American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158 – 163

Osteonecrosis of the jaws due to bisphosphonate use.
A review of 60 cases and treatment proposals
Christos Magopoulos, DDSa,4, Georgios Karakinaris, DDSa, Zisis Telioudis, DDSb,
Konstantinos Vahtsevanos, MD, DDSb, Ioannis Dimitrakopoulos, MD, DDSa,
Konstantinos Antoniadis, MD, DDSa, Sideris Delaroudis, MDc
a

Department of Oral and Maxillofacial Surgery, Aristotle University, Thessaloniki, Greece
b
dTheagenioT Cancer Istitute, Thessaloniki, Greece
c
Department of Endocrinology, Military Hospital, Thessaloniki, Greece
Received 15 May 2006

Abstract

Purpose: Bisphosphonates are compounds used in the treatment of various metabolic and malignant bone diseases. In the last two and a half years, there has been a striking increased referral of patients with exposed necrotic jawbone, mostly after several teeth extractions. The only clinical feature common in all patients was the use of bisphosphonates in the treatment of bone diseases.

Patients and methods: We performed a retrospective multicentric study of 60 patients with necrotic bone lesions of the jaws of various extent from July 2003 to October 2005. The necrotic bone involved the maxilla (37%), the mandible (50%), or both (13%). The bisphosphonate administered was mostly zoledronate. The management of the patients included cessation of bisphosphonate therapy for more than 6 months, long-term antibiotics, hyperbaric oxygen administration in some cases, and various surgical restorative procedures.

Results: The implementation of the treatment protocol in 7 patients so farlead to high cure rates, whereas surgical restoration of the defect without previous cessation of bisphosphonate therapy had discouraging results.
Conclusions: Clinicians and dentists should have in mind this new complication of bisphosphonate administration to identify and treat osteonecrosis of the jaws.
D 2007 Elsevier Inc. All rights reserved.

1. Introduction Bisphosphonates have been developed over the past
3 decades and have been used in the treatment of many skeletal disorders, such as bone metastases, osteoporosis, Paget disease, hypercalcemia of malignancy, and bone pain. Those that contain nitrogen and are the most potent ones are the N-BPs and are represented by aledronate, ibandronate, incadronate, olpadronate, pamidronate, risedronate, and zoledronate, whereas the non–N-BPs are mainly clodronate, etidronate, and tiludronate [1,2].

4 Corresponding author. Glinou 4-Pilea, 54352 Thessaloniki, Greece.
Tel.: +30 2310949711; fax: +30 2310995362.
E-mail address: mago@med.auth.gr (C. Magopoulos).
0196-0709/$ – see front matter D 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjoto.2006.08.004

The main pharmacological effect of bisphosphonates is
the inhibition of bone resorption caused by osteoclasts
decreased function, whereas other effects, like inhibition of
calcification in the treatment of hypercalcemia of malignancy and reduction of joint inflammatory reaction in the
treatment of arthritis, are of secondary magnitude.
In the last 2 years, there has been an increased referral of
patients with exposed necrotic jaw bone, diagnosed elsewhere as chronic refractory osteomyelitis of jaws mostly
after teeth extractions. Most ofthe patients had a natural
history of malignancy, and the only thing in common was the
long-term intravenous administration of bisphosphonates.
In this article, we present the clinical evaluation of this
recently described complication of the use of bisphosphonates previously called drug-induced avascular osteonecrosis or bisphosphonate-induced exposed bone of the


C. Magopoulos et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158–163

159

Table 1
Patient data
Patient no.

