Osteonecrosis of the Jaw: Risk Factors, Prevention, and Treatment for Patients on Bone-Loss Therapy

What Is Medication-Related Osteonecrosis of the Jaw?

Medication-related osteonecrosis of the jaw (MRONJ) is defined by the American Association of Oral and Maxillofacial Surgeons (AAOMS) as exposed bone in the maxillofacial region that does not heal within 8 weeks, occurring in a patient who is currently taking or has previously taken an antiresorptive or antiangiogenic medication, with no history of radiation therapy to the jaws [1]. The term replaced the older "bisphosphonate-related osteonecrosis of the jaw" (BRONJ) in 2014 after cases linked to denosumab and bevacizumab made the original name too narrow.

Patients often first notice a rough area of exposed bone along the gum line, sometimes painless, sometimes accompanied by swelling, infection, or numbness. The condition ranges from an incidental radiographic finding (stage 0) to large areas of exposed, infected bone with pathologic fracture (stage 3) [1]. Most osteoporosis-dose cases fall into stages 1 or 2 and respond to conservative management, including antimicrobial rinses and limited debridement. Surgical resection is typically reserved for stage 3 disease or cases refractory to conservative care [2].

The pathophysiology is not fully settled. Bisphosphonates and denosumab suppress osteoclast activity, which slows the normal bone turnover that repairs micro-damage in the jaw. Because the jaws experience higher remodeling rates than the appendicular skeleton, they are disproportionately affected by this suppression [3].

How Common Is ONJ in Osteoporosis Patients?

The absolute risk is low. A 2015 task force report from the American Society for Bone and Mineral Research (ASBMR) estimated the incidence of ONJ in patients taking oral bisphosphonates for osteoporosis at 1.04 to 69 per 100,000 patient-years, with most cohort studies clustering between 1 and 10 per 100,000 [4]. For comparison, the annual fracture risk in untreated postmenopausal women with a T-score of -2.5 exceeds 3,000 per 100,000. The risk-benefit ratio favors treatment by a wide margin.

Oncology patients face a different equation. Those receiving monthly intravenous zoledronic acid (4 mg) or denosumab (120 mg) for bone metastases develop ONJ at rates of 1% to 15%, depending on treatment duration and concurrent chemotherapy [5]. This is roughly 100 times the osteoporosis-dose risk, a distinction that often gets lost when patients search for ONJ online and encounter oncology statistics.

Duration matters. A Kaiser Permanente cohort study (N = 572,606) found that the ONJ risk on oral bisphosphonates was 0.1% after 4 or more years of use, compared to nearly undetectable rates in the first 2 years [6]. The AAOMS position paper states: "For patients receiving antiresorptive therapy for osteoporosis, the risk of developing ONJ is low, and the benefits of therapy in reducing the risk of osteoporotic fractures far outweigh the risk of developing ONJ" [1].

Which Drugs Carry the Highest Risk?

Not all antiresorptive agents carry equal ONJ risk. The distinction rests on dose, route, potency, and skeletal half-life.

Intravenous zoledronic acid at oncology doses (4 mg monthly) carries the highest documented risk, with the FREEDOM extension trial (denosumab arm, N = 4,550) reporting a cumulative ONJ incidence of 5.2 per 10,000 participant-years over 10 years of denosumab 60 mg every 6 months [7]. Oral alendronate at standard doses (70 mg weekly) sits at the lower end. A meta-analysis of 18 observational studies calculated a pooled incidence of 0.024% for oral bisphosphonates vs. 1.16% for intravenous oncology-dose bisphosphonates [8].

Denosumab at the osteoporosis dose (60 mg subcutaneously every 6 months) shows ONJ rates comparable to oral bisphosphonates. In the FREEDOM trial (N = 7,808), only 2 adjudicated ONJ events occurred in the denosumab group during the initial 3-year period [9]. One notable difference: because denosumab does not incorporate into bone, its antiresorptive effect is fully reversible within 6 months of the last injection. This may allow a recovery window before elective dental surgery, a pharmacologic advantage bisphosphonates do not share.

Risk Factors That Increase ONJ Probability

The strongest predictor is invasive dental work. A systematic review by Yamazaki et al. (2019) found that tooth extraction preceded 52% to 61% of all MRONJ cases across study populations [10]. Other established risk factors include:

Drug-related factors. Higher cumulative dose, longer duration of therapy, intravenous route, and concurrent corticosteroid use all increase risk. Patients on both antiresorptive therapy and antiangiogenic agents (bevacizumab, sunitinib) face compounding suppression of bone and vascular repair [5].

