Eur Spine J (2001) 10 : S96S101

DOI 10.1007/ s005860100282

T. Albrektsson
C. Johansson

Received: 15 February 2001

Accepted: 3 March 2001
Published online: 30 June 2001
Springer-Verlag 2001

T. Albrektsson () C. Johansson
Department of Biomaterials/
Handicap Research, P.O. Box 412,
SE 405 30 Gothenburg, Sweden
e-mail: tomas.albrektsson@hkf.gu.se

O R I G I N A L A RT I C L E

Osteoinduction, osteoconduction
and osseointegration

Abstract Osteoinduction is the

process by which osteogenesis is induced. It is a phenomenon regularly
seen in any type of bone healing
process. Osteoinduction implies the
recruitment of immature cells and
the stimulation of these cells to develop into preosteoblasts. In a bone
healing situation such as a fracture,
the majority of bone healing is dependent on osteoinduction. Osteoconduction means that bone grows
on a surface. This phenomenon is
regularly seen in the case of bone
implants. Implant materials of low
biocompatibility such as copper, silver
and bone cement shows little or no

Introduction
The terms osteoinduction, osteoconduction and osseointegration are frequently, but not always correctly, used
terms in many orthopaedic papers. To give but one example of incorrect terminology, arthroplasties are commonly
claimed to be osseointegrated based only on radiographic
evidence, despite the fact that the resolution of radiography alone is too poor to determine whether an implant is
osseointegrated or not. The aim of this paper is to first
briefly explain and define these terms and then to look
at them in some detail. Osteoinduction, osteoconduction
and osseointegration are now the subject of much discussion, e.g. in connection with bone morphogenic proteins
(BMP), bone growth factors and direct bone anchorage,
respectively. Suggested definitions of the terms osteoinduction, osteoconduction and osseointegration read as follows:

osteoconduction. Osseointegration is
the stable anchorage of an implant
achieved by direct bone-to-implant
contact. In craniofacial implantology,
this mode of anchorage is the only
one for which high success rates
have been reported. Osseointegration
is possible in other parts of the body,
but its importance for the anchorage
of major arthroplasties is under debate. Ingrowth of bone in a porouscoated prosthesis may or may not
represent osseointegration.
Keywords Osteoinduction
Osteoconduction Osseointegration

Osteoinduction. This term means that primitive, undifferentiated and pluripotent cells are somehow stimulated to
develop into the bone-forming cell lineage. One proposed
definition is the process by which osteogenesis is induced
[43].
Osteoconduction. This term means that bone grows on a
surface. An osteoconductive surface is one that permits
bone growth on its surface or down into pores, channels
or pipes. Wilson-Hench [43] has suggested that osteoconduction is the process by which bone is directed so as to
conform to a materials surface. However, Glantz [18] has
pointed out that this way of looking at bone conduction is
somewhat restricted, since the original definition bears little or no relation to biomaterials.
Osseointegration. This was first described by Brnemark
and co-workers [12]. The term was first defined in a paper
by Albrektsson et al. [4] as direct contact (at the light mi-

S97

croscope level) between living bone and implant. Osseointegration is also histologically defined in Dorlands Illustrated Medical Dictionary as the direct anchorage of an
implant by the formation of bony tissue around the implant without the growth of fibrous tissue at the boneimplant interface. Since the histological definitions have
some shortcomings, mainly that they have a limited clinical application, another more biomechanically oriented definition of osseointegration has been suggested: A process
whereby clinically asymptomatic rigid fixation of alloplastic materials is achieved, and maintained, in bone during
functional loading [46]. The rigid fixation of an implant
in orthopaedic praxis can be determined using radio-stereophotogrammetic (RSA) techniques and, at least in craniofacial implantology, resonance frequency analysis (RFA)
[28].