Diagnosis

Bisphosphonates

Osteonecrosis

Management

Outcome

1
2
3

Multiple myeloma
Prostate cancer
Breast cancer

Mandible
Mandible
Mandible and maxilla

Protocol 1
Protocol 1
Protocol 1

Complete healing
Complete healing
Complete healing

4
5

Multiple myeloma
Breast cancer

Mandible
Maxilla

Protocol 1
Protocol 2

Complete healing
Complete healing

6
7
8

Lung cancer
Fibrous dysplasia
Multiple myeloma

Maxilla
Maxilla
Mandible

Protocol 2
Protocol 2
Protocol 3

Complete healing
Complete healing
Recurrence

9
10
11

Multiple myeloma
Multiple myeloma
Multiple myeloma

Mandible
Mandible and maxilla
Mandible

Protocol 3
Protocol 4
Protocol 4

Recurrence
Recurrence
Recurrence

12
13
14

Breast cancer
Breast cancer
Prostate cancer

Mandible
Mandible
Maxilla

Protocol 4
Protocol 4
Protocol 4

Recurrence
Recurrence
Recurrence

15

Multiple myeloma

Maxilla

Protocol 4

Recurrence

16
17
18
19

Breast cancer
Breast cancer
Multiple myeloma
Prostate cancer

Maxilla
Mandible
Mandible
Mandible and maxilla

Protocol
ProtocolProtocol
Protocol

5
5
5
5

Recurrence
Recurrence
Recurrence
Recurrence

20

Multiple myeloma

Mandible

Protocol 5

Recurrence

21
22
23

Multiple myeloma
Multiple myeloma
Breast cancer

Mandible
Maxilla
Mandible and maxilla

Protocol 5
Protocol 5
Protocol 5

Recurrence
Recurrence
Recurrence

24

Multiple myeloma

Maxilla

Protocol 5

Recurrence

25

Breast cancer

Mandible

Protocol 5

Recurrence

26

Multiple myeloma

Maxilla

Protocol 6

Stabilized necrosis

27

Multiple myeloma

Mandible

Protocol 6

Stabilized necrosis

28
29

Multiple myeloma
Multiple myeloma

Mandible
Maxilla

Protocol 6
Protocol 6

Stabilized necrosis
Stabilized necrosis

30
31
32

Multiple myeloma
Breast cancer
Breast cancer

Mandible
Mandible
Maxilla

Protocol 6
Protocol 6
Protocol 6

Stabilized necrosis
Stabilized necrosis
Stabilized necrosis

33

Breast cancer

Mandible and maxilla

Protocol 6

Stabilized necrosis

34

Prostate cancer

Maxilla

Protocol 6

Stabilized necrosis

35
36
37
38

Multiple
Multiple
Multiple
Multiple

Maxilla
Mandible
Mandible
Mandible

Protocol
Protocol
Protocol
Protocol

6
6
6
6

Stabilized
Stabilized
Stabilized
Stabilized

necrosis
necrosis
necrosis
necrosis

39
40
41
42

Multiple myeloma
Multiple myeloma
Breast cancer
Breast cancer

Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
PamidronateZoledronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Ibandronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Ibandronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate

Maxilla
Maxilla
Mandible
Mandible and maxilla

Protocol
Protocol
Protocol
Protocol

6
6
6
6

Stabilized
Stabilized
Stabilized
Stabilized

necrosis
necrosis
necrosis
necrosis

myeloma
myeloma
myeloma
myeloma


160

C. Magopoulos et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158–163

Table 1 (continued )
Patient no.

Diagnosis

Bisphosphonates

Osteonecrosis

Management

Outcome

43
44
45

Prostate cancer
Multiple myeloma
Multiple myeloma

Maxilla
Mandible
Maxilla

Protocol 6
Protocol 6
Protocol 6

Stabilized necrosis
Stabilized necrosis
Stabilized necrosis

46
47
48
49

Multiple myeloma
Amyloidosis
Multiple myeloma
Multiple myeloma

Mandible
Mandible
Maxilla
Maxilla

Protocol
Protocol
Protocol
Protocol

Stabilized
Stabilized
Stabilized
Stabilized

50
51

Breast cancer
Breast cancer

Maxilla
Mandible and maxilla

Protocol 7
Protocol 7

Progressive necrosis
Progressive necrosis

52

Neuroendocrine cancer

Mandible

Protocol 7

Progressive necrosis

53
54Multiple myeloma
Multiple myeloma

Mandible
Maxilla

Protocol 7
Protocol 7

Progressive necrosis
Progressive necrosis

55
56

Lung cancer
Breast cancer

Mandible
Mandible

Protocol 7
Protocol 7

Progressive necrosis
Progressive necrosis

57
58

Breast cancer
Breast cancer

Mandible and maxilla
Maxilla

Protocol 7
Protocol 7

Progressive necrosis
Progressive necrosis

59
60

Multiple myeloma
Multiple myeloma

Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Ibandronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate
Zoledronate
Pamidronate
Zoledronate

