatoes. Insectic. & Acaric. Tests
11:189-191.
Carson WG, Nakakihara H, Trumble JT, Vath V (1987) Spider mite and aphid
control in chemical trials on strawberries. Insectic. & Acaric. Tests 12:82.
Carson WG, Trumble JT, Nakakihara H (1986) Effect of insecticides on lepidop-
terous larvae and leafminers on celery. Insectic. & Acaric. Tests 11:109-110.
Chandler LD (1985) Response of Liriornyza trifolii (Burgess) to selected insecti-
cides with notes on hymenopterous parasites. Southwest. Entornol. 10:228-235.
Childers CC (1985) Citrus rust mite control (CRM). Insectic. & Acaric. Tests
10:57-58.
Childers CC, Sorrell RW (1983) Citrus rust mite control (CRM). Insectic. &
Acaric. Tests 8:34-35.
Childers CC, Sorrell RW (1984) Citrus red mite control (CRM). Insectic. &
Corbitt TS, Green A St J, Wright DJ (in press) Relative potency of avermectin B 1
against larval stages of Spodoptera littoralis Boisd., Heliothis armigera Hubu.

and Heliothis virescens F. (Lepidoptera: Noctuidae). Crop Protect.
Dunbar DM, Dybas RA, Norton JA (in press) Abamectin 0.15 EC (ZEPHYR® ):
Miticide/insecticide for spider mite control on San Joaquin Valley cotton.
Beltwide Cotton Insect Research and Control Conference Proceedings.
Dybas RA (1985) Synergistic avermectin combination for treating plant pests.
U.S. Patent 4,560,677.
Dybas RA, Green A St J (1984) Avermectins: their chemistry and pesticidal
activity. 1984 British Corp Protection Conference, Pests and Diseases 9B-
3:947-954.
Dybas RA, Hilton NJ, Babu JR, Preiser FA, Dolce GJ (1989) Novel second
generation avermectin miticides and insecticides for crop protection. Topics in
industrial microbiology (Demain, Somkuti Hunter-Cevera, and Rossmoore,
eds.), chapter 23, 201-210 Elsevier, Amsterdam.
Elmer HS, Morse JG (1984) Citrus red mite pesticide efficacy tests in the San
Joaquin Valley. Insectic. & Acaric. Tests 9:87-88.
Elmer HS, Morse JG (1985) Citrus red mite pesticide efficacy tests in the San
Joaquin Valley. Insectic. & Acaric. Tests 10:59-60.
French JV (1984) Rust mite control on Texas citrus. Insectic. & Acaric. Tests
9:71.
French JV (1986) Spider mite control on Texas citrus. Insectic. & Acaric. Tests
11:98-99.
Glancey BM, Lofgren CS, Williams DF (1982) Avermectin B\a: effects on the
ovaries of red imported fire ant queens (Hymenoptera: Formicidae). 1. Med.
Entomol. 19:743-747.
Grafton-Cardwell EE, Hoy MA (1983) Comparative toxicity of avermectin B\ to
the predator Metaseiulus occidentalis (Nesbitt) (Acari: Phytoseiidae) and the
spider mites Tetranychus urticae (Koch) and Panonychus ulmi (Koch) (Acari:
Tetranychidae) per. J. Econ. Entomol. 76:1216-1220.
Green A St J, Dybas RA (1984) Avermectin B\: Control of mites on ornamentals.
1984 British Crop Protection Conference, Pests and Diseases 11A-7: 1129-1133.
Green A St J, Heijne B, Schreurs J, Dybas RA (1985a) Serpentine leafminer
(Liriomyza trifolii) (Burgess) control with abamectin (MK-936) in Dutch orna-
mentals, a review of the processes involved in the evolution of the use
directions, and a summary of the results of phytotoxicity evaluations. Med.
Fac. Landbouww. Rijksuniv. Gent 50/2 b:603-622.
Green A St J, Heijne B, Schreurs J, Dybas RA (1985b) Twospotted spider mite
(Tetranychus urticae Koch) control with abamectin (MK-936) on roses. Med.
Fac. Landbouww. Rijksuniv. Gent 50/2 b:623-632.
Greenblatt JA, Norton JA, Dybas RA, Harlan DP (1986) Control of the red
imported fire ant with abamectin (AFFIRM® ), a novel insecticide, in individual
mound trials. 1. Agric. Entomol. 3:233-241.
Grout TG, Morse JG (1986) Insect growth regulators: promising effects on citrus
thrips (Thysanoptera: Thripidae). Can. Ent. 118:389-392.
Hare J (1986) Citrus red mite pesticide efficacy tests in Riverside. Insectic. &
Hilton NJ, Dybas RA (in press) Abamectin (ZEPHYR® ): A novel avermectin
miticide/insecticide for spider mite control on cotton. Beltwide Cotton Insect
Research and Control Conference Proceedings.
Howitt AJ, Hays AR (1986) Pear, full season psylla block test. Insectic. & Acaric.
Tests 11:71-72.
Hoy MA, Cave FE (1985) Laboratory evaluation of avermectin as a selective
acaricide for use with Metaseiulus occidentalis (Nesbitt) Acarina: Phyto-
seiidae). Exper. & Appl. Acarol. 1:139-152.
Iwata I, MacConnell JG, Flor JE, Putter I, Dinoff TM (1985) Residues of
avermectin Bla on and in citrus fruits and foliage. 1. Agric. Food Chem.
33:467-471.
Leigh T, Wynholds P, Bentley W, Arrendondo C (1988) Insect pest status in the
San Joaquin Valley in 1987. Cal. Cot. Prog. Rep. March: 8-9.
Lemke LA, Kissam JB (1987) Evaluation of various insecticides and home
remedies for control of individual red imported fire ant colonies. 1. Entomol.
Sci. 22:275-281.
Lofgren CS, Williams CF (1982) Avermectin B1a: highly potent inhibitor of
reproduction by queens of the red imported fire ant (Hymenoptera: Formici-
dae). 1. Econ. Entomol. 75:798-803.
MacConnell JG, Demchak RJ, Preiser FA, Dybas RA (in press) A study of the
relative stability, toxicity, and penetrability of abamectin and its 8,9-oxide. 1.
Agric. Food Chem.
McCoy, CW, Bullock RC, Dybas RA (1982) Avermectin B1: a novel miticide
active against citrus mites. Proc. Fla. State Hort. Soc. 95:51-56.
Mink JS, Luttrell RG (1987) Chemical control of the fall armyworm on cotton.
Insectic. & Acaric. Tests 12:236-237.
Mizell RF, Schiffhauer DE, Taylor JL (1986) Mortality of Tetranychus urticae
Koch (Acari: Tetranychidae) from abamectin residues: effects of host plant,
light, and surfactants. 1. Entomol. Sci. 21:329-337.
Moar WJ, Trumble JT (1987) Toxicity, joint action, and mean time of mortality of
dipel 2X, avermectin BJ, neem and thuringiensin against beet armyworms
(Lepidptera: Noctuidae). 1. Econ. Entomol. 80:588-592.
Mollet JA, Nelson RD, Norton JA, Dybas RA (1986) Abamectin for control of
spider mites on greenhouse carnations and roses Entomological Society of
America National Conference, Reno, NV Dec. 7-11.
Morse JG, Bellows TS, Gaston LK, Iwata Y (1987) Residual toxicity of
acaricides to three beneficial species on California citrus. 1. Econ. Entomol.
80:953-960.
Morse JG, Brawner OL (1983) Citrus thrips pesticide efficacy tests in California.
Insectic. & Acaric Tests 8:37-39.
Morse JG, Brawner OL (1986) Toxicity of pesticides to Scirtothrips citri (Thysa-
noptera: Thripidae) and implications to resistance management. 1. Econ.
Entomol. 79:565-570.
Morse JG and Brawner OL (1987) Citrus thrips control trial in California. Insectic.
& Acaric Tests 12:91-92.
Morse JG, Elliott WG (1985) Citrus red mite pesticide efficacy tests in Riverside.
Insectic. & Acaric. Tests 10:55-56.
Morse JG, Elmer HS, Brawner OL (1984) Evaluation of unregistered materials for
control of citrus thrips in California. Insectic. & Acaric Tests 9:91-92.
Morse JG, Grout TG, Brawner OL (1986) Citrus thrips control trials in California.
Insectic. & Acaric Tests 11:101-102.
Oetting RD (1987) Laboratory evaluation of the toxicity and residual activity of
abamectin to Ecinothrips americanus. 1. Agric. Entomol. 4:321-326.
Parrella MP (1983) Evaluations of selected insecticides for control of permethrin-

resistant Liriomyza trifolii (Diptera: Agromyzidae) on chrysanthemum.
J. Econ. Entomol. 76:1460-1464.
Parrella, MP, Robb KL, Virzi JK, Dybas RA (1988) Analysis of the impact of
abamectin on Liriomyza trifolii (Burgess) (Diptera: Agromyzidae). Can. Ent.
120:831-837.
Payne JA, Dutcher J, Ellis HC (1985) Performance of pecan insecticides: past,
present and future. Insectic. & Acaric Tests 10:63-65.
Putter I, MacConnell JG, Preiser FA, Haidri AA, Ristich SS, Dybas RA (1981)
Avermectins: novel insecticides, acaricides and nematicides from a soil micro-
organism. Experientia 37:963-964.
Riedl H, Shearer PW (1987) Pear, insect control with abamectin. Insectic. &
Acaric Tests 12:66-67.
Robb KL, Parrella MP (1986) Western flower thrips control. Greenhouse Grower,
September, 94-98.
Robb KL, Virzi JK, Parrella MP (1985) Control of twospotted spider mites on
chrysanthemum, California. Insectic. & Acaric Tests 10:284.
Royalty RN, Perring TM (1987) Comparative toxicity of acaricides to Aculops
lycopersici and Homeopronematus anconai (Acari: Eriophydae, Tydeidae).
J. Econ. Entomol. 80:348-351.
Royer TA, Edelson JV (1987) Spider mite and cotton aphid control. Insectic. &
Royer TA, Edelson JV, Cartwright B (1985) Effect of various soil and foliar
applied insecticides on leafminers in bell peppers. Insectic. & Acaric Tests
10:127.
Sanderson JP, Barnes MM (1983) Control of spider mites on almonds. Insectic. &
Schuster DJ (1984) Insect control on fresh market tomatoes in West-Central
Florida. Insectic & Acaric Tests 9:201-202.
Schuster DJ (1987) Insect control on fresh market tomatoes in West-Central
Florida. Insectic & Acaric Tests 12:157-158.
Schuster DJ, Everett PH (1983) Response of Liriomyza trifolii (Diptera: Agro-
myzidae) to insecticides on tomatoes. J. Econ. Entomol. 76:1170-1174.
Schuster DJ, Everett PH (1987) Insect control on tomato in Southwest Florida.
Insectic & Acaric Tests 12: 156.
Schuster DJ, Taylor JL (1987) Residual activity of abamectin against Liriomyza
trifolii (Diptera: Agromyzidae). Fla. Entomol. 70:351-354.
Trumble JT (1983a) Pesticidal impact on leafminers and parasites in celery.
Insectic & Acaric Tests 8:125-126.
Trumble JT (1983b) Suppression of lepidopterous larvae and leafminers on
tomato. Insectic & Acaric. Tests 8:150.
Trumble JT (1985a) Impact of pesticides and leafminers and their associated
parasites on tomatoes. Insectic & Acaric Tests 10: 166.
Trumble JT (1985b) Integrated pest management of Liriomyza trifolii: influence of
avermectin, cyromazine, and methomyl on leafminer ecology in celery. Agric.
Ecosystems Environ. 12:181-188.
Trumble JT, Moar, WJ, Babu JR, Dybas RA (1987) Laboratory bioassay of the
acute and antifeedant effects of avermectin B1• J. Agric. Entomol. 4:21-28.
Trumble JT, Nakakihara H (1984) Twospotted spider mite control on straw-
berries. Insectic & Acaric Tests 9:62-63.
Tuttle DM, Mullis CH (1984) Control of carmine spider mite on cotton. Insectic &
Acaric Tests 9:278.
Waddill V (1982) Evaluation of insecticides for control of insect pests of tomato.
Insectic & Acaric Tests 7: 126.
Williams DF (1983) The development of toxic baits for the control of the red
imported fire ant. Fla. Entomol. 66:162-171.
Williams DF (1985) Laboratory and field evaluation of avermectin against the
imported fire ant. Southwest Entomol. (Suppl.) No. 7:27-33.
Wright DJ, Loy A, Green A St J, Dybas RA (l985a) The translaminar activity
of abamectin (MK-936) against mites and aphids. Med. Fac. Landbouww.
Rijksuniv. Gent 50I2b:595-601.
Wright DJ, Roberts ITJ, Androher A, Green A St J, Dybas RA (1985b) The
residual activity of abamectin (MK-936) against Tetranychus urticae (Koch) on
cotton. Med. Fac. Landbouww. Rijksuniv. Gent 50I2b:633-637.
Wright JE, Jenkins IN, Villavaso, EJ (1985) Evaluation of avermectin Bl
(MK-936) against Heliothis spp. in the laboratory and in field plots and against
the boll weevil in field plots. Southwest. Entomol. (Suppl.) No. 7:11-16.
Wynholds PF, Leigh TF (1982) Control of spider mites on cotton. Insectic &
Acaric Tests 7:165.
Wynholds PF, Leigh TF (1986) Control of spider mites on cotton. Insectic &
Zehnder GW, Trumble JT (1984) Insect control on celery. Insectic. & Acaric.
Tests 9:89-90.
CHAPTER 21
Use of Ivermectin in Humans

Bruce M. Greene, Kenneth R. Brown,
and Hugh R. Taylor
I. Introduction
II. Treatment of Onchocerciasis
A. Phase I Studies
B. Phase II Studies
C. Phase III Studies
D. Phase IV Studies
E. Ivermectin Effect on the Parasite O. volvulus
F. Ivermectin Effect on Transmission and as a Prophylactic Agent
G. Conclusion and Present Clinical Usage
III. I vermectin for Uses in Filarial Infections other than Onchocerciasis
A. Ivermectin in the Treatment of Filariasis Caused by Wuchereria
bancrofti
B. Ivermectin Use in Patients Infected with More than One Filarial
Species
C. Ivermectin Treatment for Human Nematodes
I. Introduction
Ivermectin (Campbell et al. 1983) has been extensively tested in human
onchocerciasis, and is now considered the drug of choice. In a single
yearly dose, it suppresses microfilariae in the skin and eyes, and in most
infected persons prevents disease progression. It also shows considerable
promise for treatment of lymphatic filariasis and strongyloidiasis.
II. Treatment of Onchocerciasis

A. PHASE I STUDIES
The first reports of the use of ivermectin against Onchocerca volvulus

infection in humans were published in 1982 (Aziz et al. 1982a; Aziz et al.
1982b). In a crossover study on a small number of lightly infected people
in Senegal, they found that a single dose of ivermectin exhibited microfi-
312 B.M. Greene. K.R. Brown. and H.R. Taylor
laricidal activity; it was efficacious at remarkably low doses, ranging from

