Medication-Related Osteonecrosis of the Jaw: A Systematic Review of Case Reports and Case Series

Abstract

Medication-related osteonecrosis of the jaw (MRONJ) is a rare condition, typically seen in patients receiving antiresorptive or antiangiogenic drugs. This study aims to synthesize reports and case series of MRONJ regarding sociodemographic and clinical characteristics and to evaluate the effectiveness of the treatments applied. Following PRISMA guidelines, a search for case reports and case series was carried out in the PubMed-Medline database until March 2024. A total of 88 articles were included in this review, in a total of 151 cases. The key findings reveal that females were the most affected individuals (71% of the cases) with the average age at diagnosis being 66.27 years with a standard deviation of ±13.03. Stage 2 was the most observed stage, in 43% of cases, and zoledronic acid was the most commonly used drug (32% of cases). The oral route was the most common route of administration, in 26% of cases, with an average administration duration of 60.88 months (standard deviation ± 50.92). The mandible was the most commonly affected anatomical location (in 60% of cases). Amoxicillin + clavulanic acid (875 mg + 125 mg) and chlorhexidine (0.12%) were the most used antibiotics and mouthwash, with 16% and 26% of the cases, respectively. Surgical debridement was the most common surgical procedure, in 32% of cases, while the use of an L-PRF membrane was the most prevalent unconventional treatment, in 21% of cases. This study identified a statistically significant relationship between unconventional treatments and the cure of MRONJ (p < 0.001), indicating the need for further research to confirm these results.

1. Introduction

Medication-related osteonecrosis of the jaw (MRONJ) is a rare and debilitating multifactorial condition often associated with antiresorptive and antiangiogenic agents, used in the treatment of bone conditions and complications, such as osteoporosis, Paget’s disease, hypercalcemia, and metabolic bone lesions associated with malignancy [1,2,3].

According to the American Association of Oral and Maxillofacial Surgeons (AAOMS), the diagnosis of MRONJ requires the presence of three criteria simultaneously: current or previous treatment with antiresorptive or antiangiogenic agents, presence of visible or probe-able exposed bone in the maxillofacial region for more than eight weeks, and absence of a history of radiotherapy or metastasis in the jaws [1,2,4,5,6].

The AAOMS recognizes three classes of medications associated with MRONJ: bisphosphonates, RANKL ligand inhibitors, and antiangiogenic medications [1,2]. Bisphosphonates, derived from inorganic pyrophosphates, inhibit bone mineralization and resorption, promoting the apoptosis of osteoclasts [7]. Denosumab, a monoclonal antibody, binds to RANKL, inhibiting the formation and function of osteoclasts [6]. Although through different antiresorptive mechanisms, both drugs cause a decrease in the rate of bone turnover, leading to the accumulation of unresorbed hypermineralized bone [8]. In turn, antiangiogenic drugs inhibit the formation of new blood vessels, possibly leading to ischemia and hypoperfusion [6].

In 2003, Marx et al. reported the first cases of osteonecrosis of the jaws related to the use of bisphosphonates, and osteonecrosis of the jaw was subsequently recognized as an adverse effect of bisphosphonate treatment [2,6,7,9,10]. In 2010, Aghaloo et al. reported cases of osteonecrosis of the jaws related to the use of denosumab. Following this and other similar reports, and in order to include all the drugs implicated in osteonecrosis of the jaws, the AAOMS proposed changing the nomenclature from “Bisphosphonate-related osteonecrosis of the jaw” (BRONJ) to “Medication-related osteonecrosis of the jaw” (MRONJ) [2,6,7,9,10].

According to the AAOMS, the diagnosis of MRONJ is mostly clinical; however, the support of complementary diagnostic exams is advantageous in the diagnosis and evaluation of the progression of the disease, namely orthopantomography, computed tomography, and magnetic resonance imaging [8].

When it comes to recognized risk factors of MRONJ, antiresorptive medications such as bisphosphonates or denosumab and antiangiogenic medications such as aflibercept and sunitinib are associated with an increased risk of developing MRONJ [11]. It is necessary to evaluate the route of administration and duration of the treatment, with the risk being considered dose- and time-dependent, that is, as the dose increases over a longer period, the risk of developing osteonecrosis increases [2,11]. Several studies report that, among patients diagnosed with MRONJ, tooth extraction was a predisposing event in 62% to 82% of cases [11]. In turn, the presence of pre-existing periodontal or periapical infection in patients treated with antiresorptive or antiangiogenic medications is also a recognized risk factor for the development of MRONJ [2,11]. Finally, the development of MRONJ is more common in the mandible (75%) than in the maxilla (25%) and can occur in both jaws (4.5%) [2,11].

