Potential role of comprehensive dental care in preventing medication related osteonecrosis of the jaw (MRONJ): a single centre study

Abstract

Objective

Medication-related osteonecrosis of the jaw bones (MRONJ) is a well-known complication of antiresorptive and antiangiogenic drugs. Since the first report, more occurrences of MRONJ have been described worldwide. Dental extraction has been described by many studies as one of the risk factors for MRONJ. Comprehensive dental care (CDC) is a preventive dental method provided to patients prior to drug commencement. This study aims to determine the association between CDC and MRONJ. Patients and methods. A retrospective analysis was performed on 75 medical records of patients on antiresorptive and/or antiangiogenic drugs between February 2018 and May 2021. Demographics and clinical and radiographic data were collected. Univariate and multivariate analyses were performed to determine the factors associated with MRONJ.

Results

Of the 75 patients who met the inclusion criteria, 11 (14.7%) developed MRONJ. Three out of 11 patients (27.2%) developed MRONJ spontaneously, while eight patients (72.8%) developed it after trauma from dentures or dental extractions. Following a binary logistic regression analysis, the lack of CDC was identified as a significant predictor of MRONJ. Patients who did not receive CDC had an odds ratio of 8.64 (95% confidence interval (CI): 1.27–58.62, p = 0.03). However, dental extraction did not show a statistically significant association with MRONJ in both the univariate and multivariate analyses.

Conclusion

CDC before treatment with antiresorptive and antiangiogenic drugs is a potentially effective preventive method for reducing MRONJ. Dental extraction was not a significant factor in relation to MRONJ.

Introduction

Medication-related osteonecrosis of the jaw (MRONJ) is a serious adverse reaction to antiresorptive and antiangiogenic drugs. These drugs (e.g., zoledronate, ibandronate, alendronate, and risedronate from the bisphosphonate group, denosumab from the receptor activator of the nuclear factor kappa-B ligand (RANKL) inhibitor, sunitinib from the tyrosine kinase inhibitor group, and bevacizumab from the monoclonal antibodies) [1, 2] cause progressive bone destruction in the jaw region and affect the mastication, speech, and general well-being of a patient. Clinically, MRONJ appears as an area of exposed bone or bone that can be probed through an intraoral or extraoral fistula(e) in the maxillofacial region that lasts for more than 8 weeks in patients who are on the drugs, but do not have a history of radiation therapy to the head and neck or a metastasis in the jaw bones [1, 2].

Symptoms, such as swelling, pain, fistula formation, and pus discharge in the oral cavity or on the skin with an unpleasant smell, are often observed. As they progress, more serious complications, such as pathological fractures or sepsis, could occur, which could be life-threatening [3, 4].

The MRONJ prevalence is 0.043% among patients treated with oral bisphosphonates, 1.03% among those on intravenous bisphosphonates, and 3.64% in those on high-dose denosumab [5]. A recent population-based study in Japan showed that the MRONJ prevalence is 0.06% among the osteoporosis group and 1.47% among the cancer group, while the incidence was 22.9 per 100,000 person-years and 1,231.7 per 100,000 person-years for the osteoporosis and cancer groups, respectively [6].

Dental extraction is one of the most important risk factors for the development of MRONJ. The incidence of MRONJ following a dental extraction in the osteoporosis cohort is 2.7% per person-year (95% CI 1.6–4.6), while for the bone metastases cohort, it is 26.4% per person-year (95% CI 20.4–34.2) [7].

Contrary to the majority of findings, some authors argued that it was the concurrent infection, and not the tooth extraction itself that predisposed patients to MRONJ [4, 8–10]. This contention is bolstered by animal studies demonstrating the potential role of inflammatory or infectious dental pathology in the mechanism of the MRONJ development [11, 12].

Soutome et al. (2021) advocated dental extractions in patients with teeth that can be a source of infection, even after commencing antiresorptive and antiangiogenic drugs to lower the risk of MRONJ development [4].

To reduce the occurrence of infection that could eventually progress to MRONJ, preventive dental procedures were advocated by many studies [13–16]. The preventive dental procedure, also known as the primary prevention procedure by Mauceri et al. (2022), is a series of dental procedures performed before and during the treatment with antiresorptive and antiangiogenic agents [13].

