Alveolar bone and the bisphosphonates


Alastair N Goss
Professor and Director
Oral and Maxillofacial Surgery Unit
The University of Adelaide
Adelaide SA 5005


Bisphosphonate associated osteonecrosis of the jaws (ONJ) usually commences at the alveolus. Comparison is made between the structure and function of long bones and alveolar bone and the differing susceptibilities of the bisphosphonates at these different sites are explored. Current concepts of the causation of ONJ are discussed. The clinical implications of these findings to dentists managing periodontal conditions are presented.

Abbreviations and acronyms

bone mineral density


computerized tomography


serum beta cross laps assay


osteonecrosis of the jaws


Healthy alveolar bone is an essential component of an intact dentition. Chronic periodontitis, which leads to progressive destruction of alveolar bone, is the most common cause of tooth loss in the over fifties.1,2 Once the teeth are lost the alveolar bone disappears. Alveolar bone is subject to a number of local and systemic conditions which will result in its progressive loss. Consideration of the complex interplay of most of these factors is beyond the scope of this paper but is covered elsewhere in the supplement.3 An important systemic factor is generalized bone diseases and their treatment, with some, such as osteoporosis, being very common.4

Five per cent of all Australians suffer from osteoporosis and the involvement will increase with age, particularly in females. It is estimated that over 50 per cent of all females over the age of 60 suffer osteoporosis. A number of preventive and therapeutic strategies exist to medically manage osteoporosis.5 A key drug group involved is the bisphosphonates which are effective in stabilizing osteoporosis and reducing the risk of low impact fractures in the elderly. An unfortunate complication of these drugs is the development of osteonecrosis of the jaws (ONJ) particularly following alveolar bone invasive dental treatment. Accordingly, given the large number of patients being treated for osteoporosis, there is a strong likelihood of dentists and periodontists treating a patient at risk for this complication.

This review examines the key elements of the interaction between alveolar bone, the bisphosphonates and ONJ.

Alveolar bone

Alveolar bone exists to support the teeth. Its structure varies between individuals and generally it gets denser with age.6 Broadly, there is a dense bone wall near the gingivae and then the middle portion of the tooth root. There are larger marrow spaces near the tooth apex.7 The alveolar bone walls at the attachment of the periodontal membrane have a cribiform structure with open channels. The bone structure follows that of bone structure throughout the body with cortical bone containing osteons and Haversian systems. New bone is formed in a lamellar structure by osteoblasts with the osteocytes being incorporated within the bone. Older bone, or bone in the path of erupting or moving teeth is resorbed by osteoclasts. In keeping with all bone in the body, alveolar bone is a dynamic structure with the bone constantly remodelling and adapting to functional needs.

The key question however, is whether alveolar bone is exactly the same as the long bones or whether it is subtly different. Alveolar bone develops as a membrane bone whereas the limbs and vertebrae develop as endochondral bones. The mandible is of neural crest origin whereas the limbs and vertebral column are of mesodermal origin.8 There are minor phenotypic differences between osteoblasts depending on their site of origin and anatomical location, which can be demonstrated biochemically.9 Membrane bone osteoblasts also have an increased rate of cell division as compared to iliac crest osteoblasts.10

Osteoclasts are derived from mononuclear precursor cells which migrate from the bone marrow via the vasculature to the bone site. Their function is dictated largely by interaction with the osteoblasts in the area. There are biochemical differences between osteoclasts of membrane bone origin and long bone origin.11 There are also differences in behaviour between giant cell tumours of the jaws and of the long bones.12 The overall rate of turnover of alveolar bone is 10 times greater than that of the long bones.13

Thus, there are key differences in behaviour between the alveolar bones and the long bones. This is summarized in Table 1.