The AAOMS introduced a staging system in 2009, updated in 2022, to characterize the clinical presentation of MRONJ: patients at risk of developing MRONJ, stage 0, stage 1, stage 2, and stage 3 [5,8,11]. Patients at risk are asymptomatic and are undergoing treatment, or have been previously treated, with antiresorptive or antiangiogenic drugs, without evidence of necrotic bone. In stage 0, there is no evidence of necrotic bone, but patients present with nonspecific symptoms and clinical and imaging findings. In stage 1, it is possible to observe exposed and necrotic bone, or bone that can be probed through a fistula, in asymptomatic patients with no evidence of infection or inflammation, and it is also possible to observe radiographic findings characteristic of stage 0. In stage 2, it is possible to observe exposed and necrotic bone, or bone that can be probed through a fistula, in symptomatic patients, with evidence of infection or inflammation, and it is also possible to observe radiographic findings characteristic of stage 0. In stage 3, patients present exposed and necrotic bone, or bone capable of being probed through a fistula, with evidence of infection and at least one of the following signs: exposed necrotic bone extending beyond the region of the alveolar bone (lower border and ramus of the mandible, maxillary sinus, or zygomatic bone), pathological fracture, extraoral fistula, oroantral communication, and osteolysis extending to the lower border of the mandible or floor of the maxillary sinus [5,8,11].

The treatment of MRONJ is challenging and there is no consensus regarding the most appropriate management strategy [1,12]. According to the AAOMS, the main goals of treatment are to eliminate pain, control infection, and minimize the occurrence or progression of osteonecrosis [12]. The choice between surgical and non-surgical therapy must be specific and adapted to the needs of each patient. Nonsurgical strategies, such as administration of oral antiseptics and antibiotics, are useful, especially in patients with comorbidities that preclude surgery. Surgical therapy, including sequestrectomy, surgical debridement, and resective surgery, is a viable option, with high success rates [11].

For stage 0, management focuses on conservative care: educating the patient on oral hygiene, using symptomatic treatments like analgesics and antibiotics if needed, and applying chlorhexidine 0.12% rinses. Invasive dental procedures should be avoided to prevent disease progression. In stage 1, conservative management is recommended, including the use of a chlorhexidine rinse and regular follow-ups every 8–12 weeks. Systemic antibiotics are not typically needed unless infection occurs. Emerging treatments like low-level laser therapy (LLLT) may help with pain and inflammation. For stage 2, treatment recommendations include antibiotics (amoxicillin/clavulanate or clindamycin), chlorhexidine rinses, and analgesics. Limited debridement is one of the recommendations, but extensive surgery should be avoided. Hyperbaric oxygen therapy (HBOT) might aid healing, though its effectiveness is still debated. The most recent recommendations for stage 3 include broad-spectrum antibiotics, extensive surgical debridement or resection, and possible jaw reconstruction with flaps or grafts. Chlorhexidine rinses and intensive pain management along with infection control are critical. Adjunctive therapies like Leukocyte- and Platelet-Rich Fibrin (L-PRF) or Bone Morphogenetic Proteins (BMPs) may support healing, though evidence for these is mixed [11].

There is not yet sufficient evidence to recommend adjuvant therapies, namely L-PRF membranes, hyperbaric oxygen therapy, ozone therapy, use of vitamin E, pentoxifylline, and teriparatide, which is why these should not be recommended as the basis of MRONJ treatment. A multidisciplinary approach is essential, and collaboration between oncologists, rheumatologists, dentists, and oral and maxillofacial surgeons is essential for a comprehensive evaluation of the patient diagnosed with MRONJ [11].

The clinical practice of antiresorptive drug holidays to mitigate MRONJ risk in patients undergoing dentoalveolar surgery remains controversial. The historical use of a drug holiday was intended to decrease the prevalence of MRONJ subsequent to the performance of high-risk surgical procedures. The concern regarding this practice is the loss of efficacy of antiresorptive therapy with the development of skeletal-related events and fragility fractures. Factors for consideration may include disease-related risk (cancer vs. osteoporosis), drug-dosing frequency, duration of therapy, comorbidities, other medications (especially chemotherapy, steroids, or antiangiogenics), degree of underlying infection/inflammation, and extent of surgery to be performed. A special concern should be considered for suspending RANKL inhibitors in osteoporosis patients [11].

Several studies have demonstrated a rebound increase in bone resorption following the discontinuation of denosumab, resulting in an increased risk of multilevel vertebral fractures. If demosumab is to be suspended, the timing and duration of the holiday should be optimized in order to minimize this risk. The planned dentoalveolar surgery can be completed 3–4 months following the last dose of denosumab when the level of osteoclast inhibition is waning. It can then be reinstituted 6–8 weeks post-surgery. This management strategy minimizes the length of the drug holiday while maintaining a favorable environment for bone healing [11].

