SIMPLE DIPSTICK ASSAY FOR THE DETECTION OF SALMONELLA

TYPHI-SPECIFIC IgM ANTIBODIES AND THE EVOLUTION OF THE IMMUNE

RESPONSE IN PATIENTS WITH TYPHOID FEVER
MOCHAMMAD HATTA, MARGA G. A. GORIS, EVY HEERKENS, JAIRO GOOSKENS, AND HENK L. SMITS
Department of Microbiology, Hasanuddin University, Makassar, Indonesia; KIT Biomedical Research, Royal Tropical
Institute/Koninklijk Instituut voor de Tropen, Amsterdam, The Netherlands
Abstract. Application of a dipstick assay for the detection of Salmonella typhi-specific IgM antibodies on samples
collected from S. typhi or S. paratyphi culture-positive patients at the day of admission to the hospital revealed the
presence of specific IgM antibodies in 43.5%, 92.9%, and 100% for samples collected 46 days, 69 days, and > 9 days
after the onset of fever, respectively. The mean sensitivity for samples collected an average of 6.6 days after the onset
of fever was 65.3%. Culture was positive in 65.9% of the cases with a final clinical diagnosis of typhoid fever. Testing
of paired serum samples from culture negative patients with a final clinical diagnosis of typhoid fever resulted in staining
of the dipstick in 4.3% of the samples collected at the day of admission to the hospital and in 76.6% of the samples
collected one week later, thereby provided strong supporting evidence of typhoid fever by demonstrating seroconversion
in a large proportion of the patients. The dipstick assay may thus also be useful for the serodiagnosis of culture-negative
patients with clinical signs and symptoms consistent with typhoid fever. The advantages of the dipstick assay are that the
result can be obtained on the same day allowing a prompt treatment, that only a small volume of serum is needed, and
that no special laboratory equipment is needed to perform the assay. The stability of the reagents of the dipstick and the
simplicity of the assay allows its use in places that lack laboratory facilities.
INTRODUCTION
Typhoid fever is still an important health problem in many
developing countries. Worldwide, an estimated 17 million
cases occur annually. Most of this burden occurs among citi-
zens of low-income countries, particular those in South East
Asia, Africa, and Latin America. The clinical diagnosis of this
condition is considered to be unreliable.
1
A definite diagnosis
is obtained when the etiologic agent, Salmonella typhi, is iso-
lated from bone marrow or blood.
2
Facilities to perform this
complicated and time-consuming procedure are usually not
available in endemic areas. In these problematic situations
the Widal test can be used to aid the clinical diagnosis. How-
ever, many limitations lead to difficulties in the interpretation
of the Widal test. Results of the Widal test have demonstrated
to vary between different areas and in time, due to variation
in background levels as well as a result of variation in the
quality of the antigen.
312
The need for a rapid and inexpen-
sive laboratory test for early and accurate diagnosis of pa-
tients with typhoid fever has prompted the exploration of a
variety of serologic and antigen detection methods, including
counter immunoelectrophoresis,
13,14
enzyme-linked immuno-
sorbent assay (ELISA),
1518
dot immunoassay,
16,1921
hemag-
glutination,
22
and coagglutination.
23
However, these assays
are not very easy to perform, not rapid, require special equip-
ment or skills, or depend on electricity and on refrigeration
for storage of components. None of these assays has yet
reached widespread use.
To fulfill the need for a simple and rapid laboratory test we
have developed a dipstick assay for the detection of IgM an-
tibodies for S. typhi. The dipstick assay is a simplified version
of the ELISA that can be used without the need for special
equipment or electricity. The assay uses stabilized compo-
nents that can be stored for more than two years outside the
refrigerator. The ELISAs for typhoid fever have found to be
superior to the Widal test.
15,24
In this study, the clinical utility
of the dipstick assay was evaluated in an endemic area in
Indonesia and on a collection of serum samples from Kenya.
The dipstick assay is based on methodology also used in
dipstick assays developed for the serodiagnosis of leptospiro-
sis
25
and brucellosis.
26
The assay format has proven to be well
suited for use in clinical settings that lack laboratory facilities
to perform the more complicated standard laboratory tests.
2730
MATERIALS AND METHODS
Patients. The following groups were included in the study.
