Bisphosphonates and jaw osteonecrosis

Zosia Chustecka

May 20, 2005

May 20, 2005

Iowa City, IA - As the FDA issues another warning about jaw osteonecrosis associated with the use of 2 injectable bisphosphonates used in cancer patients, an oral surgeon warns that these cases may be "only the tip of the iceberg" and says that, because they are so widely used, "even oral bisphosphonate may yet prove to be of clinical concern [ 1 ]." The warning is issued by Dr John Hellstein (clinical professor of oral pathology, University of Iowa, Iowa City), in the May 2005 issue of the Journal of Oral and Maxillofacial Surgery and publicized in a press release issued by the American Association of Oral and Maxillofacial Surgeons.

This concern is echoed by osteoporosis expert Prof Gordon Strewler (Beth Israel Deaconess Hospital, Boston, MA), who says, "We may be on the leading edge of a wave of cases." He tells rheuma wire that more than 800 cases have now been reported to the FDA and says, "I certainly do not think this issue is trivial. . . . We may have to rethink the use of high-dose bisphosphonates in cancer—mainly multiple myeloma and breast cancer with bone metastases. The dose used orally in osteoporosis is much lower, but there have been a few cases reported on alendronate and risedronate." He also emphasizes that the effect of these drugs persists even after the patient stops treatment: "They are not cleared from the bone for years."

However, Dr Nelson Watts (Bone Health and Osteoporosis Center, University of Cincinnati, OH) tells rheuma wire : "I have not seen any osteonecrosis of the jaw in my patients with osteoporosis, nor have any of the colleagues whom I have spoken with about it. I doubt that the risk is particularly great in patients being treated for osteoporosis, but 'better safe than sorry' would apply to patients who have ongoing dental problems."

Warning issued for injectable products

The warning from the FDA has been issued in conjunction with Novartis about its 2 injectable products, pamidronate (Aredia) and zoledronic acid (Zometa), both of which are used in cancer patients for the treatment of hypercalcemia of malignancy and for bone metastases, mainly in patients with multiple myeloma and breast cancer but also in some other solid tumors. However, zoledronic acid (as Aclasta, Novartis) is also in late clinical trials for osteoporosis, for which it is thought to have huge potential, as it may be effective with only 1 injection annually; it is also in clinical trials in Paget's disease and rheumatoid arthritis (RA).

The risk-benefit considerations are different for people with malignancy compared to people being treated for a benign bone disease.

The latest warning, issued earlier this month, is directed at dental health professionals and recommends that cancer patients receive a dental examination before starting intravenous treatment with either of these drugs and avoid invasive dental procedures while on the treatment [ 2 ]. An earlier warning and a "Dear doctor" letter, issued in September 2004, alerted all prescribers to reports of jaw osteonecrosis in patients taking these drugs [ 3 ]; at that stage, Novartis said it was aware of 500 cases. There had also been an earlier update on the product labels in August 2003.

The issue of jaw osteonecrosis associated with pamidronate and zoledronic acid was also discussed at a meeting of the FDA Oncologic Drugs Advisory Committee on March 4, 2005 [ 4 ]. FDA official Dr RichardPazdur asked the meeting to focus only on that issue; although there have been anecdotal reports of jaw osteonecrosis with oral bisphosphonates used in osteoporosis, there are much fewer data on this, he said, and "the risk-benefit considerations are different for people with malignancy compared with people being treated for a benign bone disease."

First spotted by oral surgeons

The complication was first spotted by oral surgeons. Dr Robert Marx (Miami, FL) reported a series of 36 patients in September 2003 [ 5 ]. All but 1 of these patients were on pamidronate, zoledronate, or both for the indication of hypercalcemia related to multiple myeloma or metastatic breast cancer; 1 patient was reportedly taking the drug for osteoporosis. Most patients presented with painful exposed avascular bone in the mandible, the maxilla, or both; in most of the patients (28/36, 77.7%) it followed removal of a painful tooth, but in 8 patients it developed spontaneously. Dr Salvatore Ruggierro and colleagues (Long Island Jewish Medical Center, New Hyde Park, NY) reported a series of 63 cases last year [ 6 ]. Most of these patients were taking pamidronate or zoledronate and had multiple myeloma (28 patients), breast cancer (20 patients), or other cancers, but there were also 7 patients being treated for osteoporosis—5 patients taking oral alendronate, 1 patient on oral risedronate, and 1 patient on oral alendronate as well as zoledronic acid. These 7 patients had no of history malignant disease or chemotherapy exposure, the researchers note.