Table 1.   Comparison of long bone to alveolar bone and the effects of the bisphosphonates
TissueLong boneAlveolar boneBisphosphonate effect on alveolar bone
Neural crest
Osteoblasts9,10Divide slowlyDivide faster
Increased bone depression
Osteocytes8Respond slowly to mechanical stressRespond faster to mechanical stressIncreased bone depression
Osteoclasts11,12Site dependent differences
Haemopoietic origin
Some osteoblast fusion
Less resorptive activity
More nuclei
Giant cell tumours rarer but aggressive
Site dependent differences
Haemopoietic origin
? more osteoblast fusion
Greater resorptive activity
Small cell
Giant cell tumours common and less aggressive
Increased bone depression
Bone turnover13Slowx10 faster than long bonesIncreased bone depression
Protection from injuryDeeply covered in soft tissue
Uncommonly exposed
Mucoperiosteal cover only
Commonly exposed
Alveolus more prone to exposure to bacterial contamination
Angiogenesis32,33Low vascularHigh vascularDecreased healing capacity
Keratinocytes34Not involved in bone healingInvolved in mucoperiosteal healingDecreased healing capacity

Systemic bone diseases and alveolar bone

The most common bone disease is osteoporosis.4,5 This occurs most frequently in post-menopausal females but also occurs in men and younger individuals on steroids.14 It is a chronic progressive disorder in which bone resorption exceeds formation. The resulting decreased bone mass and loss of microarchitecture results in increased susceptibility to bone fractures. The bone fractures may be spontaneous or low impact types.15 There are community impacts with resulting financial burdens for the state and families as well as the affected individuals. For example, females over the age of 70 years with an osteoporotic hip fracture have a 25 per cent chance of being dead within 12 months.

Osteoporosis is primarily diagnosed by a bone mineral density (BMD) test which compares the bone density of the lumbar vertebrae and hip to standard values obtained from young white American females. One standard deviation from the mean is normal, two standard deviations is osteopenia and three standard deviations is osteoporosis. The presence or absence of fractures also contributes to the diagnostic stage.16 BMD is not a sensitive test and requires a minimum of 12 months between tests to show effect. A serum assay for the presence of C-telopeptide fragments of collagen-I which are generated by osteoclastic activity (CTX test or serum cross laps assay) is a measure of bone turnover which can also be employed, particularly to show the effects of the hormones and the bisphosphonates on bones.17 It will show changes within six weeks with higher activity during active bone loss. Prevention and management of osteoporosis is multifactorial and has recently been reviewed in Australia by the National Prescribing Service.5 A key element is the bisphosphonates which are a group of anti-resorptive drugs. Oral bisphosphonates, most commonly alendronate and risedronate, are extensively used with over 300 000 prescriptions issued for oral bisphosphonates in 2007.

Most bone physicians are of the view that “osteoporosis does not occur in the head and neck”. This is clinically true as spontaneous or low impact fractures of the facial bones do not present to bone physicians. Maxillofacial surgeons usually ascribe facial fractures in the elderly to general frailty, poor balance and advanced alveolar resorbtion rather than primarily to osteoporosis. However, a number of studies have looked at the associations between osteoporosis and periodontal disease. These show generalized osteoporosis is associated with tooth loss in otherwise healthy post-menopausal females.18–20 However, a reciprocal study which examined tooth loss as a predictor of osteoporosis, found no direct association.21 Similarly, generalized osteoporosis is associated with alveolar resorption in the edentulous.22

Paget’s disease is a common metabolic disease of bone which mainly affects the middle-aged and elderly. It is mostly asymptomatic but may present as pain, bone enlargement and low impact fractures. Imaging, particularly by CT, is valuable with bone scans showing areas of activity. A CTX test will show high levels when active resorption is occurring. Oral or intravenous bisphosphonates are useful for controlling the symptoms.

