J Oral Maxillofac Surg

65:415-423, 2007

Nature and Frequency of

Bisphosphonate-Associated Osteonecrosis
of the Jaws in Australia
Tony Mavrokokki, BDS,* Andrew Cheng, BDS,†
Brien Stein, MBBS, FRACP,‡ and
Alastair Goss, DDSC, FRACDS(OMS), FICD§
Purpose: The purpose of this study is to estimate the frequency and describe the clinical characteristics
of patients diagnosed with bisphosphonate-associated osteonecrosis of the jaws (ONJ) in Australia.
Materials and Methods: Cases of ONJ were identified in 2004 and 2005 primarily by a postal survey
of Australian Oral and Maxillofacial Surgeons (OMS) with additional cases from other dental specialists
and the Commonwealth of Australia Adverse Drug Reaction Committee (ADRAC). The clinical charac-
teristics were recorded. The frequency of ONJ cases was estimated from prescription and dental
extraction data. Univariate and bivariate statistics were calculated.
Results: One hundred fifty-eight cases of ONJ were identified. These were primarily in patients with
bone malignancy (72%) and the main trigger was dental extraction (73%). The reported number of cases
varied between different Australian States with the highest frequency being reported in the States with
the best integrated health systems. The frequency of ONJ in osteoporotic patients, mainly on weekly oral
alendronate was 1 in 2,260 to 8,470 (0.01% to 0.04%) patients. If extractions were carried out, the
calculated frequency was 1 in 296 to 1,130 cases (0.09% to 0.34%). The total dose of oral alendronate at
the onset of ONJ was 9,060 (!7,269) mg. The frequency of ONJ for Paget’s disease cases was 1 in 56 to
380 (0.26% to 1.8%). If extractions were carried out, the calculated frequency of ONJ was 1 in 7.4 to 48
(2.1% to 13.5%). The frequency of ONJ in bone malignancy cases, treated with mainly intravenous
zoledronate or pamidronate was 1 in 87 to 114 (0.88% to 1.15%). If extractions were carried out, the
calculated frequency of ONJ was 1 in 11 to 15 (6.67% to 9.1%) The total dose of pamidronate was 3,285
(!2,530) mg and zoledronate 62 (!54.28) mg at the onset of ONJ. The median time to onset of ONJ was
12 months for zoledronate, 24 months for pamidronate, and 24 months alendronate.
Conclusions: Before the prescription of bisphosphonates for bone disease the patient should be made
dentally fit so that the need for subsequent dental extractions is minimized. Appropriate informed
consent for the risk of ONJ for different bisphosphonates, for osteoporosis, and malignancy both in
general and in particular for dental extractions can be provided using this data.
© 2007 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 65:415-423, 2007

Bisphosphonate-associated osteonecrosis of the jaws have been presented worldwide.5-15 It is evident that
(ONJ) was first reported in the North American liter- ONJ associated with bisphosphonates is uncommon
ature in 2003.1-4 Since that time multiple case reports relative to the wide prescription of these medications

*Staff Dentist, Oral and Maxillofacial Surgery Unit, Adelaide Den-
tal Hospital and University of Adelaide, Adelaide, South Australia.
†Registrar, Oral and Maxillofacial Surgery Unit, Royal Adelaide
Hospital, Adelaide Dental Hospital and University of Adelaide, Ad-
elaide, South Australia.
‡Medical Oncologist, Ashford Cancer Center, Ashford, and Uni-
versity of Adelaide, Adelaide, South Australia.
§Professor and Director, Oral and Maxillofacial Surgery Unit,
Royal Adelaide Hospital, Adelaide Dental Hospital, Adelaide, and
Professor of Oral and Maxillofacial Surgery, School of Dentistry,
University of Adelaide, Adelaide, South Australia.
Professor Goss has acted as a consultant to Novartis (Interna-
tional and Australia) and to Merck (International and Australia).
Dr A. Mavrokokki has acted as a consultant to Novartis (Australia)
and Dr A. Cheng to Merck (Australia).
Address correspondence and reprint requests to Dr Goss: Oral
and Maxillofacial Surgery Unit, The University of Adelaide, Ad-
elaide, Australia, 5005; e-mail: oral.surgery@adelaide.edu.au
© 2007 American Association of Oral and Maxillofacial Surgeons
0278-2391/07/6503-0008$32.00/0
doi:10.1016/j.joms.2006.10.061