The implementation of primary prevention procedures reduces MRONJ and ensures that patients are informed of the potential drug risks and benefits [14]. The intervention consists of regular scaling, use of antiseptic mouthwashes, adjustment of dentures, education on the importance of good oral hygiene, and correction of systemic risk factors (smoking, alcohol use, and uncontrolled diabetes) [14–16].

In our hospital, the primary prevention procedure is known as comprehensive dental care (CDC). We constructed the following research objectives to assess the association between the explanatory factors and MRONJ. The primary objective is to determine the association between CDC and MRONJ. The secondary objective is to explore other factors associated with MRONJ, particularly dental extraction.

The hypotheses for this study are as follows:

• i)Performing CDC prior to and before the commencement of antiresorptive and antiangiogenic treatments can reduce the MRONJ occurrence.
• ii)Dental extraction may or may not have a positive association with MRONJ.

Patients and methods

The study protocol adhered to the ethical guidelines of the Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects and received ethical approval from the Research Ethics Committee (REC), Universiti Kebangsaan Malaysia (UKM) (reference: UKM PPI/111/8/JEP2020-791) dated 25 January 2021. Informed consent was waived by the REC UKM in view of the retrospective nature of the study. All procedures being performed were part of the routine care.

Characteristics of the patients’ records

This study was a retrospective record review conducted at the Oral and Maxillofacial Clinic, Hospital Canselor Tuanku Muhriz (HCTM). Records of adult patients dated from 1 January 2018 to 31 December 2021 were reviewed. The patients were selected from the clinic’s database. The selection was based on the inclusion and exclusion criteria presented here. The inclusion criteria were as follows: i) presence of complete records from patients who were on treatment with the antiresorptive and/or antiangiogenic drugs; and ii) at least a radiograph must be available when the MRONJ diagnosis was made. For our long-term patients, radiographs prior to the diagnosis were available as the baseline. However, for the newly referred patients, only diagnostic radiographs were accessible. Therefore, in this study, we ensured that at least one radiograph was available for assessing the radiographic features in relation to MRONJ.

On the contrary, the records of patients with a history of i) radiation therapy to the head and neck area or/and ii) recurrence or metastatic tumor to the jaw bones were excluded from the study.

The sample size was calculated using the epidemiological statistics application, Open Epi, which was available online at https://www.openepi.com/ [17]. The reference from Soutome et al. (2018) [8] was used, and the calculation yielded 68 samples at a 95% CI.

Data collection

The data were collected using a research tool comprising the clinical and radiological information. The tool included all the following independent and dependent variables:

Independent variables

• i)Demographic data included age, gender, and ethnic groups.
• ii) Habits, including smoking and alcohol consumption.
• iii)iii) Diagnosis of the primary disease that necessitates the prescription of the antiresorptive and/or antiangiogenesis. The primary diseases were categorized as the osteometabolic (e.g., osteoporosis) or oncology (e.g., multiple myeloma or metastatic cancer).
• iv)Drug-related factors, including a) name and type, b) single versus multiple drug intake, c) route of administration, and d) duration of the drug intake.
• v)Dental trauma factors, including all forms of dental treatment prior to MRONJ onset.
• vi)Comprehensive dental care, including both invasive and non-invasive dental treatment prior to starting the medication and only non-invasive dental care during treatment with antiresorptive and antiangiogenic agents. The CDC is normally started within 2 weeks following the patients’ referral to our clinic and continued regularly throughout the treatment with the drugs of concern.

Dependent variable

MRONJ

The clinical description of MRONJ was according to the definition of the American Association of Oral and Maxillofacial Surgeons (AAOMS) Position Paper 2014 and 2022 [1, 2]. The dental panoramic tomogram (DPT) with or without other imaging was used to aid the diagnosis and clinical staging of MRONJ. The bony changes observed in the MRONJ cases were classified according to Berg et al. (2016) [18] and Cardoso et al. (2017) [19].

Statistical analysis

IBM SPSS Statistics version 23.0 was used for data entry and analysis (Armonk, NY: IBM Corp.). Descriptive statistics on the characteristics of the study factors, as well as the outcome (MRONJ) were reported in the form of frequencies, means, standard deviations, medians, and interquartile ranges (IQRs). The association between the independent factors and MRONJ was assessed using Pearson’s Chi-squared test or Fisher’s exact test. The significance level was set at p < 0.05.