Actually, there is insufficient evidence to conclude that the use of the other interventions investigated would reduce the risk of MRONJ or would improve the healing of MRONJ [13].

2. Materials and Methods

This systematic review was carried out, according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) recommendations [14].

2.1. Search Strategy

An electronic search for case reports and case series was carried out, until March 2024, in the PubMed-Medline database, using the keywords “medication” AND “related” AND “osteonecrosis” AND “jaw”. The search was limited to human studies and articles written in the English language, and no time limit was set.

The following search was carried out: (“medication” AND “related” AND “osteonecrosis” AND “jaw”) AND (“Case Reports”[Publication Type] OR “Case Series”[Publication Type]) AND (“english”[Language]).

2.2. Selection Criteria

Reports and case series reported in the literature on medication-related osteonecrosis of the jaw were selected.

Letters to the editor, systematic or literature reviews, correspondence, and book chapters were excluded. Articles published in a language other than English and animal studies and articles whose text was not fully available were also excluded.

All cases that presented factors that could alter the immune response and, therefore, affect the healing process were also excluded in order to evaluate solely the influence of the medication on the development of osteonecrosis, namely poorly controlled diabetes mellitus (or without evidence of being controlled), rheumatoid arthritis, multiple myeloma, psoriatic arthritis, osteoarthritis, toxiphilic, alcoholic and smoking habits, and ingestion of corticosteroids and/or immunosuppressants [3,8]. Finally, cases in which patients underwent head and neck radiotherapy were also excluded, in order to exclude the possibility of osteoradionecrosis of the jaw. Therefore, all cases that included possible etiological factors of osteonecrosis of the jaw other than the drugs under study were excluded.

The selection, reading, and analysis of the articles included in this systematic review were performed by one researcher (F.F.) and were carried out manually. Articles that raised questions about their inclusion were reviewed by a second investigator (A.M.).

2.3. Data Extraction

Information regarding the parameters under study was collected manually. The articles were read, and the respective information was, in the first step, recorded in a Microsoft Excel 365 MSO spreadsheet (version 2403).

2.4. Study Risk of Bias Assessment

In order to evaluate the quality of the articles, the evaluation forms of case reports and case series from Joanna Briggs Institute–University of Adelaide were used [15].

2.5. Data Collected

Information was collected regarding the sex and age of the patients, stage of the disease, medication associated with the development of MRONJ (active ingredient, route of administration, and time of administration), anatomical location of osteonecrosis, therapy of MRONJ (antibiotic, mouthwash, surgery, and other treatments), and, finally, follow-up time from the beginning of treatment for the disease until its cure. All data were entered into the Microsoft^® Excel^® program for Microsoft 365 MSO (version 2403).

2.6. Data Analysis

Statistical analysis was performed using the SPSS^® program, version 28.0 (IBM, Armonk, NY, USA), and included descriptive statistics measures (absolute and relative frequencies) and inferential statistics (Fisher’s exact test).

3. Results

From the Boolean search in the PubMed-Medline database, using the aforementioned keywords, 178 results were obtained. In total, 144 articles were selected based on reading of the title. In the articles whose titles raised doubts, the respective abstracts were read; thus, 10 articles were selected. The 154 articles then selected were analyzed by the reviewer, and only those articles that had the full text available and that met the selection criteria were included. Therefore, 88 articles were included in this systematic review, corresponding to 62 case reports and 26 case series (Appendix A).

Figure 1 represents the flowchart with the systematization of the selection process.

Figure 1. PRISMA flow diagram of search results and screening process.

This systematic review included 151 cases of MRONJ for study. Not all articles included information on all the defined variables. For each variable, the actual number of cases providing the respective information is specified in the text.

3.1. Sex

Information regarding the sex of the patients was available in 150 cases. It was observed that females were the most affected, with 71% of cases (n = 107). In turn, males represented 28% of cases (n = 43), making a ratio of 2.49:1 (Figure 2).

Figure 2. Distribution of the population by sex (relative frequency and absolute frequency).

3.2. Age

Information regarding age was available in 146 cases. The average age at diagnosis was 66.27 years, with a standard deviation of ±13.03. The minimum age was 23 years, and the maximum age was 89 years.

In relation to females, the average age was 67.23 years, with a standard deviation of ±13.07, corresponding to a minimum age of 27 years and a maximum age of 89 years. In males, the average age was 64.17 years (standard deviation ± 12.87), the minimum age was 23 years, and the maximum age was 82 years.