Patients from Indonesia with clinical suspicion of typhoid
fever. Blood and serum samples were collected from clinically
suspected typhoid patients on admission at the Hasanuddin
University Hospital of Makassar, South Sulawesi, Indonesia
at three primary health care centers in Makassar, and at a
District Hospital in Gowa. Gowa district is located about 30
km south of Makassar. A total of 245 patients were included
in the study. The mean age of the patients was 21.4 years
(range 750). They had a mean temperature on admission
of 38.0C (range 37.040.5C) and a mean duration of
illness on admission of 6.6 days (range 419). The main
clinical symptoms were hepatomegaly (19.2%), confusion
(17.6%), and splenomegaly (13.5%). Abdominal pain was re-
ported in 3.3% of the patients. Gastrointestinal bleeding,
jaundice, and rash were observed in less than 1%. Follow-up
samples (n 192) collected an average of one and two weeks
after the initial sample could be collected from 86 (35.1%)
patients.
Hospital control group from Indonesia. Acute-phase serum
samples were collected from a group of 259 hospitalized pa-
tients from Makassar with a final diagnosis other than typhoid
fever. The final diagnosis was hepatitis for 82 patients, lep-
tospirosis for 35 patients, malaria for 74 patients, upper re-
spiratory tract infection for 48 patients, and pyrexia of un-
known origin for 20 patients. Their mean age was 32 years
(range 270). They had a mean temperature on admission
of 38.2C (37.140.1C) and a mean duration of illness on
admission of eight days (range 421).
Healthy controls from Indonesia. Serum samples were col-
lected from 194 healthy school children (mean age 11.7
Am. J. Trop. Med. Hyg., 66(4), 2002, pp. 416421
Copyright 2002 by The American Society of Tropical Medicine and Hygiene
416
years, range 1014) from a village located in Gowa district,
Sulawesi, Indonesia.
Hospital control group from a non-endemic area. Samples
were collected from patients from the Netherlands with vari-
ous diseases including infection with human immunodeficien-
cy virus (n 20), hepatitis A (n 10), hepatitis B (n 9),
syphilis (n 20), malaria (n 20), toxoplasmosis (n 11),
meningitis (n 10), meningococcal meningitis (n 10),
Lyme borreliosis (n 20), hantavirus infection (n 20),and
an autoimmune disease (rheumatoid arthritis n 10, sys-
temic lupus erythematosus n 20).
Case patients and controls from Kenya. Serum samples
were collected at St. Marys Hospital, Mumias, Kenya from
patients in the early stage of the disease with fever including
19 patients with a blood culture proven S. typhi infection,
eight patients with a blood culture-proven S. enteritidis infec-
tion, six patients with a blood culture-proven S. typhimurium
infection, and 185 patients with a final diagnosis other than
enteric fever.
Ethical considerations. The project was approved by the
review boards of the participating institutes and informed
consent for participation in the study was obtained from all
participants or their parents/guardians.
Culture. Blood culture was performed for each of the group
of patients with clinically suspected typhoid fever from Indo-
nesia and for all patients with fever from Kenya. The blood
culture was performed by inoculation of 15 ml of bile broth
(Merck, Rahway, NJ) with 5 ml of freshly collected blood.
Cultures were incubated for 24 hr at 37C. A 1 ml culture
sample was then plated on Salmonella Shigella agar. After
incubation for 24 hr at 37C, colonies were examined by
Gram staining and tested biochemically to identify S. typhi-
positive cultures.
31
Widal test. The Widal test procedure using O antigen was
performed according to the manufacturers protocol (Murex
Biotech,Ltd., Dartford, UK). Briefly, two-fold serial dilutions
(1:201:1,280) of the serum sample were prepared. One drop
(25 l) of the O antigen suspension was added to each tube
containing the diluted sample. Antigen and serum were
mixed and incubated at 50C. Tubes were checked for agglu-
tination after 4 hr. According to routine diagnostic criteria, a
titer 1:320 was considered positive for the samples tested in
Indonesia. In Kenya, a titer 1:160 was considered positive.
Dipstick assay. The dipstick consists of a strip of nitrocel-
lulose membrane containing a 2 mm-wide line of immobilized
antigen as detection band and a separate line of immobilized
anti-human IgM antibody as reagent control that is adhered
to a rigid backing. The antigen was prepared from a culture of
a recent isolate of S. typhi from Indonesia. The culture was
grown in LB broth and the antigen was prepared by heating
a washed and 15 concentrated bacterial cell suspension of a
three-day old culture for 30 min at 95C. Cell debris was
removed by centrifugation. The supernatant was diluted five
times and blotted in 2 mm-wide lanes onto a nitrocellulose
membrane by incubation for 2 hr at 40C. At the end of the
incubation the lanes of the blotting apparatus were rinsed
with phosphate-buffered saline (PBS) to remove excess anti-
gen. Blotted strips were rinsed with PBS, blocked with 3%
skim milk, rinsed again, and allowed to dry. The non-
enzymatic detection reagent consists of a monoclonal anti-
human IgM antibody conjugated to a colloidal suspension of
Palanyl red.