Binary logistic regression analysis was used in the model with a dichotomous dependent variable. This analysis was used to examine the association between the independent and dependent (outcome) variables. The selection of variables to be included in the logistic regression model included those with a significance level of p < 0.25 [20], as well as those with clinical significance. Potential confounding factors were added to the baseline model to assess their effect on the overall model performance. The quality of the model was examined for goodness of fit using the Hosmer–Lemeshow test and the classification table. Obtaining a nonsignificant result in the Hosmer–Lemeshow test indicates a fit model. In the classification table, models with a percentage of the overall correctly classified closer to 100% were considered more accurate.

Results

Between 1 January 2018 and 31 December 2021, 200 patients attended the Oral and Maxillofacial Surgery Clinic of UKMMC. These patients were mostly referred from the hematology, oncology, orthopedic, obstetrics, and gynecology clinics of HCTM, as well as from other clinics and medical centers in Klang Valley, Malaysia. Three types of referrals were observed. The first type was an ideal referral, where the patients were referred to our clinic prior to the commencement of antiresorptive and antiangiogenesis treatment. The second type was caused by toothache or dental disease-related issues. For this referral type, the patients were already on antiresorptive and antiangiogenesis drugs. The patients might or might not have had CDC prior to drug commencement. Finally, the third referral type was for the management of an established case of MRONJ. Similar to the second type, the patients might or might not have had CDC.

Of the 200 patients’ records, only 75 fulfilled the inclusion criteria and had complete and clear data related to the independent and dependent variables. Among the 75 records, 11 were confirmed with MRONJ.

Establishment of the MRONJ diagnosis

All 11 cases were diagnosed through clinical and radiographic examinations.

The causes of MRONJ are categorized into two main groups as follows: i) spontaneous; and ii) dental or trauma-related. Of the 11 patients, eight patients (72.7%) had a history of dental extraction or trauma from the prosthesis, and three patients (27.3%) denied any history of dental trauma. Six patients (54.5%) had MRONJ in the mandible; three (27.3%) had MRONJ in the maxilla; and the remaining two cases (18.2%) had bone necrosis in both jaws. Based on the staging described in the AAOMS’ Position Paper 2014, 10 patients were classified as Stage 2, while one patient was classified as Stage 3. Ten patients had single drug, that is, four on intravenous zoledronate, one patient on subcutaneous denosumab, four on oral alendronate, one on oral ibandronate, and one patient on intravenous zoledronate and subcutaneous denosumab. The treatment duration varied between 23 and 168 months (median: 36 months, IQR: 27) before they were diagnosed with MRONJ. The reason of referrals for the majority of patients was MRONJ management (n = 9, 81.8%). Two patients (18.2%) had dental extractions in our center and developed MRONJ. In relation to imaging, a DPT and an intraoral periapical radiograph were the most commonly used imaging modalities to diagnose MRONJ. In this study, the DPTs were examined by a postgraduate student and the oral and maxillofacial surgery consultants through digital imaging, Medweb on an HP computer (model: Compaq dc7800 Convertible Minitower, serial number: SGH7450LWG).

The bone change findings are as follows:

• i)osteolytic changes (n = 10; 27.0%)
• ii)bone sequestration (n = 7; 18.9%)
• iii)focal bone sclerosis (n = 9; 24.3%)
• iv)erosion of the cortical bone (n = 5; 13.5%)
• v)prominent IAN (n = 1; 2.7%)
• vi)PDL widening (n = 3; 8.1%)
• vii)thickening in the lamina dura (n = 1; 2.7%)

When the image was doubtful or unreliable, computed tomography (CT) was arranged. The CT scan report was prepared by the radiologists.

Table 1 below shows the characteristics of the 11 cases of MRONJ presented in this study.

Table 1.

