Prediction of magnetic resonance imaging scoring system for differentiating leiomyomas from leiomyosarcomas: A cross-sectional study

Objectives The aim of the study was to evaluate the preoperative magnetic resonance imaging (MRI) features that best distinguish leiomyomas (UM) from leiomyosarcomas (LMS) and to predict UMs preoperatively using an MRI scoring system. Methods This was a cross-sectional study conducted at a tertiary center. Between 2013 and 2023, 109 patients who underwent myomectomy or hysterectomy with pelvic MRI in the six months before the operation and had a histopathologically confirmed report of UM or LMS were included in the study. The cases were classified as Group 1 (UM; n=101, 92.6%) and Group 2 (LMS; n=8, 7.4%) according to histopathology reports. Non-normally distributed variables were analyzed using the Mann-Whitney U test. Categorical data were analyzed using the Chi-square test and Fisher's exact test. Receiver Operating Characteristic (ROC) analysis was performed to calculate the cut-off value, specificity and sensitivity of MRI scoring system. Logistic regression analysis were used to evaluate MRI features. The p value considered statistically significant was <0.05. Results Six of the 15 MRI features were statistically different between the groups. Total MRI predictive score was 3 in the UM group and 4.5 in the LMS group. There was a statistical difference between the groups (p<0.01). Total score lower than 7 was calculated as the cut-off point prediciting UM (sensitivity 98%; specificity 38%; PPV 95.5%; NPV 60%, p<0.01). Conclusion This preoperative MRI scoring system demonstrates predictive capability in distinguishing between UMs and LMSs.