32,33
Briefly, the monoclonal antibody was labeled
with a washed suspension of Palanyl red in 10 mM phosphate
buffer containing 2.7 mM NaCl. The concentration of the dye
suspension was adjusted such that a 1:500 dilution had a spec-
trophotometric absorbance of 520 nm (A
520
) of 1. The con-
jugate was subsequently blocked with 30% bovine serum al-
bumin in 5 mM NaCl. After blocking, the conjugate was sus-
pended in 32.3 mM phosphate buffer containing 125 mM
NaCl., 6% trehalose, and 1.67% bovine serum albumin. Fi-
nally, the suspension was lyophilized for preservation.
The dipstick assay was performed by incubation of a wetted
dipstick in a mixture of 5 l of serum and 250 l of detection
reagent for 3 hr at room temperature. At the end of the
incubation the dipsticks were thoroughly rinsed with water,
and dried. The staining intensity of the antigen band was then
graded by comparison with a colored reference strip. The test
result is scored as negative when no staining was observed, 1+
when a weak staining was observed, and 2+, 3+, or 4+ when a
moderate-to-strong staining was observed.
Statistical analysis. Difference in the immune response be-
tween groups for age, fever, and duration of illness at time of
sampling, and for result of culture were analyzed by linear
regression analysis using the SPSS (SPSS, Inc., Chicago, IL)
computer package.
RESULTS
A final diagnosis of enteric fever due to either S. typhi or S.
paratyphi was made for 179 of the 245 suspected patients from
Indonesia. The final diagnosis was based on a positive blood
culture for 118 (65.9%) patients and on clinical symptoms and
signs consistent with typhoid or paratyphoid fever for 61
(34.1%) patients. Salmonella typhi was isolated from the cul-
tures of 112 patients and S. paratyphi from six of them. An
alternative final clinical diagnosis was made for 64 patients.
The final diagnosis for the latter group of patients was malaria
(23 cases), hepatitis (20 cases), upper respiratory tract infec-
tion (11 cases), dengue hemorrhagic fever (1 case), and py-
rexia of unknown origin (9 cases). Three patients died before
a final diagnosis could be made.
A positive result in the dipstick assay for samples collected
at the time of hospitalization was obtained for 73 (65.2%) of
the S. typhi culture-positive patients and for four (66.6%) of
the S. paratyphi culture-positive patients (Table 1). A positive
result in the dipstick assay was also obtained for some
(13.1%) of the blood culture negative patients with a final
clinical diagnosis of typhoid fever. None of the samples from
the patients with a final diagnosis other than enteric fever
gave a positive result. These results indicate a sensitivity for
the dipstick assay of 65.3% for samples collected at the time
of admission to the hospital from the S. typhi and S. paratyphi
culture-positive patients (Table 2). The sensitivity for samples
collected at admission for the total group of culture-positive
and -negative patients with a final diagnosis of enteric fever
was 47.5%.
The sensitivity of the dipstick increased with the duration
of fever and was, as calculated for the group of culture-proven
patients, 43.5% for patients with 46 days of fever prior to
admission to the hospital and laboratory testing, 92.9% for
patients with 79 days of fever, and 100% for patients with >
9 days of fever (Table 2). For the combined group of culture-
positive and culture-negative patients with a final diagnosis of
typhoid or paratyphoid fever, the sensitivity was lower in par-
DIPSTICK ASSAY FOR THE SERODIAGNOSIS OF TYPHOID FEVER 417
ticular for samples collected during the first days with fever
(Table 2). The increase in sensitivity of the dipstick assay with
the duration of fever correlated with an increase in the stain-
ing intensity of the antigen band of the dipstick. The staining
intensity was graded moderate (2+) to strong (4+) for 13.3%,
68%, and 85.7% of the positive samples collected from the
patients with 46, 79, and > 9 days of fever, respectively. A
1+ staining intensity was recorded for the remaining dipstick-
positive samples.