Paget’s disease involves the jaws in approximately 20 per cent of advanced cases, mainly the maxilla, sometimes the maxilla and mandible and only occasionally the mandible alone. The main feature is alveolar enlargement with spreading of the teeth in dentate individuals and broader large alveolar processes in the edentulous, where the patient’s dentures no longer fit. Radiographically, there is a mixed radiopacity and radiolucency or “cotton wool” appearance. Histologically, there is a wide variation ranging from dense inactive cortical bone to rapid bone turnover and vascularization. A typical “mosaic bone” pattern is seen in both the alveolus and other bony sites reflecting the histological changes caused by the increased turnover.23

Malignancies which metastasize to the bones such as the adenocarcinomas of the breast, prostate and gut will result in osteolytic activity, both from direct effects of invading malignant cells and from paracrine secretion of factors which activate osteoclasts. These can be imaged radiographically, by a CT or bone scan. The CTX level is high. Treatment is by chemotherapeutic agents, steroids and intravenous bisphosphonates. Multiple myeloma is a primary malignant neoplasm of bone marrow. The tumour consists of neoplastic plasma cells which destroy the bones, particularly of the axial skeleton. In addition, there is increased bone resorption from paracrine secretion of cytokines such as IL-6. The CTX level is high. Treatment is by chemotherapeutic drugs, steroids, intravenous bisphosphonates, localized radiotherapy and bone marrow transplants.

Metastatic tumours to the bones and multiple myeloma do not primarily involve the alveolus but settle in the marrow, commonly near the mandibular angle.23 Metastasis may enlarge to involve the alveolus, hence any loosening of teeth, mandibular sensory change and radiographic bone changes, particularly in patients with known adenocarcinomas, must be treated with suspicion that they are a metastatic deposit. It is uncommon that the jaws are involved alone with widespread bone involvement being the norm.

The bisphosphonates

The bisphosphonates and their implications to dentistry have been extensively reviewed.24 Their development and therapeutic use has also been described in detail in the medical literature.5,25

In brief, the nitrogen containing agents come as both oral and intravenous preparations. The most common oral preparations are alendronate (Fosamax) and risedronate (Actonel) which are primarily used for osteoporosis. These are usually taken weekly. The intravenous preparations, pamidronate (Aredia, Pamisol) and zoledronate (Zometa, Aclasta) are considerably more potent and are used mainly to manage bony metastatic disease and multiple myeloma. These are usually administered monthly. Both oral and intravenous agents are used as indicated for Paget’s disease. Intravenous bisphosphonates are used in a single dose to reverse hypercalcaemia, usually secondary to bone metastatic malignancy, hyperparathyroidism or the humoral effects of neoplasia. Recently, an intravenous form of zoledronic acid (Aclasta) has been introduced to the Australian market for the treatment of osteoporosis, and is now approved as an annual treatment.

The mechanisms of action of the bisphosphonates in bone metabolism are complex and multifactorial. They act almost exclusively on bone because of their specific affinity to bone where they are deposited in newly formed bone and close to osteoclasts. Although the time in the circulation is short, 30 to 180 minutes, once incorporated into bone they can persist for up to 10 years.26,27 Different types of bisphosphonates have differing affinities to bone with the rank order from greatest to least being zoledronate, alendronate, ibandronate, risedronate, etidronate and clodronate.28

Once in the bone they directly affect mononuclear activity, which is the parent cell of osteoclasts, they disrupt osteoclast mediated, bone resorption and increase apoptosis of osteoclasts.29,30 This in turn reduces bone deposition by osteoblasts.31 The net effect of this is to reduce bone resorption and bone turnover.

Angiogenesis is reduced by depression of blood flow and a marked decrease in vascular endothelial growth factor.32,33 Epithelial keratinocytes are also inhibited.34 The net effect of these actions is to reduce healing capacity.

The bisphosphonates were extensively tested before they were released onto the market with a wide margin of safety. However, in 2003 the first reports of the severe complication of osteonecrosis of the jaws was reported in America31,35–37 and Australia.38 More recently, a wide range of other complications including atrial fibrillation,39 severe and incapacitating bone pain40 and spontaneous femur fracture41 have been reported.