Osteoinduction and its importance for bone healing

In addition to the differentiated bone cells, i.e. osteoblasts,
osteoclasts and osteocytes, bone and adjacent tissues contain a number of less differentiated cells. These undifferentiated cells are of utmost importance for proper bone
healing or anchorage of an implant, since they can be recruited to form osteoprogenitor cells [45] and, with time,
develop into differentiated bone cells (Fig. 1). With the correct stimulus (the inductive agent), an undifferentiated mesenchymal cell can be transformed into a preosteoblast, a
process which constitutes bone induction. The classical papers describing bone induction at various host sites were
published a long time ago [20, 25, 40]. These authors used
gall bladder epithelium, alcohol extracts of bone and transplants to muscles or the anterior chamber of the eye, re-

Fig. 1 At the time of injury, adequate cells for bone repair are both
undifferentiated and differentiated bone cells. The majority of newly
formed bone depends on the undifferentiated cells that are induced
to become preosteoblasts

Fig. 2 The best way to demonstrate whether a specific agent is osteoinductive is to inject it into a soft tissue pouch, where bone formation does not occur under normal conditions. BMP-7 induced
bone formation 19 days after injection into a subcutaneous site in
a rat. Toluidine blue. Bar, 100 m

spectively, to demonstrate heterotopic bone formation. The

safest way to demonstrate whether a particular agent is osteoinductive or not is still to inject it into a heterotopic bed
such as a muscle pouch and to analyse any potential bone
formation (Fig. 2). Inductive agents naturally function in
bone surroundings too, but it is difficult to differentiate
between bone induction and bone conduction in an orthotopic site.
More modern research into osteoinduction dates back
to Urists experiment in the mid-1960s [39]. Demineralised
bone was used as an osteoinductive agent. Later, Urist et
al. [41] isolated a soluble glycoprotein called BMP as the
inductive agent. The BMP belong to the transforming
growth factor (TGF)--family of growth factors. There are
at least 15 different BMP [34], of which BMP-2 and
BMP-7 seem to be particularly interesting. To date, a great
number of research projects involving various types of
BMP are being conducted (for reviews, see [24, 34]). BMP
are naturally released in response to trauma or at bone remodelling and are the only known inducive agents [26].
However, physical stimuli such as stress or types of electrical signals otherwise applied have been regarded as, directly or indirectly, influencing bone induction [10, 13,
14, 44].
Osteoinduction, i.e. the recruitment of immature cells
and the stimulation of these cells to develop into preosteoblasts, is a basic biological mechanism that occurs
regularly, e.g. in fracture healing and implant incorporation. Even if pre-existing osteoblasts (i.e. before the injury) may help to form new bone, it is generally agreed
that such pre-existing cells only contribute a minor portion of the new bone needed in a fracture-healing situation
[16, 17]. According to Frost [16, 17] (Fig. 3), the inevitable
bone, marrow and soft tissue injury triggers the subsequent
repair by sensitising different types of surviving cells. Si-

S98

Fig. 3 According to Frosts

theory, injury triggers off a
healing response by the release
of growth factors and sensitising of cells. This is a primitive
healing response with stimulation of many different types of
cells

multaneously, the injury releases local, biochemical and

biophysical messengers that help cells to respond and that
guide them to respond in the proper manner. Some of these
messengers guide the differentiation and organisation of
cells, while others provide mitogens. This initial part of the
healing response thus includes osteoinduction, a process
that starts immediately after the injury and is very active
during the first week thereafter, even though the action of
the newly recruited preosteoblasts is not obvious until several weeks later, in the callus stage.