Mandible
Mandible

Protocol 7
Protocol 7

Progressive necrosis
Progressive necrosis

jaws [3]. We suggest a therapeutic protocol for these
patients, and in the discussion that follows, we put under
speculation the mechanism that leads to this complication.
2. Patients and methods
From July 2003 to October 2005, 60 patients referred
to the Department of Oral and Maxillofacial Surgery
of Aristotle University and dTheagenioT Cancer Institute of
Thessaloniki with the diagnosis of refractory osteomelitis of
the jaws subsequent to their main oncological history of one
of the following diseases: multiple myeloma (32 cases,
53.33%), breast cancer (18 cases, 30%), prostate cancer
(5 cases, 8,33%), lung cancer (2 cases, 3.33%), neuroendocrine cancer (1 case, 1.67%), fibrous dysplasia (1 case,
1.67%), and amyloidosis (1 case, 1.67%) (Table 1). The
mean patient age at thetime of presentation was 61 years
(range, 47–76 years). The sex ratio was 1:1.14 (28 men and
32 women). The exposed bone was subsequent to nonhealing extraction sockets (41 cases, 68.33%) or chronic
denture trauma (9 cases, 15%), and it was a spontaneous
procedure in 10 cases (16.67%). The typical presenting
symptoms were localized swelling and pain, with exposed
necrotic bone involving the maxilla (37%), the mandible
(50%), or both (13%) (Table 1).
Most of the patients were treated elsewhere for chronic
refractory osteomyelitis. The treatment included nearly all

6
6
6
6

necrosis
necrosis
necrosis
necrosis

the current therapeutic protocols for osteomyelitis with
conservative (antibiotics and hyperbaric oxygen [HBO]
and/or surgical (debridement-sequestrectomy) procedures.
None of the patients had previously received radiation
therapy to the head and neck region, but almost all received
chemotherapy in accordance to the specific type of tumors
that had already been ceased by the time of bisphosphonate
therapy initiation. The only thing in common was the
bisphosphonate therapy administered to all patients in the
treatment of the—metastatic or not—osteolytic bone
lesions. The bisphosphonates were administered intravenously, mostly zoledronate (33 cases, 55%), pamidronate
followed by zoledronate (24 cases, 40%), and pamidronate
followed by ibandronate and finally by zoledronate (3 cases
5%) (Table 1). The duration of bisphosphonate therapy
was 6 to 60 months and was administered in month or
40-day dosages.
The treatment protocols followed included
1.

Cessation of bisphosphonate therapy more than
6 months, HBO therapy, surgery,antibiotics
(4 patients);
2. Cessation of bisphosphonate therapy more than
6 months, surgery, antibiotics (3 patients)
3. Cessation of bisphosphonate therapy less than
6 months, surgery, antibiotics (2 patients)
4. HBO therapy, surgery, antibiotics (6 patients)


C. Magopoulos et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158–163

Fig. 1. Maxillary osteonecrosis in patient 5.


6.

Surgery, antibiotics (10 patients)
Cessation of bisphosphonate therapy, antibiotics,
surveillance (24 patients);
Antibiotics, surveillance (11 patients) (Table 1).

161

Fig. 3. Mandibular osteonecrosis in patient 1.

3. Results

Surgical treatment included sequestrectomy or debridement of varying extent until viable, bleeding surgical bone
margins were found. Hyperbaric oxygen therapy was
administered in 10 cases (16.67%) with or without
previous cessation of bisphosphonate therapy, and multiple
biopsies and cultures from the sites of the lesions were
performed. The antibiotic administration included clindamycin for the first 2 weeks, amoxycillin and clavulanic
acid for another 2 weeks, and penicillin G for a period that
depends on the microbiology culture of the defect. Other
conservative procedures such as 0.12% clorohexidine
antiseptic mouthwash and teeth restoration were part of
the treatment protocol when needed. The mean follow-up
period was 5 months (2–8 months).

All the biopsy specimens taken from the area of necrotic
bone including both soft and hard tissue were negative for
malignancy, whereas culture results revealed normal oral
flora with sporadic cases of Actinomyces israeli , Escherichiacoli , and Bacteroides melaninogenicous .
For the first 7 patients, the implementation of protocols
1 and 2 resulted in complete wound healing (Figs. 1-5).
In the next 18 patients, treatment following protocols 3,
4, and 5 lead to bone necrosis recurrence of minor,
moderate, and wide extent, respectively. In the following
24 patients, treatment protocol 6 resulted in bone
necrosis stabilization with no further bone exposure. For
those patients, surgery will be the next step as soon as
bisphosphonate therapy cessation reaches 6 months. Lastly
the implementation of the treatment protocol 7 in 11 patients
had destructive results and lead to the progression of
bone necrosis (Table 1). It is worth mentioning that

Fig. 2. Complete wound healing after treatment in the same patient.