Bisphosphonate associated osteonecrosis of the jaws

Bisphosphonate associated osteonecrosis of the jaws (ONJ) has been reviewed in detail in the Medications in Dentistry supplement to the Australian Dental Journal.24 Since the initial descriptions of the condition, and more detailed reviews, there have been numerous reports of ONJ with the condition becoming increasingly well characterized. Usually, ONJ commences in the alveolar bone and then spreads into the jaws. By definition, ONJ is an area of exposed bone in the jaws of more than eight weeks duration. Important exclusions are presence of malignancy at the site and that the jaws have not been irradiated.42 The condition is most common in elderly, medically compromised patients on intravenous bisphosphonates but may occur in medically well patients on oral bisphosphonates. The key concern with ONJ is that once established it is usually longstanding, debilitating to the patient and refractory to treatment.43

When ONJ was first described in 2003, it was thought to be rare, an avascular necrosis of jaw bone which related to the underlying bone disease and other jaw bone necroses and somewhat indirectly to the bisphosphonates. Considerable research both clinical and basic since that time has focused on the true nature of ONJ.

Firstly, it is not rare, although compared to the very large numbers of patients on the bisphosphonates, it is not common either. There is a difference between the relatively small number of patients on intravenous bisphosphonates for malignancy where the risk is high and the very large numbers of patients on oral bisphosphonates where the risk is much lower. However, a small percentage of a very large number is a large number of affected individuals. Estimates of risk for the oral bisphosphonate vary from 1 in 10 000 to 1 in 100 000.42 However, both of these estimates are overall population estimates and don’t take into account the role of trigger factors. The most common trigger is a dental extraction and if this is taken into account then the risk increases to the order of 1 in 1000.44 The more detailed the capture of all cases of ONJ, then the higher the risk. In South Australia, where it is straightforward to capture all cases of ONJ then the risk of ONJ was of the order of 1 in 296 for patients having an extraction.44 In more recent studies with good capture, the overall risk was reported as 0.03 per cent or 1 in 3333 but this did not factor in the extraction risk.45 The extraction risk was factored into a San Diego study which had the highest reported rate at 4 per cent to date.46 Thus, for patients on oral bisphosphonates having dental extractions there is an identifiable risk, probably of the order of 1 in 300 which given the very large number of patients on oral bisphosphonates, means large numbers of ONJ sufferers.

With intravenous bisphosphonates though there is less discussion of the risk, it is agreed it is high. Estimate range is 1 to 10 per cent of dental extractions for patients on intravenous bisphosphonates.44 Once established the condition is usually longstanding and refractory.

The triggers for the development of ONJ are sometimes spontaneous but usually involve bone invasive procedures, particularly tooth extractions. The various published estimates of the percentage involvement of trigger factors are presented in Table 2. There is controversy as to the involvement of periodontal factors and it is difficult to determine whether it is the presence of periodontal disease or the effect of bone-exposing periodontal treatment, such as periodontal surgery. In the authors’ experience, periodontal disease was commonly present but this was the indication for extractions, which was the trigger.47 We have not seen any cases initiated by periodontal treatment such as scaling or root planing but have seen cases relating to dental implants.48

Table 2.   Precipitating factors for osteonecrosis of the jaws
AuthorSpontaneousExtractionAdvance periodontitisPeriodontal flap surgeryImplantsEndodonticsDentures
  1. Yarom et al.69 also identified heavy tobacco use as a risk factor for ONJ. Smoking is a known risk factor for periodontal disease, dry socket and implant failure.