Local factors. Pre-existing dental or periodontal disease, ill-fitting dentures causing mucosal trauma, and anatomical sites with thin mucosa (mandibular tori, mylohyoid ridge) are common local contributors. The mandible is affected roughly twice as often as the maxilla, likely because of its denser cortical bone and lower blood supply [1].

Systemic factors. Diabetes, smoking, immunosuppression, and chemotherapy all impair wound healing. A retrospective cohort from the University of Southern California (N = 2,292 patients on antiresorptive therapy) found that diabetes increased ONJ risk by 2.7-fold after adjusting for drug type and duration [11].

Dr. Salvatore Ruggiero, one of the lead authors of the AAOMS position paper, has noted: "The most effective strategy for reducing ONJ risk is a thorough dental evaluation before starting antiresorptive therapy, combined with a maintenance program that minimizes the need for future invasive procedures" [1].

Prevention: The Pre-Treatment Dental Visit

A dental screening before starting bisphosphonates or denosumab is the single most effective ONJ prevention measure. The 2022 AAOMS update recommends completing all necessary extractions, implant placements, and periodontal treatments before the first dose, then allowing mucosal healing for 2 to 3 weeks [1].

For patients already on therapy, the approach depends on urgency and drug pharmacology. If a tooth extraction becomes necessary while on denosumab, some clinicians delay the procedure until 4 to 5 months after the last injection, performing the extraction near the end of the dosing interval when osteoclast suppression is at its nadir. Evidence supporting this "end-of-cycle" timing is observational, not randomized, but the biological rationale is sound [12].

For patients on oral bisphosphonates, the older recommendation of a "drug holiday" 2 months before surgery has fallen out of favor. The AAOMS no longer endorses a mandatory drug holiday for osteoporosis-dose oral bisphosphonates prior to dental procedures because the level of osteoclast suppression at these doses is modest, and the evidence for benefit is weak [1]. Prophylactic antibiotics and atraumatic surgical technique are more consistently supported.

Post-procedure monitoring is straightforward. Patients should be seen 1 week and 4 weeks after extraction to confirm mucosal closure. Any exposed bone persisting beyond 8 weeks warrants specialist referral.

Atypical Femoral Fractures: The Other Rare Complication

Atypical femoral fractures (AFF) share the same underlying mechanism as ONJ: excessive suppression of bone remodeling. These are low-energy, transverse fractures of the femoral shaft that occur in areas of cortical bone subjected to high mechanical stress. The ASBMR task force revised its case definition in 2014, requiring the fracture to be located along the femoral diaphysis from just distal to the lesser trochanter to just proximal to the supracondylar flare [13].

The absolute risk is small but increases with duration. A Swedish national registry study (N = 12,777 women with subtrochanteric or femoral shaft fractures) calculated an AFF incidence of 11 per 100,000 person-years among bisphosphonate users, compared to 2 per 100 to 000 in non-users [14]. After more than 5 years of continuous bisphosphonate use, relative risk rose to 3- to 10-fold above baseline, depending on the study.

Prodromal thigh or groin pain is the key warning sign. Approximately 70% of AFF patients report weeks to months of aching in the anterior thigh before fracture [13]. Any patient on long-term bisphosphonates who develops new thigh pain should undergo full-length femoral radiographs, and clinicians should examine both legs because bilateral involvement occurs in up to 28% of cases [14].

The relationship between AFF and drug holidays is well studied. Risk decreases rapidly after bisphosphonate discontinuation. In the Kaiser Permanente cohort, AFF risk dropped by roughly 70% within 2 years of stopping alendronate [15]. This declining risk profile is a primary justification for bisphosphonate drug holidays after 5 years of oral therapy or 3 years of intravenous zoledronic acid, as recommended by the ASBMR and the Endocrine Society [16].

Denosumab Discontinuation and Rebound Vertebral Fractures

Stopping denosumab is not straightforward. Unlike bisphosphonates, which persist in bone for years after discontinuation, denosumab's effect on osteoclast suppression reverses completely within 6 months of the last dose. Bone turnover markers rebound to above-baseline levels, and bone mineral density (BMD) returns to pre-treatment values within 12 to 18 months [17].