5 to 30ltg/kg. This report was followed by further dose-finding studies in
Paris (Coulaud et al. 1983; Coulaud et al. 1984), in which a number of
African immigrants, most of whom were also lightly infected, were
treated with various doses up to 200 ltg/kg. Although microfilariae
persisted or reappeared in some people, the drug showed good efficacy
overall; doses up to 200 ltg/kg were found to be safe and well tolerated.
A third Phase I study was performed in Tamale, Ghana (Awadzi et al.
1984; Awadzi et al. 1985). Four different doses of ivermectin-50, 100,
150, and 200 ltg/kg-were compared for efficacy in an open study. The
data suggested that ivermectin was probably safe and effective; 50 ltg/kg
was less effective than the higher doses. Although some patients ex-
hibited mild transient changes, including postural hypotension, iver-
mectin showed minimal toxicity. Without an adequate control group,
however, the data are difficult to interpret, complicated further by the fact
that the study population came from northern Ghana, where there had
been 10 years of vector control and the microfilaria counts, even without
treatment, were falling.
B. PHASE II STUDIES
The Phase I clinical data made it obvious that ivermectin was a very
promising drug (Editorial 1984). This led to a series of Phase II studies,
which were conducted as double-blind, controlled clinical trials. They
focused particularly on assessing invermectin's safety and efficacy as
compared to the standard treatment for onchocerciasis, diethylcarbam-
azine (DEC); the control group received a placebo. The 4 studies used
essentially the same rigorous protocol: subjects were hospitalized, treated
with single-dose ivermectin (200 ltg/kg) or a I-week course of DEC, and
examined every day-some of the observations being done every 4 to 6
hours. Patients also had detailed eye exams, including fluorescein angio-
graphy. Each study had 10 to 15 subjects per group. Phase II studies were
conducted in Liberia, Senegal, Ghana, and Mali.
The trial in Liberia (Greene et al. 1985) showed that a single dose of
ivermectin was at least as effective in reducing the microfilaria counts as
DEC given daily for 8 days; also, ivermectin caused very little reaction as
compared to DEC. Subsequent results showed that the marked reduction
of skin microfilaria counts seen following ivermectin treatment persisted
to 12 months (Taylor et al. 1986). These results were confirmed in the
other 3 Phase II studies (Lariviere et al. 1985; Diallo et al. 1986; Awadzi
et al. 1986), Ivermectin was clearly microfilaricidal, but examination of
excised nodules showed that it did not kill the adult worms (Greene et ai,
1985), It did, however, produce a transient interruption of microfilarial
release (Schulz-Key et al. 1985).
Ivermectin treatment produced either no or a very mild clinical reaction
21. Use of Ivermectin in Humans 313
(Mazzotti reaction). The absence of a Mazzotti reaction is highly signifi-

cant: this is the limiting factor to widespread use of DEC (Taylor 1984).
To quantitate this observation, a clinical grading system, which scores 7
separate clinical signs and then combines them in an aggregate clinical
reaction score, was used. The clinical reaction scores ofthe Liberia Phase
II patients treated with DEC were approximately 7.5 on Day 2. Typically,
a patient with a score of 5 or 6 felt unwell and would be confined to bed.
Ivermectin induced only a mild reaction, and the placebo group had no
reaction. The differences between the ivermectin group and the DEC
group were statistically significant.
The ocular changes that occur with DEC treatment of onchocerciasis
can lead to permanent loss of vision, thus restricting its clinical utility
(Taylor 1984; Bird et al. 1980). In the Phase II trials, detailed ocular
examinations were performed serially over a 12-month period, (Taylor et
al. 1986; Dadzie et al. 1987). DEC treatment caused a marked increase in
living and dead microfilariae in the cornea, punctate opacities, and
limbitis during the first week of therapy. Ivermectin (200 ILg/kg) had no
such effect, and resulted in a long-term reduction in intraocular microfi-
lariae comparable to that seen with DEC (Taylor et al. 1986). The Phase II
trials thus showed that ivermectin appeared to induce few ocular compli-
cations, to be better tolerated, and to be more effective than DEC as a
microfilaricidal agent in treating onchocerciasis.
C. PHASE III STUDIES
Phase III trials assessed the efficacy of 3 different dosages of the drug
(100, 150, and 200 ILg/kg) and collected further information on safety.
These trials were all double-blind, and, again, patients were followed for 1
year. Each group had at least 50 patients; all Phase III participants were
hospitalized for at least 2 days after treatment. At this point, the Liberian
study is the only one of the 6 published. Other studies following similar
protocols were undertaken in Ghana, Mali, Cote d'Ivore, Togo, and
Guatemala.
The 12-month follow-up data ofthe Liberian study show that, by Day 3,
each of the 3 doses of ivermectin produced quite a marked drop in skin
microfilaria counts (White et al. 1987). This decrease was more marked at
Day 3 for the 200 ILg/kg group than for the 100 ILg/kg dose, suggesting a
dose response at this early time point. At 3 months, however, the counts
for the 3 different doses were indistinguishable; they were still indistin-
guishable at 6 months and at 12 months.
In the Phase III trials, the clinical reaction score was assessed as it had
been in the Phase II studies. In the Liberian study, each of the 3 treatment
groups showed an increase in the clinical reaction score over the first few
days following treatment (White et al. 1987). Because there were now 50
people in each group (as opposed to only 10 in each group in the Phase II
314 B.M. Greene, K.R. Brown, and H.R. Taylor
study), this increase was statistically significant for both 200 and
150 ILg/kg compared to the placebo group. The reaction score for the
100 ILg/kg group was not significantly different from the placebo group.
The reaction was relatively mild in each case (except for 1 patient out of
150 who missed 1 day of work) and had essentially resolved by Day 4.
Again, no serious ocular lesions were precipitated by treatment, and the
ocular status was markedly improved 1 year after treatment (White et al.
1987). Similar results from the other Phase III trials have been reported
informally (Lariviere et al. 1986; Mossinger et al. 1987; Hussein, Bird,
and Jones 1987; Ranque et al. 1986).
The Phase III studies reconfirmed that ivermectin was both safe and
effective and induced only a minimal Mazzotti reaction without a
significant ocular reaction. These studies suggested that 150 ILg/kg was
probably the optimal dose, in terms of its antiparasitic activity and lack of
side effects.
At the end of the I-year follow-up in the Liberian Phase III study,
participants were reallocated for a dose-interval study. At 12 months, the
patients in the 3 treatment groups (100, 150, or 200 ILg/kg of ivermectin)
were rerandomized so that, in each treatment group, half the persons
were retreated with the same dose of ivermectin they had received
initially, and the other half were given placebo (Greene et al. 1987). The
original placebo group received 150 ILg/kg on a 6-monthly basis. At 24
months, all patients were again given ivermectin; the dose being the same
as they had received initially. This study provided information on 3
different retreatment schedules: 6-monthly treatments with 150 ILg/kg
with a I-year follow-up; annual treatment with 100, 150, or 200 ILg/kg with
a 2-year follow-up; and 2-yearly treatment with the 3 doses, also with a
2-year follow-up.
The 2-year follow-up data suggest that, even 24 months after initial
treatment, there is still a marked effect on skin microfilaria counts at each
treatment dose (Figure 21.1). Yearly retreatment with 150 ILg/kg,
however, appears to be an optimal regime for reducing microfilarial
counts (Figure 21.2) and for safety. Importantly, the clinical reaction seen
after retreatment at 12 months was less than that seen after the initial
treatment.
The ocular examination data at 24 months for the 3 different treatment
groups are shown in Figure 21.3. There was a marked improvement in the
ocular status of each group after treatment (Newland et al. 1988). Even
without annual retreatment, the ocular status showed marked improve-
ment over conditions prior to the initial treatment, 24 months earlier.
Ivermectin was also found to be safe and effective in patients with severe
ocular involvement (Taylor et al. unpublished). Interestingly, the im-
provement in ocular disease appeared to be even more prolonged than the
effect on the skin snip counts.
0--0 1OO1"J/1<9
.......
........
I5OI"JIKt
2OOI"J/KQ
Placebo at 12 months
20
II
l
-
2
c
II
/t,
I 12
~
J 8 / ........._- j
/" /
" t-
"" .. '
---;::.-<";>(,
--~i
1·1
'1
,::.--- I
V' ,/ _ '
4 I
OL---~~---L------ ____ ~ ________ ~ __________ ~
I 12 18 24
t t
Time (Months)
FIGURE 21.1. Skin microfilaria density over 24 months, expressed as microfi-

lariae/mg (mf/mg) skin. The broken line after 12 months shows the response of
patients given only a single dose of ivermectin. The solid line shows the response
of patients retreated at 12 months. Vertical bars show 95% confidence intervals.
D. PHASE IV STUDIES
Normally, these data, based on treatment of approximately 1400 persons,

would be considered sufficient for drug registration. Because ivermectin
would be used on a large scale in developing areas and therefore would
not be subjected to the normal postmarketing surveillance, however, it
was decided that a number of formal Phase IV trials were necessary,
involving treating tens of thousands of persons with ivermectin yearly and
following them for the occurrence of severe adverse reactions. Such
Phase IV studies are currently underway in Liberia (where almost 8000
persons have been treated twice without severe adverse reactions;
Pacque et al. unpublished), the Volta River Basin Area of West Africa,
Cameroon, Nigeria, and Guatemala. Preliminary results from all
- - - Historical Control - - - 12 monthly treatment

1985-1986 050poQ/KQ)
1985-1986
- - 6 monthly treatment
24 050~/Ko)
1986-1987
c: 20
~
(J)
co
- "\
E
....... 16
~
c:
0
\
,'\,"\ .
cu
E
-
u 12
.~
cu
E " \
0
cu \\
,\
"
(!) 8 \
\\
\,
,\,
4
\\.11 . .-:::."':'..::-.,:_ .......-'1
\. \ { II
r.-· ------ I
o~--------~----------~-----------
6 12
Time (Months)
FIGURE 21.2. Comparison of the effect on skin microfilaria density load 6-monthly
and 12-monthly treatment with ivermectin 150 I-'g/kg.
studies-which have now treated more than 70,000 persons-indicate no

unexpected major adverse effect of ivermectin (unpublished). Unex-
plained hypotension did occur in about 1 in 1000 to 1 in 10,000 persons,
but this was reversible and also has been reported following DEC
administration for O. volvulus infection.
E. IVERMECTIN EFFECT ON THE PARASITE O. VOLVULUS
Ivermectin's effect on the adult worm was examined in detail in both the
Phase II and Phase III trials. Ivermectin did not affect the viability of the
adult worm-male or female. In some studies, there appeared to be
slightly fewer adult males in the ivermectin nodules, but this was not
statistically significant (Greene et al. 1985; Schulz-Key et al. 1985). There
3 100 meg/kg -RIl

- - - Placebo
2 I: 95% C.l.
O~---L----~---------~---------~---------~
4 3 6 12 18 24
3 150 meg/kg
4
200 meg/kg
3
2
T
o 3 6 12 18
TIME (MONTHS)
FIGURE 21.3. Ocular reaction index results over 24 months. The broken line after
12 months shows the response of patients given only a single dose of ivermectin.
The solid line shows the response of patients retreated at 12 months.
was a dramatic effect on the uterine microfilariae, however; by 2 months,

microfilarial production totally ceased, and large numbers of degenerate
microfilariae accumulated in the uterus (Schulz-Key et al. 1985, 1986).
Because microfilariae are not released, they degenerate. By 6 to 12
months, normal production of microfilariae resumes, at least in some of
the worms. This effect on the adult female is probably responsible for the
delayed reappearance of microfilariae in the skin after ivermectin treat-
ment. In contrast, DEC kills the microfilariae very effectively but has no
residual effect on adult worms, which continue to produce microfilariae
normally (Albiez et al. 1988) The rate of reappearance of microfilariae in
the skin after DEC is, therefore, different from that after ivermectin.
Pretreatment levels are usually reached 1 to 2 years after DEC (Duke
1968), whereas after treatment with ivermectin, the microfilaria counts in
skin snips are still low at 1 or even 2 years.
F. IVERMECTIN EFFECT ON TRANSMISSION AND AS A

PROPHYLACTIC AGENT
Studies in Liberia (Cupp et al. 1986) suggested that ivermectin-treated

infected people are much less able to transmit the parasite. Although the
study was small and not perfectly controlled, the number of microfilariae
taken up by biting black flies was greatly reduced; so was the number of
subsequent developing infective larvae. This suggested that community-
based treatment with ivermectin might reduce or abolish transmission of
new infection. A recent report from the Ivory Coast (Prod'hon et al. 1987)
confirms and extends these findings.
Because there is no drug available to prevent O. volvulus infection,
there was considerable interest as to whether ivermectin could be used as
a prophylactic agent. A study was undertaken in chimpanzees experimen-
tally infected with infective larvae (Taylor et al. 1988). Animals were
given ivermectin (200 #Lg/kg) on the day of inoculation and 28 days later,
times selected to coincide with the presence of 3 and 4 larval stages.
Unfortunately, the incidence of patent infection in the chimpanzees
suggests that ivermectin has no demonstrable effect when given at the 4
larval stage and possibly only a partial effect at the 3 stage, indicating that
it will not be effective as a prophylactic drug in humans.
G. CONCLUSION AND PRESENT CLINICAL USAGE
In October 1987, just over 5 years after the first report on ivermectin's use
in onchocerciasis was published, the French government approved the
drug's use in humans (Lindley 1987). Although not a macrofilaricide-and
therefore not a permanent cure for onchocerciasis-ivermectin given as a
single oral dose of 150 #Lg/kg annually shows promise of revolutionizing
the treatment of this major disease of humankind, which previously was
essentially untreatable.
Ivermectin is now considered the drug of choice for onchocerciasis. It
should not be taken during pregnancy nor by lactating females within 3
months of delivery. It induces a minimal reaction (Mazzotti-like) in 5% to
15% of infected adults and a more significant reaction in about 1%. Its
safety is unproven in children under the age of 5 and in persons with
meningitis, sleeping sickness, and other CNS disorders. Its use was
21. Use ofIvermectin in Humans 319
associated with hypotension, usually occurring within 24 hours of admin-

istration in about 1 in 1000 persons treated in savannah West Africa. The
infrequent problems associated with ivermectin usage do not, however,
preclude its widespread use for onchocerciasis control.
III. Ivermectin for Uses in Filarial Infections other

than Onchocerciasis
A. IVERMECTIN IN THE TREATMENT OF FILARIASIS CAUSED BY
WUCHERERIA BANCROFT[
Four studies have been performed in patients with bancroftian filariasis.

Diallo and colleagues (1987) studied single-dose ivermectin therapy at a
level of either 50 JLg/kg (7 patients) or 100 JLg/kg (9 patients). Both dose
levels cleared microfilariae from the blood within 3 days of therapy; the
authors concluded that 100 JLg/kg might be the more effective dose.
In a further dose-ranging study of ivermectin in bancroftian filariasis in
Polynesia, Roux, Cartel, and Azia (1987) demonstrated the superiority of
100, 150, or 200 JLg/kg over a 50 JLg/kg single dose, but they were unable
to establish superiority of any 1 of the 3 higher doses over the other 2. In a
third study, performed in India (Kumaraswami et al. 1988), in patients
with bancroftian filariasis, all of the dose levels chosen (25, 50, 100, or
200 JLg/kg) were efficacious in clearing the blood of all patients treated.
This efficacy was evaluated at 5 to 12 days, but in most patients
microfilaremia recurred by 3 months after dosing. At 6 months there was
no correlation between dose level and efficacy. In this study there
appeared to be a good correspondence between high levels of microfilare-
mia, the higher doses used (100 and 200 JLg/kg), and the frequency and
severity of adverse reactions. The latter were similar to those frequently
seen in patients treated with DEC.
Finally, in a comparative study of DEC and 2 dose levels (1.0 mg or
6.0 mg) of ivermectin reported by Ottesen and colleagues (1987), iver-
mectin was superior to DEC in rapidly clearing blood microfilaria, and
equivalent to DEC in decreasing the microfilaria counts over the first 3
months; ivermectin was inferior to DEC in terms of sustained response at
6 months. Furthermore, adverse reactions reported for ivermectin were
qualitatively and quantitatively similar to those for DEC; the higher dose
of ivermectin was associated with a higher frequency of adverse reactions
and was also correlated with the microfilarial level prior to treatment.
Because ofthe compliance problem with DEC, the investigators conclude
that ivermectin given as a single dose would be superior to the standard
course of DEC treatment; they propose that it should be considered the
best candidate to replace DEC in control programs. Additional dose
finding studies are now underway.
B. IVERMECTIN USE IN PATIENTS INFECTED WITH MORE THAN

