Taking computed tomography (CT) scans before treatment with bone resorption inhibitors may help identify patients at risk of developing medication-related osteonecrosis of the jaw (MRONJ), according to a news release from Hiroshima University in Japan.
Furthermore, measurements of both cancellous and cortical jawbone features may predict which patients later develop MRONJ, according to the release dated August 7.
“This study demonstrated the potential to predict the development of MRONJ using image features extracted from CT images acquired prior to the administration of bone resorption inhibitors,” Masaru Konishi, a lecturer at Hiroshima University Hospital and the author of the study, said in the release.
The study examined pretreatment multidetector CT scans from 94 patients, including 44 who later developed MRONJ in the lower jaw and 50 who remained free of the condition for at least two years after starting bone resorption inhibitor therapy, which is given to patients with conditions that cause excessive bone loss or high blood calcium. Researchers used radiomic measurements from two areas of the jaw: cancellous bone alone and a region containing both cancellous and cortical bone.
Machine-learning models generally performed better when cortical bone was included in the analysis. Using the Random Forest model, the average area under the curve (AUC) increased from 0.741 with cancellous bone alone to 0.931 when both bone types were evaluated, according to the results.
Also, the support vector machine model improved from an AUC of 0.706 to 0.917, while the Multilayer Perceptron model increased from an AUC of 0.802 to 0.921. These results indicate strong predictive ability within the study population, because an AUC of 1 represents perfect discrimination and 0.5 represents chance-level performance.
In one analysis of the combined bone region, the multilayer perceptron model correctly identified 88.6% of patients who later developed MRONJ and 88.6% of those who did not. Overall, the findings suggest that evaluating both cancellous and cortical bone on pretreatment CT scans may improve the ability to identify patients at risk for MRONJ, according to the release.
However, the study had limitations. The study’s retrospective design and limited information on dental factors may have affected the consistency and accuracy of the findings regarding the MRONJ risk.
Further research should focus on identifying specific imaging features, such as bone quality and microarchitecture, and evaluating patients’ dental health before MRONJ develops, according to the release.
