April 1, 2007



A 54-year-old man is seen by his internist and dentist on consecutive days because of new oral sores. Has never had similar lesions. Known IgG myeloma complicated by hypercalcemia; for 3 years, has been taking dexamethasone for his myeloma, and pamidronate, 90 mg by monthly intravenous infusion, producing good control of hypercalcemia. No known myeloma involvement of skull or jaw. Denies trauma to the area. No recent extraction or periodontal surgery. Does not wear tongue ring or other intra-oral device.


Afebrile man, not cachectic or in pain. No hepatosplenomegaly or lymphadenopathy. Two symmetrically located intra-oral yellow areas that have been denuded of mucosa; each is surrounded by an erythematous rim. Each arises just above the floor of the mouth at the base of the tongue, abutting the gums. Neither sore contains surface exudate; no matter extrudes when it is touched with an instrument; neither one fluctuant.

The retractor has helped immensely by moving the tongue out of the way-a step as easily taken by a physician with a tongue blade as by a dentist with the kind of dental mirror illustrated. Relatively uniform erythema surrounds the area of mucosal loss; the redness is punctuated by two faint yellow areas that may represent earlier lesions or satellite lesions. Nothing resembles a vesicle at any stage, including pustule formation, drying, unroofing, or resolution. The erythema is not as intense as in a conventional high-grade inflammatory lesion, nor is there clear-cut bleeding.

A diagnosis of osteonecrosis due to bisphosphonates was made.1-7

Additional observations included abundant saliva, eliminating any sicca element in causation. There were no fillings in the immediately adjacent teeth and the teeth were not mobile; thus, a periapical abscess at the base of the tooth root was unlikely. Small portions of tooth roots-seen to the reader's right-hinted at a probable history of current or past periodontal disease. There was no bleeding or pus about the root of any tooth, specifically including those contiguous with the lesion.


The diagnosis incorporated knowledge of the predisposition to this lesion in persons receiving high-dose bisphosphonate therapy; the locale; lesion morphology; and the lack of compelling features for several other differential diagnoses.

Osteoradionecrosis represents another key differential diagnosis that can be excluded by the history. Exposed necrotic alveolar bone can develop in persons who receive high-dose radiotherapy to the mouth or jaw, often for squamous cell cancer, though sometimes for other malignant neoplasms or even for metastases at this site. Hyperbaric oxygen therapy may promote bone revascularization in radionecrosis of bone, but bisphosphonate-related osteonecrosis does not improve with hyperbaric oxygen therapy,6 at least in part because the bisphosphonate has a half-life in bone that is measured in years. Thus, withdrawal ends only the application of new noxious stimulus, not the ongoing injury from extant, persistent intra-osseous drug.

Among other diagnostic possibilities are aphthae (aphthous ulcers). These should occur on unattached mucosa rather than as here, at a point where the epithelium is bound tightly to underlying tissues.8

In bound-down keratinized oral mucosa, infection with herpes simplex virus type I (HSV-I) would become a more substantial consideration; it cannot be excluded by the lack of intact vesicles, since these can be missed at a single "snapshot in time." However, the bilateral discontinuous large lesions, the rings of pale rather than bright erythema, and the absence of severe pain made this prospect fade.

Trauma on bound-down keratinized mucosa could result, classically, from a denture-rubbing injury, but this patient had no oral appliance, and also was considered a reliable reporter who denied any significant trauma.

Jaw abscess is less likely because of the paucity of inflammatory findings, the absence of pain, and the lack of nearby fillings or broken or decayed teeth that might predispose to deep infection.

Myelomatous involvement could be considered, but this pattern is not familiar in the literature, and ulceration is not expected (Figure 1). The inclusion of a case4 that proved to have a localized myeloma deposit (a plasmacytoma) that led to pathologic jaw fracture, in a paper on bisphosphonate-induced osteonecrosis,9 might broaden one's differential diagnosis, but it also clouds the distinctive presentation of bisphosphonate-induced osteonecrosis, which typically does not lead to fracture.9


Standard dental-medical co-management of this condition has evolved rapidly. Today it is suggested though not universally agreed that one should stop the bisphosphonate. All concur on providing topical and systemic measures, chiefly for hygiene and for control of any complicating-not causative-infection, as described fully in other literature.1-7,9,10 Lesional morphology has varied, and some cases have been considerably more disfiguring than this one (see the vivid illustrations in references 2 to 7 and in 9). Apart from effects of superinfection of ulcers, the mechanism of lesion variation has not yet been elucidated.

The oral lesions of the patient illustrated first worsened though remaining asymptomatic, such that the bone was more fully exposed and became mobile (Figure 2). By this point, the exposed bone has evolved from the yellow of necrosis at diagnosis, to white to cream-color, presumably because the necrotic bone became a sequestrum (separated from the viable) and was shed; what we see now is not the characteristic brown of macroscopic myelomatous involvement (Figures 3 through 6). Nor does it have a macroscopic brown and ratty appearance more typical of longstanding necrotic bone as illustrated by Nase and Suzuki7; such necrotic bone tends to show mottling radiographically as new bone tissue is produced adjacent to areas of infarction.

Two months later still, the area of former sequestrum had healed and became covered with intact epithelium (Figure 7) and the patient was doing well, a state maintained at least through his most recent follow-up 8 months later. Often, the exposed bone in such bisphosphonate-induced osteonecrosis does not follow this sequence but rather remains in situ, exposed.