ONE FILARIAL SPECIES
Richard-Lenoble and colleagues (1988) treated patients who were infected

with Loa loa as well as o. volvulus; some of these also had M. perstans
infection. A single dose of ivermectin at 200 JLg/kg had no significant
effect on the M. perstans in the blood, produced a transitory drop in
microfilaremia due to Loa loa and, as expected, produced a dramatic drop
in counts of O. volvulus in the skin. Of greater importance was the high
level of safety: no multiply infected patients developed serious adverse
reactions related to ivermectin treatment.
C. IVERMECTIN TREATMENT FOR HUMAN NEMATODES
Nalin and colleagues (1987) reported efficacy results from studies of 294
infected patients treated with ivermectin. Mean percent cures were:
S. stercoralis, 82%; A. lumbricoides, 100%; E. vermicularis, 84%; and
T. trichiuris, 69%.
Naquira and colleagues (unpublished data on file at Merck Sharp &
Dohme Research Laboratories) have studied 100 patients ranging in age
from 11 to 74 who were infected with Strongyloides stercora lis . Diagnosis
was based on the presence of S. stercora lis larvae in the stool; ivermectin
was given either as a single oral dose (50, 100, 150, or 200 JLg/kg) or as a
single dose of 100 or 200 JLg/kg on each of 2 successive days. Follow-up
stool examinations were performed at least once between days 17 to 24
and between days 30 to 38 after dosing.
Patients who received a total of 50 ro 100 JLg/kg ivermectin had a
response rate which ranged from 69% to 75%, whereas the patients who
received a total of 150 JLg/kg or more had a cure rate of 88% to 100%.
Based on this study, ivermectin in a dose of 150 JLg/kg would appear to be
highly effective against S. stercoralis. Even for patients who were not
cured, the change in larval counts in the stool was considered to be very
beneficial.
Studies of other parasites in these patients and 131 additional persons
without strongyloides were performed; cure rates obtained for human
hookworms (either Necator or Ancylostoma) were unsatisfactory even
though hookworm egg counts in stools were reduced. Even at
200 JLg/kg/day, the 30-day cure rate for patients with hookworm was only
22%; at 30 days patients who were not cured had only a 42% drop in egg
counts in the stool. The mean cure rate in all patients at 30 days
(regardless of dose) was 85% for Trichuris trichiura, 100% for Ascaris
lumbricoides, and 85% for Enterobius vermicularis.
The preceding data demonstrate ivermectin's high degree of efficacy
and suggest its possible use as a replacement for thiabendazole in treating
Strongyloides stercoralis. Ivermectin could not be recommended as a
broad-spectrum agent for monotherapy of human roundworm infestation,

however, because it fails to effectively control human hookworms.
REFERENCES
Albiez EJ, Newland HS, White AT, Kaiser A, Greene BM, Taylor HR, Buttner
DW (1988) Chemotherapy of onchocerciasis with high doses of diethylcarbama-
zine or a single dose of ivermectin: microfilaria levels and side effects. Trap.
Med. Parasit. 39: 19-24.
Awadzi K, Dadzie KY, Schulz-Key H, Gilles HM, Fulford AJ, Azia MA (1986)
The chemotherapy of onchocerciasis XI. A double-blind comparative study of
ivermectin, diethylcarbamazine, and placebo in human onchocerciasis in
Northern Ghana. Ann. Trap. Med. Parasit. 80:433-442.
Awadzi K, Kadzie KY, Schulz-Key H, Haddock DRW, Gilles HM, Aziz MA
(1984) Ivermectin in onchocerciasis (letter). Lancet 2:2921.
Awadzi K, Dadzie KY, Schulz-Key H, Haddock DRW, Gilles HM, Aziz MA
(1985) The chemotherapy of onchocerciasis X. An assessment of four single
dose treatment regimes of MK-933 (ivermectin) in human onchocerciasis. Ann.
Trap. Med. Parasit. 79:63-78.
Aziz MA, Diallo S, Diop 1M, Lariviere M, Porta M (l982a) Efficacy and tolerance
of ivermectin in human onchocerciasis. Lancet 2:171-173.
Aziz MA, Diallo S, Lariviere M, Diop 1M, Porta M, Gaxotte P (l982b) Ivermectin
in onchocerciasis. Lancet 2:1456-1457.
Bird AC, EI-Sheikh H, Anderson J, Fuglsang H (1980) Changes in visual function
and in the posterior segment of the eye during treatment of onchocerciasis with
diethylcarbamazine citrate. Brit. 1. Ophthalmol. 64: 191-200.
Campbell WC, Fisher MH, Stapley EO, Albers-Schonberg G, Jacob TA (1983)
Ivermectin: a potent new antiparasitic agent. Science 221:823-828.
Coulaud JP, Lariviere M, Aziz MA, Gervais MC, Gaxotte P, Deluol AM, Cenac J
(1984) Ivermectin in onchocerciasis. Lancet 2:526-527.
Coulaud JP, Lariviere M, Gervais MC, Gaxotte P, Aziz MA, Deluol AM, Cenac J
(1983) Treatment of human onchocerciasis with ivermectin. Bull. Soc. Pathol.
Exot. Filiales 76:681-688.
Cupp EW, Bernardo MJ, Kiszewski AE, Collins RC, Taylor HR, Aziz MA,
Greene BM (1986) The effects of ivermectin on transmission of Onchocerca
volvulus. Science 231 :740-742.
Dadzie KY, Bird AC, Awadzi K, Schulz-Key H, Gilles HM, Aziz MA (1987)
Ocular findings is a double-blind study of ivermectin versus diethylcarbamazine
versus placebo in the treatment of onchocerciasis. Brit. 1. Ophthalmol.
71:78-85.
Diallo S, Aziz MA, Lariviere M, Diallo JS, Diop-Mar I, N'Dir 0, Badiane S,
Py D, Schulz-Key H, Gaxotte P, Victorius A (1986) A double-blind comparison
of the efficacy and safety of ivermectin and diethylcarbamazine in a placebo
controlled study of Senegalese patients with onchocerciasis. Trans. Roy. Soc.
Trop. Med. Hyg. 80:927-934.
Diallo S, Aziz MA, Ndir 0, Badiane S, Bah IB, Gaye 0. (1987) Dose-ranging
study of ivermectin in treatment of filariasis due to Wuchereria bancrofti.
Lancet 1:1030.
Drummond RO (1985) Effectiveness of ivermectin for control of arthropod pests

of livestock. Southwest. Entomol. (Suppl.) 7:34-42.
Duke BO (1968) The effects of drugs on Onchocerca volvulus I. Methods of
assessment, population dynamics of the parasite and the effects of diethylcar-
bamazine. Bull. WHO 39:137-146.
Editorial (1984) Ivermectin in onchocerciasis. Lancet 2: 1021.
Greene BM, Taylor HR, Cupp EW, Murphy RP, White AT, Aziz MA, Schulz-Key
H, D'Anna SA, Newland HS, Goldschmidt LP, Auer C, Hanson AP, Freeman
SV, Reber EW, Williams PN (1985) Comparison of ivermectin and diethylcar-
bamazine in the treatment of onchocerciasis. New Engl. l. Med. 313:133-138.
Greene BM, White AT, Newland HS, Keyvan-Larijani E, Dukuly F, Gallin MY,
Aziz MA, Williams PN, Taylor HR (1987) Single dose therapy with ivermectin
for onchocerciasis. Trans. Assoc. Amer. Phys. C:131-138.
Hussein S, Bird A, Jones BR (1987) Ocular lesions seen in phase III trial of
ivermectin therapy of onchocerciasis in Togo. Trop. Med. Parasit. 38:67.
Kumaraswami V, Ottesen EA, Vijayasekaran MD, UmaDevi S, Sxaminathan M,
Aziz MA, Sarma GR, Prabhakar R, Tripathy SP (1988) Ivermectin for the
treatment of Wuchereria bancrofti filariasis. lAMA 259:3150-3153.
Lariviere M, Abelloos J, Aziz M, Beauvais B, Bamba M, Ferly-Therizol M,
Basset D, Basset A, Derouin F, Sarfati C, Bosseboeuf C, Toure Y, Song D,
Gaxotte P (1986) A double-blind study of ivermectin and placebo in onchocerci-
asis. Proceedings of the IXth International Congress of Infectious and Parasitic
Diseases, Munich, p 548.
Lariviere M, Aziz MA, Weimann D, Ginoux J, Gaxotte P, Vinotain P, Brauvae B,
Derouin F, Schulz-Key H, Barret D, Sarfati C (1985) Double-blind study of
ivermectin and diethylcarbamazine in African onchocerciasis patients with
ocular involvement. Lancet 2: 174-177.
Lindley D (1987) Merck's new drug free to WHO for river blindness programme.
Nature 329:752.
Mossinger J, Schulz-Key H, Helling G, Adjamgba A, Klager S (1987) Emergence
of Onchocerca volvulus microfilariae from skin biopsies of patients treated with
ivermectin. Trop. Med. Parasit. 38:69.
Nalin D Aziz M, Neu D, Ruiz-Palacios G, Naquira C (1987) Efficacy ofivermectin
(MK-933) against intestinal nematode parasites. Abstract A-101. Annual Meet-
ing of the American Society for Microbiology, Atlanta, p 17.
Newland HS, White AT, Greene BM, D'Anna SA, Keyvan-Larijani E, Aziz MA,
Williams PN, Taylor HR (1988) The effect of single dose ivermectin therapy on
human O. volvulus infection with onchocercal ocular involvement. Brit.
l. Ophthalmol. 72:561-569.
Newland HS, White AT, Greene BM, Keyvan-Larijani E, Taylor HR (in press)
The effect of repeated treatment with ivermectin in onchocerciasis. Report of
Third General Assembly of the International Agency for Prevention of
Blindness.
Ottesen EA, Kumaraswami V, Vijayasekaran MA, Aziz R, Prabhakar R, Tripathy
SP (1987) Ivermectin and diethylcarbamazine (DEC) in a blinded, placebo-
controlled trial to treat Wuchereria bancrofti (Wb) filariasis. Proceedings of the
36th Annual Meeting of the American Society of Tropical Medicine and
Hygiene, Los Angeles, p 255.
Prod'hon J, Lardeux P, Bain 0, Hebrard G, Prod'hom JM (1987) Ivermectine et
modalites de la reduction de l'infection des simulies. Ann. Parasitol. Hum.

Compo 68:590-598.
Ranque P, Aziz M, Pichard E, Ginoux J, Cozette P, Rivaud C, Sacko A, Coulibaly
S, Gaxotte P (1986) A double-blind study of ivermectin and placebo in
onchocerciasis in Mali. Proceedings of the IXth International Congress of
Infectious and Parasitic Diseases, Munich, p 140.
Richard-Lenoble D, Kombila M, Rupp EA, Pappaylion ES, Gaxotte P, Njuiri C,
Aziz MA (1988) Ivermectin in loiasis and concomitant O. volvulus and
M. perstans infections. Am. J. Trop Med Hyg 39:480-483.
Roux J, Cartel JL, Aziz MA (1987) A dose-ranging study of ivermectin for the
treatment of microfilaremic filariasis due to Wuchereria bancrofti in French
Polynesia. Abstract 395. 36th Annual Meeting of the American Society of
Tropical Medicine and Hygiene, Los Angeles.
Schulz-Key H, Greene BM, Awadzi K, Lariviere M, Klager S, Dadzie Y, Aziz
MA (1986) Efficacy of ivermectin on the reproductivity of female Onchocerca
volvulus. Trop. Med. Parasit. 37:89.
Schulz-Key H, Klager S, Awadzi K, Diallo S, Greene BM, Lariviere M, Aziz MA
(1985) Treatment of human onchocerciasis: the efficacy of ivermectin on the
parasite. Trop. Med. Parasit. 36:20.
Taylor HR (1984) Recent developments in the treatment of onchocerciasis. Bull.
WHO 509-515.
Taylor HR, Murphy RP, Newland HS, White AT, D'Anna SA, Keyvan-Larijani
E, Aziz MA, Cupp EW, Greene BM (1986) Comparison of the treatment of
ocular onchocerciasis with ivermectin and diethylcarbamazine. Arch. Ophthal-
mol. 104:863-870.
Taylor HR, Trpis M, Cupp EW, Brotman B, Newland HS, Soboslay PT, Greene
BM (1988) Ivermectin prophylaxis against experimental Onchocerca volvulus
infection in chimpanzees. Am. J. Trop. Med. Hyg 39:86-90.
White AT, Newland HS, Taylor HR, Erttmann KD, Keyvan-Larijani E, Nara A,
Aziz MA, D' Anna SA, Williams PN, Greene BM (1987) Controlled trial and
dose-finding study of ivermectin for treatment of onchocerciasis. J. Infect. Dis.
156:463-470.
ApPENDIX I
The Determinative Method for

Assaying Ivermectin Residues in
Tissue and Plasma
G. v. Downing
A. Summary
B. Reagents
C. Solutions
D. Apparatus
E. Chromatographic Apparatus
F. Standards for the Determinative Assay
G. Procedure for the Determinative Assay of Tissues
H. Suitable Stopping Places for Tissue Assays
I. Analytical Procedure for Asay in Plasma
A. Summary
The tissue sample is homogenized with acetone-water, and dihydro B 1a is
extracted from the tissue with isooctane. The isooctane is evaporated,
and solvent-solvent distributions into acetonitrile out of hexane and into
hexane out of acetonitrile-water are performed. Solvent is removed and
fluorescence produced by heating at 95° C with the imidazole-derivatizing
reagent. Mter adding chloroform, the reaction mixture is washed through
a silica gel SEP-PAK, the solvent is removed, and HPLC analysis
(fluorescence detection) on Zorbax CI8 is performed on the residue
dissolved in methanol. A flow diagram of the assay is shown in Figure
AI. I.
Proteins are precipitated from plasma by addition of acetone and then
water. The ivermectin dihydro B la is extracted with ethyl acetate from
the acetone-water-plasma mixture. The solvent is removed and fluo-
rescence produced by heating at 95° C with the derivatizing reagent. Mter
adding chloroform, the reaction mixture is washed through a silica gel
SEP-PAK, the solvent is removed, and the HPLC analysis (fluorescence
detection) on Zorbax CI8 is performed on the residue dissolved in
methanol.
Homogenize 5g t i _ Extract Into evaporate Dissolve In methanol centrtfuge Dissolve In
with -a-Iweter l - - laooctane end cool -20"C evaporate hexane
~ -
,
A
Exnctlnto Welhwlth concentrate Extract Into eveporate Dissolve In

• KetonItrtte r---- hexane add weter hexane 1.0 mI methanol
-- --
t
B
0.5 mI ellquot AdcI ..egent end Dilute with chlorofonn eveporate Dissolve In Inject Inlo LC
heet85-100·C r---- end pass through methanol ~ with IIU0f'8scenc:e
eveponte Sep·Pe~
for 1 hour cletec:llon
t
C
FIGURE ALl. Flow diagram of determinative assay. A, Band C are possible stopping points during the assay.
326 G.V. Downing
B. Reagents
1. Acetic anhydride-Fisher or Baker, reagent grade.
2. Acetone-Mallinckrodt, Nanograde.
3. Acetonitrile-Fisher, HPLC grade; Mallinckrodt, Nanograde; or
Burdick & Jackson, distilled in glass.
3. Chloroform-Burdick & Jackson, distilled in glass (1% ethanol
preservative).
5. Dimethyl Formamide (DMF)-Baker or Fisher, reagent grade; or
Mallinckrodt, AR.
6. Ethyl Acetate-Burdick & Jackson, distilled in glass; or Fisher,
pesticide grade.
7. Hexane-Burdick & Jackson, distilled in glass; or Mallinckrodt,
Nanograde.
8. Methanol-Burdick & Jackson, distilled in glass; or Mallinckrodt,
Nanograde.
9. Methylene Chloride-any grade available.
10. 1-Methylimidazole-99%, Aldrich Chemical Company.
11. Nitrogen-the equivalent of Matheson extra-dry compressed gas.
12. Sodium Chloride-Baker or Fisher, reagent grade.
13. Sylon-CT-Supelco, Inc.
14. 2,2,4-Trimethylpentane-isooctane, Burdick & Jackson, distilled in
glass; or Mallinckrodt, Nanograde.
15. Water, Filtered Millipore-Distilled water is treated with a Milli-Q
system and then filtered through a Millipore Type HA 0.451Lm disc.
C. Solutions
1. Derivatizing Reagent-Mix 0.9 ml of DMF with 0.3 ml of acetic
anhydride and 0.2 ml 1-methylimidazole. Make up just before use.
2. Acetone/Water-50% (v/v).
3. 95% Methanol/Water-Make 100 ml of filtered Millipore water to 2
liters with methanol. Mix. With the aid of vacuum, filter mixture
through Millipore FH 0.51Lm filters. Deaerate by slowly bubbling
nitrogen through for 10 minutes.
D. Apparatus
1. Balance-analytical, capable of weighing 1 mg accurately.
2. Balance-capable of weighing 5 g accurately into an approximately
50-g tube.
3. Bath, water-variable temperature 40° to 80° C.
4. Bath, oil-95° to 100° C.
5. Centrifuge, IEC Model HN-S-II, with 6-place rotor IEC #958 and 15-
Appendix I. Assaying Ivermectin Residues in Tissue and Plasma 327
and 50-ml cups. The centrifuge is run at 2000 to 2500 rpm. The
centrifuge used gives ca. 700 to 750 x g maximum centrifugal force
(RCF).
6. Centrifuge tubes-glass, 15 to 50 ml with polyethylene stoppers to fit.
7. Centrifuge tubes-50-ml polyproplyene, Corning#25331 (used only
for storing standard solutions).
8. Centrifuge tubes, 15 ml, silylated approximately once every 2 months
(used only for the derivatization reaction). Tube stoppers must fit
tightly. Fill each tube to the top with Sylon-CT. Let stand 20 minutes.
Immediately and quickly rinse, first thoroughly with toluene and then
with methanol. Fill with methanol. Let stand 20 minutes, rinse
thoroughly with acetone, and dry. All glassware used should be
completely free of all acidic and alkaline residues. (These tubes
should be cleaned by hand by first soaking in methylene chloride
immediately after use and then in detergent for at least several hours,
followed by thorough rinsing with hot water, distilled water, and
acetone before thorough drying. Variations in the washing regimen
are not recommended, since some analysts have had poor results
when the standard washing method was not followed.)
9. Dispensing pipettors-lO, 15, and 20 ml.
10. Filtration unit, complete with Millipore FHLP 04700 filter.
11. Freezer-capable of reaching temperatures of -200 C.
12. Gloves-disposable PVC from Fisher Chemical.
13. Graduate cylinder-25 ml, l00ml.
14. Graduate cylinder-2000 ml with ground-glass stopper.
15. Homogenizer, Polytron-Brinkmann Instruments, #27-11-200-5 with
PTA lOS generator #27-11-330-3.
16. Pipets-disposable, Pasteur.
17. Pipets-graduated 1.0 ml and 2.0 ml.
18. Pipets-volumetric 0.5 ml, 1 ml, 2 ml, 3 ml, 4 ml, and 5 ml.
19. Repipet-5 to 10 ml volumes.
20. SEP-PAK-silica cartridge, Part No. 51900-Waters Associates.
21. Spatula-stainless steel.
22. Syringe-50ILI, 250ILI, and 5 ml and 10 ml.
23. Tape-#13 2956, ~ inch from Ace Scientific.
24. Ultrasonic bath-Sonogen Automatic Cleaner-Branson Model 520
or equivalent.
25. Vortex mixer.
A Beckman-Altex Model IIOA liquid chromatographic pump complete
with Waters Wisp Autosampler and Kratos-Schoeffel Instruments Model
FS950 fluorescence detector with a I-millivolt recorder was used. A
328 G. V. Downing
51Lm, 4.6-mm ID, CI8 standard Brownlee Labs guard column (Spheri-5
RP-18 OD-GU, obtained from Rainin Instrument Co., Inc.) was used
before the analytical column. This guard column is replaced monthly
unless the pressure reaches 2000 psi, in which case it is replaced
immediately.
The ivermectin H2Bla derivative has a retention time of about 12 to 13
minutes when the indicated conditions are used.
Conditions: 15 cm length x 4.6 mm ID Zorbax ODS-CI8 column.

