Abstract
OBJECTIVES: To compare transvaginal sonography (TVS) and magnetic resonance imaging (MRI) with intraoperative measurement (IOM) using a rectal probe in the estimation of the location of rectosigmoid endometriotic lesions, i.e. lesion-to-anal-verge distance (LAVD), and to compare two different MRI techniques for measuring LAVD. METHODS: This was a prospective single-center observational study that included women undergoing surgery for symptomatic rectosigmoid endometriosis by discoid (DR) or segmental (SR) resection from December 2018 to December 2019. TVS and MRI were performed presurgically for each participant to evaluate LAVD, and the measurements on imaging were compared with IOM using a rectal probe. Clinically acceptable difference and limits of agreement (LoA) between TVS and MRI compared with IOM were set at ± 20 mm. Two different measuring methods for MRI, MRI(Center) and MRI(Direct) , were proposed and evaluated, as there is currently no guideline to describe deep endometriosis on MRI. Bland-Altman plots and LoA were used to assess agreement of TVS and both MRI methods with IOM. Systematic and proportional biases were assessed using paired t-test and Bland-Altman plots. RESULTS: Seventy-five women were eligible for inclusion. Twenty-eight women were excluded, leaving 47 women for the analysis. Twenty-three DR and 26 SR procedures were performed, with both procedures performed in two women. The Bland-Altman plots showed that there were no systematic differences between TVS or MRI(Center) when compared with IOM for all included participants. MRI(Direct) systematically underestimated LAVD for lesions located further from the anal verge. TVS, MRI(Center) and MRI(Direct) had LoA outside the preset clinically acceptable difference when compared with IOM. LAVD was within the clinically acceptable difference from IOM of ± 20 mm in 70% (33/47) of women on TVS, 72% (34/47) of women on MRI(Center) and 47% (22/47) of women on MRI(Direct) . CONCLUSIONS: TVS should be the preferred method to estimate the location of a rectosigmoid endometriotic lesion, i.e. LAVD, as it is more available, less expensive and has a similar accuracy to that of MRI. Estimating LAVD can be relevant for planning colorectal surgery for rectosigmoid endometriosis. © 2022 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.