415
416
for benign and malignant bone disease.7,14 Bisphospho-
nates have proved very effective in bone disease man-
agement with considerable easing of pain and minimi-
zation of complications in the population treated.16-21
However, when dental treatment is carried out, in
particular on older medically compromised patients
being treated for bone malignancy with potent nitro-
gen containing bisphosphonates, there is a risk of
osteonecrosis of the jaws. Beyond these broad gener-
alizations, however, there are many unknowns.1-15
Bisphosphonates are prescribed widely in different
doses, potencies, and durations for a range of bone
diseases including osteoporosis, Paget’s disease, or
bone malignancy. The registered indications for use in
Australia (and many other countries) are metastatic
bone disease from breast, hormone refractory pros-
tate cancer, and multiple myeloma. Extended trials
were conducted before the introduction of bisphos-
phonates to clinical practice but no cases of ONJ were
detected.17-21
The main reported initiating factor for ONJ is dental
extraction although periodontal disease and denture
trauma have been implicated.2,3,5,7,14,15 Some appar-
ently spontaneous cases have also been reported.2
Dental diseases are ubiquitous and the probability
of patients having dental trauma and extractions is
high. ONJ presents a new challenge to the dental
profession as previously bone diseases were periph-
eral to the area of interest. There is now a common
bisphosphonate-related condition where either the
lack of dental treatment or simple dental treatment
such as extractions may result in a long-term intrac-
table condition for which there is no single immedi-
ately effective treatment.7,14,15,22 This has led to the
development of preventative proposals to optimize
oral health and avoid dental extractions and other jaw
surgery. These proposals are currently based on ex-
pert panel advice and are not evidence-based.23,24
Evidence needs to be obtained and some clinical trials
have been started. An important unknown is the fre-
quency of ONJ. Generally, when one looks at the
frequency of the bone conditions and the widespread
use of bisphosphonates, ONJ seems rare. General fre-
quency figures of the wide range of 1 in 10,000 for
osteoporosis to 1 in 10 for multiple myeloma have been
reported.10,14,24,25 The key issue, however, is the fre-
quency related to dental interventions, in particular
extractions and possibly the duration of bisphospho-
nate treatment.26,27 This frequency not only has im-
portant clinical implications, it also has important
medico-legal implications in those countries with ad-
vanced legal systems.28
Australia has advanced medical, dental, and legal
systems. Although the island continent is huge in
area, its population is small at 20.3 million. The bulk
of the population, however, is concentrated in the
BISPHOSPHONATE-ASSOCIATED ONJ
major metropolitan capital centers for each State,
with much smaller centers scattered through the rural
areas. The first bisphosphonate-associated ONJ cases
were reported in 2003.29 Since that time there has
been a detailed case series on ONJ in South Austra-
lia.7,14 The State of South Australia is unusual in that
although it has a vast area (2.3 times the state of
Texas), Adelaide is the single major metropolitan cen-
ter. Seventy-five percent of the population of South
Australia reside within 15 kilometers of the Adelaide
Central Business District. All specialist medical and
dental services are based in Adelaide with 2 major
groups of hospitals and a single Oral and Maxillofacial
Surgery service with close links between the public
and private medical and dental systems. Hence it was
straightforward in early 2004 to communicate warn-
ings about the risk of ONJ to key stakeholders and to
locate all cases within the State.14 Good State-based
prospective data collection has been developed for
ONJ.7,14 This has been broadened to include all of
Australia. Australia has a good national health system
and the total number of prescriptions for bisphospho-
nates is known.30 It also has a well-developed but
voluntary adverse drug reporting system8 and a na-
tional database on dental health. A national dental
interview survey showed that in 2002, 62.5% of Aus-
tralians received dental treatment and of these 15.4%
had dental extractions.31 These features have been
used in this study.
The purpose of this study is to identify as many
cases as possible of bisphosphonate-associated ONJ in
Australia. The demographics, drug type, and manage-
ment of the cases and in particular the frequency
related to different types of bisphosphonate, bone
disease, and dental extractions are presented.
Materials and Methods
A 2-page survey form with a reply-paid envelope
was sent to all of the Australian members of the
Australian and New Zealand Association of Oral and
Maxillofacial Surgeons (ANZAOMS). The cover letter
indicated the reasons for the survey and gave recipi-
ents the options of not participating, simply giving
the number of ONJ cases that they had treated, or
providing details of the cases treated. A working def-
inition of ONJ is an area of exposed bone in the
jawbones that fails to heal within 6 weeks in patients