Mobile phase-5% water in methanol (v/v).
Column temperature-30° C.
Flow-1.8 mil minute (usual pressure 1000 psi, 500 to
2000 psi acceptable).
Detector Settings:
Excitation lamp-FSA 110, standard 365 nm.
Standard Kratos Flowcell-FSA 210.
Excitation filter-FSA 403, 365 nm band filters.
Emission filter-418 nm.
Sensitivity range-0.2 A.
Time constant-about 6.
F. Standards for the Determinative Assay

For tissue assays using the most sensitive scale, accurate aliquots of 50,
100, 150,200, and 250 ILl of ca. 500 ng/ml standard solution of dihydro B la
in methanol (working standard solution) were added to silylated 15-ml
centrifuge tubes, and the samples were evaporated to dryness at 60° C
with nitrogen purge. For plasma assays, accurate aliquots of 20, 40, 60,
80, and 100 ILl of working standard solution were added to silylated tubes
and evaporated to dryness. Mter reaction, SEP-PAK treatment, etc.,
these derivatized samples are redissolved in I ml of methanol to give
25, 50, 75, 100, and 125 ng/ml standard solution equivalents for tissue
assays and 10, 20, 30, 40, and 50 ng/ml standard solution for plasma
assays.
Starting with solutions of about 0.045% (w/w) dihydro B la in propylene
glycol, stock solutions of about 10 micrograms/ml can be prepared by
weighing accurately about 1 gram of the solution into a 50-ml volumetric
flask. Methanol is used to dilute to the mark, and the solution is
thoroughly mixed. Storage of this stock solution and all dilutions of same
should be in polypropylene tubes at -20° C.
Analysts attempting the ivermectin residue analysis for the first time
are advised to run a standard curve first before attempting the complete
method.
Procedure for the Determinative Assay of Tissues

Homogenization
All Tissues
1. Weigh accurately 5.0 grams of sample into a 50-ml centrifuge tube.
Liver, Muscle, Kidney, and Dose Site

2. Add 15 ml of 50% acetone/water to the tube from a dispensing
pipettor.
3. Homogenize for 15 to 20 seconds or shorter time with a Polytron
setting of 7. Homogenize until a homogeneous-looking mixture re-
sults.
4. Put a clean 50-ml tube in place containing 15 ml of isooctane from a
dispensing pipettor.
5. Homogenize 5 to 10 seconds and pour the isooctane into the 50-ml
tube with the homogenate. Clean the generator probe between
samples with isooctane, distilled water, and acetone.
Fat Only
2. Add 15 ml of the acetone/water and 10 ml ofisooctane to the tube and
homogenize until a homogeneous mixture results.
3. Rinse probe with 10 ml isooctane in a clean 50-ml tube.
4. Pour isooctance into the 50-ml tube with the homogenate.
5. Add 1 gram of solid sodium chloride to the tube containing the
homogenate.
Extractions
All Tissues
6. Stopper the tube, shake well for 1 minute, and centrifuge for 10
minutes.
7. Transfer the upper or isooctane layer to a second 50-ml centrifuge
tube with a disposable pipet. Completely avoid the lower layer.
8. Break up the plug by using a vortex mixer and/or shaking; add 15 ml
of clean isooctane, repeating the extraction and combining isooctane
layers.
9. Evaporate the combined isooctane layers to a small volume (or
ca. 5 ml for fat) using an 80° C bath and nitrogen purge.
10. Repeat the extractions with 2 more passes of 15 ml each of isooctane,
adding the isooctane in each case to the same tube which had the
previous 2 isooctane layers.
330 G.V. Downing
11. Again evaporate as far as possible using the 80° C bath and purge.
12. Add 6 ml of methanol to all samples except fat and dissolve or
resuspend completely with an ultrasonic bath and/or vortex mixer.
13. Place all samples except fat in a refrigerator until thoroughly cooled
(at this point the sample is best left overnight in the refrigerator or
freezer).
Fat Only
12. Either add 2 ml of hexane and S ml of acetonitrile to the still hot and
molten fat or freeze overnight at -20° C. For samples frozen
overnight, remelt in a 80° C bath and add the hexane and acetonitrile
the next day.
13. Shake the still warm mixture thoroughly for ca. 1 minute and
immediately centrifuge.
14. Cool in ice water until the lower or fat layer congeals and the upper or
acetonitrile layer can be decanted completely into a fresh IS-ml
centrifuge tube.
IS. Repeat the melting, extraction with S ml of acetonitrile, and transfer
into the same IS-ml tube.
16. Add 2 ml of hexane to the IS-ml tube, shake thoroughly, centrifuge,
and remove the upper, hexane layer to waste by disposable pipet or
by aspiration. Go to Step 22.
Liver, Muscle, Kidney, and Dose Site

14. Centrifuge the cold SO-ml tube for S minutes, and decant off the clear
supernatant to a clean IS-ml centrifuge tube.
IS. Wash the residue in the SO-ml tube with 1 ml of methanol using a
vortex mixer, centrifuge, and decant off into same IS-ml tube. (If
solids are decanted in either step, recentrifuge the IS-ml tube and
decant into a new IS-ml tube to get a clear solution to continue the
assay with.)
ISa.For muscles and dose sites only, do a 2-ml hexane extraction of the
residue solids and add to the combined methanol.
16. Using a 60° C bath and nitrogen purge, evaporate off all the methanol.
17. By repipet add 3 ml of hexane to the tube previously containing the
methanol and use an ultrasonic bath to remove all material from the
walls of the centrifuge tube.
18. Add by repipet 4 ml of acetonitrile and repeat the ultrasonic treatment
briefly.
19. Shake thoroughly, centrifuge S minutes, and move the lower or
acetonitrile layer to a clean IS-ml tube.
20. Reextract the hexane layer with a second 4 ml of acetonitrile and
combine acetonitrile layers.
21. Extract the combined acetonitrile layers with 1 ml of hexane, centri-
fuge to clear the layers, and move the upper, hexane layer to waste by
disposable pipet or by aspiration.
All Tissues
22. Evaporate the acetonitrile solution to 1.0 ml using a ca. 60° C bath and
nitrogen flush.
23. If the acetonitrile is less than 1 ml, make up to 1 ml with fresh
acetonitrile.
24. Use an ultrasonic bath to get a homogeneous mix if necessary.
25. Add 4 ml of distilled water (2 ml for kidney samples) and 5 ml of
hexane; shake ca. 1 minute and centrifuge for 5 minutes.
26. Move the upper, hexane layer to a clean 15-ml centrifuge tube using a
disposable pipet. Avoid the lower layer.
27. Repeat with 5 ml and, then, 4 ml hexane extractions combining all 3
hexane extracts.
28. Evaporate to dryness (or as near dryness as possible) using a 40° C
bath and nitrogen purge. (To get the sample more completely dry,
more heat may need to be applied at the end of the evaporation. The
bath may go up to ca. 80° C at this point.)
29. Redissolve the residue in exactly 1.0 ml of methanol using a vortex
mixer and ultrasonic bath.
30. Mix thoroughly and centrifuge for 5 minutes.
Derivatization
31. Pipet exactly 0.5 ml of the solution from Step 30 into the bottom of a
silylated 15-ml centrifuge tube. Use a O.5-ml volumetric pipet.
32. Completely evaporate the methanol carefully, using at 60° C bath and
nitrogen purge. Avoid spattering.
33. With a 1.0-ml graduated pipet, add 0.1 ml of freshly made acetic
anhydride-methylimidazole-DMF mix to each tube and a series of
standard tubes (see standard section F preceding).
34. Stopper, vortex briefly, and centrifuge for a few seconds.
35. Tape the stoppers in place and put all samples and standards in a
well-stirred 95° C oil bath for 1 hour. The bottoms of the tubes should
be about 1 inch below the oil surface.
36. Remove the tubes, wash the oil off the outside of the tubes with
acetone in a wash bottle, and cool briefly. The residue should be
black.
37. Add about 1 ml of chloroform to each tube and vortex to mix.
38. Wash a SEP-PAK cartridge with 3 to 4 ml of chloroforM, using a
syringe to force the liquid through.
39. Add the sample by disposable pipet to the syringe and force it through
the cartridge.
332 G.V. Downing
All Tissues
40. Wash the centrifuge tube 3 times with 1 ml each of chloroform into the
syringe and through the cartridge.
41. Elute the column with a further 8 to 9 ml of chloroform (with aliquots
of 3, 3, 2, or 3 ml).
42. Collect all the chloroform eluate from the SEP-PAK in a 15-ml
centrifuge tube.
43. Evaporate the chloroform off with a 60° C bath and nitrogen purge.
Get the residue completely dry.
HPLC Analysis
44. Pipet 0.5 ml of methanol (or other suitable quantity) into the tube and
use a vortex mixer and ultrasonic bath to completely dissolve the
residue.
45. Centrifuge briefly and load WISP inserts if autosampler is used.
46. Inject 50 ILl of the supernatant of each sample and standard into the
HPLC.
47. Measure the peak heights at the retention time of the derivative of
dihydro-B 1a as indicated by the standards.
48. Generate a standard curve by performing a linear regression analysis
of response (peak height) versus concentration (ng/ml) for standards.
The curve should be linear with a regression coefficient of 0.97 or
better. Use the regression equation to determine the concentration of
unknowns with an observed peak height.
Calculate the sample concentration as follows:
VI VI
ppb = (ng/ml)s x V2 x DIG = V2 x (ng/ml)s x DI5
Where: (ng/ml)s for the unknown is determined from the standard

curve.
VI = ml sample dissolved in at end of assay (Step 44).
V2= ml of sample taken to make derivative (Step 31).
D= dilution of sample at end of assay or 1 if no dilution is
made.
G= grams of sample taken = 5.
H. Suitable Stopping Places for Tissue Assays

The following stopping places allow storage for a maximum of 4 days:
1. In a refrigerator at 4° C.
a. Only in methanol after Step 12.
2. In a freezer at -20° C.
a. After Step 11 for fat only.
b. After Step 30, before derivatization.
c. After Step 36, the derivatization.

d. After Step 44, dissolution of the residue in methanol.
If stopping places other than the above are needed, assay stability at
said stopping place should be demonstrated first.
I. Analytical Procedure for Assay in Plasma

1. Pipet 5 ml of plasma into a 50-ml centrifuge tube, pipet in 7.5 ml of
acetone, shake briefly, and let stand 15 minutes.
2. Pipet 7.5 ml of distilled water and 15 ml of ethyl acetate into the tube
and shake by hand moderately for 1 minute. Shaking should be
thorough enough to allow extraction but not violent enough to cause
emulsions.
3. Centrifuge thoroughly and move the upper, ethyl acetate layer to a
second 50-ml tube as completely as possible, using a disposable pipet.
Move absolutely no lower phase.
4. Thoroughly break up the plug in the bottom of the first tube by
shaking or with the wide end of a disposable pipet and repeat the ethyl
acetate extraction a second time. Combine the ethyl acetate layers.
5. Evaporate the ethyl acetate completely with a nitrogen purge in a 75° C
water bath.
6. Pipet exactly 1 ml of methanol into the dry tube.
7. Dissolve the residue as much as possible by using an ultrasonic bath
with intermittent shaking. (The methanol must contact all the tube.)
8. Centrifuge and pour the supernatant solution into a 15-ml centrifuge
tube and pipet exactly 0.5 ml of this solution into the bottom of a
silylated 15-ml centrifuge tube.
9. Evaporate completely to dryness with a nitrogen purge and 60° C
bath.
10. Using a I-ml graduated pipet, add 0.1 ml of derivatizing reagent to
samples and standards, stopper, and use a vortex genie to dissolve the
residue.
11. Centrifuge briefly, tape the stoppers, and put unknowns and standards
together in a well-stirred 95° C oil bath for 1 hour.
12. Remove tubes and rinse the oil off the outsides with acetone. The
residue should be black. (If the residue is not black, the remaining
sample from Step 8 can be used to repeat the derivatization with the
next set of standards.)
13. Add ca. 1 ml of chloroform to the residue in each tube.
14. Use a vortex genie to thoroughly mix.
15. Wash a silica SEP-PAK cartridge with 3 to 4 ml of chloroform, using a
syringe to force the liquid through and discard to waste.
16. Add the chloroform sample to the syringe with a disposable pipet and
force the liquid through the cartridge.
17. Wash the centrifuge tube 3 times with ca. 1 ml each of chloroform into
. : i., ·-'-i-'-i- 1 c_
L!- -r;'-;-t- ~'8¥!-i';
:..H-1':'1:: '_"j_,.J .- H+=tl:··t ·ITj,+,'·Hi-'!+' i#-~"1-:" ~.--i-: '~'i'+" i,' ., ,,'-, -~,',; ,-'+
, ,,!_~,_L .•- -.-+.... ' ".,-~t-J:.t;-, r~W-':-d'F' T'T;'
fffi"-++;-+t-t-'. .i·+=m+~'-Ltt +!.J
D-irri--rl-H-t-r-i-h .+~.- --~-r-l:t-t-: """'1 '-",~.-,-+-
,! 1-' ·d· i-' -t- . f'I~'i;-:!t " " ,_. t+t-;-H-j ,-tt , - , T T!'--,'" - -t'l-': "-,-;- .I-t+- ~-:--!-'-
4~'Trt- -t+ -." 1_'--+ - _.h- -It, '-i-+J-r " ~~ ..~ -,-t--""~-'-i ''', :-l't-:-
:.',]-,::1+ !E'" ' .'
I I I I
-"t.' " L ,'
1-:·:1"+: : ' - -r-7TT Determinative assay - Itf+ 41-, ::.r T , ~
;~~;.~...,..; r:!i '+:rt..J~ Lll'~ -~H +H-' .+H:;'-8-
j T '- ~- -j+ - +,-+-H--fW - i , IT
':- .'--;=rl= -. I i - ' -I,i-t-ritt+ + H- .,. , , -:-r-
!j- " ' - =I-+- . L:- - f-W-:- . ...: +H+t-
r ,: ,
tt '; , ; -t+
'_:': --1- LL
-+~ H-
HI , " B _.j ' .. - - ~
.-
H++ , +~-
ti-L,-<-+-
,,'+ _L.,'-+
tl' t- '
H-
+ . .q*l=i
'- ". :::_~-;:-. 2 . ' 0. L!.+.~$:+FH-W
-- _ -h+ , '
j-t H- 1 T Ll-,' 1- -i-i- ttR-l±t-!-;-,
h -f ':,+ -;- i : ;'-" ,r~I~:· ~r :i1
-+H+ ..l- ~- - - .) ,++-1+++' -4=1: - '-i-j -llit
- ' j - - _. ffi~~ ~-' , " '+ ++ -,~-++
+-'h"';'" L1"
.J.-i-!± -H+ -H+ - +'-1-+--o.::
- - j - --'-4- • - L ... _ to+- r -1- -+-t. --. - +-t ,--I--t-r-- !ll
~+,- ~I-j+ *', , =t -clttlfflu+ r 4--:- -1- ~+tt + =+ =.n -l±r± "1--'+l:I±P+b.·-ri-~--1 Ettr
rH...;-t i..., ' ' r- -r+: ++--l~-l-.' i-! ~-- :H~ " r ; " . ; , - _. -j". ., ' , "
:t~~ + - r1~t~ r:i+l~~ t-i:+++- J-1+= ·:H++-lql++lq.::' l1+~ ·nti= +FI:++ffiFf-t-:'::ffi+-rlf+H:p:++~*~;·++::q· i-

n:f" ", +- - -H H-t++i-H ' "t.E .;tm' ,'r-;-' tl+,-i-L
fl' t- -: .. cf· ~,~1fti" ttrt-i+
-H'::;+I-HI-~ tt± -: ,!.j.4-t-t++-f-jt+Jt+!+f-t+I~ttt~ .. !+It~tH±r ,1:11 -T', '-W-I-
' '_~c j_L '" ""L.1...:-r-+-oo-