"Typical presentation is a nonhealing tooth-extraction socket or exposed jawbone, both of which appear to be refractory to conservative debridement and antibiotic therapy," Ruggierro et al comment. Typical presenting symptoms were pain and exposed bone at the site of a previous tooth extraction, but 9 of the 63 patients (14%) had no history of a recent dentoalveolar procedure and nevertheless presented with spontaneous exposure and necrosis of the alveolar bone. In addition, 6 patients had radiographic signs of osteolysis before teeth were removed, which suggests involvement of the alveolar bone before extraction. Ruggierro et al also reported that cessation of bisphosphonate treatment did not have a major impact: 5 patients had persistent bone necrosis and even developed new regions of exposed bone, despite being removed from bisphosphonate therapy by their oncologists.

These reports prompted Dr Brian Durie (Cedars-Sinai Outpatients Cancer Center, Los Angeles, CA), a cofounder of the International Myeloma Foundation (IMF) patient advocacy group, to conduct a survey, recruiting a total of 1203 patients via cancer websites and email lists. Durie reported the findings at the American Society of Hematology in December 2004 and has posted the presentation on the IMF website [ 7 ]. Of 904 patients with myeloma who responded, 62 reported jaw osteonecrosis and 54 reported "suspicious findings" suggesting osteonecrosis, such as bone erosions, bone spurs, or exposed bone in the jaw. Of 299 with breast cancer who responded, 13 reported osteonecrosis and 23 reported suspicious findings.

Commenting on this survey to rheuma wire , Strewler notes that the results are preliminary, the responses were anonymous, and there was no follow-up to confirm diagnosis. "The overall incidence is strikingly high in the respondents to the survey," but he adds that symptomatic patients were more likely to respond, and so there are reasons to doubt this frequency data. But he also notes that the data show a sharply rising incidence after months of treatment with bisphosphonates and no relationship to other drugs being used in treatment.

When he was presenting results from the survey at the FDA Advisory meeting in March, Durie commented on the time frame involved. The average length of time patients had taken zoledronic acid before reporting osteonecrosis was 18 months (although some cases were seen after 6 and 12 months), and the average length of time with pamidronate was 6 years. "There is a coincidence of the time frames related to the time since these drugs have been on the marketplace," he commented. Zoledronic acid was approved for hypercalcemia of malignancy in August 2001 and had a broad bone-metastases indication added in February 2002; pamidronate was approved by the FDA for use in hypercalcemia of malignancy in 1991, multiple myeloma in 1995, and osteolytic bone metastases in breast cancer in 1996.

At the FDA hearing, Novartis said that its pamidronate product, Aredia, had been used by 1.9 million patients since 1991 but added that it was no longer used much in the US, having been replaced by generic pamidronate. The company said 1 million patients have used zoledronic acid since 2001 and that use of this drug had been "increasing recently." As of February 22, 2005, Novartis had 875 reports of jaw osteonecrosis. On the basis of these spontaneous reports and the patient numbers involved, company researcher Dr Diane Young (vice president of clinical development, oncology, Novartis), an oncologist by training, came up with an incidence rate of 0.03% but commented that "this was certainly an underestimate." At the other extreme, the web-based survey from Durie gives an incidence of 6.2%, which is "likely to be an overestimate," she said. Going back over reports from controlled clinical trials (with follow-up from 5 to 18 months), Novartis found 2 cases in 1334 patients on pamidronate and 4 cases out of 2730 patients on zoledronic acid, giving an estimated incidence of around 0.15%. In addition, Young mentioned an ongoing review of patient case histories at the MD Anderson Cancer Center, Houston, TX; preliminary results show 18 cases from 963 patient reviews, giving an incidence of 1.9%.

These patients are in pain, they smell bad, they can't eat, and there's nothing I can do to fix it.

Young told the meeting that there was "an important range of severity" in this complication, and she commented that the cases she had personal experience with required only intermittent antibiotics and the patients continued on bisphosphonate therapy. However, oral surgeon Dr Felice O'Ryan (Kaiser Permanente, Northern California) described several very severe cases, including 1 woman who had a draining fistula on the outside of her face. She described the bone necrosis as spreading "like fire" across the jaw and said, "Nobody can tell me this is not an insignificant problem." She also noted that although patients stopped bisphosphonate therapy, not 1 case has shown healing or regression, but some became worse, and in her hands the complication has been unresponsive to antibiotics and topical therapies. "These patients are in pain, they smell bad, they can't eat, and there's nothing I can do to fix it."