The age group with the highest number of MRONJ cases was 60 to 70 years old (n = 50) (Figure 3).

Figure 3. Distribution of the population by age (absolute frequency).

3.3. Stage

The MRONJ stage was specified in 106 cases, according to the staging system proposed by AAOMS. Stage 1 was verified in 5% of cases (n = 7), stage 2 in 43% of cases (n = 65), and stage 3 in 23% of cases (n = 34). Thus, it is concluded that in the studied population, stage 2 of MRONJ was the most observed (Figure 4).

Figure 4. Distribution of the population by MRONJ stage (relative frequency and absolute frequency).

3.4. Medication Associated with Osteonecrosis: Active Ingredient

Information regarding the active ingredient associated with the development of MRONJ was available in 146 cases. The most used active ingredient was zoledronic acid, in 32% of cases (n = 60). The second most used active ingredient was alendronic acid, with 21% of cases (n = 40), followed by denosumab represented by 19% of cases (n = 36). Ibandronic acid was used in 4% of cases (n = 8) and bevacizumab in 3% of cases (n = 5). The other active ingredients each represent 1% of cases (n = 1 or n = 2). Of these, simvastatin and arsenic trioxide stand out, which, according to the literature, are not active ingredients commonly associated with MRONJ [16,17] (Figure 5).

Figure 5. Distribution of the population by active ingredient administered (relative frequency and absolute frequency).

It is important to highlight that the medications under study were not always used alone, with cases in which at least two active ingredients were administered to the same patient. In these cases, it was impossible to assertively determine whether just one of the active ingredients administered was associated with the development of MRONJ, or whether it was a combination of several.

3.5. Medication Associated with Osteonecrosis: Route of Administration

The route of administration of the active ingredients in question was available in 71 cases.

The most common route of administration was oral, with 26% of cases (n = 50), followed by intravenous, with 24% of cases (n = 46). In 5% of cases (n = 9), the active ingredient was administered subcutaneously, and there were 2 cases in which the active ingredient associated with MRONJ was administered through intravitreal administration (Figure 6).

Figure 6. Distribution of the population by route of administration (relative frequency and absolute frequency).

3.6. Medication Associated with Osteonecrosis: Administration Time

Information regarding the time between the first administration of the active ingredient and the diagnosis of MRONJ was available in 99 cases. The average administration time was 60.88 months, the minimum time passed was 1 month, and the maximum time was 252 months, with a standard deviation of ±50.92.

Of the cases in which the administration time was specified, it was observed that in a larger number of cases (n = 35), the administration time was between 34 and 67 months (Figure 7).

Figure 7. Distribution of the population by administration time (absolute frequency).

It should be noted that, in cases where more than one active ingredient that could be associated with the development of MRONJ was administered and which, therefore, had different administration intervals, only the first administration was considered, that is, the longest period of administration.

3.7. Anatomical Location

The anatomical location of osteonecrosis was available in 147 cases. The mandible was the most affected location, with 60% of cases (n = 91). In turn, the maxilla was affected in 32% of cases (n = 48). Finally, it was observed that, in only 5% of cases, both jaws were affected (Figure 8).

Figure 8. Population distribution by anatomical location of lesions (relative frequency and absolute frequency).

3.8. Targeted Treatment for Osteonecrosis (Antibiotics, Mouthwash, Surgery, and Others)

In general, the MRONJ therapy instituted included administration of antibiotics, use of mouthwash, performance of surgical procedures, and use of unconventional treatments for MRONJ. In 85% of cases, antibiotics were administered (n = 128); in 77% of cases, surgical procedures were performed (n = 117); in 61% of cases, mouthwash was administered (n = 92); and, finally, in 46% of the cases, non-conventional treatments were carried out (n = 70) (Figure 9).

Figure 9. Distribution of the population by therapy instituted (relative frequency and absolute frequency).

It is important to note that the types of treatments presented were used concomitantly in some cases, at least in combinations of two, and, in certain cases, all treatments presented were applied.

3.9. Targeted Treatment for Osteonecrosis: Antibiotics

Information regarding the antibiotic administered for the treatment of MRONJ was available in 116 cases. Of these, amoxicillin + clavulanic acid (875 mg + 125 mg) was the most administered antibiotic, in 16% of cases (n = 20). It is followed by amoxicillin + clavulanic acid (875 mg + 125 mg) administered together with metronidazole (500 mg) in 6% of cases (n = 8). Clindamycin was administered, alone and at an unknown dose, in 3% of cases (n = 4) and at a dose of 300 mg, alone, in 2% of cases (n = 3) (Figure 10).