-L-i...J..-
"r:·5i+··,-'-j-t-l-t-I-h-t.. B1-+'-l-f+ IJ::J:+i1++ H:a.B,o.
- 1 ~ -+ H . ~~, _q:ttti-t:::tt - t-t"H-F- ~T, ,
@ U+tt
. i '- - -HL l
8, --;tt;,: -. . - 14J::t + -: +t- LLH-'- -H--- - li-ri- H- .+- :-h-H-I-· rt l fJ-l
_', " .. ' '_ 2.. • ,.LLLL .' , L I ' ~ u....u.
ai' -' .b. nTT TT"1 ,,+,. ~
'I" :+~ '" '.....:..J..~LLLLL ._LJ...L . L+ + C.w.. _ .-I-l-I-_I.J._.I...LLLt_L.~iLLLl.! ~++
--!"T,-
t-r- -, : ..:..:::...:--r:-T- _l __-L
,; I , . l~.'-L ~~~I ;.-+..l}.....l...!_~.~~--:-,__ ~
+-+ . . ±. _. '..:-.. ··.·+-i '-:-J.~.-I-1-+-=1 f-t-~+.;.--....... -.-+·-l~--·"";-~-·-.··;""-'-t· -,-,-+ .. !- •• -.-. . .•• , . . . . . -- ••.•.• • ~:-: ," __ . . _
t±ttt~~J. ,+H·-i-'-+t· '- ' ' .L-.. -h·1-t- -r-ri-t+ -!i++-H'

r- '-r'-
+, ,\A. .-i--j-H-' rr 1-+ - ... _f.llI
-H'T-+-
i . .;':'~. ' - '~-
'l-~.~~;' . ,- -t-++
J.~ -t' - --
,+~p+,. J i+ t:H=t+t··! -I+H-H+ ,-H-t-rt-:
, I-
L.LL InJections . , - ! . ! . _ H- -r- - ""i-r+I:t~+I+I-..
-,-rL:... -"'-r-Thi"--',I-H++-, --11"1+,' + ..1 ,
·'··H·'·! h
" '--=t· ~i,"rh"
'1-1--1'+
I . I .' •••• , ••.• ' , . , . . , • I I . , I I
. .; t-' 't..--I'll
1l_:...-V
FIGURE AL2. Chromatographic tracings of (A) standard and (B) plasma sample containing ivermectin.
the syringe and through the cartridge, using the same disposable
pipet.
18. Elute the cartridge with a further 8 to 9 ml of chloroform (aliquots of 3,
3, 2, or 3 mI).
19. Collect all the chloroform, leaving the SEP-PAK from Steps 16, 17,
and 18 (eluent) in a IS-ml centrifuge tube.
20. Evaporate the chloroform completely off with a 60° C bath and
nitrogen purge. The residue must be completely dry.
21. Pipet exactly 0.5 ml of methanol (or other larger volumes for
high-assay plasmas) into the tube and use a vortex mixer and
ultrasonic bath to completely dissolve the residue.
22. Centrifuge and inject SO ILl of sample from each sample and each
standard in turn into the HPLC unit using the WISP Autosampler.
23. Measure the peak heights of the derivative at the retention time
indicated by the standards.
24. Calculate results as described in the tissue method starting at Step 48,
substituting ng/ml for ppb and S ml for G in the equation.
ApPENDIX II
Confirmatory Assay for Assaying

Ivermectin Residues in Liver
G. V. Downing
A. Summary of the Confirmatory Residue Method

B. Reagents
C. Solutions
D. Apparatus
F. Standards for the Confirmatory Assay
G. Procedure for the Confirmatory Method
A. Summary of the Confirmatory Residue Method

The tissue is homogenized with acetone-water and the dihydro B 1a is
extracted with isooctane with the tissue present. Following removal of the
isooctane, solvent-solvent distributions into acetonitrile out of hexane
and into hexane out of acetonitrile-water are performed. Solvent is
removed from aliquots of the resulting solution, and monosaccharide and
aglycone derivatives are formed. Extractions into methylene chloride/
hexane/isobutyl alcohol out of methanol/water are followed by solvent
removal and fluorescence production by heating at 95° C in imidazole
reagent. Extractions are performed out of methanol/water into hexane/
isobutyl alcohol. Solvent is removed and HPLC analysis performed on
Zorbax C18. Confirmation results when peaks are observed at appropriate
levels for H 2B 1a itself and the monosaccharide and aglycone derivatives.
B. Reagents
1. Acetic anhydride-Baker or Fisher, reagent grade.
2. Acetone-Mallinckrodt, Nanograde.
3. Acetonitrile-Fisher, HPLC grade; or Mallinckrodt, Nanograde.
4. Dimethyl Formamide (DMF)-Baker or Fisher, reagent grade.
5. Hexane-Burdick & Jackson, distilled in glass; or Mallinckrodt,
Nanograde.
Extract Into
Add IUIfurIc DIsecIM In G.1 mI
DIeeoMtIn 0.5 mt aliquot Lelstenelst { Methylene chloride Enporste
2.0 mI meIhenoI KId rugenta Hexe.,. rugent and hNt
".ponIte Iodrytubn RTovemlght oHlOhent
IDA IS-100°C 1 hour
and teke thInI .. aliquot
f
I
Add 1 mlMeoH InJect Into

and extract Into Enpcnte DIaoIw In 0.5 mI LCwlth
oHIOIvMt MeoHforLC FIuoreecence __1kIn
{Henne
IDA
~
-~ -
FIGURE All. I. Flow diagram of confirmatory assay. B is a possible stopping point during the assay.
338 G. V. Downing
6. Isobutyl alcohol-Burdick & Jackson, distilled in glass.

7. Isopropanol-Burdick & Jackson, distilled in glass.
8. Methanol-Burdick & Jackson, distilled in glass; or Mallinckrodt,
Nanograde.
9. Methylene Chloride-any grade available.
10. Methylene chloride-Burdick & Jackson, distilled in glass.
11. I-Methylimidazole-99%, Aldrich Chemical Company.
12. Nitrogen-the equivalent of Matheson extra-dry compressed gas.
13. Sylon-CT-Supelco, Inc.
14. 2,2,4-Trimethylpentane-isooctane, Burdick & Jackson, distilled in
glass.
15. Water, doubled distilled-Distilled, deionized water is redistilled in
an all-glass apparatus.
C. Solutions
1. Derivatizing Reagent-Mix 0.2 ml of I-methylimidazole with 0.3 ml of
acetic anhydride and 0.9 ml of DMF .Make up just before use.
2. Acetone/Water-50% (v/v).
3. Methanol/Water Mobile Phase (90/I0)-Make 200 ml of doubly dis-
tilled and filtered water to 2 liters with methanol. Deaerate by slowly
bubbling nitrogen through for 10 minutes. Make other water content
mobile phases correspondingly.
4. 40% Methylene Chloride-Hexane-Isobutyl Alcohol-Fill a 500-ml
stoppered graduate to the 200 mark with methylene chloride. Add
hexane to the 500-ml mark and, then, 20 ml of isobutyl alcohol. Mix.
5. 1% Isobutyl Alcohol in Hexane-Make 5 ml of isobutyl alcohol to 500
ml in a graduated cylinder. Mix.
6. 1% Sulfuric Acid in Isopropanol-Pipet 0.5 ml of sulfuric acid carefully
into about 40 ml of isopropanol in a 50-ml volumetric flask. Mix. Dilute
to the mark with isopropanol and thoroughly mix. Make fresh just
before use.
7. 1% Sulfuric Acid in Methanol-Pipet 0.5 ml of sulfuric acid carefully
into about 40 ml of methanol in a 50-ml volumetric flask. Mix. Dilute to
the mark with methanol and thoroughly mix. Make fresh just be-
fore use.
D. Apparatus
1. Balance-analytical, capable of weighing 1 mg accurately.
2. Balance-capable of weighing 5 g accurately into an approximately
60-g tube.
Appendix II. Assaying Ivermectin Residues in Liver 339
3. Bath, water-variable temperature 40° to 80° C.

4. Bath, oil-95° to 100° C.
5. Centrifuge, IEC Model HN-S-II, with 6-place rotor IEC 958 and 15-
and 50-ml cups. The centrifuge is run at 2000 to 2500 rpm. The
centrifuge used gives ca. 700 to 750 X g maximum centrifugal force
(RCF).
6. Centrifuge tubes, glass, 15- and 50-ml with polyethylene stoppers
to fit.
7. Centrifuge tubes-50-ml polypropylene, Corning 25331 (used only for
storing standard solutions).
8. Centrifuge tubes, 15 ml, silylated approximately once every 2 months
(used only for the derivatization reaction). Tube stoppers must fit
tightly. Fill each tube to the top with Sylon-CT. Let stand 20 minutes.
Immediately and quickly rinse, first thoroughly with toluene and then
with methanol. Fill with methanol. Let stand 20 minutes, rinse
thoroughly with acetone, and dry. All glassware used should
be completely free of all acidic and alkaline residues. (These
tubes should be cleaned by hand by first soaking in methylene
chloride immediately after use and then in detergent for at least sev-
eral hours each, followed by thorough rinsing with hot water, dis-
tilled water, and acetone before thorough drying. Variations in
the washing regimen are not recommended, since some analysts
have had poor results when the standard washing method was not
followed.)
9. Dispensing pipettors-lO, 15, and 20 ml.
10. Gloves-disposable PVC from Fisher Chemical.
11. Freezer-capable of reaching temperatures of -20° C.
12. Graduate cylinder-25 ml.
13. Graduate cylinder-500 ml.
14. Homogenizer, Polytron-Brinkmann Instruments, 27-11-200-5 with
PTA lOS generator 27-22-330-3.
15. Parafilm-American Can Co.
16. Pipets-disposable.
17. Pipets-graduated 1,2,5, and 10 ml.
18. Pipets-volumetric 0.5 ml, 1 ml, 2 ml, 3 ml, 4 ml, and 5 ml.
19. Reciprocating shaker, variable speed-Eberbach, J. T. Baker Catalog
8287-E30 or equivalent.
20. SEP-PAK, silica catridge, Part No. 51900-Waters Associates.
21. Spatula-stainless steel.
22. Syringe-50 ILl and 250 ILL
23. Tape-13 2956, 112 inch from Ace Scientific.
24. Ultrasonic bath-Sonogen Automatic Cleaner-Branson Model 520
or equivalent.
25. Vortex mixer.
340 G.V. Downing
A Beckman-Altex Model 110A liquid chromatographic pump complete
with Valco sample valve with syringe-loading sample loop or Waters
Wisp Autosampler and Kratos-Schoeffel Instruments Model FS950 fluo-
rescence detector with a I-millivolt recorder is used. A 51Lm, 4.6-mm 10,
C18 standard Brownlee Labs guard column (Spheri-5 RP-18 OD-G,
obtained from Rainin Instrument Co., Inc.) is used before the analytical
column. This guard column is replaced monthly unless the pressure
reaches 2000 psi, in which case it is replaced immediately.
Conditions: 15 cm lengthx4.6 mm 10 Zorbax ODS-CI8 column.
Mobile phase-IO% water in methanol (v/v).
Column temperature-43° C.
Flow-1.8 mIl minute (usual pressure 1000 psi, 500 to
2000 psi acceptable).
For FS950:
Excitation Lamp-FSA 110, standard 365 nm.
Standard Kratos Flowcell-FSA 210.
Excitation Filter-FSA 403, 365 nm band filters.
Emission Filter-418 nm.
Sensitivity Range-0.2 A or higher.
Retention Time-14 minutes (or greater with Modifications for I, the
parent compound)
Time Constant-about 6.
F. Standards for the Confirmatory Assay

The objective is to have standards for the parent (I), aglycone (A), and
monosaccharide (M) at approximately the concentration observed in the
determinative assay or at the RM. To achieve this, dilute an aliquot of
standard ivermectin solution in methanol with methanol.
The final concentration of the solution should be 10 times the level to be
detected (±10%). This results from the following equation:
C=5xL
where: C= the weight of ivermectin in 2 ml of the diluted standard.
5= number of grams of liver sample used.
L= level of analyst wishes to determine in ppb or ng/g.
Because the unknowns are dissolved in 2 ml before aliquoting into I, M,

and A, the standard is figured on the basis of 2 ml. Two 0.5 ml aliquots of
the diluted standard solution should be run through each part of the assay
(M, A, and I). The M and A standards should start in the procedure at the
Step 4 evaporation. The I standard should start at Step 15.
G. Procedure for the Confirmatory Method

1. Follow the method described in the determinative assay procedure up
through Step 28.
2. Take the residue up in exactly 2 ml of methanol, using vortex mixer
and ultrasonic bath.
3. Mix thoroughly and centrifuge for 5 minutes.
4. Pipet exactly 0.5 ml of the solution from Step 3 into each of 2 fresh
15-ml tubes and blow completely to dryness with nitrogen in a 70° C
bath. The dried samples generally look like a small drop of oil. All
solvent must be completely removed. Store the remaining 1 ml from
Step 3 in a freezer until later.
5. Add 0.1 ml of 1% sulfuric acid in methanol to 1 ofthe 2 samples (the A
sample) and 0.1 ml of 1% sulfuric acid in isopropanol to the second
(the M sample).
6. Vortex mix the sample for 10 seconds, thoroughly ultrasound it, and
repeat the vortex mixing.
7. Let stand at room temperature for 16-18 hours (overnight in the
dark).
8. To all (A) samples and (A) standards at one time, add 0.9 ml of
methanol, using a 5-ml graduated pipet.
9. Add the 40% methylene chloride-hexane-isobutyl alcohol solvent
mixture to the 7-ml mark and mix.
10. Add 4 ml of distilled water, using a lO-ml graduated pipet, shake for 1
minute, and centrifuge for 5 minutes.
11. Move the mixed solvent upper phase by disposable pipet to a fresh
silylated 15-ml tube. Move as much upper phase as possible but
absolutely no lower phase.
12. Repeate the extraction of the lower phase with a second 6 ml of the
mixed solvent and combine the extracts in the silylated tube.
13. Do the same extractions with the (M) samples, except that the shaking
must be only moderate and the centrifuging should be for 10 or more
minutes (otherwise emulsions may result).
14. Into a third 15-ml silylated tube, pipet another 0.5 ml of the sample
from the freezer stored in Step 4 (the I sample).
15. Blow the (M) and (A) extracts and the (I) sample completely to
dryness. Again, all solvent must be removed and no moisture picked
up. A bath temperature of 40° C is suitable for the (M) and (A) samples
and temperatures up to 70° C for the (I) sample.
16. Add 0.1 ml of derivatizing reagent to all (M), (A), and (I) dry tubes and
all standards, vortex mix 10 seconds, tape the stoppers in place,
centrifuge briefly, and place in a 95-100° C oil bath for 1 hour, all at
the same time.
17. Cool to room temperature, add 0.9 ml of methanol using a 5-ml
graduated pipet, and mix.
18. Add 1% isobutyl alcohol in hexane to the 7-ml mark and mix.
19. Add 4 ml of water from a graduated pipet, shake 1 minute, and
centrifuge for 5 minutes.
20. Move the upper phase as completely as possible to a fresh 15-ml tube.
Move no lower phase.
21. Repeate the extraction with a second 6 ml and combine extracts.
22. Blow completely to dryness with a nitrogen flush and a 40° C bath. At
the end of this evaporation, the temperature can be allowed to rise to
60-70° C.
23. Take up residue in exactly 0.5 ml of methanol, using vortex mixing
and ultrasound. In all above (17-23), the (M) samples should be
handled first and placed on the LC before handling the (A) or (I)
samples. After extraction, protect the (M) samples from light as much
as possible. (If the (A) and (I) samples are to be delayed signifcantly,
they should be stored in freezer at - 20° C before adding the
methanol.)
24. Centrifuge and inject 50 ILl of the clear phase (M samples) onto the
HPLC with the modified mobile phase doing the first standard of a
type, the unknowns, and finally the second standard, in that order.
25. Do the same with the (A) and (I) samples.
26. Examine the unknown sample chromatograms for the presence of the
(A), (M), and (I) peaks at the same elution time as the standard peak.
(That is, compare the unknown (A) to standard (A), etc.)
27. Average the 2 peak heights for each standard (A, M, and I) and
calculate each unknown as a precent of that standard average.
28. A value of 60% or more of all 3 (A, M, and I) is proof that that
particular level of ivermectin is present in the tissue (liver).
ApPENDIX III
List of Registrations
J. Di Netta
Ivermectin was released for registration by Merck & Co., Inc., in 1981
and first registered in France as IVOMEC Injection for Cattle that same
year.
Since that time ivermectin has been approved for use in over 60 countries.
It is currently registered for use in cattle, sheep, horses, goats, swine,
dogs, camels, reindeer, and bison.
The following lists the products and the countries in which ivermectin is
registered.
IVERMECTIN APPROVALS
IVOMEC' byection for Cattle
ALGERIA GUATEMALA PHILIPPINES
ARGENTINA HOLLAND POLAND
AUSTRALIN HONDURAS PORTUGAL
AUSTRIA HUNGARY ROMANIA
BELGIUM INDIA S. AFRICA
BRAZIL IRELAND SPAIN
BULGARIA ISRAEL SUDAN
CANADA ITALY SWEDEN
CHILE JAMAICA SWITZERLAND
COLOMBIA JAPAN THAILAND
COSTA RICA JORDAN TRINIDAD
C.S.S.R KENYA TUNISIA
DENMARK LUXEMBOURG TURKEY
ECUADOR MALAYSIA U.K.
EGYPT MEXICO URUGUAY
ELSALVADOR MOROCCO U.S.A.
FINLAND NEW ZEALAND U.S.S.R.
FRANCE NORWAY VENEZUELA
G.D.R. PANAMA YUGOSLAVIA
GERMANY PARAGUAY ZIMBABWE
GREECE PERU
IVOMEC' Oral Solution for Cattle
FRANCE NEW ZEALAND U.K.
HOLLAND SOUTHAFRICA
Appendix III. List of Registrations 345
IVOMEC' Paste for Cattle