This patient did not have recrudescent hypercalcemia even long after the bisphosphonate was stopped. This could reflect enduring benefit from the drug that remained embedded in bone; or that the underlying multiple myeloma had remitted and the trigger to abnormal bone turnover had thus been shut off.


The issue of whether patients need a drug holiday from conventional bisphosphonates has been debated, and there is not yet a consensus.2,3,10 Any sound conclusion about this topic is made more difficult by the years-long half-life of the drug in bone, such that any bone toxicity of any nitrogen-containing bisphosphonates might well persist and worsen even after a long drug holiday. Yet healing has occurred in many patients just as illustrated in this case, notwithstanding levels of the drug in bone that have not appreciably fallen in the interval.

Further unexplained mysteries remain: for instance, why does a process that causes death of bone result not in a purely osseous defect, but in sloughing of the overlying mucosa? Why does the inhibition of osteoclasts, which are "bone-chewing" cells, lead to death of bone? On the face of it, this is not sensible even if the drug has led to an imbalance between bone formation and resorption. One would envision unduly hard bones akin to those in a native disorder featuring lack of osteoclasts, namely osteopetrosis, yet the pathology is not this at all.

Why is the jaw selectively affected? Speculative but reasonable hypotheses abound--eg, that the jaws receive much more frequent if low-level traumatic damage from mastication than the bumps to which other bones are subjected, and that their cloak of mucosa is thin. That mucosa is in turn bathed in high concentrations of bacteria, some of which are pathogenic. Why does the pattern of involvement not accord with the relative vascularity of the maxilla versus that of the mandible? No scientifically proven conclusions exist.2-7,9,10


Injudicious publicity in the lay media led to a flurry of concern that every patient who takes alendronate might suffer a like fate. However, the rate of osteonecrosis from this bisphosphonate is exceptionally low.2-7,9,10 Most cases have resulted from the use of the much higher doses characteristic of treating painful bony lesions and hypercalcemia in multiple myeloma and in breast cancer with metastases.

Dental procedures such as extractions have preceded the development of bisphosphonate-induced osteo-necrosis in a substantial subset of cases, and they seem to have a relationship with the "dry socket with non-healing" that is known in the dental literature and less familiar to physicians. Some papers emphasize wound non-healing as a potential additional issue in patients who receive bisphosphonates and have dental surgical procedures.5

Despite the immensity of our ignorance, several principles of management enjoy relative consensus:

•For the patient about to begin conventional bisphosphonate therapy for prevention or treatment of osteoporosis, good dental hygiene and regular professional dental care seem to be sufficient.

•For those about to embark on higher doses, whether for Paget's disease of bone or for malignancy-related use, prior close consultation with a dentist is mandatory.

•If extractions or other oral surgical procedures are required, they should be completed, and healing complete if time permits, before commencement of high-dose bisphosphonate therapy.

•If extractions or other procedures are indicated in a person already receiving high doses of a bisphosphonate, a drug holiday should be initiated and the procedure deferred, if medically feasible, for some time thereafter.

We stand at the dawn of even the faintest understanding about this phenomenon, important both in its own right and for the insights it may yield about the biology of bone, the interaction between bone and adjacent soft tissue, and the necessary cooperation and collegiality of physician and dentist.

Schneiderman H, Goldblatt R. Osteonecrosis of jaw from high-dose bisphosphonate therapy. CONSULTANT. 2007;47: 413-423.


REFERENCES:1. Treister N, Woo SB. Images in clinical medicine. Bisphosphonate-associated osteonecrosis of the jaw. N Engl J Med. 2006;355:2348.
2. Bilezikian JP. Osteonecrosis of the jaw-do bisphosphonates pose a risk? N Engl J Med. 2006;355:2278-2281.
3. Woo SB, Hellstein JW, Kalmar JR. Narrative [corrected] review: bisphosphonates and osteonecrosis of the jaws. Ann Intern Med. 2006;144:753-761.
4. Ruggiero SL, Mehrotra B, Rosenberg TJ, Engroff SL. Osteonecrosis of the jaws associated with the use of bisphosphonates: a review of 63 cases. J Oral Maxillofac Surg. 2004;62:527-534.
5. Markiewicz MR, Margarone JE 3rd, Campbell JH, Aguirre A. Bisphosphonate-associated osteonecrosis of the jaws: a review of current knowledge. J Am Dent Assoc. 2005;136:1669-1674.
6. Melo MD, Obeid G. Osteonecrosis of the jaws in patients with a history of receiving bisphosphonate therapy: strategies for prevention and early recognition. J Am Dent Assoc. 2005;136:1675-1681.
7. Nase JB, Suzuki JB. Osteonecrosis of the jaw and oral bisphosphonate treatment. J Am Dent Assoc. 2006;137:1115-1119; quiz 1169-1170.
8. Schneiderman H. Recurrent aphthous ulcer (canker sore) and its mimics. Consultant. 1991(September);31(9):35-36.
9. Migliorati CA, Casiglia J, Epstein J, et al. Managing the care of patients with bisphosphonate-associated osteonecrosis: an American Academy of Oral Medicine position paper. J Am Dent Assoc. 2005;136:1658-1668.
10. Black DM, Schwartz AV, Ensrud KE, et al. Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial. JAMA. 2006;296:2927-2938.