Figure 10. Distribution of the population by antibiotic administered—relative frequency and absolute frequency (unk. dose = unknown dose).

3.10. Targeted Treatment for Osteonecrosis: Mouthwash

Information about the mouthwash administered for the treatment of MRONJ was available in 140 cases. The most commonly used mouthwash was chlorhexidine, although its concentration was not specified, in 38% of cases (n = 35). When the concentration was specified, it was observed that 0.12% chlorhexidine was used more than 0.2% chlorhexidine, representing 26% of cases (n = 24) and 14% of cases (n = 13), respectively. A concentration of 0.12% chlorhexidine was also administered together with hydrogen peroxide, at 3% or at an unknown concentration, in 1% of cases (n = 1) in both situations. In turn, 0.2% chlorhexidine was administered together with a nystatin solution, of unknown concentration, in 1% of cases (n = 1) (Figure 11).

Figure 11. Distribution of the population by mouthwash administered—relative frequency and absolute frequency (unk. % = unknown concentration).

3.11. Targeted Treatment for Osteonecrosis: Surgical Procedure

The type of surgical procedure performed in the treatment of MRONJ was specified in 146 cases. Surgical debridement alone was the most common surgical procedure, accounting for 32% of cases (n = 37). Sequestrectomy follows, performed alone, in 22% of cases (n = 26). These two surgical procedures were also performed together in 13% of cases (n = 15). Mandibular resective surgery was the only surgical option chosen in 6% of cases (n = 7). This type of surgery was performed on the lower jaw together with sequestrectomy in 3% of cases (n = 4) and on the upper jaw, also together with sequestrectomy, in 2% of cases (n = 2) (Figure 12).

Figure 12. Distribution of the population by surgical procedure (relative frequency and absolute frequency).

3.12. Other Treatments Aimed at Osteonecrosis

Among the different types of non-conventional treatments administered, the L-PRF membrane stands out, applied in 21% of cases (n = 15). Furthermore, teriparatide injections, at an unspecified dose, were administered in 7% of cases (n = 5). In the same percentage of the population, a combination of pentoxifylline + tocopherol was administered, in unknown doses, and amniotic membrane was applied independently. Of the other types of non-conventional treatment, ozone therapy and negative pressure wound therapy stand out, each applied separately in two cases (Figure 13).

Figure 13. Distribution of the population by other treatments implemented (relative frequency and absolute frequency).

3.13. Follow-Up Time from the Start of MRONJ Treatment until Its Cure

Information regarding the follow-up time from the beginning of implementation of MRONJ treatment until its cure was available in 131 cases. Of these, no cure for the disease was observed in 40% of the studied population (n = 61). The minimum follow-up time from the start of treatment to cure was 1 month and the maximum time was 120 months. The average follow-up time was 18.56 months, with a standard deviation of ±18.13 (Figure 14).

Figure 14. Distribution of the population by follow-up time (relative frequency and absolute frequency).

3.14. Relationship between Established Therapy and Cure of MRONJ

It was possible to relate the type of treatment instituted (antibiotic, mouthwash, surgical procedure, and others) and the existence or non-existence of a cure for MRONJ, through the application of Fisher’s exact test, analyzing the exact two-sided significance value obtained. Therefore, it is concluded that, in cases where p ≤ 0.05, the relationship between the two variables is statistically significant.

Regarding the relationship between the administration of antibiotics and obtaining a cure, it is observed that, of the 128 cases in which antibiotics were administered, there was a cure in 74 cases (57.8%) (Table 1). The relationship between these two variables resulted in a value of p = 0.359 (>0.05), concluding that the relationship between the administration of antibiotics and the achievement of a cure is not statistically significant (Table 2).

Table 1. Count in relation to the relationship between two variables (antibiotic administration and obtaining a cure).

			Cure		Total
			No	Yes
Antibiotics	No	Count	7	16	23
		% within Antibiotics	30.4%	69.6%	100.0%
		% within Cure	11.5%	17.8%	15.2%
		% of Total	4.6%	10.6%	15.2%
	Yes	Count	54	74	128
		% within Antibiotics	42.2%	57.8%	100.0%
		% within Cure	88.5%	82.2%	84.8%
		% of Total	35.8%	49.0%	84.8%
Total		Count	61	90	151
		% within Antibiotics	40.4%	59.6%	100.0%
		% within Cure	100.0%	100.0%	100.0%
		% of Total	40.4%	59.6%	100.0%

Table 2. Application of Fisher’s test in the relationship between two variables (administration of antibiotics and achievement of cure).