who are on bisphosphonates for bone disease. The
absence of malignancy at the site of the lesion and the
absence of osteoradionecrosis32 are important exclu-
sions. This definition and the nature of ONJ had been
discussed widely in the oral and maxillofacial surgeon
(OMS) community before the survey. The diagnosis of
ONJ was made by the respondent OMS and there was
no independent adjudication of cases. After 6 weeks a
MAVROKOKKI ET AL
reminder was sent out to non-respondents and subse-
quently an individual phone follow-up was carried
out.
From responses to the survey it was apparent that
there were some other cases that had been managed
by other dental specialists and these clinicians were
contacted directly for information. The Common-
wealth of Australia Adverse Drug Reaction Advisory
Committee (ADRAC) was contacted and they pro-
vided details of cases voluntarily reported to them by
prescribers.
These data elements were recorded on a non-net-
worked personal computer. Comparisons of age, gen-
der, drug, and disease were made to exclude double
counting. The data were checked to confirm that the
cases recorded were consistent with the diagnosis of
ONJ.
ESTIMATION OF ONJ FREQUENCY
The total number of prescriptions for all bisphos-
phonates in Australia was obtained from Medicare
Australia of the Commonwealth of Australia.30 The
number of patients receiving the drug was calculated
by dividing the number of prescriptions for osteopo-
rosis and Paget’s disease by 9 as the compliance of the
monthly oral prescription is estimated at 75% (M.
Attia, Merck Sharp Dohme–Australia, personal com-
munication, March 2006). The total number of pa-
tients receiving bisphosphonates for malignancy was
more difficult to calculate as part receive it via Medi-
care-funded prescriptions30 and some from State gov-
ernment-funded oncology sources. These data were
obtained from a combination of Federal and State
government pharmacy sources (C. Doecke, Royal
Table 1. FREQUENCY CALCULATIONS
Denominator – No. of patients
● Prescription data30
● State prescription data for oncology patients (33%)
● Reduction for compliance and mortality, 75% benign
and 50% malignant
Numerator – No. of ONJ cases
Minimum
● Survey data " ADRAC data
Maximum
● South Australian data extrapolation7,14
Frequency
● ONJ overall
● Disease state
Extraction data
● Trigger 73%
● 9.6% Patients having extractions31
Frequency ONJ by extractions
● Overall
● Disease state
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil-
lofac Surg 2007.
417
Adelaide Hospital Pharmacy, personal communica-
tion, July 2006).30 Malignancy patients on average
received 6 doses per year, this relating to compli-
ance and morbidity. The frequency of ONJ per pre-
scriptions and per person was calculated both overall
and then further subdivided as to type of bone disease
and type of bisphosphonate prescribed. The percent-
age of Australians receiving dental treatment in a
given year and the percentage requiring dental extrac-
tions was obtained from the Dental Research Unit of
the Australian Institute of Health of the Common-
wealth Government of Australia. This was used to
determine the number of patients in Australia requir-
ing extractions.31 The frequency of ONJ related to
dental extractions was then calculated. The informa-
tion used to determine the frequency of ONJ is sum-
marized in Table 1. To account for the attrition of
patients treated for malignancy where the total num-
ber of prescriptions is likely to underestimate the
number of people exposed, data from 1 community-
based center were gathered to see if current practice
mirrors the data found in clinical trials.33,34
The South Australian subset,7,14 which was pro-
spectively collected through a single center, showed
a higher frequency than elsewhere in Australia and
the South Australian findings were extrapolated to the
Australian population.
Results
One hundred thirteen questionnaires were mailed
to all of the clinically active members of ANZAOMS,
with 94 (83%) responding. One hundred and three
cases of ONJ were identified with completed ques-
tionnaires being provided for 78 cases. Many ques-
tionnaires were incomplete. Hence the variable n
numbers reported in the tables. The survey was com-
pleted in September 2005. Only a few cases were
reported to occur before September 2003, hence the
study period was 2 years. A further 6 cases were
identified as under the care of other dental specialists
with 4 cases having detailed records. It was found that
22 cases had identical age, gender, bone disease, and
drug prescription profiles when the dental clinician
data were compared with the ADRAC data. These
were considered duplicates and only counted once.
After duplicate exclusion ADRAC had reports of 49
cases, with age, gender, drug, and bone disease being
identified for 32. A total of 158 cases, with details on
114, forms the basis of this study.
The demographics of the ONJ cases are presented
in Table 2. This shows that predominantly patients
with ONJ were older, with malignant bone disease
and on potent nitrogen-containing bisphosphonates.
The presentation and treatment outcomes of ONJ are
presented in Table 3. Most cases were ongoing at the
418 BISPHOSPHONATE-ASSOCIATED ONJ

