Abstract / Summary
Goal: To determine whether an expanded apparent diffusion coefficient (ADC) can differentiate between benign and malignant uterine tumors that are rich in microcysts. Method: Retrospective evaluation of 160 patients with pathologically confirmed endometrial lesions containing multiple microcysts who underwent preoperative 3.0T pelvic MRI (including DWI at b = 0, 800 s/mm 2 ). According to post-operative histology, the subjects were divided into a benign group (n = 118: 52 cystic endometrial polyps, 41 cystic submucosal leiomyomas, 25 submucosal adenomyomas) and a malignant group (n = 42: 15 FIGO Stage I uterine adenocarcinomas, 27 FIGO Stage I uterine carcinosarcomas). ADC metrics and ADC histogram features of the two groups were compared, and ROC curves for diagnostic accuracy were calculated. Result: All ADC indicators in the benign group (ADC, rADC, ADCmean, ADCmax, ADCmin, plus the 10th, 25th, 50th, 75th and 90th percentiles) were significantly higher than those in the malignant group (all P < 0.01). ROC curve analysis showed that all ADC parameters had moderate or higher diagnostic performance in differentiating benign from malignant uterine lesions with multiple microcysts. The AUCs for rADC, ADCmean, Q50, and Q25 were 0.928, 0.927, 0.916 and 0.902, respectively. Pairwise comparisons using the DeLong test showed no statistically significant differences among these ROC curves (all P > 0.05), indicating that rADC, ADCmean, Q50, and Q25 possess statistically equivalent diagnostic performance. Conclusion: ADC quantitative parameters have important clinical value in differentiating benign from malignant uterine lesions with multiple microcysts. rADC, ADCmean, Q50 and Q25 all showed good discriminative ability, and their 95% CIs overlapped, indicating similar diagnostic performance. The optimal diagnostic threshold of rADC was 1.37, and rADC may serve as one of the reference indicators for preoperative non-invasive differentiation of such lesions, providing an objective basis for individualised surgical planning.