Radiographic Findings of MRONJ

A retrospective study was conducted using records and panoramic radiographs from the databases of the Hospital Santa Catarina and University of Sagrado Coração, São Paulo, Brazil, encompassing a period of 12 years (2003–2015). The inclusion criterion was the patient presenting MRONJ associated with bisphosphonate therapy. Diagnosis and staging were based on the publication by the AAOMS of its 2014 guidelines [1 (link)]. Patients undergoing head and neck radiotherapy were excluded from this study. The data was collected from the patient's medical records as follows: age, gender, type of systemic disease, type of BP, duration of BP treatment, site of the MRONJ, drug administration protocol, and MRONJ staging. The aim of this study was to identify the radiographic findings of MRONJ and correlate them with the AAOMS clinical staging system. Stage 0 included patients with clinical signs of osteonecrosis other than exposed bone. Stage 1 included patients with exposed necrotic bone but no signs of infection. Stage 2 included patients with exposure of necrotic bone together with signs and symptoms of infection, and stage 3 included patients with exposed necrotic bone and an extraoral fistula, sequestration, or mandibular fracture. Radiographic features consisted of the use of panoramic radiography.
Radiographic evaluation was performed by two calibrated examiners (1 and 2). For the evaluation, the arches were divided into sextants (1, 2, 3: maxilla and 4, 5, 6: mandible) based on a previously described methodology [5 (link)]. Osteolysis (OT), cortical bone erosion (EC), bone sclerosis focal (FS) and diffuse sclerosis (DS), bone sequestration (BS), thickening of the lamina dura (TD), prominence of the inferior alveolar nerve canal (IAN), persisting alveolar sockets (SK), and the presence of pathological fracture (PF) were investigated. MRONJ staging and the patient's medical history were also recorded for correlation with the radiographic findings. Data from the measurements were organized in Excel tables (Microsoft Office Excel, Redmond, WA, USA) and submitted to SigmaPlot software (SigmaPlot, San Jose, CA, USA) version 12.3. The agreement between the different factors evaluated by the examiners (1 and 2) was interpreted by a Kappa inter-rater test. For the association between nominal variables we used the statistical test Chi-square and Fisher's exact statistical tests. Pearson correlation coefficients (nominal variables) and Spearman correlation (ordinal variable) were used for correlations. The agreement between the AAOMS staging system and the radiographic findings for the detection of bone disease was evaluated by calculating the proportion of patients in each AAOMS stage. Data were analyzed regarding normal distribution (Shapiro-Wilk test and equal variance assumption) and subsequentl the one-criterion analysis of variance test (Score Factor) was adopted with the radiographic findings.