	Value	df	Asymptotic Significance (2-Sided)	Exact Sig. (2-Sided)	Exact Sig. (1-Sided)
Pearson Chi-Square	1.118 a	1	0.290
Continuity Correction b	0.684	1	0.408
Likelihood Ratio	1.151	1	0.283
Fisher’s Exact Test				0.359	0.205
N of Valid Cases	151

^a 0 cells (0.0%) have expected count less than 5. The minimum expected count is 9.29. ^b Computed only for a 2 × 2 table.

From the relationship between the administration of mouthwash and the achievement of a cure, it can be seen that, of the 92 cases in which mouthwash was administered, a cure was achieved in 54 cases (58.7%) (Table 3). The application of Fisher’s test between the two variables mentioned resulted in a value of p = 0.865, concluding that the relationship between the administration of mouthwash and the achievement of a cure is not statistically significant (Table 4).

Table 3. Count in relation to the relationship between two variables (mouthwash administration and obtaining a cure).

			Cure		Total
			No	Yes
Mouthwash	No	Count	23	36	59
		% within Mouthwash	39.0%	61.0%	100.0%
		% within Cure	37.7%	40.0%	39.1%
		% of Total	15.2%	23.8%	39.1%
	Yes	Count	38	54	92
		% within Mouthwash	41.3%	58.7%	100.0%
		% within Cure	62.3%	60.0%	60.9%
		% of Total	25.2%	35.8%	60.9%
Total		Count	61	90	151
		% within Mouthwash	40.4%	59.6%	100.0%
		% within Cure	100.0%	100.0%	100.0%
		% of Total	40.4%	59.6%	100.0%

Table 4. Application of Fisher’s test in the relationship between two variables (administration of mouthwash and achievement of a cure).

	Value	df	Asymptotic Significance (2-Sided)	Exact Sig. (2-Sided)	Exact Sig. (1-Sided)
Pearson Chi-Square	0.080 a	1	0.777
Continuity Correction b	0.013	1	0.909
Likelihood Ratio	0.081	1	0.777
Fisher’s Exact Test				0.865	0.456
N of Valid Cases	151

^a 0 cells (0.0%) have expected count less than 5. The minimum expected count is 23.83. ^b Computed only for a 2 × 2 table.

From the relationship between the application of surgical procedures and obtaining a cure, it is observed that, of the 117 cases in which surgery was performed, 72 were cured (61.5%) (Table 5). The application of Fisher’s test revealed a value of p = 0.429 (p > 0.05); therefore, the relationship between surgery and a cure is not statistically significant (Table 6).

Table 5. Count in relation to the relationship between two variables (surgical procedures and obtaining a cure).

			Cure		Total
			No	Yes
Surgery	No	Count	16	18	34
		% within Surgery	47.1%	52.9%	100.0%
		% within Cure	26.2%	20.0%	22.5%
		% of Total	10.6%	11.9%	22.5%
	Yes	Count	45	72	117
		% within Surgery	38.5%	61.5%	100.0%
		% within Cure	73.8%	80.0%	77.5%
		% of Total	29.8%	47.7%	77.5%
Total		Count	61	90	151
		% within Surgery	40.4%	59.6%	100.0%
		% within Cure	100.0%	100.0%	100.0%
		% of Total	40.4%	59.6%	100.0%

Table 6. Application of Fisher’s test in the relationship between two variables (surgical procedures and achievement of a cure).

	Value	df	Asymptotic Significance (2-Sided)	Exact Sig. (2-Sided)	Exact Sig. (1-Sided)
Pearson Chi-Square	0.809 a	1	0.369
Continuity Correction b	0.491	1	0.483
Likelihood Ratio	0.801	1	0.371
Fisher’s Exact Test				0.429	0.241
N of Valid Cases	151

^a 0 cells (0.0%) have expected count less than 5. The minimum expected count is 13.74. ^b Computed only for a 2 × 2 table.

Relating the performance of other treatments with obtaining a cure, it is observed that, of the 70 cases in which non-conventional treatments were applied, there was a cure in 52 cases (74.3%) (Table 7). The application of Fisher’s test revealed a value of p < 0.001; therefore, since p < 0.05, it is concluded that there is a statistically significant relationship between the application of unconventional treatments and the achievement of a cure (Table 8).

Table 7. Count in relation to the relationship between two variables (other treatments and obtaining a cure).

			Cure		Total
			No	Yes
Others	No	Count	43	38	81
		% within Others	53.1%	46.9%	100.0%
		% within Cure	70.5%	42.2%	53.6%
		% of Total	28.5%	25.2%	53.6%
	Yes	Count	18	52	70
		% within Others	25.7%	74.3%	100.0%
		% within Cure	29.5%	57.8%	46.4%
		% of Total	11.9%	34.4%	46.4%
Total		Count	61	90	151
		% within Others	40.4%	59.6%	100.0%
		% within Cure	100.0%	100.0%	100.0%
		% of Total	40.4%	59.6%	100.0%

Table 8. Application of Fisher’s test in the relationship between two variables (other treatments and achievement of a cure).