To further investigate the antibody response, follow-up
samples collected an average of one and two weeks after
hospital admission were tested. Follow-up samples were col-
lected from 39 S. typhi culture-positive patients and from 47
culture-negative patients with a final diagnosis of typhoid fe-
ver (Table 3). For the group of culture-positive patients, a
positive dipstick result was obtained for 76.9% of the initially
collected samples and seroconversion was observed for eight
(20.5%) patients at subsequent collections. In contrast, the
initially collected samples from only two (4.3%) of the blood
culture-negative patients tested positive while seroconversion
was observed for 37 (78.7%) patients of this group. A nega-
tive result for all three serum samples was obtained for eight
patients with a negative blood culture. The samples from the
culture-negative patients with typhoid fever were collected an
average of two days earlier than the samples from the culture-
positive patients. Regression analysis indicated that this dif-
ference in time of sample collection contributed to the lower
percentage of dipstick positive result for the initially collected
samples of the culture-negative group and thus to the higher
seroconversion rate observed for this group. Age and the
degrees of fever were not related to the difference in results
of the dipstick assay for the two groups.
The specificity of the dipstick assay as calculated for the
group of suspected patients with a final diagnosis other than
typhoid or paratyphoid fever was 100% (Table 1). The high
specificity was confirmed by testing samples collected in an
endemic area froma hospital population with a final diagnosis
other than typhoid fever and by testing samples from healthy
schoolchildren from Indonesia. None of the 258 sera from the
hospital control group tested positive in the dipstick assay.
Two samples from the school children survey tested weakly
positive. The other 192 (99%) samples were negative. Cross-
reactivity also was not observed for the 180 samples from
patients with various diseases from a non-endemic area.
The sensitivity of the Widal test at the routinely used cut-
off level of 1:320 for samples collected at hospital admission
was 60.7% for the S. typhi culture-positive patients and 83.3%
for the S. paratyphi culture-positive patients (Table 1). The
specificity was 88.4%.
In a separate study performed in Kenya on serum samples
collected in the early stage of the disease from culture-proven
patients, a positive result in the dipstick assay was obtained in
samples from 14 (73.7%) of 19 patients with an S. typhi in-
fection, and from five (62.5%) of eight patients with an S.
enteritidis infection. Samples from six patients with an S. ty-
phimurium infection gave a negative result. Staining was also
observed for seven of 185 patients with a final diagnosis other
than typhoid fever, indicating a specificity of 96.2%. In the
study performed on the samples collected in Kenya, a strong
cross-reactivity was observed for a sample from a patient with
a positive blood culture for Escherichia coli. The Widal test
was performed on part of the samples from Kenya and gave
a relatively low sensitivity (66.6%) and specificity (67.0%) at
the routinely used cut-off level of 1:160.
TABLE 2
Results of the dipstick assay and Widal test for samples collected at
hospitalization and stratified according to the duration of
the illness
Duration of
fever (days)
No. of patients with a positive dipstick result
(%)/total no. of patients
Patients with a final clinical diagnosis of typhoid
or paratyphoid fever
Culture positive Culture positive and negative
Dipstick assay
46 30 (43.5)/69 33 (28.9)/114
79 26 (92.9)/28 28 (70.0)/40
>9 21 (100)/21 24 (96.0)/25
Total 77 (65.3)/118 85 (47.5)/179
Widal test
46 33 (47.8)/69 39 (34.2)/114
79 21 (75.0)/28 25 (62.5)/40
>9 19 (90.4)/21 21 (84.0)/25
Total 73 (61.8)/118 85 (47.5)/179
TABLE 1
Results of the dipstick based on the results of culture and the final clinical diagnosis for samples collected at the time of hospitalization
No. of patients with a positive result (%)/total no. of patients
Dipstick assay Widal
Clinically suspected typhoid and paratyphoid patients (Indonesia)
Final diagnosis of typhoid or paratyphoid fever 85 (47.5)/179 85 (47.5)/179
Salmonella typhi culture positive 73 (65.2)/112 68 (60.7)/112
S. paratyphi culture positive 4 (66.6)/6 5 (83.3)/6
Culture negative 8 (13.1)/61 12 (19.7)/61
Final clinical diagnosis other than typhoid or paratyphoid fever 0 (0)/64 10 (15.6)/64
Hospital controls (Indonesia) 0 (0)/259 NT*
School children (Indonesia) 2 (1)/194 NT
Hospital controls (The Netherlands) 0 (0)/180 NT
Clinically suspected typhoid and paratyphoid patients (Kenya)
S. typhi culture positive 14 (73.7)/19 16 (84.2)/19
S. enteritidis culture positive 5 (62.5)/8 6 (75.0)/8
S. typhimurium culture positive 0 (0)/6 0 (0)/6
Culture negative 7 (3.8)/185 33 (33.0)/100
* NT not tested.
