Uterine neoplasms: Magnetization transfer analysis of MR images

PURPOSE: To determine whether the magnetization transfer effects seen at fast spin-echo (SE) magnetic resonance (MR) imaging provide additional information in the diagnosis of uterine neoplasms. MATERIALS AND METHODS: MR images were obtained in 58 subjects with a normal uterus, 27 patients with cervical carcinoma (17 before therapy; 10 after irradiation), 10 with endometrial carcinoma, and five with endometrial hyperplasia. The magnetization transfer ratios (MTRs) were calculated by measuring the signal intensity on the central section of multisection fast SE images (15 sections) and that on a single-section fast SE image (5,000/136 [repetition time msec/echo time msec]). The additional imaging time required for MTR measurement was 4 minutes. RESULTS: Conspicuity of lesions and normal zonal anatomy was improved with single-section fast SE. The mean MTR in untreated cervical carcinoma (37.6% ± 4.2 [standard deviation]) showed a statistically significant increase relative to that in irradiated cervical carcinoma (24.5% ± 5.1, P < .01), normal cervical stroma (23.8% ± 6.7, P < .01), or normal endocervix (24.2% ± 7.3, P < .01). The mean MTR in endometrial carcinoma (34.5% ± 3.6) was significantly higher than that in endometrial hyperplasia (25.5% ± 1.5, P < .01) or normal endometrium (24.3% ± 3.6, P < .01). CONCLUSION: Single-section fast SE imaging provides better lesion contrast than multisection fast SE imaging and enables quantification of the MTRs in uterine neoplasms.