Etiology and clinical and radiological features of MRONJ patients

Case number	Trauma history	Location in the jaw bone	Staging based on AAOMS (2022)	Single vs multiple & Type of drugs	Route of administration	Duration of treatment	Reason of referral	Radiographic features [Berg et al. (2016) [18] and Cardoso et al. (2017) [19]
1	Trauma from prosthesis	Maxilla	Stage 2	Single Bisphosphonate (Zolendronate)	IV	72 months	Management of MRONJ	i) osteolytic change ii) focal bony sclerosis iii) persistent alveolar socket
2	Trauma from prosthesis	Maxilla & mandible	Stage 2	Single Bisphosphonate (Zolendronate)	IV	31 months	Management of MRONJ	i) osteolytic change ii) focal bony sclerosis
3	Spontaneous	Maxilla	Stage 2	Single Bisphosphonate (Zolendronate)	IV	49 months	Management of MRONJ	i) focal bony sclerosis ii) pdl space widening
4	Spontaneous	Mandible	Stage 2	Single Bisphosphonate (Alandronate))	Oral	168 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum
5	Dental extraction	Maxilla	Stage 2	Single Bisphosphonate (Zolendronate)	IV	36 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum
6	Dental extraction	Mandible	Stage 3	Multiple Bisphosphonate (Zolendronate) & RANKL inhibitor (Denosumab)	IV & SC	25 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum iii) focal bony sclerosis
7	Dental extraction	Maxilla & mandible	Stage 2	Single RANKL inhibitor (Denosumab)	SC	23 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum iii) focal bony sclerosis iv) erosion cortical bone v) IAN prominence vi) pdl space widening
8	Dental extraction	Mandible	Stage 2	Single Bisphosphonate (Alandronate))	Oral	36 months	Dental extraction due to mobile teeth	i) osteolytic change ii) focal bony sclerosis iii) erosion cortical bone iv) pdl widening
9	Dental extraction	Mandible	Stage 2	Single Bisphosphonate (Alandronate))	Oral	36 months	Dental extraction	i) osteolytic change ii) bone sequestrum iii) focal bony sclerosis iv) erosion cortical bone
10	Spontaneous	Mandible	Stage 2	Single Bisphosphonate (Ibandronate)	Oral	58 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum iii) focal bony sclerosis iv) erosion cortical bone v) lamina dura thickening
11	Dental extraction	Mandible	Stage 2	Single Bisphosphonate (Alandronate))	Oral	36 months	Management of MRONJ	i) osteolytic change ii) bone sequestrum iii) focal bony sclerosis iv) erosion cortical bone

Abbreviations: pdl Periodontal ligament, IAN Inferior alveolar nerve, IV Intravenous, SC Subcutaneous, MRONJ medication-related osteonecrosis of the jaw

Clinical characteristics of the patients

The clinical characteristics of 11 MRONJ and 64 non-MRONJ patients are described as follows, and the association between the clinical characteristics and MRONJ was analyzed using Pearson’s Chi-squared test or Fisher’s exact test (Table 2).

Table 2.

Demographic, social and clinical characteristics of the sample

Characteristics	MRONJ, n(%)	Without MRONJ, n(%)	Total
Demographic
Age (years)
Median (IQR)	69.0 (14.0)	68.5 (12.0)	69.0 (12.0)
Mean ± SD	68.6 ± 7.9	67.7 ± 11.6	67.9 ± 11.1
Min	57.0	21.0	21.0
Max	79.0	89.0	89.0
			p value (Pearson chi-square test or Fisher’s exact test)
Age category (years)
≤ 60	3(27.3)	10(15.6)	0.39
> 60	8(72.7)	54(84.4)
Gender
Male	4(36.4)	24(37.5)	1.00
Female	7(63.6)	40(62.5)
Ethnic group
Malay	6(54.5)	44(68.8)	0.46
Chinese	4(36.4)	16(25.0)
Indian	1(9.1)	4(6.3)
High-risk habits
Smoking
Yes	2(18.2)	10(15.6)	1.00
No	9(81.8)	54(84.4)
Alcohol consumption
Yes	0 (0.0)	3 (4.7)	1.00
No	11 (100.0)	61(95.3)
Primary disease
Oncology	6(54.5)	37(57.8)	1.00
Osteometabolic	5(45.5)	27(42.2)
Drug factors
Drug type & single vs combination
Anti-resorptive	10(90.9)	45(70.3)	0.75
Anti-angiogenesis	0 (0.0)	3(4.7)
Anti-resorptive & Anti-angiogenesis	0(0.0)	2(3.1)
Anti-resorptive & Anti-resorptive	1(9.1)	14(21.9)
Route of administration
IV or SC	5(45.5)	22(34.4)	0.64
Oral	5(45.5)	27(42.2)
IV or SC & oral	1 (9.1)	15(23.4)
Duration of drug intake (months)
≤ 24	1 (9.1)	28(43.8)	0.04*
> 24	10 (90.9)	36(56.2)
Dental factors
Dental extraction
Yes	5(45.5)	26 (40.6)	1.00
No	6(54.5)	38 (59.4)
Comprehensive Dental Care (CDC)
Before commencement of medication
Yes	2(18.2)	42(65.4)	0.01*
No	9(81.8)	22(34.4)