HATTA AND OTHERS 418
DISCUSSION
In our study in Indonesia, blood culture confirmed the final
diagnosis of typhoid fever in 112 patients with clinical sus-
pected typhoid fever. Culture demonstrated an infection with
S. paratyphi in six other patients. A final diagnosis of typhoid
fever also was made on clinical signs and symptoms for 61
suspected patients despite of a negative culture result. A final
diagnosis other than typhoid fever was made for 64 suspects
based on the evolution of the disease and additional labora-
tory testing. These results demonstrate that the early clinical
symptoms and signs of typhoid fever are not very specific and
that further clinical and laboratory investigation is needed to
come to a final diagnosis.
Culture is an accurate method for the diagnosis of typhoid
fever for blood samples drawn early in the disease. More than
70% of the cases will be confirmed when multiple blood
samples are tested.
34,35
However, the use of antibiotics and
the often low amount of bacteria present in the blood may
hamper the result. The amount of blood used to inoculate the
culture also may affect the detection rate of culture. Ideally,
at least 10 ml of blood is needed to inoculate the culture.
Reluctance to donate a sufficiently large blood volume is
common in many countries. For this reason, 5 ml of blood was
used to inoculate the cultures. Culture has a number of other
limitations as well. Culture takes a minimum of 23 days to
get a result, is expensive, and requires specific laboratory fa-
cilities and trained staff to perform the culture and the sero-
logic and biochemical testing that is needed for identification.
Based on the final clinical diagnosis, the detection rate of the
culture was 65.9% for single blood samples drawn an average
of six days after the onset of fever. The use of antibiotics
before hospital admission could well have contributed to the
relatively low sensitivity of the culture procedure.
36
The dipstick assay was demonstrated to have a high speci-
ficity. The specificity was 100% for patients with clinical sus-
picion of typhoid fever from an endemic area in Indonesia.
The fact that it usually takes about 57 days after the onset of
fever before detectable antibody levels are present contrib-
utes to the limited sensitivity of this serologic detection
method in that period of the disease. The sensitivity was
65.3% for samples drawn at hospitalization from blood cul-
ture-positive patients with on average 6.6 days of fever and
clearly varied with the duration of the disease from 43.5% for
patients with 46 days of fever to 92.9% for patients with 79
days of fever and to 100% for patients with more than nine
days of fever. Ideally, serologic testing for infectious diseases
should be performed on paired serum samples as demonstra-
tion of seroconversion or an increase in titer provides stron-
ger evidence of the infection than demonstration of an el-
evated antibody level in a single serum sample. Testing of
paired samples increased the sensitivity of the dipstick. In-
deed testing of a second serum sample collected on average
after one week of hospitalization increased the sensitivity
from 76.9% for samples collected at the time of hospitaliza-
tion to 83.1% for the culture-positive patients. Seroconver-
sion also was observed for the majority of the blood culture-
negative patients with clinical typhoid fever. The sensitivity
for this group of patients was 4.3% for samples collected at
the time of hospitalization and 76.6% for samples collected
one week later. Possibly the group of culture-negative pa-
tients with a final clinical diagnosis of typhoid fever included
a number of patients who did not have typhoid fever, explain-
ing the somewhat lower response in the dipstick assay as well
as in the Widal test. One also could speculate that antibiotic
use before hospitalization or the dose of infection might have
affected the result of culture and contributed to a slower de-
velopment of the immune response in the culture-negative
group. It also is possible that they were cases who presented
earlier in the course of their illness.
Cross-reactivity in the dipstick assay could be expected in
case of septicemia with other salmonellae species sharing an-
tigen determinants with S. typhi. Salmonella typhi shares the
somatic or O antigen type 9 with a number of other salmo-
nellae species of Salmonella group D
1
organisms including S.
enteritidis. In addition S. typhi shares the O antigen type 12
with group A and B organisms including S. paratyphi A, and
S. paratyphi B and S. typhimurium, respectively. The results
of the present study demonstrated cross-reactivity in the dip-
stick assay in patients with a positive blood culture for S.