Marx et al.67
 n = 119
25.2% 37.8%28.6%11.2%    3.4%0.8%0
Carter, Doeke, Goss68
 n = 5
 n = 11
Mavrokokki et al.44
 n = 91

The nature of ONJ is now considered not an area of avascular necrosis nor an osteomyelitis but an inability of the alveolar bone to respond to injury. When a tooth is extracted there is a deep and narrow wound in the bone which is contaminated by bacterially infested saliva. Normally, the local immune reaction and the bone react swiftly to repair the wound. Macrophages and other inflammatory cells counter the bacterial contamination, osteoclasts remove any damaged bone, osteoblasts form new bone and the epithelium regrows over the wound. The bisphosphonates inhibit all of these functions with a greater effect on alveolar bone (with its high bone turnover, increased vascularity, more susceptible osteoclasts, osteoblasts, mononuclear cells and keratinocytes) as compared to the endochondral bones. This increased effect in the alveolus has been neatly encapsulated in the statement that “the correct dose of bisphosphonate for the long bones is an overdose for the alveolar bone of the jaws”43 (Table 1).

The dose of bisphosphonates varies with the rate of administration and type with zoledronic acid being the most potent and highest affinity to bone and thus not surprisingly the highest rate of ONJ.44 The median dose to onset was 12 months and the authors have seen cases occurring following three injections at monthly intervals.44 The two main oral bisphosphonates are alendronate and risedronate. Different rates of development of ONJ have been reported with more risk with alendronate. This is more than the difference in market share. Alendronate has a greater bone affinity than risedronate.28 The recommended weekly dose of alendronate at 70 mg weekly is almost double the potency of the recommended dose of 35 mg risedronate. It was also initially felt that the oral bisphosphonate needed to have been prescribed for a considerable time for the cumulative dose to have built up to where ONJ may occur, commonly a window period of three years was suggested. However, the median period for the onset of ONJ from alendronate was 24 months in the large Australia-wide study.44 Thus, ONJ can and will occur in a shorter period. The shortest time seen by the authors was with a dental implant which lost integration within 12 weeks of commencement of alendronate.48

An insight as to why there is such variable time of onset from commencement of oral bisphosphonates to ONJ was gained from the recent CTX study.49 There was wide variation in the degree of decreased bone turnover in 215 patients presenting for tooth extractions; some (37 per cent) had very low bone turnover, which put them in the risk zone for ONJ, some (35 per cent) were in a more appropriate therapeutic range for the bisphosphonates and the remainder were in the normal range. This would appear to be beyond just a matter of compliance and didn’t correlate with age, gender, medical comorbidity, bone disease or duration of treatment. Thus not surprisingly, different individuals have different dose and duration drug requirements. There is also good evidence that many osteoporotic patients don’t need to be on bisphosphonates forever. The results of five years therapy and ceasing were similar to 10 years continuance of therapy.50

Lastly, and most importantly, there is the issue of whether bisphosphonates cause ONJ. Initially, the association was considered weak and this view is still supported by some medical authorities and parts of the pharmaceutical industry. Clinically however, ONJ has a unique presentation and only occurs in patients on bisphosphonates with the only exception being the historical condition of “phossy jaw”. The chemistry of the phosphorous agents used in the munitions and match industry in the late 1800s and early 1900s has been compared to the chemistry of the bisphosphonates. Essentially the yellow phosphorous, to which patients with “phossy jaw” were exposed to, is converted to potent amino bisphosphonates by natural chemical reactions in the body. These amino bisphosphonates are almost identical to current pharmaceutical bisphosphonates.51 Animal studies have shown that dogs on oral alendronate for three years show increased bone matrix necrosis as compared to controls. These animal studies, however, did not have extractions so ONJ was not seen.52 The corresponding author developed a research protocol for a sheep model with intravenous bisphosphonates with staged extractions but could not obtain funding for the bisphosphonates.

Thus, currently the direct association between ONJ and the bisphosphonates is indisputable but the precise mechanism of causation is suggested but not fully documented.

Case study

The various clinical, radiographic and tissue features presented in this review are illustrated by an ONJ case managed by the Oral and Maxillofacial Surgery Unit at The University of Adelaide.