Abbreviations: IQR Interquartile range, SD Standard deviation, IV Intravenous, SC Subcutaneous

1. Demographic factors.

The study involved 28 (37.3%) men and 47 (62.6%) women with a median age (IQR) of 69.0 (12.0) years (range 21–89 years). Most patients were Malays (n = 50, 66.7%), followed by Chinese (n = 20, 26.7%) and Indians (n = 5, 6.7%).

Age, gender, and ethnic groups did not show a statistically significant association with MRONJ (p > 0.05).

• 2)Demographic factors

Only two patients in the smoking group developed MRONJ. All patients who consumed alcohol did not develop MRONJ.

Both smoking and alcohol did not have a statistically significant association with MRONJ (p > 0.05).

• 3) Primary diseases indicated for treatment with antiresorptive and antiangiogenesis.

Of the 11 patients with MRONJ, six (54.5%) were diagnosed with oncology diseases, while five had osteometabolic diseases (45.5%). These diseases did not show a statistically significant association with MRONJ (p > 0.05).

• 4) Drug characteristics.

The MRONJ-related drugs included antiresorptive comprising bisphosphonates and the receptor activator of nuclear factor-κB ligand (RANKL) inhibitors and antiangiogenic drugs. The drug characteristics are as follows: i) name/type of drug and ii) single versus multiple drugs, iii) route of administration, and iv) treatment duration.

The most commonly reported drug used are bisphosphonates (i.e., zoledronate, alendronate, pamidronate, and ibandronate, and denosumab as the RANKL inhibitor). The patients who developed MRONJ were on either single or double antiresorptive drugs. None of the patients who were on antiangiogenic therapy, either as a single drug or in combination with antiresorptive therapy, developed MRONJ.

The drugs were administered intravenously (IV), subcutaneously (SC), or orally.

Five patients in the oral group (45.5%), five in the IV/SC group (45.5%), and one under both IV/SC and oral route groups (9.1%) developed MRONJ.

In relation to the drug duration, the minimum duration of drug usage was 2 months, and the maximum was 196 months, with a median of 36 months (IQR, 36).

Among the factors investigated in this category, only the drug duration showed statistically significant results. The univariate association showed that the duration of drug intake of less and more than 24 months had a significant association with MRONJ (p = 0.04).

• 5) Dental factors.

Dental extraction was performed before and during treatment with MRONJ-related medication. As shown in Table 2, the proportions of patients who underwent dental extraction and those who did not were nearly equivalent within the MRONJ group, and the association was not statistically significant (p = 1.00).

Meanwhile, CDC showed statistically significant results with MRONJ (p = 0.01).

Further analysis using binary logistic regression was performed. Factors, such as i) duration of medication taken, ii) single vs multiple drugs, iii) dental extraction (yes or no), and iv) received CDC or not, were entered into the model. The potential confounding factor was the age of the patients.

The final model revealed the following results (Table 3):

• i)CDC had a statistically significant (p = 0.03) association with MRONJ. Those patients who did not undergo CDC showed an odds ratio (OR) of 8.64 (95% CI: 1.27–58.62) of having MRONJ.
• ii)Patients who took the medication more than 24 months had the OR of 10.91 (95% CI: 0.83–144.02) of developing MRONJ, and the association was close to being significant (p = 0.07).
• iii)Patients who were on multiple combined drugs had 0.17 times the likelihood of developing MRONJ [OR 0.17(95% CI: 0.01–2.36), p = 0.19].
• iv)The odds ratio was similar between having a dental extraction and no dental extraction in association with MRONJ [OR 1.15, 95% CI: 0.21–6.43), p = 0.88].

Table 3.