Table 2. DEMOGRAPHICS OF ONJ

tions this represents 2.34 million repeat prescriptions
and 0.49 million prescriptions for new patients. Over
N (%) the 2-year period the total prescriptions for non-re-
peat patients was 2.83 million, with 2.74 million for
Age and gender (n $ 114)
Age (average, 66.8; range, 39-99 yr) osteoporosis, 27,300 for Paget’s disease, and 61,500
Gender for bone malignancy.30
Female 63 (55) Patients prescribed oral bisphosphonates for osteo-
Male 51 (45) porosis and Paget’s disease do not always complete
Bone disease (n $ 114)
the course with the compliance rate being estimated
Osteoporosis 26 (23)
Paget’s disease 6 (5) at 75%. Hence the number of patients on oral bisphos-
Bone metastasis 51 (45) phonates for osteoporosis is estimated at 304,900 and
Multiple myeloma 31 (27) Paget’s disease at 3,030.
Bisphosphonate (n $ 114) Patients with bone malignancy usually have the
Zoledronate* 43 (38)
drug administered intravenously but a drop-out rate
Alendronate 30 (26)
Pamidronate* 20 (18) relates to morbidity and mortality. The total number
Pamidronate*/zoledronate* 13 (11) of patients with bone malignancy from the Federal
Risedronate 2 (2) and State sources was 12,970.30
Clodronate 2 (2) The total number of patients identified with ONJ
Alendronate/risedronate 2 (2)
was 158, with osteoporosis 36, Paget’s disease 8, and
Pamidronate*/alendronate 1 (1)
Zoledronate*/ibandronate 1 (1) bone malignancy at 114. This gives an overall fre-
quency of 1 in 2,030, with 1 in 8,470 for osteoporosis,
*Intravenous agents.
1 in 380 for Paget’s disease, and 1 in 114 for bone
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil- malignancy cases.
lofac Surg 2007.
In Australia, 62.5% of the population received den-
tal care in 2002 and 15.4% had 1 or more teeth
time of completion of the survey. Six cases (7%) died
of their malignancy during the survey period.
Chi-square tests were carried out between paired
Table 3. CLINICAL CHARACTERISTICS AND
variables from Tables 2 and 3 and most were not TREATMENT OUTCOMES OF ONJ
statistically significant. There was a weak correlation
between alendronate and mild/moderate ONJ as com- N (%)
pared with pamidronate and zoledronate and severe Bone Involved (n $ 89)
ONJ (P # .06). Mandible only 57 (64)
The relationship of ONJ cases to the different Maxilla only 24 (27)
bisphosphonates and different bone diseases is pre- Maxilla and mandible 8 (9)
sented in Table 4. The average dose and range of Trigger for ONJ (n $ 112)
Extraction 82 (73)
bisphosphonate to the time of onset of ONJ is pre- Denture 6 (5)
sented in Table 5. The time of onset of ONJ is plotted Spontaneous 23 (21)
in Figure 1. The duration of administration to onset Mandible fracture 1 (1)
time was plotted against the cumulative frequency in Severity (n $ 75)
each treatment group. This shows that there is a Mild* 22 (29)
Moderate† 9 (12)
significant delay in onset for the majority of cases Severe‡ 44 (59)
(median onset at 12 months with zoledronate, 24 Treatment (n $ 82)
months with pamidronate, and 24 months with alen- Nonsurgical 22 (27)
dronate). Furthermore there is a significantly shorter Curretage only 32 (39)
time to onset with zoledronate (P $ .0015 by log-rank Major surgical 28 (34)
Outcome (n $ 83)
testing of survival curves). The response and fre- Resolved 19 (23)
quency by different Australian States is presented in Ongoing 58 (70)
Figure 2. Death from malignancy 6 (7)
The total number of prescriptions for bisphospho- *Exposed bone and low grade pain only.
nates in the year July 2003 to June 2004 was 2.34 †Exposed bone, severe pain and intermittent soft tissue infec-
million, of these 2.26 million (96.6%) were for osteo- tion.
‡Extensive sequestration, severe continuous pain ! pathologic
porosis, 27,900 (1.2%) for Paget’s disease and 51,500 fracture, orocutaneous fistula, severe neck infection, and weight
(2.2%) were for bone malignancy.30 In July 2004 to loss.
June 2005, there were 2.83 million prescriptions and Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil-
given the ongoing nature of bisphosphonate medica- lofac Surg 2007.
MAVROKOKKI ET AL 419

Table 4. ONJ CASES STRATIFIED BY BISPHOSPHONATE AND BONE DISEASE (N ! 114)

Bone Multiple Total Total

Drug Osteoporosis Paget’s Metastasis Myeloma Benign Malignant Total

Alendronate 22 3 5 – 25 5 30
Zoledronate* – – 27 16 – 43 43
Pamidronate* – 2 9 9 2 18 20
Pamidronate*/zoledronate* – – 8 5 – 13 13
Risedronate 2 – – – 2 – 2
Clodronate – – 1 1 – 2 2
Alendronate/risedronate 2 – – – 2 – 2
Alendronate/pamidronate* – 1 – – 1 – 1
Zoledronate*/ibandronate – – 1 – – 1 1
*Intravenous agents.
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxillofac Surg 2007.

extracted. This indicates that 9.6% of the population
have an extraction in any given year. Of the total
patients on bisphosphonates 30,800 would have had
an extraction. Most of these 29,270 have osteoporo-
sis, 290 Paget’s disease, and 1,245 with bone malig-
nancy.
The overall frequency of extraction-related ONJ is 1
in 270, for osteoporosis 1 in 1,130, for Paget’s disease
1 in 48, and 1 in 15 for bone malignancy.
These frequency figures are summarized in Table 6.
They represent the minimum risk of ONJ for patients
on bisphosphonates overall, for specific bone condi-
tions and for the ONJ risk of patients on bisphospho-
nates needing dental extractions.
The South Australian experience was different as
this was a prospective capture of all of the known
cases14 in the State. Of 25 cases, 10 were in osteopo-
rotics, 4 in Paget’s disease, and 11 with bone malig-
nancy. The population of South Australia is 1.5 million
giving a population frequency of 1 in 60,000 people.
If this experience is projected to the total Australian
population of 20.3 million people then one would
expect 338 cases of ONJ with 135 in osteoporotics,
54 Paget’s, and 149 with bone malignancy. When this
is compared with the patients on bisphosphonates
then the overall frequency is 1 in 930, with 1 in 2,260
for osteoporotics, 1 in 56 Paget’s, and 1 in 87 for bone
malignancy. Dental extractions as a trigger for 73%,
would result in 247 cases overall, 99 osteoporotics, 39
Paget’s, and 109 bone malignancy cases being related
to extractions.
The overall maximum frequency of extraction-re-
lated ONJ is 1 in 125, with 1 in 296 for osteoporosis,
1 in 7.4 for Paget’s, and 1 in 11 for bone malignancy.
These figures are summarized in Table 6.
Discussion
This study shows that bisphosphonate-associated
ONJ is a recent and severe condition in Australia.
Given the increasing frequency of osteoporosis as the
population of Australia progressively ages then the
use of oral bisphosphonates is likely to increase. Sim-
ilarly the bisphosphonates have been shown to be an
effective drug in the management of bone malig-
nancy. ONJ is an ongoing and probably increasing