A 47-year-old male agriculture worker first presented to his general medical practitioner at a town over 1000 km from Adelaide, with severe back pain. This failed to respond to standard musculoskeletal management and a radiographic anomaly was identified in his spine. Following referral to the Royal Adelaide Hospital, he was found to be suffering an advanced aggressive adenocarcinoma of the prostate with local, regional and widespread bone metastasis. Treatment by the multidisciplinary prostatic malignancy group, which consists of urologists, medical and radiation oncologists and palliative care specialists, was aimed at aggressive systemic palliation to limit spread, control pain and maintain quality of life. Curative treatment was not possible.

The initial treatment was chemotherapy, irradiation to shoulder, spinal and pelvic bone metastases and pain control. Prior to commencement of intravenous bisphosphonate, he was referred to the Oral and Maxillofacial Surgery Unit for assessment. He presented with severe right mandibular pain which was diagnosed as an acute exacerbation of chronic periodontitis (Fig 1a).

Figure 1.

 Sequential OPG radiographs. (a) At presentation showing periodontal bone loss and bifurcation involvement of the right mandibular molars. (b) 20 months later and 8 months after the onset of ONJ. Large right lingual sequestrum involving the right premolars. (c) 29 months after presentation showing ONJ involving the right hemimandible from angle to parasymphysis. Submandibular abscess. (d) 30 months after presentation showing resected area. Further sequestration has occurred at the parasymphysis.

The right first and second mandibular molars were extracted under local anaesthetic with antibiotic prophylaxis and primary closure. The socket healed uneventfully. He was commenced on monthly 4 mg zoledronic acid (Zometa) infusions.

Twelve months later he presented again with right mandibular pain and bone exposure of the lingual plate near the premolars. The CTX value was below 70 pg/ml. ONJ was diagnosed and symptomatic treatment commenced. This controlled the jaw pain.

At 20 months he presented with a large 4 cm × 1.5 cm sequestrum of the right mandibular lingual plate which was removed along with the first and second premolars (Fig 1b).

The resulting bone wound closed and was again comfortable. The bisphosphonate infusions were ceased.

At 29 months he returned with a severe right submandibular abscess with pus discharge through the skin and trismus (Fig 1c).

The right hemimandible was necrotic. The CTX value had improved to 272 pg/ml since the bisphosphonate had been ceased. Treatment consisted of incision and drainage of the submandibular abscess and prolonged intravenous antibiotics. Jaw resection was discussed but declined by the patient.

Initial pain and infection control was obtained but at 30 months, or one month following the severe infection, the abscess with orocutaneous pus discharge returned. The necrotic hemimandible was resected to bleeding bone at the angle and symphysis (Fig 1d). This controlled the pain and infection but with residual functional and aesthetic problems. A further 1 cm × 1 cm sequestrum spontaneously discharged into the mouth but the skin remained intact.

Possible future reconstructive surgery was discussed but this was declined.

Current oncologic management is palliative only as all chemotherapeutic and radiation treatments have been exhausted. In total he received 24 mg zoledronic acid from six infusions over six months (Royal Adelaide Hospital Pharmacy data). This commenced after his first extractions (Fig 1a) and ceased before the first major sequestration (Fig 1b). He had received no bisphosphonates for 20 months prior to the jaw resection. He had not received any radiotherapy to the head and neck but had been on several immunosuppressive drugs over three years. The patient indicated that the adverse effects of the ONJ on his quality of life as similar or greater than the effect of the bone mets in his spine, hip and pelvis.

The histopathology of his resected hemimandible is presented in Figs 2 and 3.

Figure 2.

 Macroscopic sections of the mandible. (a) Normal right mandible at the first molar. Note the relation of the alveolar bone to the tooth, the thick cortical plate with a distinct marrow space. (b) Resected specimen at the right canine. The marrow space is obliterated with dense bone. This is the effect of bisphosphonate on the mandible. The lingual plate has been lost and there is limited ONJ. (c) Resected specimen at the premolar, molar area. Alveolus lost. Mixed area of very dense bone, necrotic bone and attempts at bone repair.