Association between CDC and duration of medication taken with MRONJ (controlled for age)

Factors	Odds ratio (95% CI)	p value
CDC before medication
Yes	1.00	0.03
No	8.64(1.27–58.62)
Duration of medication taken
≤ 24 months	1.00	0.07
> 24 months	10.91(0.83–144.02)
Single vs multiple drugs
Single drug	1.00	0.19
Multiple drugs	0.17(0.01–2.36)
Dental extraction
No	1.00	0.88
Yes	1.15(0.21–6.43)

*Controlled for age

i) Hosmer Lemeshow test (²(6)=1.36, p=0.97)

ii) Percentage of accuracy: 84.0%

The model below was considered to be a good fit with i) the Hosmer–Lemeshow test (²(6) = 1.36, p = 0.97) and ii) 84.0% percentage of accuracy.

Discussion

This study reveals two key findings: patients who did not undergo CDC were at higher odds of developing MRONJ, while those who had dental extractions did not exhibit an increased likelihood of the condition.

Our results showed that patients who did not undergo CDC before antiresorptive and antiangiogenic treatment had eightfold higher odds of developing MRONJ compared to those who had CDC. This significant finding was in line with the results from previous studies [15, 21–27]. Providing CDC prior to the commencement of antiresorptive and antiangiogenic treatment proved to be an effective modality for preventing MRONJ [15, 21–27]. Preventive measures reduce the risk of disease onset, facilitate early diagnosis, lessen severity, and improve the quality of life [28]. Patients who had preventive dental care showed 50% reduction in the risk of developing MRONJ compared to those who did not have one [22]. Ripamonti et al. (2009) advocated measures, such as complete radiographic assessment with appropriate dental treatment, which resulted in a reduction in MRONJ from 3.2% to 1.3% [29].

A recent study from Otsuru et al. (2024) showed that patients with fewer teeth, apical lesions, periapical osteosclerosis, and local infection were more likely to develop MRONJ [30]. Their study was very detailed and analyzed patient by patient, jaw by jaw, and tooth by tooth. Their work involved 172 patients, 329 jaws, and 3734 teeth. Bacci et al. (2022) showed that most patients who developed MRONJ had not had a dental check-up before starting drug therapy, and the risk of developing MRONJ was 36.62% [26].

In our hospital, the CDC is advocated before and throughout the course of treatment with the antiresorptive and antiangiogenic agents. The CDC includes:

• i)invasive and non-invasive dental treatment (i.e., dental extraction, periodontal therapy, dental restorations, and oral hygiene enforcement) BEFORE the administration of antiresorptives and antiangiogenesis;
• ii)conservative dental treatment (i.e., periodontal therapy, dental restorations, and oral hygiene enforcement) DURING the treatment with antiresorptives and antiangiogenesis

If a dental extraction cannot be avoided while patients are undergoing treatment with the medication of concern, it is performed by a dental surgeon or a specialist trained to recognize signs of compromised healing. This includes identifying “dry” sockets or sockets with minimal bleeding immediately after extraction and teeth that are difficult to extract, which may require an extended duration. For cases with very little bleeding, curettage to promote bleeding is performed. For cases involving difficult extractions, a prescription of oral amoxicillin 500 mg with clavulanic acid 125 mg [9] (i.e., Augmentin®) thrice a day for 2 weeks is recommended. This antibiotic is also prescribed to patients with local infection. In addition, during or following a dental extraction, a full-thickness mucoperiosteum elevation for primary closure is totally discouraged because it will compromise the blood supply to the area [31, 32].

Our second main finding is related to dental extraction, which showed no difference in the odds of developing MRONJ between those who had and did not have a dental extraction. Dental extraction is frequently reported as one of the strongest risk factors for MRONJ [33]. Vahtsevanos et al. (2009) showed that dental extraction had an 18-fold relative risk of developing MRONJ [34]. In the clinical study of Saad et al. (2012), tooth extraction was considered a predictor in 61% of the confirmed MRONJ cases [3]. Tooth extraction and periodontal disease are reported to be major risk factors for MRONJ [33]. Because of this evidence, clinicians tried their best to avoid dental extraction during the treatment period with antiresorptive and antiangiogenic agents.