Table 5. AVERAGE DOSE OF BISPHOSPHONATE TO

THE ONSET OF ONJ*

Average Dose (mg) and

Drug No. Range (mg)

Zoledronate 26 62 (4-240)
Alendronate 19 9,060 (900-26,100)
Pamidronate 14 3,285 (630-8,640)
*Most common single drug regimens (#3 cases of ONJ and not FIGURE 1. Cumulative frequency of osteonecrosis in 59 patients in
possible to calculate the dosage for combinations). whom duration of therapy was known, grouped by therapy.
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil- Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil-
lofac Surg 2007. lofac Surg 2007.
420
FIGURE 2. Responses by Australian State (n $ 109).
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxil-
lofac Surg 2007.
challenge that needs to be understood and strategies
developed to prevent its occurrence.
The patients identified in Australia were similar to
those described previously in South Australia7,14 and
the United States.1-6,15 Patients were mainly older
with an average of 67 years and predominately with
malignant bone disease (72%), the mandible was
more involved (64%) and the maxilla (27%), or in both
jaws (9%). The most common trigger for bisphospho-
nate-associated ONJ was dental extractions (73%).
The Australian series had a lower incidence from
periodontal disease, trauma to tori, or implant-related
cases than the American series.5,15,22,24 Most Austra-
lian cases were severe (59%) and most were ongoing
(70%). There was an equal spread of nonsurgical
(27%), curettage (39%), and major surgical treatment
(34%). This probably reflected uncertainty as to the
appropriate method of treatment, and with the ex-
ception of the South Australian series, most oral and
maxillofacial surgeons had experience of only a few
cases, ranging from 0 to 8.
In South Australia, arrangements were put in place
early so that all 25 cases reported in that State were
seen at the single Oral and Maxillofacial Surgery Unit
Table 6. MINIMUM AND MAXIMUM FREQUENCY OF ONJ OVERALL FOR DIFFERENT BONE CONDITIONS AND IF
EXTRACTION IS CARRIED OUT
Frequency of ONJ Per Patient Overall
On bisphosphonates
Min 1/2,030
Max 1/950
On bisphosphonates having extractions
Min 1/270
Max 1/125
Mavrokokki et al. Bisphosphonate-Associated ONJ. J Oral Maxillofac Surg 2007.
BISPHOSPHONATE-ASSOCIATED ONJ
at the major teaching hospital. Beside the bone ma-
lignancy-associated cases of ONJ, the South Australian
group also reported 10 of 25 (40%) osteoporotic cases
on weekly oral bisphosphonates, mainly alendronate,
in this study. The explanation of this is that the Oral
and Maxillofacial Surgery Unit in South Australia also
conducts a large extraction clinic for public patients,
these are predominantly older, more medically com-
promised patients and have more extractions as com-
pared with the private sector. Early in 2004 a directive
was sent out to all the government-funded community
dental clinics in South Australia that patients on
bisphosphonates requiring extractions should be re-
ferred to the specialist OMS unit. Hence a large group
of osteoporotic patients on bisphosphonates requir-
ing extractions were seen and postextraction healing
evaluated carefully. It was found that osteoporotic
patients on oral bisphosphonates usually have a less
severe form of ONJ that is likely to eventually resolve
over a year. This is particularly so if the bisphospho-
nates are withdrawn.7,14
It is noted that the distribution of South Australian
cases is different to the United States studies. The
particular circumstances of health delivery allow for
complete capture of all cases from minor to major,
which is more difficult in larger and more compli-
cated populations and health centers. Although cases
were not independently adjudicated, all were seen by
several experienced clinicians.
A detailed understanding of the frequency of the
complication of ONJ is important so that medical
practitioners who prescribe bisphosphonates for os-
teoporosis and bone malignancy can give their pa-
tients appropriate advice and choice before com-
mencing bisphosphonate treatment. An important
aspect is the patient’s oral health before commence-
ment of bisphosphonates. Dentists need to under-
stand the issues to prepare patients before the com-
mencement of bisphosphonate therapy. In particular,
patients who are already on bisphosphonates who
require dental extractions need to be provided in-
formed consent about the risk of ONJ. Established
cases of ONJ require specialist management.
% Osteoporosis % Paget’s % Malignancy %
0.05 1/8,470 0.01 1/380 0.26 1/114 0.88
0.10 1/2,260 0.04 1/56 1.8 1/87 1.15
0.37 1/1,130 0.09 1/48 2.1 1/15 6.67
0.80 1/296 0.34 1/7.4 13.5 1/11 9.1
MAVROKOKKI ET AL
To calculate the frequency of ONJ in the Australian
population involves a combination of a number of
known facts and assumptions (Table 1). The first