Figure 3.

 Histology. (a) Normal alveolar bone (A), periodontal ligament (P), cementum (C) and dentine (D). (b) Alveolar bone (A) and periodontal attachment (P) as well as cementum (C) and dentine (D) of the right mandibular canine from the resected specimen (decalcified specimen). (c) Normal trabecular bone from the mandible showing osteoblasts lining the periphery of the bone, osteocytes within intraosseous lacunae and multinucleated osteoclasts (decalcified specimen). (d) Resected specimen showing areas of bisphosphonate affected bone; note the general acellularity.

The bisphosphonates as a treatment to preserve alveolar bone

The possibility that bisphosphonates may be a useful treatment in periodontics was contemplated in the 1990s and early 2000s. It was considered that bisphosphonate use may prevent bone loss around teeth and implants and possibly stimulate new bone formation. In a state of the art review in the Journal of Periodontology in 2002 it was stated that bisphosphonate use might provide a “potentially exciting avenue for future exploration”.53 Essentially, the concept was that as the bisphosphonates inhibit bone resorbtion and preserve or improve the quality of bone as measured clinically by reduced lower impact fracture risk and it was thought that they might reduce tooth loss.54 A number of studies investigating this concept were established although they often were involving different populations, using different doses and types of bisphosphonates and using different outcome measures.55–57

In one study, 40 otherwise healthy post-menopausal females were semi-randomized with one group taking a standardized dose of alendronate (20 mg/day for six months) and the other group were given a placebo.58 Standardized blinded investigation of a range of periodontal indices was undertaken. It was found that no subject lost any teeth in the study, however the study was only of six months duration. The study found that the alendronate group experienced a significant reduction in probing pocket depth (p = 0.0024), a significant increase in alveolar bone height (p = 0.000008) and a significant decrease in tooth mobility (p = 0.009) as compared with the placebo group. There was also a better gain in the clinical attachment level in the alendronate group although this was not statistically significant.58

Thus, one could reasonably speculate that if the trial had continued longer, as both osteoporosis and chronic periodontal disease are conditions measured in years not months, then the alendronate group may have demonstrated less tooth loss than control.58

These studies all ceased once ONJ was described in the literature. One would not obtain ethical approval for such a study if the primary indication for giving the patient a bisphosphonate was for their alveolar bone. This, however, would not stop a study of patients requiring an oral bisphosphonate for medically proven osteoporosis who were then subsequently measured for the state of their oral health. This could either be against a placebo or matched with osteoporotic patients treated without bisphosphonates. The studies would need to be of long duration with large numbers and careful documentation.

Similar studies were conducted for dental implants. It was proposed that by improving the quality of alveolar bone then this would increase the survival of dental implants.59 Studies have also been conducted where the bisphosphonate is incorporated into the surface coating of the implant to deliver a localized increase of bisphosphonate and thus bone activity.60 The results of large long-term clinical trials of bisphosphonate coated implants have not been published.

There have also been a number of studies on the impact of dental implants being placed in patients who were on oral bisphosphonates for bone disease. Initial studies appear to show no impact on prospective studies but these were of relatively small numbers and of short duration.56,61 Individual cases of implant loss in patients on bisphosphonates have been reported.62,63

A detailed study has recently been completed which looked at all of the implants placed in South Australia and then identified those cases with bisphosphonate associated problems.48 It was found that out of a total of 28 000 implants in 16 000 patients, there were seven clear-cut cases of bisphosphonate associated loss. Three patients, who were already on bisphosphonates when the implants were placed, did not achieve integration. The remaining four patients were patients with successful implants who were subsequently placed on bisphosphonates. Progressive loss of integration and loss of the implant occurred. The overall risk was calculated at 1 in 114 (0.88 per cent patients).48

Recommendation for dental treatment for patients on bisphosphonates

There have been numerous consensus guidelines for the management of patients on bisphosphonates.64,65 At this stage, most are expert opinions based on experience and literature review rather than evidence from prospective studies. Such studies are difficult as given the relatively low numbers of similar cases with malignancy on IV bisphosphonates but with high risk and the large numbers of cases on oral bisphosphonates with a relatively low risk, then there are issues with developing studies of sufficient power. There are also ethical issues, particularly with the high risk IV bisphosphonates.