Osteoconduction and its importance for bone healing

Bone growth on an implant surface depends on the action
of differentiated bone cells. These cells may originate either in pre-existing preosteoblasts/osteoblasts that are activated by trauma or in cells recruited from primitive mesenchymal cells by osteoinduction [16, 17]. In the practical
situation, therefore, osteoconduction (Fig. 4) depends to a
fairly large extent on previous osteoinduction. The debate
concerning whether or not a particular biomaterial acts as
an osteoinductor may be slightly academic, since the injury at placement is sufficient to recruit previously undifferentiated bone cells.
Various types of bone growth factors are necessary for
bone formation. Furthermore, bone growth, including bone
conduction, does not occur without a proper blood supply.
Albrektsson [1] studied bone conduction and remodelling
in vivo and came to the conclusion that so-called full vascularisation was necessary for bone formation. It is there-

Fig. 4 In biomaterials science, osteoconduction means growth of

bone on the surface of a foreign material, as seen in the lower part
of this titanium screw implant (arrows). Distance between thread
peaks, 600 m

S99

fore not surprising that the principal action of many growth

factors is both mitogenic and angiogenic [37]. Growth factors that regulate bone tissue in one way or another include
insulin-like growth factor (IGF I, II), fibroblast growth
factor (FGF), TGF- and platelet-derived growth factor
(PDGF). The IGF are also called somatomedins. The growth
factors are small proteins that serve as signalling agents
for cells [37] (see the paper by Lind, this volume, for a
more detailed discussion of various growth factors).
However, in the case of implants, bone conduction is
not only dependent on conditions for bone repair, but also
on the biomaterial used and its reactions. Bone conduction
is not possible on certain materials such as copper and silver [3]. However, bone conduction is seen with biomaterials not regarded as ideal from the point of view of biocompatibility, such as stainless steel [22] and obviously
materials of high biocompatibility such as commercially
pure (c.p.) titanium. Bone conduction on implants may be
quantified. There is a significant difference in the amount
of bone that grows on seemingly similar materials such as
c.p. titanium and titanium 6-aluminum 4-vanadium [23].
However, the clinical implications of this difference remain
unknown.

Osseointegration of implants
Brnemark, who introduced this term, suggested the
spelling osseointegration instead of osteointegration,
and the original spelling is preferred in this paper. Osseointegration is not an isolated phenomenon, but instead
depends on previous osteoinduction and osteoconduction.
Thus materials that are too toxic to allow osteoconduction
will not be osseointegrated either. However, many materials show at least some bone attachment, which has inspired
bone pathologists to regard osseointegration as a simple
foreign body reaction [15], whereas more clinically oriented
scientists have rejected such a view. Osseointegrated implants have undergone a real breakthrough in oral and
craniofacial implantology, yielding excellent functional results, in contrast to alternatively anchored implants, which
have generally shown very poor success rates [6, 12, 35,
36]. Even if initial osseointegration is dependent on bone
induction and conduction, the term implies that the bone
anchorage is maintained over time. Cylindrical implant
designs (without threads), rough plasma-sprayed surfaces
and overloading represent factors that may lead to secondary failure of osseointegration [2, 4].
The ultrastructure of the bonetitanium interface in osseointegration demonstrates an amorphous layer from 20
40 to 500 nm thick. Some investigators [5] have described
collagen and calcified tissue in this zone, whereas others
[32] have failed to verify these findings. This zone is too
narrow to be seen at the light microscope level of resolution. At the light microscope level, direct bone contact,
osteogenesis and bone resorption occur simultaneously

Fig. 5 Simultaneous bone formation and resorption at the interface between bone (B) and a commercially pure titanium (c.p. Ti)
implant. There are three cavities arranged in a horizontal line in the
middle of the figure. In the left cavity, red dominates, i.e. positive
staining for acid phosphatase meaning active bone resorption. In
the middle cavity, blue dominates, i.e. positive staining for alkaline
phosphatase, meaning active bone formation. In the right cavity,
there is red and blue staining, i.e. both bone formation and resorption. Bar, 100 m