This statement was challenged by several authors who showed that it was the concurrent infection, and not the tooth extraction that was the risk factor for the MRONJ development [8–10]. Dental infection, as the reason for extraction, increased the osteonecrosis risk in the osteoporosis (OR 22.77; 95% CI: 2.85–181.62; p = 0.003) and bone metastasis cohorts (OR 2.72; 95% CI: 1.28–5.81; p = 0.010) [7]. Anaerobic bacteria representative of periodontal microflora, mainly Porphyromonas, Lactobacillus, Tannerella, Prevotella, Actinomyces, Treponema, Streptococcus, and Fusobacterium, were shown to be associated with MRONJ [35]. In addition, a study evaluated the adhesion of different strains of Staphylococci and Pseudomonas to the hydroxyapatites in association with and without pamidronate [36]. The adhesion of Staphylococci and Pseudomonas on the hydroxyapatite disks coated with pamidronate were seven and three times higher than the uncoated disks, respectively [36]. This was supported by an animal study that showed that bones injected with bacteria are prone to MRONJ [37]. Moreover, a large animal (minipig) study showed the occurrence of MRONJ in infected extraction sockets, areas of marginal periodontitis, periapical pathology, and areas of food entrapment [38].

Another study confirmed that there was no difference in the bone and mucosal healing in patients who had dental extraction while being on bisphosphonate compared to patients who did not receive bisphosphonate treatment [39].

The median age for the MRONJ group in this study was 69.00 (IQR 14.0) years, with a range between 57 and 79 years of age. Srivastava et al. reported an age range between 65 and 69 years old in their systematic review [40]. In relation to gender, this study showed a higher percentage of female being diagnosed with MRONJ; however, there was no statistically significant difference between the genders. Another study from Malaysia also showed similar findings on the demographic characteristics [41]. This was in agreement with a systematic review that found no statistically significant difference between males and females [40].

Our results revealed that the mandible had the highest involvement (54.5%), followed by the maxilla (27.3%) and both jaws (18.2%). The higher mandibular involvement was also shown by Jeong et al. (2017) [42]. However, other studies showed a predominance of maxillary MRONJ compared to mandibular MRONJ [4, 43].

Apart from clinical observation and precaution, radiological assessment plays an important role in showing evidence of early and established stages of MRONJ. The AAOMS Position Papers in 2014 and 2022 outlined the radiological features in the early stage of MRONJ (i.e., Stage 0) [1, 2]. These included alveolar bone loss or resorption not related to chronic periodontal disease, trabecular pattern sclerotic bone change, no new bone in the extraction sockets, islands of osteosclerosis in the alveolar bone and/or the surrounding basilar bone, thickening of the lamina dura, sclerosis, and decreased size of the periodontal ligament space [1, 2, 44].

In this study, osteolytic changes were the most frequently detected radiological signs, followed by focal bone sclerosis and bone sequestration, among others. Cardoso et al. (2017) reported bone sclerosis as the most frequent sign, followed by osteolysis. The other radiological signs were shown less frequently [19].

There were three patients (27%) in this study who did not give any trauma history prior to the MRONJ onset. Most of the MRONJ sites involved an edentulous ridge, and the patients were not denture users. Hence, we classified this MRONJ type as “spontaneous.” Cases of MRONJ that could not be associated with any predisposing factor are categorized as “spontaneous” and associated with certain anatomical sites, including the tori, exostoses, and mylohyoid ridges [43, 45]. Khominsky and Lim (2018) debated that prominent bony protuberances would not be suitable for classification as “spontaneous” because these sites are susceptible to the risk of trauma [45]. They further explained that there is a possible low-grade trauma or a consistent microtrauma around these anatomical structures that will eventually result in injuring the thin mucosa, exposing the underneath bone [45]. Two of the three patients in this study did not have bony prominence. Clinical palpation also did not show any irregularity or prominence of the jaw bones. Spontaneous MRONJ in these cases occurred on the resorbed edentulous ridge and on the palate with no palatal torus. One of the patients showed “spontaneous” MRONJ at multiple sites over a period of 3 years. Altogether, she had six areas of Stage 1 and 2 MRONJ in her oral cavity at a different time phase. All sites showed signs of healing, with two sites healed following surgery and the rest healed without any surgical intervention.

The question is, how can this type of MRONJ be prevented from occurring? To date, there are no studies discussing preventive or therapeutic strategies for spontaneous MRONJ [45].

