The study aims to assess the impact of bowel preparation on MRI scans in diagnosing rectosigmoid endometriosis (RSE) and to examine the level of agreement among investigators, including radiologists and clinicians with varying levels of MRI experience. Fifty-eight patients underwent MRI scans, with 34 showing RSE and 24 without involvement, according to TVUS as the reference standard. The scans were conducted both with and without a bowel preparatory protocol (BPP). The highest concordance was observed in the group 4, which included patients with RSE and BPP, with a Kappa value of 0.47-0.52. In contrast, for groups with MRI scans without the BPP, there was no consistency among the investigators (Kappa <0.20). The sensitivity, specificity, positive predictive value, and negative predictive value were 0.72, 0.47, 0.72, and 0.47 without BPP, and increased to 0.82, 0.74, 0.85, and 0.74, respectively, after BPP. The results reveal that the concordance between investigators in detecting RSE via MRI scan improves with the use of BPP. Additionally, the success rate of MRI and TVUS for detecting RSE can be comparable, but only if bowel preparatory techniques are utilized during MRI scans.
Read full abstract