assumption is that when ONJ occurs, general medical
and dental practitioners would refer the patient to an
oral and maxillofacial surgeon. Although this practice
is not universal this is the usual referral pattern. The
number and contact address of oral and maxillofacial
surgeons in Australia who are members of ANZAOMS,
is known as this organization represents over 90% of
all registered OMS specialists in Australia. The retro-
spective postal questionnaire was sent to all active
ANZAOMS members in Australia. There was a variable
response with the largest number of ONJ cases iden-
tified in South Australia and Victoria and the lowest
from New South Wales. A similar pattern has been
noted in previous surveys unrelated to ONJ.35 The
response to the questionnaire showed varying de-
grees of detail as is usual for retrospective clinical
questionnaires. This was calculated by using the ac-
tual response rate for each question, hence the vari-
able n numbers presented in the tables.
It was also indicated by some respondents that they
knew that a few other specialists had patients with
ONJ and these were contacted and included. ADRAC
receives voluntary reports from medical, dental, and
pharmaceutical sources, also has an Australian data-
base on this adverse reaction. Their data were ac-
cessed but was limited to age, gender, drug, and
condition treated, without greater detail. When the
questionnaire and ADRAC data were compared, 22
cases were removed as duplicates. By these means an
estimate of the number of cases of ONJ in Australia
was determined.
The total number of prescriptions for bisphospho-
nates per year is available from the Federal govern-
ment agency responsible for prescriptions in Austra-
lia,30 with additional data from State and Federal
sources for bone malignancy cases (C. Doecke, per-
sonal communication). This is divided into the differ-
ent types of bisphosphonates and also whether they
were prescribed for osteoporosis, Paget’s disease or
bone malignancy.
The number of patients for whom the bisphospho-
nates are prescribed is not readily available but was
calculated with an adjustment for compliance, and in
the case of malignancy, for patient attrition from
death or disease progression. By this means, the num-
ber of cases per prescription and per patient was
calculated. The estimated number of 304,900 (2004
to 2005) and 251,000 (2003 to 2004) patients on
bisphosphonates for osteoporosis is consistent with
the most recent pharmaceutical company estimate of
275,000 patients on alendronate (M. Attia, personal
comminication).
421
The number of malignancy cases for 2004 to 2005
was 12,97030 that correlates with the pharmaceutical
company estimate (V. Ferrari, Novartis–Australia, per-
sonal communication, May 2006).
Shortly after the finalization of the survey the au-
thors published a review for Australian general dental
practitioners14 and since that time more cases in gen-
eral dental practice have been brought to the authors’
attention. ADRAC has had a further 69 cases, in addi-
tion to the 49 reported in this study, in the period
September 2005 to May 2006. Of 118 cases, 64 (54%)
related to zoledronic acid. After a recent national
media report, health professionals and patients with
further cases have contacted the corresponding au-
thor.36
Although the frequency of ONJ per patient with
bone disease is important because the condition is
triggered primarily by dental extraction and dental
disease, the frequency of ONJ to extraction is the key
variable. Australia has good data on dental disease and
its management.31 The frequency of extractions does
not vary greatly over the adult age range. If one
assumes that patients with bone disease have similar
oral health to the overall population then the assump-
tion that in any one year 9.6% of patients will have a
dental extraction is a reasonable estimate.31 This
would be particularly so for osteoporotic and Paget’s
disease patients. Patients with malignancy are less
likely to seek elective dental care in the active phase
of their disease. Once they are in remission and feel
better then they may seek dental care, this may be a
factor in the later presentation of ONJ.
The key question confronting a dentist about to
extract a tooth for a patient on bisphosphonates is,
what is the risk of ONJ? The overall frequency of ONJ
after extraction is at minimum the risk determined
from the survey and more likely maximum from the
extrapolation to Australia of the South Australian data.
Health professionals have an obligation under the
State and Federal law to provide patients with in-
formed consent about their treatment.37,38 The con-
dition of bisphosphonate-associated ONJ was not
known before 2003. Since then there has been little
information about the true frequency of ONJ and
hence the risk, particularly relating to extractions. To
