Evaluation and comparison of the accuracy of transvaginal ultrasound and MRI for the diagnosis of deep rectal endometriosis: A cross-sectional study.

Abstract

To determine and compare the value of magnetic resonance imaging (MRI) and transvaginal sonography (TVS) in the diagnosis of rectal endometriosis. In this cross-sectional study, 555 patients with rectal endometriosis, who had undergone MRI and TVS before laparoscopic operation, were included. The sensitivity, specificity, and accuracy of these two imaging modalities were evaluated and compared based on histopathologic reports and three different kinds of rectal endometriosis surgeries (shaving as a first group and disk and segmental resection methods combined as a second group). Sensitivity, specificity, accuracy, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of ultrasound in the diagnosis of rectal endometriosis were, respectively, 69.55% (65.4%-73.4%), 91.30% (72%-98.9%), 70.4% (66.4%-74.17%), 11.48% (9.77%-13.43%), 99.46% (98%-99.86%), 8.0 (2.12-30.1), and 0.3 (0.28-0.4). These values were 51.37% (47.1%-55.6%), 79.17% (57.8%-92.9%), 52.53% (48.3%-56.6%), 6.64% (5.39%-8.16%), 98.26% (96.26%-99.2%), 2.47 (1.13-5.4), and 0.6 (0.49-077) for MRI. Even though ultrasound had better accuracy for detection of superficial rectal endometriosis (Group 1) (P< 0.001), the sensitivity, specificity, and accuracy of both imaging modalities in diagnosis of deep rectal lesions (Group 2) were almost identical. TVS should be considered as a first-line modality for diagnosis of rectal endometriosis, mainly because of its greater availability, lower cost, and higher accuracy.