[31] (Fig. 5). From a purely biomechanical viewpoint,

Skalak and Zhao [33] have demonstrated that when a hole
slightly smaller than the implant diameter is prepared for
implant placement, force-fitting stress increases installation torque and initial stability can be induced at a similar
magnitude as seen with roughened implants.
Oral implants retrieved from patients despite remaining stability have shown that there does not seem to be
100% bone attachment. Implants retrieved after clinical
function for up to 17 years showed an average of 7080%
bone contact with an absolute minimum of 60% [7]. Functioning osseointegrated implants demonstrate interfacial
bone density similar to that of the bone in which the implant was implanted [38]. Even if long-term functioning
osseointegrated implants show what seems to be similar
bone tissue reactions, osseointegration might be able to be
achieved more rapidly than otherwise observed. Such potentially accelerated osseointegration has been indicated
by results from experiments with hydroxyapatite coating
[19], using intermediary roughened implants [42], after hyperbaric oxygen treatment [30] or by using anodised c.p.
titanium with artificially enhanced oxide layers [9]. Acceleration of osseointegration may depend on the removal
of negative tissue conditions or optimisation of the biomaterial rather than on an actual increase in the rate of bone
response.
Much less attention has been paid to the possibility of
establishing osseointegration in orthopaedic surgery than
in oral and craniofacial surgery. The original notion that
polymerised bone cement may be histologically osseointegrated has not been confirmed in more recent investigations [29]. Histological sections to reveal true bone-to-im-

S100

loading. It is known that interfacial implant movement of

more than 150 m will inevitably lead to soft tissue formation instead of bone, for instance [11]. Even if one- or
two-point bone contact can be demonstrated, this need not
represent actual osseointegration of the entire implant.
Screw-type implants inserted using a modified minimally traumatising technique have been convincingly osseointegrated, e.g. in hip arthroplasties [8] and interphalangeal implants [27] or vertebral screws (Fig. 6). However, whether or not osseointegration will become as important a type of anchorage in orthopaedics as in oral and
craniofacial implantology will depend on the reported longterm clinical results of this type of anchorage.
Fig. 6 Hydroxyapatite-coated vertebral screw in a goat. Osseointegration is evident. Bar, 1000 m (Courtesy of Dr. B. Sandn,
Uppsala University)

plant contact need to be quite thin (of the order of 10

20 m) to really reveal osseointegration. Thicker sections
have a shadow effect [21] that make it impossible to state
whether or not true direct bone contact has been achieved.
Apart from poor resolution, this is the reason why common radiographs are of little value in the diagnosis of osseointegration. The question is whether it is really possible to establish osseointegration of conventional orthopaedic
arthroplasties with the combined use of less biocompatible materials, interfacial heat due to curing bone cement,
drilling or reaming without a cooling agent and too rapid

Conclusion
Osteoinduction, osteoconduction and osseointegration are
interrelated, but not identical phenomena. Osteoinduction
is part of normal bone healing and is responsible for the
majority of newly formed bone, e.g. after a fracture or the
insertion of an implant. The implant itself may be osteoinductive, but this is not a prerequisite for bone induction.
Osteoconduction is a term now usually used in conjunction with implants. Osteoconduction and osseointegration
both depend not only on biological factors, but also on the
response to a foreign material. The osteoconductive response may be rather short lived, but successful osseointegration maintains its bone anchorage over a long period.

References
1. Albrektsson T (1980) The healing of
autologous bone grafts after varying
degrees of surgical trauma. J Bone
Joint Surg [Br] 32:403410
2. Albrektsson T (1993) On long-term
maintenance of the osseointegrated response. Aust Prosthodont J 7 [Suppl]:
1524
3. Albrektsson T (1995) Principles of osseointegration. In: Hobkirk JA, Watson
K (eds) Dental and maxillofacial implantology. Mosby-Wolfe, London,
pp 919
4. Albrektsson T, Brnemark PI, Hansson
HA, Lindstrm J (1981) Osseointegrated titanium implants. Requirements
for ensuring a long-lasting, direct bone
anchorage in man. Acta Orthop Scand
52:155170
5. Albrektsson T, Brnemark PI, Hansson
HA, Ivarsson B, Jnsson U (1982)
Ultrastructural analysis of the interface
zone of titanium and gold implants.
Adv Biomaterials 4:167177