Fig. 4. Orthopantomographic view of mandibular osteonecrosis in the
same patient.




162

C. Magopoulos et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158–163

Fig. 5. Complete wound healing after treatment in the same patient.

meticulous oral hygiene had good impact on amelioration of
the disease.
4. Discussion
The high incidence of osteonecrosis—especially maxillary—in patients who are receiving bisphosphonates and
have not previously received radiation therapy in the head
and neck region is in contrast with the current knowledge.
Marx et al [3 ] first reported osteonecrotic lesions of the jaws
similar to those of osteoradionecrosis, which they called
drug-induced avascular osteonecrosis, implying that the
destruction of the vascular complexity of the jaws mediated
by the use of bisphosphonates is responsible fornecrosis
and secondary infection of the bone matrix.
Based on the apoptotic mechanism induced by nitrogenbisphosphonates (N-BPs) (especially zoledronate that was
administered to all patients) concerning both osteoclasts and
keratinocytes and on the observation that the osteonecrotic
lesion in all cases involved only the attached gingiva, we
came up with the hypothesis that the combination of
osteoclast and keratinocyte apoptosis, which reduce and
destruct the immune keratinocyte barrier of oral mucosa,
consists a possible pathway that leads to osteonecrosis of the
jaws [5,6]. In addition, the secondary infection established
which in some cases is rather offensive (A israeli , E coli , B
melaninogenicous ), produces a low pH environment (high
acidity) where bisphosphonates may be released from bone
surfaces, giving rise to locally high concentration of
bisphosphonates in solution or as calcium salts. This
enhances the hypothesis of the increased apoptotic rate of
keratinocytes of the attached gingiva and could explain in a
way the reason why osteonecrosis occurs only in the oral
and maxillofacial region and not elsewhere in the body
skeleton [7].
Little is known about the bisphosphonate-induced
compromise in vascularity of the bone matrix of the jaws
through underexpression of vascular endothelial growth

factor (VEGF) and endothelial cells’ apoptosis [8]. Further
investigation may give enough evidence for the explanation
of the correlation of VEGF underexpression and osteonecrosis of the jaws.
The management of these patients with bisphosphonateinduced osteonecrosis of the jaws is quite challenging. In a
review of 63 cases of osteonecrosis ofthe jaws associated
with the use of bisphosphonates [9], HBO therapy administered in 2 patients before undergoing surgical treatment
had no significant positive impact on the necrotic process.
In addition, in the same study, the cessation of bisphosphonate therapy alone did not reveal any improvement of
the bone necrosis. Recently, it has been suggested
minimally invasive procedures, antibiotics’ administration
along with 0.12% clorohexidine antiseptic mouthwash
without resolution of the exposed bone [10]. The authors
also pointed out the similarity of osteonecrosis of the jaws
with the bone defects encountered in osteopetrosis, which
we think is not the same. Osteopetrosis is a dysplasia of
endochondral ossification, affecting primary spogniosa
caused by an inherited osteoclastic dysfunction and not
apoptosis of normal functioning osteoclasts. Osteopetrosis
is characterized by failure in resorption and remodeling of
primary (immature) spogniosa by osteoclasts, resulting in
the accumulation of calcified cartilage matrix packing the
medullary cavity, which is radiologically quite obvious
while reflected by dense osteosclerosis [11]. Conversely
in osteonecrosis of the jaws, such radiological features
are not encountered in the entire bone, neither mandible
nor maxilla. Therefore, we believe that the exposed bone
can be debrided to a viable bone margin provided that
cessation of bisphosphonate therapy would allow complete
wound healing. Several other recommendations for the
surgical treatment of this entity have been made with good
results [12-14].
In our study, cessation of bisphosphonate therapy for
more than 6 months followed bydebridement/sequestration
of the defect with concomitant antibiotics administration
resulted in complete wound healing for the first 7 patients.
Reversely, debridement/sequestration of the necrotic bone
without prior cessation of bisphosphonate therapy was not
successful, whereas cessation of bisphosphonate therapy
alone without any invasive procedures resulted in stabilizing
the defect.
Therefore, we believe that discontinuity of bisphosphonate therapy—if possible—combined with surgical debridement to obtain clear and bleeding margins together with
long-term antibiotic therapy administration is the treatment
of choice for the osteonecrotic lesions of the jaws. Although
the exact period of drug cessation remains unclear and
difficult to define, more than 6 months of cessation seems to
be unfailing. The antibiotic administration includes clindamycin for the first 2 weeks, amoxycillin and clavulanic acid
for another 2 weeks, and penicillin G for a period that
depends on the microbiology culture of the defect.
Hyperbaric oxygen administration seems to be helpful in