	Value	df	Asymptotic Significance (2-Sided)	Exact Sig. (2-Sided)	Exact Sig. (1-Sided)
Pearson Chi-Square	11.684 a	1	<0.001
Continuity Correction b	10.575	1	0.001
Likelihood Ratio	11.939	1	<0.001
Fisher’s Exact Test				<0.001	<0.001
N of Valid Cases	151

^a 0 cells (0.0%) have expected count less than 5. The minimum expected count is 28.28. ^b Computed only for a 2 × 2 table.

It is possible to observe, in Figure 15, which non-conventional treatments applied were associated with obtaining a cure. Of these, the L-PRF membrane, the amniotic membrane, and the pentoxifylline + tocopherol combination stand out, although there have also been fewer cases in which these treatments were not associated with curing the disease. Of the non-conventional treatments applied that only showed a cure, the association of epimucosal fixation and L-PRF membrane, the association of minocycline injections and a nasolabial flap, negative pressure wound therapy, ozone therapy, and sodium hyaluronate and amino acid gels stand out.

Figure 15. Distribution of non-conventional treatments according to whether or not a cure is obtained.

In Appendix B, it is noted that the systematization of the different therapies instituted and the respective result regarding the obtainment or not of a cure, in all cases, is included.

4. Discussion

According to the results obtained, it was observed that females were more affected than males in a ratio of 2.49:1. This result is in line with the literature, which reports a higher prevalence of MRONJ in females, probably reflecting the underlying disease for which the drugs are administered, such as osteoporosis and breast cancer, which are more common in women [6,11,18]. Preventing MRONJ in these high-risk groups composed of women requires a comprehensive approach. It involves recognizing the impact of antiresorptive therapies on bone healing, emphasizing the need for coordinated dental care, pretreatment management, and ongoing education for patients and healthcare professionals about MRONJ risks and prevention strategies [11].

The average age at diagnosis was 66.27 years, with the 60 to 70 years age group being the most affected. This is in line with the literature, which indicates that ages over 65 years represent a risk factor for MRONJ [7]. In a retrospective statistical study that analyzed 70 articles, the average age of patients was 62 years [19].

According to Ojha et al. [20], stages 1 and 2 of MRONJ are the most commonly observed. However, the results of two retrospective studies demonstrate that stage 2 was the most frequently diagnosed, one with 129 patients [21] and the other with 71 patients [22].

A retrospective study by Ahdi et al. [1], which analyzed cases of MRONJ reported in the FAERS database from 2010 to 2021, identified the medications most associated with the development of MRONJ. The 10 most frequently associated drugs were, in descending order, zoledronic acid, alendronic acid, denosumab, pamidronic acid, ibandronic acid, lenalidomide, risedronic acid, sunitinib, bevacizumab, and prednisolone. These findings corroborate the results of this review, where zoledronic acid was the most used drug, followed by alendronic acid and denosumab.

A clinical case describes the development of MRONJ in a patient with no history of administration of antiresorptive or antiangiogenic drugs, corticosteroids, or head and neck radiotherapy, medicated only with simvastatin for hypercholesterolemia [16]. This case highlights the need to develop more studies on the influence of different drugs on the oral cavity.

The literature suggests a higher risk of MRONJ in patients with malignant tumors treated with intravenous bisphosphonates compared to patients treated with oral bisphosphonates for osteoporosis [2,5,8,23]. However, the results of this review show that the most common route of administration of medications was oral (26% of cases), followed by intravenous (24%). This allows us to question the premise that there is a greater risk of developing MRONJ with intravenous medications, highlighting the need to pay equal attention to oral administration of this type of medication.

The duration of treatment with antiresorptive or antiangiogenic agents until the diagnosis of MRONJ varied significantly, with the largest number of cases falling within the 34- to 67-month administration interval. A literature review that included 50 patients demonstrated an average treatment duration of 48.68 months [24]. In the present review, the average duration of treatment was 60.88 months, with a minimum time of 1 month and a maximum time of 252 months. These data, together with the fact that, in the aforementioned literature review, the majority of patients were in an advanced stage of MRONJ at the time of diagnosis, emphasize the importance of rigorous monitoring of patients medicated with antiresorptive and antiangiogenic agents in order to detect early signs of the disease.