date it has been difficult for health professionals to
provide evidence-based information. This could be
considered somewhat surprising in bone malignancy
where our frequency is about 1% and others report
higher.26 With an event frequency like this, one needs
about 3 times the number of patients to have the 95%
confidence limits exclude zero, thus 50 to 100 pa-
tients in a trial arm should be sufficient. The trials in
neoplastic diseases seemed to have sufficient power
as typical numbers enrolled in treatment arms were
180 to 220.26,27,33,34 The time-dependent risk that this
422
study and the study by Bamias et al26 find, requires us
to consider the number of patients treated at length.
This is considerably smaller, and in 2 large trials the
patients completing 2 years of therapy varied from 36
(15% of the entry cohort)33 to 65 (%30%).34 The
clinical trials thus had a substantial chance of missing
such a side effect via chance, not to consider the
possibility of missing the diagnosis in a group with
short survival and potentially many distracting symp-
toms from their malignancies.
This study indicates that the overall risk for patients
on bisphosphonates (mainly oral alendronate) for os-
teoporosis is low (Table 6). The patient generally
needs to have been on the regular weekly oral dose
for several years before ONJ occurs. The patient
needs to be warned and advised to be dentally fit. If
they need extractions or develop ONJ it is appropri-
ate for their physician to review their current osteo-
porotic status. Consideration should be given to the
withdrawal of the bisphosphonates, calcium and vita-
min D continued, and if appropriate an alternative
agent may be required.
Patients on bisphosphonates, usually intravenous
zoledronate or pamidronate for bone malignancy,
have a higher risk of ONJ (Table 6). If extractions
were carried out, then the calculated frequency of
ONJ ranges from 6.67% to 9.1%. These high risks must
be balanced against the significant benefits of bisphos-
phonates for bone malignancy. It confirms that pa-
tients being prescribed bisphosphonates for bone ma-
lignancy should be made dentally fit to minimize the
subsequent chance of extractions.
Two further considerations are raised by Table 4
and by Bamias et al.26 First, do bisphosphonates vary
in their propensity to cause ONJ? Both suggest zole-
dronate may carry a higher risk of ONJ. In the absence
of a large prolonged comparative study this is unlikely
to be shown absolutely, but further data may support
this theory. Second, duration of treatment may be an
important predictor. There is likely to be some con-
founding in this inasmuch as patients will vary when
they seek dental treatment in relation to onset of
bisphosphonate treatment. With the majority of cases
with alendronate or pamidronate occurring over 2
years after starting treatment, however, this seems
unlikely.
There is a need for further research on all aspects of
the bisphosphonates, their effect on the jaws and
extraction healing, the variation between the agents,
and the relationship between duration of therapy and
risk. This retrospective postal survey with multiple
assumptions and without independent assessment of
cases, shows the need for prospective clinical trials.
In the meantime however, if the patient is dentally fit
before commencement of bisphosphonate therapy
BISPHOSPHONATE-ASSOCIATED ONJ
and oral health maintained without extractions, then
the risk of ONJ should be reduced markedly.39
Acknowledgments
The authors thank members of ANZAOMS for their assistance in
responding to the questionnaire and to ADRAC for the data they
generously shared. The statistical advice of Dr D. Brennan and
Professor A.J. Spencer, The University of Adelaide, is acknowl-
edged. The constructive criticism and assistance of multiple bone
metabolic, oncology, pharmaceutical, and dental colleagues, both
within Australia and globally, are acknowledged.
References
1. Wang J, Goodger NM, Pogrel MA: Osteonecrosis of the jaws
associated with cancer chemotherapy. J Oral Maxillofac Surg
61:1104, 2003
2. Marx RE: Pamidronate (Aredia) and zoledronate (Zometa) in-
duced avascular necrosis of the jaws: A growing epidemic.
J Oral Maxillofac Surg 61:1115, 2003
3. Migliorati CA: Bisphosphonates and oral cavity avascular bone
necrosis. J Clin Oncol 21:4253, 2003
4. Pogrel MA: Bisphosphonates and bone necrosis. J Oral Maxil-
lofac Surg 62:391, 2004
5. Ruggiero SL, Mehrotra B, Rosenberg TJ, et al: Osteonecrosis of
the jaws associated with the use of bisphosphonates: A review
of 63 cases. J Oral Maxillofac Surg 62:527, 2004
6. Lugassy G, Shaham R, Nemets A, et al: Severe osteomyelitis of
the jaw in long-term survivors of multiple myeloma: A new
clinical entity. Am J Med 117:440, 2004
7. Carter G, Goss AN, Doecke C: Bisphosphonates and avascular
