A V Genever1, S Abdi. 1. Department of Radiology, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK. ashleigh.genever@gmail.com
Abstract
AIM: To ascertain whether magnetic resonance imaging (MRI) can be used to suggest a diagnosis of endometrial carcinosarcoma. MATERIALS AND METHODS: A retrospective imaging review was performed of the MRI images of 39 patients with endometrial carcinosarcoma. Anteroposterior and longitudinal dimensions of the uterus and measurement of the endometrial thickness were taken in the sagittal plane. The ratio of antero-posterior measurement to endometrial thickness was calculated. The same measurements were taken on 50 patients with endometrial adenocarcinoma. A two-tailed t-test and receiver operator characteristic analysis were performed on these measurements for both patient groups. RESULTS: There was a statistically significant difference between the longitudinal (p=0.0084) and anteroposterior (p=0.00036) dimensions of the uterus of the two groups. There was also a statistically significant difference between the ratios of anteroposterior dimension to endometrial thickness (p=0.00018). The optimal cut-off ratio above which the radiologist should be alerted to a potential diagnosis of endometrial carcinosarcoma was shown to be 0.63. CONCLUSION: MRI can be helpful in suggesting a diagnosis of endometrial carcinosarcoma with the help of some simple measurements.
AIM: To ascertain whether magnetic resonance imaging (MRI) can be used to suggest a diagnosis of endometrial carcinosarcoma. MATERIALS AND METHODS: A retrospective imaging review was performed of the MRI images of 39 patients with endometrial carcinosarcoma. Anteroposterior and longitudinal dimensions of the uterus and measurement of the endometrial thickness were taken in the sagittal plane. The ratio of antero-posterior measurement to endometrial thickness was calculated. The same measurements were taken on 50 patients with endometrial adenocarcinoma. A two-tailed t-test and receiver operator characteristic analysis were performed on these measurements for both patient groups. RESULTS: There was a statistically significant difference between the longitudinal (p=0.0084) and anteroposterior (p=0.00036) dimensions of the uterus of the two groups. There was also a statistically significant difference between the ratios of anteroposterior dimension to endometrial thickness (p=0.00018). The optimal cut-off ratio above which the radiologist should be alerted to a potential diagnosis of endometrial carcinosarcoma was shown to be 0.63. CONCLUSION: MRI can be helpful in suggesting a diagnosis of endometrial carcinosarcoma with the help of some simple measurements.