CANADA U.S.A.
IVOMEC' Pour-On for Cattle
ARGENTINA HOLLAND NEW ZEALAND
BRAZIL MEXICO U.K.
CANADA
IVOMEC-F' Injection for Cattle
BRAZIL IRELAND SPAIN
FRANCE MEXICO U.K.
HOLLAND PORTUGAL
IVOMEC' or ORAMEC' Liquid for Goats
BRAZIL NEW ZEALAND U.K.
HOLLAND SOUTH AFRICA
IVOMEC' or ORAMEC' Liquid for Sheep
ARGENTINA HOLLAND MOROCCO
AUSTRALIA INDIAN NEW ZEALAND
BELGIUM IRELAND SOUTH AFRICA
BRAZIL JORDAN SUDAN
CANADA KENYA U.K.
FRANCE LUXEMBOURG U.S.A.
URUGUAY
ZIMBABWE
IVOMEC' Injection for Sheep
ALGERIA DENMARK POLAND
ARGENTINA FRANCE PORTUGAL
BELGIUM HOLLAND ROMANIA
BRAZIL JORDAN SOUTH AFRICA
BULGARIA LUXEMBOURG SPAIN
C.S.S.R. MEXICO SUDAN
CANADA MOROCCO TUNISIA
CHINA PARAGUAY URUGUAY
YUGOSLAVIA
IVOMEC' Injection for Swine
AGRENTINA GERMANY PORTUGAL
AUSTRIA HOLLAND ROMANIA
BELGIUM HONG KONG SINGAPORE
BRAZIL IRELAND SOUTH AFRICA
BULGARIA ITALY SPAIN
CANADA JAPAN SWEDEN
CHILE KOREA SWITZERLAND
CHINA LUXEMBOURG TAIWAN
COLOMBIA MALAYSIA THAILAND
C.S.S.R. MEXICO U.K.
DENMARK NEW ZEALAND U.S.A.
ECUADOR NORWAY URUGUAY
FINLAND PARAGUAY VENEZUELA
FRANCE PHILIPPINES YUGOSLAVIA
G.D.R. POLAND ZIMBABWE
IVOMEC' Injection for Young Pigs
HOLLAND U.S.A.
346 J. Di Netta
HEARTGARD-30' or CARDOMEC' or CARDOTEK-30' Tablets for Dogs

AUSTRALIA HOLLAND SPAIN
CANADA JAPAN U.S.A.
FRANCE MEXICO
EQV ALAN' Paste for Horses
ALGERIA HOLLAND PARAGUAY
ARGENTINA HONDURAS PERU
BELGIUM HONG KONG PHILIPPINES
BRAZIL IRELAND POLAND
CANADA ITALY PORTUGAL
CHILE JAMAICA SINGAPORE
COLOMBIA JAPAN SOUTH AFRICA
COSTA RICA JORDAN SPAIN
DENMARK KENYA SUDAN
ECUADOR KOREA SWEDEN
EGYPT LEBANON SWITZERLAND
ELSALVADOR LUXEMBOURG THAILAND
FINLAND MEXICO TRINIDAD
FRANCE MOROCCO U.K.
G.D.R. NEW ZEALAND U.S.A.
GERMANY NORWAY U.S.S.R.
GUATEMALA PANAMA URUGUAY
VENEZUELA
EQVALAN' Liquid for Horses
CANADA MEXICO U.S.A
HOLLAND
IVOMEC' Injection for Camels
ALGERIA MOROCCO SUDAN
JORDAN
1 AVOMEC
'Trademark of MERCK & CO., Inc., Rahway, New Jersey, U.S.A.
Index
A AGRI-MEK, 288
Abamectin Airblast treatment, 202, 203
antibacterial activity, 29 (Table 2.3) See also Worker exposure.
antifungal activity, 30 (Table 2.4) Alaska, 271
biological activity, 20 (Table 1.8) Algae, toxicity, 168
clinical trials, cattle, 216-222, Algeria, 268, 344, 346
230-233 Alkylating agents, use in mutational
compared to ivermectin, 102-103 procedures, 35-36
toxicity, 103 (Table 6.5), 110 Alkylations, 15
crop protection, 288-305 Almonds, 305
environmental effects, 184-198 Alpacas, clinical trials, 271
photoinstability, 292 AI Salvador, 344
structure of, 90 (Fig. 6.1), 184 (Fig. Amblyomma americanum, 265-266
13.1) Amidine,219
toxicity studies, 102-107 Amitraz, 254
worker exposure study, 201-213 Analytical techniques, tissue residues,
See also Specific topics. 137-138, 144-145
Acarines, 288 Ancylostoma caninum, 251-252
Acetylcholine release, ivermectin, 79 Ancylostoma spp., 255, 269, 320
Acid-sensitivity, avermectins, 8 Angiostrongylus cantonensis, 265
Aculops lycopersici, 302 Anguillicola crassus, 279
Acute toxicity Anorexia, 96, 108
abamectin, 193-194 (Table 13.4) Antelope, clinical trials, 273-274
ivermectin, 92-95, 169 Anthelmintic market, 24
Aelurostrongylus abstrusus, 255 Anthelmycin, 24
AFFIRM, 289-290 Anthelvencin, 24
nonagricultural land, 289 Antifungal/antibacterial activity
Africa, 275 abamectin, 29-30 (Tables 2.3, 2.4)
Agitators, and fermentation, 47-48, Streptomyces auermitilis, 28-31
49-50 (Figs. 3.7,3.8,3.9,3.10) Aphids, 298, 300
Aglycones Aphodius, 178
avermectin, 10-12, II (Fig. 1.5), 16 Aquatic life. See Environmental
ivermectin, 12 effects of abamectin
348 Index
Argentina, 344, 345, 346 I H N MR data, 5 (Table I. I)

Armyworm, 299, 302 isolation of, 26
Ascaris Ilimbricoide.l', 74 macrocyclic lactones, I, 10
Ascaris suum, 76 mass fragmentation, 4 (Fig. 1.2)
Asia, 275 mode of action
Aspiculomycin, 24 acetylcholine release, 79
A.I'piculliris tetraptera, 262 AVM binding sites, 76-78, 80-81
Aspis cracciuora, 300 AVM effect on benzodiazepine
Aspis gossypii, 300 binding sites, 82-84
Assay method AVM effect on [3Hl GABA
apparatus, 326-327, 338-339 binding, 81-82
chromatographic apparatus, AVM stimulated neurotransmitter
327-328, 340 release, 80
determinative assay of tissues, chitin synthesis inhibition, 79-80
329-332,341-343 chloride uptake, 78-79, 84-85
determinative assay standards, 328, difficulties related to, 73-74
340 electrophysiological studies,
fat, 329, 330 74-76
liver/muscle/kidney tissues, 329, GABA binding sites, 77, 81
330 glycine binding sites, 85
plasma, 333-335 in invertebrates, 74-80
reagents, 326, 336, 338 protein kinase C activity, 85
solutions, 326, 328 retinol binding proteins,
stopping places, 332-333 interaction, 79
Ataxia, 93, 94, 96, 108, 150, 151, 152, in vertebrates, 80-85
153, 155, 157, 158, 159 most important, Bla, 3
Australia, 175, 177,345,346 photoisomerization, 9 (Fig. 1.4)
Austria, 344, 345 structure of, 55, 56 (Fig. 4.1)
Auxotropic strains, 36 naturally occurring avermectins,
A vermectins 1-3
antifungal activity, 31 Avermectin aglycone
chemical properties, 7-8 glycosyltransferase, studies with,
acid-sensitivity, 8 66,70
stability studies, 7 Avermectin B I a. See Abamectin
chemistry Avermectin B O-methyltransferase,
alkylations, 15 studies with, 66, 67-68 (Figs. 4.5,
glycoside syntheses, 15 4.9)
hydroxy group characteristics, Avermectin 5-ketoreductase, studies
9-10, 15 with, 66, 69-70 (Figs. 4.6, 4.7)
interconversions, 15 AVID, 288, 294, 295
monosaccharides and aglycones, AVOMEC, cattle, 230-233
10-\2 Axenomycins, 24
oxidations, 12-13
radiolabeled derivatives, 13, 15
reductions including ivermectin, B
13 Bancroftian filariasis, 320
l3C NMR data, 6-7 (Table 1.2) Beans, 298
discovery of, 24 Bees, toxicity, 195
dosage, 10 Beet armyworm, 302
Index 349
Beetles, dung, effect of ivermectin, [methyl-14C] methionine,

176-178 60 (Table 4.4)
Belgium, 344, 345, 346 results of studies, 62
Benomyl,203 valine, isoleucine, isobutyrate,
Benzimidazoles, 222 57, 60 (Table 4.2)
Benzodiazepine, binding sites, proposed pathway, 71 (Scheme I),
ivermectin, 82-84 72
Benzyl alcohol, 118, 122 specific inhibitors. sinefungin, 70,
Bicuculline, 78, 80, 81 72
Biexponential decay Birds, clinical trials
cattle, 114, 115 (Fig. 7.1), 116 Ascaridia. 276-277
(Fig. 7.2) Capillaria spp .• 277
sheep, 118-119 (Fig. 7.5) mites, 277-278
Bighorn sheep, clinical trials, 273 Oxyspirura sp .• 277
Bioavailability analyses Pelecitus sp .• 277
cattle, 114-115 (Fig. 7.1),117 toxicity, 195-196
(Fig. 7.3), 118 (Fig. 7.4) Bison, clinical trials, 274
dogs, 122-123 (Fig. 7.6) Blackeye peas, 301
horses, 125-126 (Fig. 7.8) Blood, 135, 136 (Table 8.3)
humans, 127-128 Bollworm, 300
sheep, 121 Boophilus decoloratus. 219
swine, 124-125 (Fig. 7.7) Boophilus microplus. 219,220,221,
Biosynthesis 231,232
bioconversion of intermediates, Bordetella hronchiseptica. 28
62-65 Bots, 223
desfurano avermectins, feeding Brain, 135
of, 63, 65 (Table 4.8) Brazil, 344, 345, 346
radioactivity distribution after Brine shrimp, 80
feeding, 64 (Table 4.8) Broad mite, 291
results of studies, 65 Brugia malayi. 276
enzyme studies, 66-70 Brugia pahangi. 264, 267, 276
avermectin aglycone Budgies, 278
glycosyltransferase, 66, 70 Buffalo, clinical trials, 274
avermectin B Buffalo flies, 174
O-methyltransferase, 66, Bulgaria, 344, 345
67-68 (Figs. 4.5, 4.9) Bush fly, 174, 175
avermectin 5-ketoreductase, 66,
69-70 (Figs. 4.6,4.7) C
incorporation of labeled precursors, C-076,25
56-62 Caenorhahditis elegans. 74, 76-77
[1_ 18°2, I-Bc] acetate and California, 213
propionate, 57, 61 (Table Caligus elongatus. 279
4.6),62 Camels, clinical trials
[14C] methionine, 60 (Table 4.3) ectoparasites, 270-271
[ 13 C] precursors, 57, 58-59 (Table endoparasites, 270
4.1) Cameroon, 315
labeled glucose, 62 (Table 4.7) Canada, 344, 345, 346
[methyl-13C] methionine, Canaries, 278
61 (Table 4.5) Cape hunting dog, clinical trials, 269
350 Index
Capillaria spp., 253, 256 Cattle feedlot environmental fate

Captan, 203 study, 171
Carbamates, 202 Celery, abamectin, 303
Carbaryl, 203 Cell mass on-line, prediction of,
Carnation foliage, 288, 295, 296, 297 45-47
Catabolite repressible, avermectin Cephalopina titillator. 271
production, 43 Cerelose, in seed media development,
Cats, clinical trials 38
Aelurostrongylus abstrusus. 255 Chabertia ovina. 231
Ancylostoma spp., 255 Chelonians, 279
Capillaria spp., 256 Cheyletiella blakei. 269
Ctenocephalides cati. 256 Chickens, 277
Nodoedres cati. 256 Chile, 344, 345, 346
Otodectes cynotis. 256 Chimpanzees. clinical studies, 318
Toxocara cati. 255 China, 345
Cattle Chitin synthesis inhibition,
clinical trials, 216-222, 230-233 ivermectin, 79-80
dosage/administration, 230 Chiarella pyrenoidosa. 168
ectoparasite activity, 232-233 Chloride
endoparasite activity, 216, 219, ions, 74, 76, 81, 82. 84, 86
220-221,231-232 uptake, ivermectin, 78-79, 84-85
grubs, 217, 219-220, 221 Chlorobenzilate, 203
lice, 218, 220. 221 Chorioptes bovis. 218, 221
mites, 218, 219 Chromatographic methods, 144-145
oral formulation, 219 Chrysanthemum foliage, 203,
screwworms, 219 208-209 (Table 14.4), 210-213,
sustained-release bolus, 221-222 294-295, 296, 301
ticks, 219 Chrysomya bezziana. 219
topical formulation, 220 Citrus crops, abamectin, 290-294
fat metabolism studies, 139-141 and beneficial organisms, 293-294
liver metabolism studies, 135-139 broad mite, 291
pharmacokinetics of ivermectin, citrus red mite, 291-292
114-119 citrus rust mite, 290-291
biexponential decay, 114, 115 citrus thrips, 293
(Fig. 7.1),116 (Fig. 7.2) decomposition of abamectin, 293
bioavailability analyses, 114-115 Cleft palate, 99, 100, 103, 109
(Fig. 7.1), 117 Clubbed forepaws, 100, 109
(Fig. 7.3); 118 (Fig. 7.4) Cnemidocoptes pilae. 278
dosage formulations and results, Cochliomyia hominivorax. 219
117-119 Cockroach, 76, 78-79
high-pressure liquid Coleoptera, 178, 179
chromatographic (HPLC) Collies, 246
method, 114 Colombia, 344, 345, 346
metabolic transformation, 114 Coma, 99
plasma concentration, 116 Convulsions, 91, 99, 105
(Fig. 7.2), 117, 118, Cooperia oncophora. 216
119 (Table 7.1) Cooperia punctata. 216
safety of ivermectin, 151-153 Copepods, 278, 279
tissue residues, 132-134 Cornea, microfilariae, 313
Index 351
Corn earworm, 299 Dictyocaulus viviparus. 216,230,231,

Costa Rica, 344, 346 232,272
Cote d'ivore, 313 Diethylcarbamazine (DEC), 159
Cotton, abamectin, 297-300 compared to ivermectin, 311-316,
aphids, 298, 300 319
lepidoptera, 299-300 Diglycerides, 52
spider mites, 297-299 22,23, dihydroavermectin Bl, 16,20
Coumaphos, 178 22,23, dihydroavermectin Bla, 3
Crenosoma striatum, 268 Dihydropicrotoxinin, 79
Crop protection Dipetalonema reconditum. 250
citrus crops, 290-294 Dipetalonema viteae. 264
cotton, 297-301 Diptera, 175, 178, 179
nut trees, 305 Dirofilaria immitis. 169, 246-250, 264,
ornamental plants, 294-297 267
pears, 303-304 See also Heartworm disease, dogs.
products used, 288-289 Dislodgeable foliar residues, worker
red imported fire ant, 289-290 exposure, 202, 209-213
strawberries, 305 DNA synthesis, unscheduled, 92
vegetables, 301-303 Dogs
Cross-feeding, mutational procedures, adverse reactions, ivermectin, 246
36-37 clinical trials
C.S.S.R., 344, 345 Ancylostoma caninum. 251-252
Ctenocephalides cati. 254-255, 256 Capillaria spp., 253
Cuterebra fontinella. mice, 265 Ctenocephalides cati. 254-255
Cyclophosphorothionate, 83-84 Demodex canis. 254
Cytodites nudus. 278 Dipetalonema reconditum. 250
Dirofilaria immitis. 246-250
D dosage/formulation, 246
Dahlia foliage, 296 Filaroides osleri. 253
Daisy foliage, 295 Otodectes cynotis. 254
Damalinia bovis. 218, 221 Sarcoptes scabiei. 253-254
Daphnia magna. 169, 170 (Table Strongyloides stercoralis. 253
11.6),171 Toxascaris leonina. 251
DDT, 219 Toxocara canis. 250-251
Deer, clinical trials, 271-272 Trichuris vulpis. 253
ectoparasites, 272-273 Uncinaria stenocephala. 252-253
endoparasites, 272 pharmacokinetics of ivermectin,
Demodex canis. 254 122-124
Denmark, 344, 345, 346 bioavailability analyses,
Dermal penetration study, abamectin, 122-123 (Fig. 7.6)
107, 108 (Table 6.8) plasma concentration, 120
Dermanyssus gallinae. 277 (Fig. 7.5), 122
Dermatobia hominis. 217, 221 tablet fQrmulation, 122-124
Destomycins, 24 safety of ivermectin, 157-159
Developmental toxicity, 98-102, 109, toxicity, 157-158
104 (Table 6.6) abamectin, 104
Dichlorvos, 178 ivermectin, 93-94, 97, 108
Dicofol, 297 See also Heartworm disease.
Dictyocaulus arnfieldi. 239 Donkeys, 239, 241
352 Index
Dopamine, 80 fate of avermectin in terrestrial