necrosis of the jaw: A possible association. Med J Aust 182:413,
2005
8. Purcell PM, Boyd AW: ADRAC report: Bisphosphonates and
osteonecrosis of the jaws. Med J Aust 182:417, 2005
9. Bagan JV, Murillo J, Jimenez Y, et al: Avascular jaw necrosis in
association with cancer chemotherapy: Series of 10 cases.
J Oral Path Med 34:120, 2005
10. Durie BGM, Katz M, Crowley J: Osteonecrosis of the jaw and
bisphosphonates. N Engl J Med 353:99, 2005
11. Schirmer I, Peters H, Reichart PA, et al: [Bisphosphonates and
osteonecrosis of the jaw]. Mund Kiefer Gesichtschir 9:239, 2005
12. Maerevoet M, Martin C, Duck L: Osteonecrosis of the jaw and
bisphosphonates. Letter to the editor. N Engl J Med 353:100,
2005
13. Hellstein JW, Marek CL: Bisphosphonate osteochemonecrosis
(bis-phossy jaw): Is this phossy jaw of the 21st century? J Oral
Maxillofac Surg 63:682, 2005
14. Cheng A, Mavrokokki A, Carter G, et al: The dental implications
of bisphosphonates and bone disease. Aust Dent J 50:S4, 2005
(suppl 2)
15. Marx RE, Sawatari Y, Fortin M, et al: Bisphosphonate induced
exposed bone (osteonecrosis/osteopetrosis) of the jaws: Risk
factors, recognition, prevention and treatment. J Oral Maxillo-
fac Surg 63:1567, 2005
16. Fleisch H: Bisphosphonates: Mechanisms of action. Endocr Rev
19:80, 1998
17. Sparidans RW, Twiss IM, Talbot S: Bisphosphonates in bone
disease. Pharm World Sci 20:206, 1998
18. Russell RGG, Rogers MJ: Bisphosphonates: From the laboratory
to the clinic and back again. Bone 25:97, 1999
19. Russell RGG, Croucher PI, Rogers MJ: Bisphosphonates: Phar-
macology, mechanisms of action and clinical uses. Osteoporos
Int 9:S66, 1999 (suppl 2)
20. Fleisch H: Bisphosphonates in osteoporosis. Eur Spine J 12:
S142, 2003 (suppl 2)
21. Wong R, Wiffen PJ: Bisphosphonates for the relief of pain
secondary to bone metastases. Cochrane Database Syst Rev
2002;(2):CD002068
22. Ruggiero S, Gralow J, Marx R, et al: Practical guidelines for the
prevention, diagnosis, and treatment of osteonecrosis of the
jaw in patients with cancer. J Oncol Pract 2:7, 2006
 MAVROKOKKI ET AL
23. Damato K, Gralow J, Hoff A, et al: Expert panel recommenda-
tions for the prevention, diagnosis, and treatment of osteone-
crosis of the jaws. (Novartis internal document). Available
at: http://www.ada.org/prof/resources/topics/topics_
osteonecrosis_whitepaper.pdf. Accessed May 2006
24. Migliorati CA, Casiglia J, Epstein J, et al: Managing the care of
patients with bisphosphonate-associated osteonecrosis. J Am
Dent Assoc 136:1658, 2005
25. Tarassoff P, Csermak K: Avascular necrosis of the jaws: risk
factors in metastatic cancer patients. J Oral Maxillofac Surg
61:1238, 2003
26. Bamias A, Kastridis E, Bamia C, et al: Osteonecrosis of the jaw
in cancer after treatment with bisphosphonates: Incidence and
risk factors. J Clin Oncol 23:8580, 2005
27. Badros A, Weikel D, Salama A, et al: Osteonecrosis of the jaw in
multiple myeloma patients: Clinical features and risk factors.
J Clin Oncol 26:945, 2006
28. Legal. Ennis and Ennis PA: Fosamax class action. http://www.
fosamax-lawyer.com/
29. Carter GD, Goss AN: Letter to the editor. Bisphosphonates and
avascular necrosis of the jaws. Aust Dent J 48:268, 2003
30. Medicare Australia. http://www.medicareaustralia.gov.au
31. Carter KD, Stewart JF: National Dental Telephone Interview
Survey. 2002 Adelaide. AUIHW, Statistics and Research Unit,
2003
The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.
423
32. Vudiniabola S, Pirone C, Williamson J, et al: Hyperbaric oxygen
in the therapeutic management of osteoradionecrosis of the
facial bones. Int J Oral Maxillofac Surg 29:435, 2000
33. Saad F, Gleason DM, Murray R, et al: Zoledronic Acid Prostate
Cancer Study Group. Long-term efficacy of zoledronic acid for
the prevention of skeletal complications in patients with met-
astatic hormone-refractory prostate cancer. J Natl Cancer Inst
96:879, 2004
34. Hortobagyi GN, Theriault RL, Lipton A, et al: Long-term pre-
vention of skeletal complications of metastatic breast cancer
with pamidronate. Protocol 19 Aredia Breast Cancer Study
Group. J Clin Oncol 16:2038, 1998
35. Brennan DS, Spencer AJ, Singh KA, et al: Practice patterns of
oral and maxillofacial surgeons in Australia: 1990 and 2000. Int
J Oral Maxillofac Surg 33:598, 2004
36. Swan N: Manuscript of recording on ‘Drug linked to death of
jawbone’ The Health Report, ABC Radio National, March 27,
2006
37. South Australian Government Legislation, Civil Liberty Act.
Section 41, 1936
38. Whittaker RV: High Court of Australia, 175 CLR 479, 1992
39. Sambrook P, Olver I, Goss AN: Bisphosphonates and osteone-
crosis of the jaw: Position statement on behalf of the Australian
and New Zealand Bone and Mineral Society, Osteoporosis
Australia, Medical Oncology Group of Australia and the Austra-
lian Dental Association. Aust Fam Physician 35:801, 2006