paratyphi or S. enteritidis. A positive result in the dipstick
assay for patients with an S. paratyphi or S. enteritidis infec-
tion in the blood would still result in an appropriate treat-
ment. No reactivity was observed when testing samples from
six patients with an S. typhimurium infection. Despite the fact
TABLE 3
Dipstick results for follow-up samples collected from Salmonella typhi culture-positive patients and from culture-negative patients with a final
diagnosis of typhoid fever
Sample
Mean no. of days
with fever (2575th
percentiles)
No. of patients with a
positive dipstick result
(%)/no. of patients
Staining intensity
Neg 1+ 2+ 3+ 4+
S. typhi culture-positive patients
1st collection 8 (511) 30 (76.9)/39 9 12 3 6 9
2nd collection 15 (1218) 32 (82.1)/39 7 6 8 8 10
3nd collection 29 (2531) 38 (97.4)/39 1 3 6 14 15
Culture-negative patients
1st collection 6 (56) 2 (4.3)/47 45 2 0 0 0
2nd collection 13 (1213) 36 (76.6)/47 11 22 9 4 1
3rd collection 27 (2527) 39 (83.0)/47 8 4 8 18 9
Total
1st collection 7 (57) 32 (37.2)/86 54 14 3 6 9
2nd collection 14 (1214) 68 (79.1)/86 18 28 17 12 11
3rd collection 28 (2528) 77 (89.5)/86 9 7 14 32 24
DIPSTICK ASSAY FOR THE SERODIAGNOSIS OF TYPHOID FEVER 419
that S. typhi and S. typhimurium share some antigens, these
results may indicate that these common antigens either are
not present in the antigen preparation used in the dipstick
assay or that these common antigens do not give rise to a
strong detectable IgM antibody response. It should be noted
although that the samples from the S. typhimurium-positive
patients were all collected during the first week of the illness,
possibly before the appearance of specific antibodies. Cross-
reactivity potentially also could be expected in patients with
previous typhoid vaccination. However, vaccination is rare in
Indonesia and in Kenya.
The sensitivity of the dipstick assay and of the Widal test
were similar, but the specificity of the dipstick assay was
higher. In this study, results of the Widal test were based on
routinely applied cut-off values. Although the specificity of
the Widal test could be improved by using a one titer step
higher cut-off value, this was unacceptable because it resulted
in a sensitivity that was too low. The value of the Widal test
recently was re-evaluated in a study performed in Vietnam.
37
The major advantages of the dipstick assay are that it assay
is easy to use, does not require special equipment or training,
and uses stabilized components. It therefore has a potential
high degree of acceptability. Culture is clearly the method of
choice in situations where laboratory facilities to perform the
assay and the required typing procedure are available. Dis-
advantages of culture are that it can be done only by few
specialized laboratories, it requires a relatively large volume
of blood collected by venipuncture, and the result is obtained
only after 23 days. A further disadvantage is that multi-
blood cultures must be performed to ensure a high sensitivity,
and to exclude contamination. Performance of the dipstick
assay does not require laboratory facilities and the result of
the dipstick can be obtained in about 3 hr. Thus, application
of the dipstick clearly offers the opportunity to start the ap-
propriate treatment at the same day the patient is admitted to
the hospital. For this reason, the dipstick assay could be of
high value even in situations where culture facilities are avail-
able. Testing of follow-up samples improves sensitivity. Test-
ing of follow-up samples in the dipstick assay proved to be
useful in culture-negative patients who had clinical signs and
symptoms consistent with typhoid fever since seroconversion
was observed in 72.3% of these patients at a average of seven
days after admission to the hospital. Therefore, the dipstick
assay not only can be useful for diagnosing typhoid fever in
health care facilities with no laboratory support capable of
performing cultures, but also can provide a meaningful addi-
tion to culture. The clinical utility of the dipstick assay was
recently also demonstrated in a study performed in Viet-
nam.
38
A sensitivity of 77% and a specificity of 95% was
calculated for the dipstick assay in that study.
Acknowledgments: We thank Drs. P. M. Tukei and P. Borus (Virus
Research Centre, Kenya Medical Research Institute, Nairobi, Kenya)
and the staff of the St. Marys Hospital (Mumias, Kenya) for gifts of
sera, Dr. B. J. van den Born and D. Sluiters for assistance with the
clinical study at the St. Marys Hospital, and the staff at the Depart-
ment of Microbiology of the Hasanuddin University Hospital,
Makassar, at the District Hospital and at the Primary Health Care
Centre at Gowa district, South Sulawesi, Indonesia for enthusiastic
and valuable support of this study. The help of Ulla Renqvist with the
statistical analysis was highly appreciated.
Financial support: This study was supported by EC grant no.
IC18CT9980381.
Authors addresses: Mochammad Hatta, Department of Microbiol-
ogy, Hasanuddin University, Makassar, Indonesia. Marga G. A.
Goris, Evy Heerkens, Jairo Gooskens, and Henk L. Smits, KIT Bio-
medical Research, Royal Tropical Institute/Koninklijk Instituut voor
de Tropen (KIT), Meibergdreef 39, 1105 AZ Amsterdam, The Neth-
erlands, Telephone: 31-20-566-5470, Fax: 31-20-697-1841, E-mail:
h.smits@kit.nl.
REFERENCES
1. Edelman R, Levine MM, 1986. Summary of an international
workshop on typhoid fever. Rev Infect Dis 25: 147150.