Proposed mechanism

In his review article, Hellstein puts forward a theory as to how the complication may arise. Bisphosphonates work by blocking bone resorption, by preventing osteoclasts from removing old bone, but don't affect the osteoblasts and their ability to build new bone. Overall, the effect is a buildup of new bone, which is beneficial in osteoporosis and other conditions where the bones have become fragile. However, in the jaw area, this disruption of the osteoclast-osteoblast axis may be detrimental, particularly in individuals who have active periodontal disease and/or oral surgical procedures that result in bone exposure and the need for remodeling to accomplish normal healing. In these cases, impairment of the osteoclasts' ability to remove and thus repair or contain areas of "diseased" bone may cause the osteoblasts to "overbuild" or "wall off" the diseased bone, resulting in a sufficient volume of necrotic bone to cause destruction. Preexisting or superimposed bacterial cofactors produce sequestration and possible osteomyelitis and prevent regeneration over the exposed bone.

Similarities with historical cases caused by phosphorus

The complication has been termed "bisphosphonate osteochemonecrosis of the jaw," but Helllstein suggests that it can also be called "bis-phossy jaw," and he argues that this term has a certain historical cachet, as the complication shows similarities to "phossy jaw" seen with white phosphorus. This complication, also known as phosphonecrosis, seen in the mid-19th century in workers from the match-making industry; although they were often young and otherwise healthy, they presented with an unusual necrosis of the jaws associated with severe pain and disfigurement. (There was also the dramatic effect of an eerie glow emanating from the mouth in a darkened environment, the result of phosphorus deposited in the bones).

White phosphorus, the cause of phossy jaw, appears to have similar skeletal effects as bisphosphonate therapy, Hellstein comments, and for this reason he believes it useful to recount some of the findings recorded about that complication. Some reports noted areas of bone exposure and sequestration, and the sequestra were often described as porous and light in weight and sometimes portrayed as either worm-eaten or pumicelike. However, this description pertains only to the sequestra, Hellstein notes. In contrast, the medullary bone was often observed to display increased trabeculations, and chronic phosphorus exposure produced a hyperostosis of the entire skeletal structure, with spongy bone transformed to sclerotic bone.

I cannot help but think that we may be witnessing only the tip of the iceberg in a possible bis-phossy jaw epidemic.

White phosphorus appeared to accumulate in the oral region, but whether this was because the oral cavity was the initial portal and thus more exposed than other regions and/or whether the faster bone turnover in the jaw resulted in phosphorus returning to the area is unknown. Of clinical importance are reports that it usually took white-phosphorus exposure of months or years before the complication was seen; in addition, the risk remained for years even after workers left their jobs, Hellstein notes. "I cannot help but think that we may be witnessing only the tip of the iceberg in a possible bis-phossy jaw epidemic," he comments.

One similarity is that both phossy jaw and the bisphosphonate jaw complications have a bacterial component in the disease process, whereas another similar complication—osteoradionecrosis (seen in workers painting luminous watch dials with radium) is different in that suppuration is uncommon. In the review of 20 cases carried out by Hellstein and colleagues, in which bisphosphonates were confirmed or suspected as cofactors, bacteria were commonly associated with the nonvital bone fragments. The submitting clinicians invariably reported purulence or suspected infection and gave a clinical impression of osteomyelitis, leading Hellstein to "believe that a bacterial component is significant in bis-phossy jaw." However, it will take time and multiple studies to see whether bacteria play a primary role or merely represent secondary infections, he adds. The cases his team reviewed and the reported observation that topical chlorhexidine has proven therapeutically beneficial seem at least to indicate that control of surface bacteria may help the body to recover denudedareas of bone.

Hellstein concludes that more research is needed to see what population groups or oral factors are the best risk predictors of the complication. "There is no doubt that bisphosphonate therapy will continue to show substantial clinical benefits and grow in use. We need to discover the ideal dosage, delivery route, and bisphosphonate for each patient category."


  1. Hellstein JW and Marek CL. Bisphosphonate osteochemonecrosis (bis-phossy jaw): is this phossy jaw of the 21st century? J Oral Maxillofac Surg 2005; 63:682-690.

  2. Latest warning from FDA to dental health professionals. The complete MedWatch 2005 Safety summary, including links to the Novartis Dear Healthcare Professional letter and revised labels for Zometa and Aredia. May 5, 2006. Available at:

  3. Novartis Oncology. Changes made to the precautions and post-marketing experience sections of the Aredia (pamidronate disodium) injection and Zometa (zoledronic acid) injection prescribing information. September 24, 2004. Available at:

  4. Department of Health and Human Services. Food and Drug Administration. Center for Drug Evaluation and Research. Oncologic Drugs Advisory Committee. March 4, 2005. Available at:

  5. Marx RE. Pamidronate (Aredia) and zolendronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003; 61;1115-1118.

  6. Ruggiero SL, Mehrotra B, Rosenberg TJ. Osteonecrosis of the jaws associated with use of bisphosphonates: a review of 63 cases. J Oral Maxillofac Surg 2004; 62:527.

  7. International Myeloma Foundation



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.