Dosage formulations and results environment, 190-191
minimum toxic dose (MTD), 94, 108 mammals, 196
no-effect level (NEL), 110 microorganisms in soil, 195
See also Specific species. plant sensitivity, 194
Downstream extraction, fermentation plant uptake, 186-187
development, 45, 52-53 risk assessment, 198-199
Draschia spp., 240 thin film degradation, 185
water photolysis studies, 184-185,
197
E Environmental effects of ivermectin
Earthworms, 168, 174 cattle feedlot environmental fate
Ecinothrips americanus, 2% study, 171
Ecuador, 344, 345, 346 dung fauna, 173-180
Edema, 150 cattle dung breeding diptera,
Edesonfilaria malayensis, 276 174-176
Edible tissues, 147 native insects, 178
See also Tissue residues, edible Onthophagus binodis, 177-178
tissues. Onthophagus gazella,
Eels, 279 176-177 (Table 12.2)
Egypt, 344, 346 role in ecosystem, 173-174
Eisenia foetida, 168 treated/untreated aged manure,
Elaphostrongylus cervi, 272 178-179
Electrophysiological studies, environmental burden,
ivermectin, 74-76 163-165 (Table 11.2)
Elephants, clinical trials, 275 environmental fate, ivermectin,
El Salvador, 346 166 (Table 11.3)
Emesis, 91, 94, 108 freshwater organisms, acute
Environmental considerations, toxicity, 169 (Table 11.5)
fermentation development, 51 safety assessment, 169-170
Environmental effects of abamectin soil
aquatic life photodegradation of ivermectin,
acute toxicity, 192-193 (Tables 167
13.1, 13.2) soil binding, 166-167, 169
chronic toxicity, 193-194 (Table toxicity to earthworms, 168
13.4) toxicity to soil microorganisms,
fate of avermectin in aquatic 167
environment, 188-189 Enzyme studies, 66-70
fish, 187-188, 193 avermectin aglycone
invertebrates, 192-193 glycosyltransferase, 66, 70
risk assessment, 197-198 avermectin B O-methyltransferase,
hydrolysis, 185 66, 67-68 (Figs. 4.5, 4.9)
soil binding, 186 avermectin 5-ketoreductase, 66,
soil metabolism, 185 69-70 (Figs. 4.6, 4.7)
soil photolysis studies, 185 Epitrimerus pyri, 304
solubility, 186 Equus asinus, 241
terrestrial life EQVALAN,158
bees, 195 injectable, 235
birds, 195-196 liquid, 234
Index 353
paste, 234 survival curves, 34-35

See also Horses, clinical trials. suspensions used, 34, 35-36
Etazolate, 81 ultraviolet (UV) light, 35
Ethylmethane sulfonate (EMS), in oxygen transfer, 47-48, 49-50
mutational procedures, 35-36 (Figs. 3.7, 3.8, 3.9, 3.10), 51-52
Euseius tularensis. 293 (Fig. 3-11)
Excretion production media development,
ivermectin, animals. 134, 163 40-41 (Table 3.4), 42-44
residues of ivermectin, 164 (Table classical approach, 40-41
11.1) medium constituent substitutions,
See also Environmental effects of 41
ivermectin. response surface methodology,
Exotic mammals. See specific species 41-43
Eye shake flash fermentation, 43
corneal microfilariae, 313 (Fig. 3-1)
ocular reaction index, 317 (Fig. scale-up strategy, 48, 51-52
21.3) (Fig. 3.11)
seed media development,
F 37-38 (Table 3.3), 39-40
Face fly, 174, 175 change in high producers, 39
Fall armyworm, 302 improved avermectin production,
Fat, 135 38 (Table 3.3), 38-39
fat pad, 136 (Table 8.3) seed medium IV, use of, 39-40
metabolism of ivermectin, in vivo stirred tanks, 44-45
systems, 139-141 viscosity profiles, 48 (Fig. 3.6)
Feces Ferns, 288
excretion of ivermectin, 134, 163, Ferrets, clinical trials, 267
164 (Table 11.1) Filarioidea. 79
See also Environmental effects of Filaroides osleri. 253
ivermectin. Finland, 344, 345, 346
Fenbendazole, 253 Fish, clinical trials, 278-279
Fenvalerate, 300 copepods, 279
Fermentation development nematodes, 279
agitators and, 47-48, 49-50 toxicity, 169 (Table 11.5), 170,
(Figs. 3.7, 3.8, 3.9, 3.10) 187-188, 193-194
Bagner/Wildman process, 52-53 Fleas, 254-255, 256
cell mass on-line, prediction of, Flour beetle, 288
45-47 Flukes, 256
components of harvested broth, 52 Flunitrazepam, 83
downstream extraction, 45, 52-53 F1uorescence-derivatization method,
environmental considerations, 51 114,121,124,137
media sterilization, 51 Fly dung, effect of ivermectin,
mutational procedures, 34-37 174-176
alkylating agents, 35-36 Fly strike, 272
auxotropic strains, 36 Foxes, clinical trials
cross-feeding, 36-37 ectoparasites, 269
detection of mutants, 36 endoparasites, 268-269
most avermectin produced, 37 France, 254, 344, 345, 346
mutants, 35 (Table 3.1), 36 Frankiniella occidentalis. 296
354 Index
Free residues, animal tissues, 132 Haematopinus suis. 226

Friedman's test, 237 Haematopinus tuberculatus. 274
Haemonchus contortus. 74, 222
G avermectin activity, 16 (Table 1.3),
GABA 19 (Table 1.6)
agonist, 84, 89, 91 Haemonchus placei. 216,231,232
binding sites, 77, 81 Hamsters, clinical trials, 266
functions of, 91 Heart, 135, 136 (Table 8.3)
Gastrophilus intestinalis. 241 HEARTGARD-30, 122, 256
Gastrophilus nasalis. 241 Heartworm disease, dogs, 122, 159,
G.D.R., 344, 345, 346 246
Genotoxicity studies, 91-92 ivermectin efficacy
Gerbils adult worms, 248
avermectin activity, 16, dosage, 249 (Table 18.2)
17-18 (Tables 1.4, 1.5) immature stages, 246-248
clinical trials, 266 microfilariae, 248-250
Germany, 344, 345, 346 prophylactic efficacy, 247 (Table
Ghana, 312, 313 18.1)
Glossina palpalis. 266, 267 See also Dogs.
Glutamate, 80 Hedgehogs, clinical trials, 268
Glycene binding sites, ivermectin, 85 Heliothis virescens. 299, 300
Glycerol formal, 114, 117, 118, 119, Heliothis zea, 299, 300
122, 124 Hemiptera triatominae, 264
Glycine, 85 Hens, 277
Glycoside syntheses, avermectins, 15 Hexachlorocyclohexane, 79
Goats, clinical trials, 224-225, 273 Hickory shuckworm, 305
injectable formulation, 225 High performance liquid
oral formulation, 224-225 chromatography (HPLC), 7, 10,
safety of ivermectin, 155 26, 114
toxicity, 155 isolation of avermectins, 26
Goldfish, 279 Holland, 213, 344, 345, 346
Greece, 344 Homeopronematus anconai. 302
Grubs, ivermectin efficacy, cattle, Honduras, 344, 346
219-220, 221 Honey bees, 77
Guatemala, 313, 315, 344, 346 Honeydew, 304
Guinea fowl, 276-277 Hong Kong, 345, 346
Guinea pigs, clinical trials Hookworms
mange mites, 266 cats, 255
ticks, 265-266 dogs, 251-253
tsetse flies, 266 human, 320-321
Gymnodia, 178 Horn fly, 174, 175, 179
Gypsophilia foliage, 295 Horses
clinical trials
H Dictyocaulus arnfieldi. 239
Habronema spp., 240 dosage/formulation, 234-235
Haematobia irritans. 174,221 Draschia spp., 240
Haematomyzus elephantis. 275 Gastrophilus intestinalis. 241
Haematopinus eurysternus. 218, 220, Gastrophilus nasalis, 241
221 Habronema spp., 240
Index 355
Onchocerca cervicalis, 240 Invertebrates

Parascaris equorum, 235-237 aquatic toxicity, 192-193
Sarcoptes scahiei, 241 ivermectin mode of action, 74-80
Strongylus edentatus, 238-239 Ireland, 344, 345, 346
Strongylus equinus, 239 Isooctane extract, preparation of,
Strongylus vulgaris, 237-238 145
pharmacokinetics of ivermectin, Israel, 344
124-125 Italy, 254, 344, 345, 346
bioavailability analyses, Ivermectin, 3
125-126 (Fig. 7.8) clinical trials
dosage formulations and results, birds, 276-278
125-126 carnivores, 268-269
safety of ivermectin, 149-151 cats, 255-256
toxicity, 151-152, 153 cattle, 216-222
Humans dogs, 246-255
bancroft ian filariasis, 319, 320 elephants, 275
Loa loa, 320 ferrets, 267
nematodes, 320-321 fish, 278-279
onchocerciasis, 311-316, 320 gerbils, 266
pharmacokinetics of ivermectin, goats, 224-225
126-129 guinea pigs, 265-266
bioavailability analyses, 127-128 hamsters, 266
dosage formulations and results, hedgehogs, 268
129 horses, 234-241
plasma concentration, mice, 262-265
127-128 (Figs. 7.10, 7.11) primates (nonhuman), 275-276
See also Onchocerciasis. rabbits, 266-267
Hungary, 344 rats, 265
Hydrofoil impeller, 47 reptiles, 279-280
Hydrolysis, abamectin, 185 rodents, 268
Hydroxy group characteristics, 9- 10, sheep, 222-224
15 swine, 225-227
Hygromycin B, 24 ungulates, 269-275
Hypoderma hovis, 217,219-220,221, l3C NMR data, 6-7 (Table 1.2)
274 derivatives, chemical structures,
Hypoderma lineatum, 217, 220, 221 137 (Fig. 8.2)
Hypoderma spp., 152 enyironmental effects, 163-179
Hypotension, 319 I H NMR data, 5 (Table 1.1)
human exposure, acceptable levels,

110
pharmacokinetic properties,
Identification/confirmation method, 113-129
147 reductions, 13
Impala, clinical trials, 273 registrations, listing of, 344-346
India, 319, 344 safety in, animal studies, 149-159
Inhibitors synthesis, by selective
of chitin synthesis, 79-80 hydrogenation of avermectin
sinefungin, 70, 72 B), 14 (Fig. I)
Interconversions, avermectins, 15 toxicity studies, 91-102
356 Index
Ivermectin (cont.) Litomosoides carinii. 264

tritiated derivative, 13, 15 Liver, 135, 136 (Table 8.3), 237, 238,
See also Individual topics. 263
IVOMEC, 159 metabolism of ivermectin
cattle, 216-222 in vitro systems, 136
goats, 224-225 in vivo systems, 137-139
sheep, 222-224 Llamas, clinical trials, 271
swine, 225-227 Loa loa, 320
Ivory coast, 318 Lobster, 74-75
Locust, 75
J Lucilia cuprina. 288
Jamaica, 344, 346 Lung, 135, 136 (Table 8.3), 237, 238,
Japan, 344, 345, 346 263
Jirds, clinical trials, 267 Lungworm, 219, 220, 222, 223, 225,
Jordan, 268, 344, 345, 346 272,273
bighorn sheep, 273
K deer, 272
Keiferia lycopersicella. 302-303 Luxembourg, 344, 345, 346
Kenya, 344, 346 Lymphoma assay, 91
Kidney, 135, 136 (Table 8.3)
Kidney worms, 225, 226 M
Kitasato Institute, 24 MA-4680, electron micrographs of,
Konigs-Knorr procedure, 15 25-26 (Figs. 2.1, 2.2)
Korea, 345, 346 MA-5856, 28-30
Macaca mulatta, 93
L Macaw, 277, 278
Lactation, 101-102 Macrocyclic lactones, I, 10
concentration of ivermectin, Malaysia, 344, 345
102 (Fig. 6.3) Mali, 312, 313
toxicity studies, 97, 101 Mange mites, 269, 269-270
Leafminer, 295, 301-302, 303 guinea pigs, 266
Lebanon, 346 rabbits, 266-267
Lepeopthirus salmonis, 279 Mansonella perstans, 320
Lepidoptera, 299-300 Manure, aged, effect of ivermectin,
Lepomis macrochirus, 169 178-179
Lernea spp., 279 Mass spectrometry/mass
Levamisole, 177, 178,222 spectrometry (MS/MS), 10, 145,
Liberia, 302, 313, 315, 318 147
Lice Maternotoxicity, 99-100, 104 (Table
cattle, 218, 220, 221 6.6), 109, 206, 207 (Table 14.3),
wolves, 269 210, 211, 213 (Table 14.6)
Limbitis, 313 Mazzotti reaction, 313, 314
Linear regression analysis, 201 Media sterilization, fermentation
Linognathus a/ricanus, 272 development, 51
Linognathus vituti, 218,220,221,230, Melanin, 36
231 Meningitis, 318
Lipophilicity, 114 Metabolism of ivermectin
Liriomyza sativae, 301 fat tissue, in vivo systems, 139-141
Liriomyza trifolii. 295-297, 301 liver
Index 357
in vitro systems, 136 Morantel, 222

in vivo systems, 137-139 Morocco, 268, 344, 345, 346
tissue residues, 132-136 Mosquitoes, 246
Metaseiulus occidentalis, 305 Mouse assay, 26
Methylation, avermectins, 15 Musca autumnalis, 174
Mexico, 344, 345, 346 Muscimol, 77, 79
Mice, clinical trials Muscle, 135, 136 (Table 8.3)
Aspiculuris tetraptera, 262 Mutational procedures, 34-37
Cuterebra fontinella, 265 alkylating agents, use of, 35-36
Hematospiroides dubius, 262 auxotropic strains, 36
microfilariae, 263-264 cross-feeding, 36-37
mites, 264 most avermectin produced, 37
Strongyloides ratti, 262 mutants
Strongyloides stercora lis , 263 morphological variants, 36
Syphacia obvelata, 263 types of, 35 (Table 3.1)
Toxocara canis, 263 survival curves, 34-35
triatomid bugs, 264 suspensions used, 34, 35-36
Trichinella spp., 263 ultraviolet (UV) light, use of, 35
Trichuris muris, 263 Myiasis, 271
Micellar formulation, 235 Mydriasis, 91, 94, 96, 97, 98, 104,
Micrococcus luteus, 28 108, 149, 155, 158, 159
Microfilariae Myobia musculi, 264
heartworm disease, 248-250 Myxin,24
mice, 263-264
N
Microgram-scale reverse isotope
Necator, 320
dilution assay (RIDA) method,
Necator americanus, 266
137-138
Nematodirus helvetianus, 216
Micronema deletrix, 241
Nematospiroides dubius, 24-25, 262
Microorganisms in soil, toxicity, 167,
Netherlands, 213
195
Neurotransmitter release, ivermectin,
Midge bites, 240
79,80
Milbemycins, 1,3, 12
New Zealand, 180, 344, 345, 346
Milk assay, 144
Nigeria, 315
Minimum toxic dose (MTD), 94, 108
Nitrification, soil, 167, 170
Mites
Nitrosoguanidine (NTG), in
abamectin activity, 292 (Table 20.1)
mutational procedures, 35-36
birds, 277-278
Nitrosomethylurethane (NMU), in
cattle, 218, 219
mutational procedures, 35
citrus crops, 290-292
Nodoedres cati, 256
foxes, 269
NOEL, for toxicological effects,
mange mites, 266-267, 269-270
210-211
mice, 264
Norway, 344, 345, 346
primates (nonhuman), 275
Notoedres cati, 267
Monkeys, toxicity studies
Notoedres douglassi, 268
abamectin, 107
Nuts, abamectin, 305
ivermectin, 94, 95, 97, 108-109
Monocrotophos, 297 o
Monosaccharides, of avermectin, Odocoileus vorginianus, 271
10-12, II (Fig. 1.5), 16 Oedemagena tarandi, 271
358 Index
Oesophagostomum columbianum, fermentation development, 47-48,

avermectin activity, 16 (Table 49-50 (Figs. 3.7, 3.8, 3.9, 3.10),
1.3), 19 (Table 1.6) 51-52 (Fig. 3-11)
Oesophagostomum radiatum, 231, 232
Oestrus ovis, 222 P
Oligomycin, 31, 80 Panama, 344, 346
Onchocerca cervicalis, 150, 240 Panonychus citri, 291-293
Onchocerca lienalis, 264 Panonychus ulmi, 304
Onchocerca volvulus. See Parafilaria bovicola, 175, 216
Onchocerciasis Paraguay, 344, 345, 346
Onchocerciasis, ivermectin, 311-316, Parascaris equorum, 235-237
320 Parelaphostrongylus tenuis, 271, 272
clinical reactions, 312-314, 318 Paris, 311
counterindications, 318-319 Paromomycin, 24
current use, 318-319 Partridges, 277
ocular disease, 314 Paste formulations, 125-126
Phases I through IV studies, Pears, abamectin, 303-304
311-316 mites, 304
prophylactic use, 318 psylla, 303-304
time span illustrations, 315-317 Pecan leaf scorch mite, 305
(Figs. 21.1, 21.2, 21.3) Pecan weevil, 305
transmission effects, 318 Pectinophora gossypiella, 300
Onthophagus binodis, 177-178 Pentobarbital, 81
Onthophagus gazella, 176-177 (Table Peppers, abamectin, 301-302, 303
12.2) Periplanata americanus, 76, 77, 78
OP compounds. See Peru, 344, 346
Organophosphates Pesticide residues, measurement of,
Ophinyssus sp., 280 202
Organophosphates, 202, 203, 219 Pharmacokinetic properties
Ornamental plants, abamectin, cattle, 114-119
294-297 dogs, 122-124
leafminer, 295 horses, 125-126
thrips species, 296-297 humans, 126-129
two-spotted spider mite, 294-295 sheep, 119-122
Ornithodoros moubata, 267 swine, 124-125
Ornithodoros savignyi, 219 See also Specific species.
Ornithonyssus sylvarium, 277 Phenothiazine, 178
OS-3153 strain, 25 Philippines, 344, 345, 346
Ostertagia, 216 Photodegradation of ivermectin, 167
Ostertagia circumcinta, 222 Photoisomerization, avermectins,
avermectin activity, 16 (Table 1.3), 9 (Fig. 1.4)
19 (Table 1.6) Photolysis, abamectin, 184-185, 197
Ostertagia ostertagia, 230, 231, 232 Phototoxicity, 170
Ostrich, 278 Phyllocoptruta o/eivora, 290-291
Otodectes cynotis, 254, 256 Physostigmine, 159
Oxidations, avermectins, 12-13 Picrotoxin, 74, 75, 78, 79, 80, 81, 83
Oxygen transfer Pigeons, 277
cell growth and oxygen uptake, Pink bollworm, 300
45 (Fig. 3.3) Pinworms, 302-303
Index 359
Plackett-Burman studies, 41 Radiolabeled derivatives,