2. Hoffman SL, Punjabi NH, Rockhill RC, Sutomo A, Rivai AR,
Overtoom R, 1984. Duodenal string-capsule culture compared
with bone-marrow, blood and rectal-swab cultures for diagnos-
ing typhoid and paratyphoid fever. J Infect Dis 149: 157161.
3. Cheong YM, Jegathsan M, 1989. Standardisation of the Widal
test. Med J Malaysia 44: 267270.
4. Clegg A, Passey M, Omena M, Karigifa K, Suve N, 1994. Re-
evaluation of the Widal agglutination test in response to the
changing pattern of typhoid fever in the highlands of Papua
New Guinea. Acta Trop 57: 255263.
5. Levine MM, Grados O, Gilman RH, Woodward WE, Solis-Plaza
R, Waldman W, 1978. Diagnostic value of the Widal test in
areas endemic for typhoid fever. Am J Trop Med Hyg 27:
795800.
6. Pang T, Puthucheary SD, 1989. False positive Widal test in non-
typhoid salmonella infections. Southeast Asian J Trop Med
Public Health 20: 163164.
7. Proklopec J, Mullerova H, Sert V, Canh DT, Radkovsky J, 1991.
Typhoid fever survey in two localities in Vietnam. J Hyg Epi-
demiol Microbiol Immunol 35: 916.
8. Saha SK, Ruhulamin M, Hanif M, Islam M, Khan WA, 1996.
Interpretation of the Widal test in the diagnosis of typhoid
fever in Bangladeshi children. Ann Trop Paediatr 16: 7578.
9. Schroeder SA, 1968. Interpretation of serological tests for ty-
phoid fever. JAMA 206: 839840.
10. Senewiratne B, Senewiratne K, 1977. Reassessment of the Widal
test in the diagnosis of typhoid. Gastroenterology 73: 233236.
11. Sommerville PC, Lewis M, Koornhof HJ, Alberts M, Alberts
HW, Raymond R, 1981. The Widal test in the diagnosis of
typhoid fever in the Transvaal. South Afr Med J 59: 851854.
12. Wicks ACB, Holmes GS, Davidson L, 1971. Endemic typhoid
fever: a diagnostic pit-fall. Q J Med 40: 341354.
13. Gupta AK, Roa KM, 1979. Simultaneous detection of Salmonella
typhi antigen and antibody in serum by counter-immuno-
electrophoresis for an early and rapid diagnosis of typhoid
fever. J Immunol Methods 30: 249253.
14. Tsang RSW, Chau PY, 1981. Serological diagnosis of typhoid
fever by counter-immuno-electrophoresis. BMJ 282: 1505
1507.
15. Beasly WJ, Joseph SW, Weiss E, 1981. Improved serodiagnosis of
Salmonella enteric fever by an enzyme-linked immunosorbent
assay. J Clin Microbiol 13: 106114.
16. Carlsson HE, Lindberg AA, Hammarstrom S, 1972. Titration of
antibodies to salmonella O antigen by enzyme-linked immu-
nosorbent assay. Infect Immun 6: 703708.
17. Nardiello S, Pizzella T, Russo M, Galanti B, 1984. Serodiagnosis
of typhoid fever by enzyme-linked immunosorbent assay de-
termination of anti-Salmonella typhi lipopolysaccharide anti-
bodies. J Clin Microbiol 20: 718721.
18. Sippel JE, Mamay HK, Weiss E, Joseph SW, Beasley WJ, 1978.
Outer-membrane protein antigens in an enzyme-linked immu-
nosorbent assay for Salmonella enteric fever and meningococ-
cal meningitis. J Clin Microbiol 7: 372378.
19. Bhutta ZA, Mansurali N, 1999. Rapid serological diagnosis of
pediatric typhoid fever in an endemic area: a prospective com-
parative evaluation of two DOT-enzyme immunoassays and
the Widal test. Am J Trop Med Hyg 61: 654657.
20. Choo KE, Davis TME, Ismail A, Tuan Ibrahim TA, Ghazali
WNW, 1999. Rapid and reliable serological diagnosis of en-
teric fever: comparative sensitivity and specificity of Typhidot
and Typhidot-M tests in febrile Malaysian Children. Acta Trop
72: 175183.
HATTA AND OTHERS 420
21. Coovadia YM, Singh V, Bhana RH, Moodley N, 1986. Compari-
son of passive haemagglutination test with Widal agglutination
test for serological diagnosis of typhoid fever in an endemic
area. J Clin Pathol 39: 680683.