We recommend a preventive approach that includes regular follow-up for all patients, including those who are edentulous. The early detection of spontaneous MRONJ is crucial to ensuring a prompt treatment. Based on the limited literature, the treatment of spontaneous MRONJ appears to be similar to that of “conventional” MRONJ [45, 46]. To date, the reported cases [45, 46] and our cases showed that the lesions were relatively small, typically Stage 1 to 2 MRONJ, and the treatment outcomes were generally uneventful.

Additionally, reports dating back to the 1990s described idiopathic oral ulceration with bone sequestration occurring at the same sites as spontaneous MRONJ [47, 48]. Similar to spontaneous MRONJ, trauma and anatomically related compromised vascularization have been proposed as the contributory factors [45–47]. Further research is needed to investigate the pathogenesis of spontaneous MRONJ and idiopathic oral ulceration with bone sequestration.

Limitations

First, as previously discussed, the retrospective nature of this study limits its ability to establish definitive cause-and-effect relationships. A research question like “Do patients with local infection who receive CDC have a higher risk of developing MRONJ compared to those without local infection who do not undergo CDC treatment?” would be more appropriately addressed through a long-term prospective cohort study. This recommendation is supported by a systematic review highlighting an unclear bias risk caused by vague baseline information on certain variables (i.e., dental infections and observational design of the studies, both of which reduced the quality of evidence) [21]. A high bias level was described for all retrospective reports and case series [49]. Therefore, the findings of this study are preliminary and unconfirmed despite showing potential.

Second, due to the study design and the limited sample size, the generalizability of the findings is restricted.

Third, we were unable to control for missing or unavailable data in the medical records. Missing data were primarily related to medication and dental factors, especially in cases where patients were referred from external institutions. Consequently, these patients had to be excluded from the analysis. For dental records, the absence of prior records and radiographs before dental extraction hindered our ability to assess the predisposing factors for periapical and periodontal infections. Additionally, the presence of other lesions in the jawbones prior to the commencement of drug therapy cannot be evaluated, introducing minimal bias to our conclusions.

Of the 31 patients without baseline radiographs, three dental panoramic tomograms revealed some radiological features during drug therapy. One patient had focal sclerosis related to tooth 35, which remained asymptomatic and was observed in radiographs taken before and during drug therapy.

Another critical issue in the data pertained to the CDC details. Part of this issue was exacerbated by the COVID-19 pandemic, which disrupted the patient’s regular dental follow-up. Consequently, from the initial pool of 200 patient records, we were only able to include 75.

Recommendation

The patients’ state of oral health determines important decision-making, particularly in the timing of dental treatment at the beginning and during subsequent reviews. The National Oral Health Survey in Adults 2010 (NOHSA 2010) showed that 88.9% of adults in Malaysia had issues related to both treated and untreated dental caries, and 94% had periodontal conditions [50]. With this background, prior to the commencement of antiresorptive and antiangiogenesis, the CDC must aim to control active periodontal and periapical lesions among Malaysian patients. Maintenance CDC should be performed 3 to 4 months, and oral health education should be re-enforced every time a patient comes to the clinic. Radiographs and cone beam CT should be routinely examined, particularly in the periapical and periodontal disease areas. The importance of oral health maintenance must be presented to the patients and family members, such that coming to the clinics frequently will not be regarded as troublesome and a burden. Like many developed countries, a guideline on MRONJ prevention and treatment must be constructed and adhered to by the team managing patients on antiresorptive and antiangiogenesis.

Conclusion

Comprehensive dental care before and during treatment with antiresorptive and antiangiogenic drugs can be an effective preventive method for MRONJ. It is very important to educate all teams and specialists involved in patient management to be aware and support this strategy. In contrast to many studies, dental extraction is not regarded as a significant factor for MRONJ.

Authors’ contributions

KA, SN, NRT, SSMY, and RR contributed to the conceptualization and design of the study. KA collected the data and performed the data description and statistical analysis under the supervision of RR. KA wrote the original draft. All authors participated in the reviewing and editing of the manuscript.

Funding

There was no funding involved.

Data availability

The dataset generated and analysed during the current study is available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study protocols were approved by the Research Ethics Committee, Universiti Kebangsaan Malaysia (UKM) (reference: UKM PPI/111/8/JEP2020-791) dated 25 January 2021. Informed consent was waived by the REC UKM in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.

Competing interests

The authors declare no competing interests.