Plants, toxicity, 186-187, 194 avermectins, 13, 15
Plasma concentration, 136 (Table 8.3) Rats, clinical trials
assay method, 333-335 Angiostrongylus cantonensis, 265
cattle, 116 (Fig. 7.2),117,118, Syphacia muris, 265
119 (Table 7.1) Trichinella spiralis, 265
dogs, 120 (Fig. 7.5), 122 Red imported fire ant, AFFIRM,
humans, 127-128 (Figs. 7.10, 7.11) effectiveness, 289-290
ivermectin, 95 (Table 6.2) Red mite, 291-292, 304
sheep, 119-120 (Fig. 7.5) Reindeer, clinical trials, 271
swine, 124 Reproductive toxicity, ivermectin,
Pneumonyssus, 275 98-102
Poland, 344, 345, 346 Reptiles, clinical trials, 279-280
Polynesia, 319 Response surface methodology,
Polyoxyethylene sorbitan monooleate, production media development,
118, 122 41-43
Polyphagotarsonemus latus, 291 Retinol binding proteins, ivermectin,
Polysorbate 80, 235 79
Polyvinylpyrrolidone, 114, 119 Rhabdochona sp., 279
Portugal, 344, 345, 346 Rhodnius prolixus, 264
Postural hypotension, 302 Rodents
Pregnancy, 318 absorption/ distribution/excretion of
Primates (nonhuman), clinical trials ivermectin, 134-135
mites, 275 clinical trials, 268
nematode infections, 275-276 fat metabolism studies, 139-141
Production media development, See liver metabolism studies, 135-139
Fermentation development tissue residues, 135, 136 (Table 8.3)
Propargite, 297 toxicity studies
Propylene glycol, 114, 117, 119, 121, abamectin, 104-107
124 ivermectin, 92-93, 94, 101, 108
Protein kinase C activity, ivermectin, Romania, 344, 345
85 Rose foliage, 295, 296
Protostrongylus, 273 Rushton radial flow impeller, 47
Protostrongylus rujescens, 223 Russet mite, 302
Psorergates ovis, 222-223, 224 Rust mite, 290-291, 304
Psoroptes cuniculi, 267
Psoroptes ovis, 218, 221, 224, 267, S
273 Salmo gairdneri, 169
Psylla pyricola, 303-304 Salmon, 278, 279
Pterolichidae, 278 Sarcoptes scabiei, 218, 221, 224, 226,
Pyrethroids, 219, 304 241,253-254,269,270
Pythons, 279-280 Schistocerca gregaria, 75
Scirtothrips citri, 293
Scolothrips sexmaculatus, 305
R Scotland 179
Rabbits, clinical trials Screwworm fly, 216
mange mites, 266-267 Screwworms, 219
ticks, 267 Sculpin, 278
tsetse flies, 267 Seagulls, 277
360 Index
Seed flasks, fermentation of culture, Sphaerocerids, 178

34 Spider mites, 297-299
Seed media development, 37-40, Spinal cord parasites, 272
38 (Table 3.3) Spodoptera eridania, 299, 300, 302
See also Fermentation Spodoptera exigua, 302
development. Squalene, 52
Senegal, 311, 312 Stable fly, 174, 175
Sepsids, 178 Staphylinids, 178
Setaria equina, 241 Stephanurus dentatus, 226
Shake flash, avermectin formation, Sterilization, media, 51
43 (Fig. 3-1) Sternostoma tracheolum, 278
Sheep Stirred tanks, fermentation
avermectin activity, 19 (Table 1.6) development, 44-45
clinical trials, 222-224 Stomoxys calcitrans, 175
ectoparasite activity, 222-223, Strawberries, abamectin, 305
224 Streptomyces avermitilis, 80, 89, 288
endoparasite activity, 222, 223 antifungal/ antibacterial activity,
injectable formulation, 223 28-31
oral formulation, 222 compounds resulting from
fat metabolism studies, 139-141 fermentation of, 1-3
liver metabolism studies, 135-139 containment of, 51
pharmacokinetics of ivermectin, cultural characteristics of, 27 (Table
119-122 2.1)
biexponential decay, 118-119 electron micrographs of,
(Fig. 7.5) 25-26 (Figs. 2.1, 2.2)
bioavailability analyses, 121 genealogy, 37 (Table 3.2)
dosage formulations and results, physiological properties of,
121 28 (Table 2.2)
plasma concentration, shake flash, avermectin formation,
119-120 (Fig. 7.5) 43 (Fig. 3-1)
safety of ivermectin, 153-155 source of, 24, 34
tissue residues, 132-134 taxonomy of, 25-26
toxicity, 153-154 See also Fermentation
Sinefungin, as inhibitor, 70, 72 development.
Singapore, 345, 346 Strongyloides fuelleborni, 275
Skin microfilariae, 314-318 Strongyloides ransomi, 226
density, 315 (Fig. 21.1) Strongyloides ratti, 262
effects, 316 (Fig. 22.2) Strongyloides stercoralis, 253, 263,
Sleeping sickness, 318 320-321
Snakes, clinical trials, 279-280 Strongyloides westeri, 241
Soil. See Environmental effects of Strongylus edentatus, 238-239
abamectin; Environmental effects Strongylus equinus, 239
of ivermectin Strongylus vulgaris, 237-238
Solenopotes capillatus, 218, 221 Strychnine, 85
Solenopsis invicta, 288-290 Subchronic toxicity, 95-98
Solubility, abamectin, 186 Sudan, 268, 344, 345, 346
South Africa, 175,344,345,346 Summer sores, 240
Southern armyworm, 20, 299 Sunlight, abamectin, effects, 292
Spain, 344, 345, 346 Superphosphate, 178
Index 361
Sustained-release bolus, 221-222 guinea pigs, 265-266

Sweden, 271, 344, 345, 346 rabbits, 267
Swine Tissue residues
clinical trials, 225-227 analytical techniques, 137-138,
ectoparasite activity, 226-227 144-145
endoparasite activity, 226 animal studies, 132-136
injectable formulation, 225-226 distribution pattern and route of
fat metabolism studies, 139-141 dose, 134
liver metabolism studies, 135-139 free residues, 132
pharmacokinetics of ivermectin, metabolism and, 133 (Table 8.l)
124-125 rodents, 135, 136 (Table 8.3)
bioavailability analyses, assay method, 324-344
124-125 (Fig. 7.7) edible tissues
dosage formulations and results, analytical methods, 144-145, 147
124-125 chemical assay, 144-148
plasma concentration, 124 depletion profiles, 147 (Table 9.3)
safety of ivermectin, 155-157 recoveries of ivermectin, 146
tissue residues, 132-134 (Tables 9.1, 9.2)
toxicity, 155-156 See also Assay method.
Switzerland, 344, 345, 346 Tobacco hornworm, 299
Syphacia muris, 265 Togo, 313
Syphacia obvelata, 263 Tomato foliage, abamectin, 301-303
Tortoises, 279, 280
Toxascaris leonina, 251
T Toxicity
Tablet formulation, dogs, 122-124 algae, 168
Taiwan, 345 bees, 195
Tampans, 216 birds, 195-196
Tapeworms, 241, 256 cattle, 151-152, 153
Teratology studies dogs, 157-158
abamectin, 104 (Table 6.6) earthworms, 168
ivermectin, 104 (Table 6.6) fish, 169 (Table 11.5), 170, 187-188,
Tetranychus cinnabarinus, 297 193-194
Tetranychus pacijicus, 305 goats, 155
Tetranychus urticae, 20 (Table 1.7) horses, 149, 151
294-295, 297-298, 302, 304, 305 microorganisms in soil, 167, 195
Thailand, 344, 345, 346 plants, 186-187, 194
Thaimycins, 24 sheep, 153-154
Thelazia, 216 swine, 155-156
Thiabendazole, 178, 275, 320 Toxicity studies
Thiacetarsamide, 269 abamectin
Thin-layer chromatography (TLC), 7, chronic studies, 103 (Table 6.7)
26 dermal penetration study, 107,
Thrips species 108 (Table 6.8)
citrus crops, 293 dogs, 104
ornamental plants, 296-297 monkeys, 107
Ticks rodents, 104-107
abamectin efficacy, 231, 232-233 teratology studies, results,
cattle, 219 104 (Table 6.6)
362 Index
Toxicity studies (cont.) Twospotted spider mite, 17, 294-295.

ivermectin 302, 304
acute toxicity, 92-95
developmental/reproductive U
toxicity, 98-102, 109 Ultraviolet methods, 145
dogs, 93-94, 97,108 use of, mutational procedures, 35
genotoxicity studies, 91-92 Uncinaria stenocephala. 252-253
lactation, 97, 10 1 Ungulates. See specific species
minimum toxic dose (MTD), 94, United Kingdom, 80, 344, 345, 346
108 Uruguay, 344, 345, 346
monkeys, 94, 95, 97, 108-109 U.S.A., 344, 345, 346
plasma concentrations, 95 (Table U.S.S.R., 344, 346
6.2)
rodents, 92-93, 94, 101, 108 V
subchronic toxicity, 95-98 Vegetables, abamectin, 301-302
teratology studies, results, armyworms, 302
104 (Table 6.6), 109 and beneficial arthropods, 303
Toxocara canis, 250-251,263,269 leafminer, 301-302
Toxocara cati, 255 pinworms, 302-303
Translocation, 166 russet mite, 302
Transuterine/transmammary twospotted spider mite, 302
transmission, 250 Venezuela, 344, 345, 346
Tremors, 91, 94, 96, 99, 104, 105, 106, Vertebrates, ivermectin mode of
108 action, 80-85
Triatomid bugs, mice, 264 Volta River Basin, 315
Tribolium confusum, 288
Trichinella spiralis, 265 W
Trichinella spp., 263 Warbles, 273
Trichodectes canis. 269 West Africa, 319
Trichostrongylus axei. 231, 232 Whipworm, dogs, 253
avermectin activity, 16 (Table 1.3), Wild boar, clinical trials, 274
19 (Table 1.6) Wolves, clinical trials, 269
Trichostrongylus colubriformis. 216, Worker exposure study, 201-213
222, 266 abamectin dissipation curve,
avermectin activity, 16-18 (Tables 208-209, 210 (Fig. 14.1)
1.3, 1.4, 1.5), 19 (Table 1.6) during airblast treatment of citrus
Trichuris muris. 263 groves, 203-208
Trichuris uulpis. 253 clothing and protection, 205, 211
Triglycerides, 52 dermal exposure, dosimeters, 204
Trinidad, 344, 346 dermal penetration, 206
Trioxabicyclooctaines, 79 dislodgeable foliar residues, 202,
Trixacarus cauiae. 266 209-213
Trout, 278 exposure model used, 201-202
Tsetse flies during harvest, 208-213
guinea pigs, 266 harvester exposure data, 213 (Table
rabbits, 267 14.6)
Tunisia, 344, 345 individual worker exposure data,
Turkey, 344 212 (Table 14.5)
Index 363
margins of safety (MOS) Y

calculations, 206-207 Yugoslavia, 344, 345
exposure data, 207 (Table 14.3)
foliar residues, 210-211, 212-213 Z
routes of exposure, 202 Zimbabwe, 344, 345
Zweig-Popendorf Factor, 202-203, Zimectrin paste, 234
208, 209, 211 Zweig-Popendorf Factor, worker
Wuchereria bancrofti, 319-320 exposure, 202-203, 208, 209, 211

Ivermectina y Abemectina

Uploaded by

Copyright:

Available Formats

Ivermectina y Abemectina

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ivermectina y Abemectina

Uploaded by

Copyright:

Available Formats

William C.

Library of Congress Cataloging-in-Publication Data

I. Ivermectin. 2. Abamectin. I. Campbell, William C. (William

© 1989 by Springer-Verlag New York Inc.

Typeset by TCSystems, Inc., Shippensburg, PA.

to the memory of the late Mr. Robert K.

Ivermectin was introduced to the marketplace as an antiparasitic drug in

The editor thanks Dr. Harold D. Hafs, Vice-President of Animal Science

Chapter 2. Isolation and Characterization of

Chapter 3. Fermentation Development and Process Improvement

Chapter 5. Mode of Action of Ivermectin

Chapter 7. Pharmacokinetics of I vermectin in Animals

Chapter 8. Metabolism and Tissue Residues

Chapter 9. Chemical Assay for Ivermectin in Edible Tissues

Chapter 10. Safety of I vermectin in Target Animals

Chapter 11. Environmental Aspects of I vermectin Usage

Chapter 12. Environmental Aspects of Use of Ivermectin and

Chapter 13. Environmental Aspects of Abamectin Use in

Chapter 14. Worker Safety Aspects of Abamectin Use in

Chapter 15. Use of Ivermectin in Cattle, Sheep, Goats,

Chapter 16. Use of Abamectin in Cattle.

Chapter 17. Use ofIvermectin in Horses

Chapter 18. Use ofIvermectin in Dogs and Cats

Chapter 19. Use of Ivermectin in Laboratory and

Chapter 20. Abamectin Use in Crop Protection

Chapter 21. Use of Ivermectin in Humans

I: The Determinative Method for Assaying Ivermectin

II: Confirmatory Assay for Assaying Ivermectin Residues

III: List of Registrations

G.W. BENZ, Ph.D.

K.R. BROWN, M.D.

BARRY C. BUCKLAND, Ph.D.

R.W. BURG, Ph.D.

WILLIAM C. CAMPBELL, Ph.D.

S.T. CHEN, Ph.D.

SHUET-HING LEE CHIU, Ph.D.

J.L. Cox, Ph.D.

GEORGE V. DOWNING, Ph.D.

R.A. DYBAS, Ph.D.

DAVID W. FINK, Ph.D.

M.H. FISHER, Ph.D.

LEA R. GORDON, D.V.M., Ph.D.

B.M. GREENE, M.D.

S.J. GROSS, Ph.D.

L.S. GROSSO, Ph.D.

B.A. HALLEY, Ph.D.

O.D. HENSENS, Ph.D.

LOVIS KAPLAN, Ph.D.

GEORGE R. LANKAS, Ph.D.

W.H.D. LEANING, B.V.Sc.

A.Y.H. Lu, Ph.D.

M. NALLIN OMSTEAD, Ph.D.

R.J. NESSEL, Ph.D.

ARTURO G. PORRAS, Ph.D.

J.M. PRESTON, B.V.M.S., Ph.D.

J.D. PULLIAM, D.V.M., Ph.D.

S.F. RICKARD, Ph.D.

R.A. RONCALLI, D.V.M.

J.M. SCHAEFFER, Ph.D.

M.D. SCHULMAN, Ph.D.