The clinical concern is rebound vertebral fractures. A post-hoc analysis of the FREEDOM trial and its extension identified multiple vertebral fractures in 5.6% of patients who discontinued denosumab after the open-label extension, compared to 0.8% who continued treatment. Patients with a prior vertebral fracture were at the greatest risk [18]. The European Calcified Tissue Society (ECTS) issued a position statement in 2017 warning that denosumab discontinuation without a follow-on antiresorptive agent "exposes patients to the risk of rapid bone loss and multiple vertebral fractures" [19].

The recommended transition strategy is to start an oral or intravenous bisphosphonate 6 months after the last denosumab injection (i.e., at the time the next dose would have been due). Zoledronic acid 5 mg IV as a single dose is the most studied approach, though the optimal timing remains debated. A randomized trial by Everts-Graber et al. (2020, N = 60) found that a single dose of zoledronic acid given 6 months after the last denosumab injection preserved BMD at the lumbar spine and total hip over 24 months, while untreated controls lost 5.2% at the spine [20].

Dr. Michael McClung, founding director of the Oregon Osteoporosis Center, has stated: "Denosumab should be considered a long-term or indefinite therapy. If discontinuation is planned, transitioning to a bisphosphonate is not optional. It is a clinical requirement" [19].

How ONJ Is Treated When It Occurs

Management depends on stage. The AAOMS staging system guides treatment intensity [1]:

Stage 0 (no exposed bone, but nonspecific symptoms or radiographic findings): systemic pain management, close monitoring, and patient education. No surgical intervention.

Stage 1 (exposed, necrotic bone without infection): chlorhexidine 0.12% oral rinse twice daily, clinical follow-up every 6 to 8 weeks, and avoidance of sharp bony edges that traumatize adjacent soft tissue.

Stage 2 (exposed bone with infection, evidenced by pain, erythema, or purulent drainage): oral antibiotics (typically amoxicillin 500 mg three times daily, or clindamycin 300 mg three times daily for penicillin-allergic patients), chlorhexidine rinses, and superficial debridement of loose sequestra. Most osteoporosis-dose ONJ cases stabilize at this stage.

Stage 3 (exposed bone with fracture, extraoral fistula, or osteolysis extending to the inferior border): surgical resection of necrotic bone with reconstruction. This is uncommon in the osteoporosis population and more often seen in oncology patients after prolonged high-dose intravenous therapy.

Resolution rates are encouraging for osteoporosis-dose cases. A retrospective Italian multicenter study (N = 567 MRONJ patients) reported complete mucosal healing in 33% and disease stabilization in an additional 51% with conservative treatment alone over a median follow-up of 24 months [21]. Surgical treatment achieved complete healing in 71% of operated cases.

Balancing Risk: When Treatment Benefits Outweigh ONJ and AFF Concerns

Fear of ONJ and AFF drives a measurable treatment gap. An analysis of U.S. Medicare claims found that bisphosphonate prescriptions fell by 50% between 2008 and 2012, even as hip fracture rates in women over 65 began rising for the first time in decades [22]. The Endocrine Society's 2019 clinical practice guideline explicitly addressed this imbalance, stating that "the benefits of pharmacological therapy in patients at high fracture risk far exceed the risks of rare complications including ONJ and AFF" [16].

The numbers make the case. A woman aged 70 with a hip T-score of -2.5 and a prior vertebral fracture has a 10-year hip fracture probability exceeding 10% without treatment. Alendronate reduces hip fracture risk by 40% to 50% based on the Fracture Intervention Trial (FIT, N = 6,459) [23]. Weighed against an annual ONJ risk of roughly 0.01%, the benefit-to-risk ratio exceeds 1,000 to 1.

Clinicians should frame these numbers in absolute terms for patients, not relative ones. Saying "bisphosphonates double your ONJ risk" is technically accurate but misleading when the baseline is 0.005%. Saying "1 in 10,000 patients per year may develop a treatable jaw condition, while 1 in 10 patients per year will fracture without treatment" communicates the same data more honestly.

Patients at the highest ONJ risk (those requiring frequent dental extractions, current smokers with periodontal disease, or those on concurrent immunosuppression) deserve individualized conversations. For the remaining 95% of osteoporosis patients, withholding antiresorptive therapy because of ONJ fear causes more harm than it prevents.