6. Albrektsson T, Brnemark PI, Hansson

HA, Kasemo B, Larsson K, Lundstrm
I, McQueen D, Skalak R (1983) The
interface zone of inorganic implants in
vivo: titanium implants in bone. Ann
Biomed Eng 11:127
7. Albrektsson T, Eriksson AR, Friberg
B, Lekholm U, Lindahl L, Nevins M,
Oikarinen V, Roos J, Sennerby L,
strand P (1993) Histologic investigations on 33 retrieved Nobelpharma implants. Clinical Implants 12:19
8. Albrektsson T, Carlsson LV, Jacobsson
M, Macdonald W (1998) The Gothenburg osseointegrated hip arthroplasty:
experience with a novel type of hip design. Clin Orth Rel Res 352:8194
9. Albrektsson T, Johansson C, Lundgren
AK, Sul YT, Gottlow J (2000) Experimental studies on oxidized implants. A
histomorphometrical and biomechanical analysis. Appl Osseointegration
Res 1:2124
10. Bassett CAL (1968) Biological significance of piezo-electricity. Calcif Tissue Res 1:252261

11. Brunski JB (1988) The influence of

force, motion and related quantities on
the response of bone to implants. In:
Fitzgerald AF (ed) Non-cemented total
hip arthroplasty. Raven, New York,
pp 719
12. Brnemark PI, Hansson BO, Adell R,
Breine U, Lindstrm J, Halln O,
hman A (1977) Osseointegrated titannium implants in the treatment of the
edentulous jaw. Scand J Plast Reconstr
Surg 11 [Suppl 16]:1175
13. Buch F (1985) On electrical stimulation of bone tissue. PhD thesis, Dept of
Biomaterials/Handicap Research, University of Gothenburg, Sweden, pp 1
139
14. Dealler SF (1981) Electrical phenomena associated with bones and fractures
and the therapeutic use of electricity in
fracture healing. J Med Eng Tech 5:
7382

S101

15. Donath K, Laass M, Gnzl HJ (1992 )

The histopathology of different foreign-body reactions in oral soft tissue
and bone tissue. Virchows Archiv [A]
420:131137
16. Frost HM (1989) The biology of fracture healing. An overview for clinicians, part I. Clin Orthop Rel Res 248:
283293
17. Frost HM (1989) The biology of fracture healing. An overview for clinicans, part II. Clin Orthop Rel Res 248:
294309
18. Glantz PO (1987) Comment. In:
Williams DF (ed) Progress in biomedical engineering, vol 4. Definitions in
biomaterials. Elsevier, Amsterdam, p 24
19. Gottlander M (1994) On hard-tissue
reactions to hydroxyapatite-coated titanium implants. PhD thesis, Dept of
Biomaterials/Handicap Research, University of Gteborg, Sweden, pp 1192
20. Huggins CB (1931) The formation of
bone under the influence of epithelium
of the urinary tract. Arch Surg 22:377
386
21. Johansson CB(1991) On tissue reactions
to metal implants. PhD thesis, Dept of
Biomaterials/Handicap Research, University of Gteborg, Sweden, pp 1203
22. Johansson C, Albrektsson T, Roos A
(1992) A biomechanical and histomorphometric comparison between different types of bone implants evaluated in
a rabbit model. Eur J Exp Musculoskel
Research 1:5161
23. Johansson C, Han CH, Wennerberg A,
Albrektsson T (1998) A quantitative
comparison of machined commercially
pure titanium and titanium-aluminumvanadium implants in rabbit bone. Int J
Oral Maxillofac Implants 13:315321
24. Korhonen LK, Vnnen K (1992)
Bone inductive molecules and enhancement of repair and regeneration.
In: Hall BK (ed) Fracture repair and
regeneration, vol 5. CRC, Boca Raton,
pp 209232