The simplest and most obvious step is to ensure that the patient is dentally healthy prior to or at the early stages of bisphosphonate administration. If they don’t require bone invasive procedures such as extractions then the risk is low. One cannot assume that candidates for bisphosphonates are dentally healthy, as the converse has been demonstrated for both osteoporosis and multiple myeloma patients so that patients need examination and treatment as required by a dentist.47 Although this step is logical and obvious to the dental profession, it has been resisted by some segments of the medical profession particularly for osteoporotic patients requiring oral bisphosphonates. Dental examination was not considered necessary by the American task force from the American Society for Bone and Mineral Research.42

If a patient is already on bisphosphonates and requires treatment then this fact must first be identified by the dental practitioner and a full history of the bone condition, the bisphosphonate type, dosage and duration established. The CTX test is a useful guide of the degree of bone suppression. Consultation with the patient’s medical practitioner is required.

The Therapeutic Guidelines: Oral and Dental, the guidelines for Australian dentists, make a clear-cut division between oral and intravenous bisphosphonates.66 Patients with malignancy and intravenous bisphosphonates are best managed by specialist practitioners with experience of these conditions. The authors have seen a number of cases of ONJ where the dentist’s defence was “the tooth had to be extracted”. When the risk of a serious intractable condition is so high, then alternate treatment such as endodontics or coronectomy must be considered before extraction. Cost should not be an issue and specialist advice should be obtained prior to extraction.

Patients on oral bisphosphonates can have extractions if strongly indicated. This requires full informed consent, antibiotic prophylaxis, careful extraction and primary closure. The CTX test will give an indication of the degree of suppression of bone turnover and can be used to monitor a drug holiday.

Specific periodontal recommendations

The defining criterion is whether the periodontal procedure requires bone invasion. Hence carefully performed scaling and cleaning with root planing are of very low risk. A decision does need to be made on the prognosis of the tooth. Can it be maintained or is it better extracted before the patient has a longer duration of bisphosphonate? An informed consent discussion should be had with the patient. Initial treatment should be to a confined area to determine the tissue response.

Periodontal flap surgery, osseous recontouring, guided tissue regeneration and placement of bone grafts and bone substitutes all carry a risk of ONJ. The precise risk is not known but is probably of a similar order to tooth extraction.44 If the planned procedure involves a response from the bone, such as bony regeneration or incorporation of grafts, then the hoped for bone reaction probably won’t occur, with resultant failure of the procedure and possible ONJ.

Placement of implants in patients on bisphosphonates does carry a small but defined risk.48 The long-term management strategy for the patient’s underlying condition needs to be discussed with the patient’s medical practitioner and informed consent of the risk needs to be provided. Patients with previously placed implants who subsequently are placed on the bisphosphonates need to be carefully monitored. In the event of integration problems, aggressive surgical salvage procedures are best avoided as these will probably hasten failure.48 Consideration of a bisphosphonate drug holiday or alternative medical management of their bone disorder should be discussed with the treating medical practitioner.


The authors acknowledge partial funding from the Australian and New Zealand Association of Oral and Maxillofacial Surgeons Research and Education Foundation.


Dr A Cheng has provided advice to Novartis (Australia). Professor A Goss has provided advice to Novartis (Australia and globally), Merck (Australia and globally) and Sanofi Aventis (Australia). He is an expert for the bisphosphonate class actions in the USA and Australia. No pharmaceutical industry funding has been received for the authors’ bisphosphonate research.