Accuracy of semiquantitative dynamic contrast-enhanced MRI for differentiating type II from type I endometrial carcinoma.

ABSTRACT

PURPOSE: To investigate type II endometrial carcinoma characterization using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and evaluate the diagnostic accuracy of semiquantitative DCE-MRI in differentiating type II from type I tumors. MATERIALS AND METHODS: Seventy-seven patients with endometrial carcinoma were retrospectively evaluated using 3T DCE-MRI. The maximum absolute enhancement of signal intensity (SImax), maximum relative enhancement (SIrel), wash-in rate (WIR), and the SImax/SI (piriformis) ratio were analyzed. To differentiate type I from type II tumors, optimal threshold criteria were established. The Mann-Whitney U-test was used for statistical comparison and receiver operating characteristic curve analysis was used to determine optimal cutoff values. RESULTS: The SIrel (P < 0.001), WIR (P < 0.0001), and SImax/SI (piriformis) ratio (P < 0.0001), but not SImax, differed significantly between type I and type II carcinomas. Cutoff values of SIrel ≥58.8, WIR ≥37.0, and SImax/SI (piriformis) ratio ≥1.55 had sensitivities of 93%, 93%, and 67%, specificities of 60%, 60%, and 79%, accuracies of 66%, 66%, and 67%, respectively, for predicting type II endometrial carcinoma. CONCLUSION: Endometrial carcinoma with strong (high level) enhancement on DCE-MRI is suggestive of type II endometrial carcinoma. Semiquantitative evaluation of DCE-MRI may be useful for differentiating type II from type I tumors.