22. Ismail A, Kader ZS, Ong KH, 1991. Dot enzyme immunosorbent
assay for the serodiagnosis of typhoid fever. Southeast Asian
Trop Med Public Health 22: 563566.
23. Lim PL, Tam FCH, Cheong YM, Jegathesan M, 1998. One-step
2-minute test to detect typhoid-specific antibodies based on
particle separation in tubes. J Clin Microbiol 36: 22712278.
24. Quiroga T, Goycoolea M, Tagle R, Gonzales F, Rodriquez L &
Villarroel L 1992. Diagnosis of typhoid fever by two serologi-
cal methods. Enzyme-linked immunosorbent assay of antili-
popolysaccharide of Salmonella typhi antibodies and Widal
test. Diagn Microbiol Infect Dis 15: 651656.
25. Gussenhoven GC, van der Hoorn MAWG, Goris MGA, Terpstra
WJ, Hartskeerl RA, Mol BW, van Ingen CW, Smits HL, 1997.
Lepto dipstick, a dipstick assay for detection of Leptospira-
specific immunoglobulin M antibodies in human sera. J Clin
Microbiol 35: 9297.
26. Smits HL, Basahi MA, Diaz R, Marrodan T, Douglas JT, Rocha
A, Veerman J, Zheludkov MM, Witte OWM, de Jong J, Gus-
senhoven GC, Goris MGA, van der Hoorn MAWG, 1999.
Development and evaluation of a rapid dipstick assay for the
serodiagnosis of acute human brucellosis. J Clin Microbiol 37:
41794182.
27. Sanders EJ, Rigau-Perez JG, Smits HL, Deseda CC, Vorndam
VA, Aye T, Spiegel RA, Weyant RS, Bragg SL, 1999. Increase
in leptospirosis in dengue-negative patients after a hurricane in
Puerto Rico, 1996. Am J Trop Med Hyg 61: 399404.
28. Sehgal SC, Vijayachari S, Sharma S, Sugunan AP, 1999. LEPTO
Dipstick: a rapid and simple method for serodiagnosis of acute
leptospirosis. Trans R Soc Trop Med Hyg 93: 161164.
29. Yersin C, Bovet P, Smits HL, Perolat P, 1999. Field evaluation of
a one-step dipstick assay for the diagnosis of human lep-
tospirosis in the Seychelles. Trop Med Int Health 4: 3845.
30. Anonymous, 2000. Leptospirosis in India. Wkly Epidemiol Rec
75: 217223.
31. Cruickshank R, 1968. Medical Microbiology, A Guide to the
Laboratory Diagnosis and Control of Infection. Edinburgh: E
& S Livingstone.
32. Gribnau T, Roeles F, van der Biezen FJ, Leuvering J, Schuurs A,
1982 The application of colloidal dye particles as label in im-
munoassays: disperse dye immunoassay (DIA) Gribnau
TCJ, Visser J, Nivard RJF, eds. Affinity Chromatography and
Related Techniques. Amsterdam: Elsevier, 411424.
33. Gribnau T, Someren A, van Dinther F, 1983 DIA-disperse dye
immunoassay. Chaiken IM, Wilchek M, Parikh I, eds. Affinity
Chromatography and Biological Recognition. Orlando, FL:
Academic Press, 275280.
34. Kamat SA, Herzog C, 1977. Typhoid: clinical picture and re-
sponse to chloramphenicol. Prospective study in Bombay
(1972). Infection 5: 8591.
35. Stuart BM, Pullen RL 1946. Typhoid: Clinical analysis of three
hundred and sixty cases. Arch Intern Med 78: 629661.
36. Gasem MH, Dolmans WM, Isbandrio BB, Wahyono H, Keuter
M, Djokomoeljanto R, 1995. Culture of Salmonella typhi and
Salmonella paratyphi from blood and bone marrow in sus-
pected typhoid fever. Trop Geogr Med 47: 164167.
37. Parry CM, Hoa NTT, Diep TS, Wain J, Chinh NT, Vinh H, Hien
TT, White NJ, Farrar JJ, 1999. Value of a single-tube Widal
test in the diagnosis of typhoid fever in Vietnam. J Clin Mi-
crobiol 37: 28822886.
38. House D, Wain J, Ho VO, Diep TO, Chinh NT, Bay PV, Vinh H,
Duc M, Parry CM, Dougan G, White NJ, Hien TT, Farrar JJ,
2001. Serology of typhoid fever in an endemic area and its
relevance to diagnosis. J. Clin. Microbiol. 39: 10021007.
DIPSTICK ASSAY FOR THE SERODIAGNOSIS OF TYPHOID FEVER 421