According to Ruggiero et al. [11], MRONJ lesions affect the mandible (75%) more than the maxilla (25%) and may involve both jaws (4.5%). Thus, the literature corroborates the results obtained, in which the mandible was the most affected location (60%), followed by the maxilla (32%) and both jaws (5%).

For the conservative treatment of MRONJ, 0.12% or 0.2% chlorhexidine gluconate represents an efficient topical bacteriostatic–bactericidal agent that works to reduce the oral bacterial population, including biofilms that promote infection [19]. In fact, according to the results obtained, the most used mouthwash was chlorhexidine, and when the information was specified, the most used concentration was 0.12%. However, oral antibiotics are the most important agents to treat the infection in osteonecrosis, and, since the infections associated with this disease are polymicrobial, broad-spectrum antibiotics are recommended, such as the combination of amoxicillin + clavulanic acid, ampicillin, metronidazole, or clindamycin [19]. This is in line with the results obtained, in which amoxicillin + clavulanic acid (875 mg + 125 mg) was the most administered antibiotic.

A systematic review by Seluki et al. [25] showed that the non-surgical approach may be useful in preventing disease progression in patients ineligible for surgery, but effective resolution of osteonecrosis should not be expected. The results of Fisher’s exact test revealed a value of p = 0.359 for the relationship between antibiotic administration and MRONJ cure and p = 0.865 for the relationship between mouthwash administration and cure, indicating that there is no relationship with a statistically significant difference between the treatments mentioned and the achievement of a cure for MRONJ.

The most common surgical procedures were surgical debridement (32%) and sequestrectomy (22%), performed alone. Although Nicolatou-Galitis et al. [3] suggest that surgical procedures are more effective in advanced cases of MRONJ, Fisher’s test showed a value of p = 0.429, indicating that the relationship between performing surgery and obtaining a cure for MRONJ is not statistically significant.

Carrying out unconventional treatments showed a statistically significant relationship with obtaining a cure for MRONJ (p < 0.001), despite the literature stating that the effectiveness of these treatments has not been proven [3]. These results suggest that non-conventional treatments may be an effective alternative for resolving MRONJ.

The relationship between carrying out conventional treatments alone and obtaining a cure was not statistically significant, particularly with regard to the administration of antibiotics, the use of mouthwash, and surgery. However, there is the possibility that combinations of conventional treatments may be effective in achieving a cure. To this end, it may be interesting to expand the study in the future in order to evaluate the statistical significance of the combinations of different treatments in obtaining a cure.

The literature regarding the time elapsed from the start of treatment for MRONJ until a cure is achieved (follow-up) is scarce and significantly heterogeneous. Mourão et al. [26] reported a mean follow-up time of 23.5 months, while a meta-analysis demonstrated a mean follow-up time of 8.7 months [27]. In the present review, the average follow-up was 18.56 months, with a minimum of 1 month and a maximum of 120 months.

Dentists should ensure thorough pretreatment dental assessments and complete any necessary invasive procedures before initiating antiresorptive or antiangiogenic therapy to reduce MRONJ risk. During therapy, they should prioritize non-invasive dental treatments, carefully assess the risks and benefits of any invasive procedures, and, in consultation with the prescribing physician, consider a temporary discontinuation of the medication. Additionally, regular dental check-ups and patient education on maintaining excellent oral hygiene and recognizing MRONJ symptoms are essential for effective prevention and early detection [11].

This systematic review highlights important findings but also underscores the limitations inherent in the current body of evidence, primarily consisting of case reports and clinical case series. These sources, while valuable, are limited by factors such as lack of randomization, variability in reports, and absence of control groups. Additionally, publication bias may skew the results toward more positive outcomes. To enhance the robustness and generalizability of the findings, there is a clear need for more well-designed, randomized clinical trials. Such studies are crucial to provide more definitive insights into the efficacy and success of treatments, ultimately guiding more effective clinical practices.

5. Conclusions

MRONJ is a complex condition, the pathophysiology of which is not yet fully understood. It is mainly associated with the use of antiresorptive and antiangiogenic drugs, but cases of MRONJ associated with other classes of drugs have been described, highlighting the need to develop more studies to clarify the pathophysiology of the disease and, thus, improve patients’ quality of life.

Based on the relationship between the type of treatment instituted and the achievement of a cure, the only statistical significance found was related to non-conventional treatments, which allows us to formulate the hypothesis that these treatments, carried out alone or associated with conservative or surgical therapies, can be effective in resolving MRONJ.

The treatment of MRONJ remains challenging, due to the lack of consensus among different authors regarding the most appropriate approach protocol and the variation in patients’ response to established therapeutic interventions, meaning that treatment must be individualized and adapted to clinical needs.