C. Magopoulos et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 28 (2007) 158–163

this direction, especially for major mandibular defects, but it
is needed to be clarified by with sufficient data.
However, critical questions remain unanswered, for
example, is there any interaction between bisphosphonates
and chemotherapy for the development of osteonecrosis? To
the best of our knowledge, there are no published data
suggesting an interaction between bisphosphonates and
chemotherapy concerning osteonecrosis of the jaws. Furthermore, chemotherapy had already been ceasedto all
patients of the present study by the time of bisphosphonate
therapy initiation. However, corticosteroids were administered at times to some of the patients with unclear
involvement in the development of bone necrosis. Further
research and prospective clinical studies are necessary to
clarify the exact mechanism of action of bisphosphonates
and the biochemical interactions between bisphosphonates
and other agents.
In conclusion, this new complication of bisphosphonate
therapy administration, such as osteonecrosis of the jaws
seems to be rather developing. The challenging management
of this entity should make clinicians reconsider the merits of
the rampant use of bisphosphonates, whereas further
investigation is needed to elucidate this complication.

References
[1] Ebetino FH, Francis MD, Rogers MJ, et al. Mechanisms of action of
etidronate and other bisphosphonates. Rev Contemp Pharmacother
1998 - 43.
[2] Rogers MJ, Gordon HL, Benford FP, et al. Cellular and molecular
mechanisms of action of bisphosphonates. Cancer 2000 - 78
[Suppl: Skeletal Complications of Malignancy].

163

[3] Marx RE. Pamidronate (AREDIA) and zoledronate (ZOMETA
induced avascular necrosis of the jaws: a growing epidemic. J Oral
Maxillofac Surg 2003 - 8.
[4] Marx RE, Stern D. Oral and maxillo-facial pathology: a rationale for
diagnosis and treatment, 1st ed Carol Stream (Ill Quintessence; 2003.
p. 36 - 8.
[5] Fleisch H, Reszka A, Rodan G, et al. Bisphosphonates—mechanism
of action. In: Bilezikian J, Raisz L, Rodan G, editors. Principles of
bone biology, vol. 2. Academic Press, 2002. p. 1361 - 78.
[6] Reszka AA, Halasy-Nagy J, Rodan GA.Nitrogen-bisphosphonates
block retinoblastoma phosphorylation and cell growth by inhibiting
the cholesterol biosynthetic pathway in keratinocyte model for
esophageal irritation. Mol Pharmacol 2001 - 202.
[7] Sato M, Grasser W, Endo N, et al. Bisphosphonate action. Aledronate
localization in rat bone and effects on osteoclast ultrastructure. J Clin
Invest 1991 - 105.
[8] Wood J, Bonjean K, Ruetz S. Novel antiangiogenic effects of the
bisphosphonate compound zoledronic acid. J Pharmacol Exp Ther
2002 .
[9] 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 2004 - 34.
[10] 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 Maxillofac Surg 2005
63:1567 - 75.
[11] Key L, Ries W. Osteopetrosis. In: Bilezikian J, Raisz L, Rodan G
editors. Principles of bone biology, vol 2. San Francisco, CA
Academic Press, 2002. p. 1217 - 27.
[12] Lenz J-H, Steiner-Krammer B, Schmidt W, et al. Does avascular
necrosis of jaws in cancer patients only occur following treatment with
bisphosphonates? J Cranio-Maxillofac Surg 2005 - 403.
[13] Ficara G, Beninati F, Rubino I, et al. Osteonecrosis of the jaws in
periodontal patients with a history of bisphosphonates treatment.
J Clin Periodontol 2005 - 8.
[14] Farrugia MC, Summerlin DJ, Krowiak E, et al. Osteonecrosis of the
mandible or maxilla associated with the use of new generation
bisphosphonates. Laryngoscope 2006 (1):115 - 20.