Evaluation of a new questionnaire for the presurgical diagnosis of bladder endometriosis

Background: The main objective was to evaluate the diagnostic accuracy of a new questionnaire for the presurgical diagnosis of bladder endometriosis in patients with a high suspicion index for this disease. Methods: We included all patients of age <40 years undergoing laparoscopy or laparotomy for chronic pelvic pain. We partially modified the American Urologic Association Symptom Index with the aim of identifying bladder endometriosis among 157 women undergoing surgery for chronic pelvic pain. All patients underwent preoperative ultrasonography; selected patients, with suspected bladder endometriosis, underwent computed tomography and cystoscopy. The physicians performing both the preoperative evaluation and surgery were blinded to the questionnaires' results. Results: A total of 127 (81%) patients had pelvic endometriosis, 14 (8.9%) had bladder endometriosis. The questionnaires' score for patients with and without bladder endometriosis was 21 +/- 8.7 and 4.6 +/- 5.7, respectively (P < 0.0001). The area under the receiver operating characteristic curve was 0.951. With a cut- off of 9, sensitivity was 93% and specificity 88%. Conclusions: The questionnaire proved to be effective in identifying bladder endometriosis, allowing a proper diagnostic work- up and surgical treatment, and minimizing the risk of recurrence. In this primary referral centre for endometriosis the prevalence of the disease was high - therefore it may achieve a lower diagnostic accuracy when evaluated on a population of women with a lower prevalence of bladder endometriosis.