25. Levander G (1938) A study of bone regeneration. Surg Gynec Obstet 67:705
714
26. Lind M (1996) Growth factors: possible
new clinical tools. Acta Orthop Scand
67:407417
27. Lundborg G, Brnemark PI (2000) Osseointegrated proximal interphalangeal
joint prostheses with a replaceable
flexible joint spacer long term results.
Scand J Plast Reconstr Surg 34:345
353
28. Meredith N (1997) On the clinical
measurement of implant stability and
osseointegration. PhD thesis, Dept of
Biomaterials/Handicap Research, University of Gteborg, Sweden, pp 1200
29. Morberg P (1991) On bone tissue reactions to acrylic cement. PhD thesis,
Dept of Biomaterials/Handicap Research, University of Gteborg, Sweden,
pp 1139
30. Nilsson LP, Granstrm G, Albrektsson
T (1987) The effect of hyperbaric oxygen treatment of bone regeneration. An
experimental study in the rabbit. Int J
Oral Maxillofac Implants 3:4348
31. Rser K, Johansson C, Donath K,
Albrektsson T (2000) A new approach
to demonstate cellular activity in bone
formation adjacent to implants. J Biomed Mater Res 51:280291
32. Sennerby L, Ericson LE, Thomsen P,
Lekholm U, strand P (1991) Structure of the bone titanium interface in
retrieved clinical oral implants. Clin
Oral Impl Res 2:103111
33. Skalak R, Zhao Y (2000) Interaction of
force-fitting and surface roughness of
implants. Clin Implant Dent Rel Res
2:219224
34. Solheim E (1998) Growth factors in
bone. Int Orthop (SICOT) 22:410416
35. Tjellstrm A, Lindstrm J, Halln O,
Albrektsson T, Brnemark PI (1981)
Osseointegrated titanium implants in
the temporal bone. A clinical study on
bone-anchored hearing aids. Am J Otol
2:303310

36. Tjellstrm A, Lindstrm J, Nyln O,

Albrektsson T, Brnemark P-I,
Birgersson B, Nero H, Sylvn C (1981)
The bone-anchored auricular epithesis.
Laryngoscope 91:811815
37. Trippel SE, Coutts RD, Einhorn TA,
Mundy GR, Rosenfeld RG (1996)
Growth factors as therapeutic agents.
J Bone Joint Surg [Am] 78:12721286
38. Tsuboi N, Sennerby L, Johansson C,
Albrektsson T, Tsuboi Y, Iizika T
(1996) Histomorphometric analysis of
bone-titanium interface in human retrieved implants. In: Ueda M (ed) Proceedings of the third international congress on tissue integration in oral and
maxillofacial reconstruction. Quintessence, Tokyo, pp 8485
39. Urist MR (1965) Bone: formation by
autoinduction. Science 150:893899
40. Urist MR, McLean FC (1952) Osteogenic potency and new bone formation by induction in transplants to the
anterior chamber of the eye. J Bone
Joint Surg [Am] 34:443452
41. Urist MR, Mikulski A, Lietze A (1979)
Solubilized bone morphogenetic protein. Proc Natl Acad Sci USA 76:
18281832
42. Wennerberg A (1996) On surface
roughness and implant incorporation.
PhD thesis, Dept of Biomaterials/
Handicap Research, University of
Gteborg, Sweden, pp 1212
43. Wilson-Hench J (1987) Osteoinduction. In: Williams DF (ed) Progress in
biomedical engineering, vol 4. Definitions in biomaterials. Elsevier, Amsterdam, p 29
44. Yasuda I (1953) Fundamental aspects
of fracture treatment. J Kyoto Med Soc
4:395404
45. Young RW (1963) Nucleic acids, protein synthesis and bone. Clin Orthop
Rel Res 26:147156
46. Zarb G, Albrektsson T (1991) Osseointegration a requiem for the periodontal ligament? An editorial. Int J
Periodont Rest Dentistry 11:8891