Learning curve for the detection of pouch of Douglas obliteration and deep infiltrating endometriosis of the rectum
Ayman Tammaa,Nadja Fritzer,Guido Strunk,Alexander Krell,Heinrich Salzer,Gernot Hudelist
DOI: https://doi.org/10.1093/humrep/deu078
Abstract:Study question: How long does it take to be proficient in diagnosing pouch of Douglas (POD) obliteration and deep infiltrating endometriosis (DIE) of the rectum with transvaginal sonography (TVS)? Summary answer: Sonographers familiar with the general use of TVS are expected to be proficient in the diagnosis of endometriosis nodules of the rectum and the detection of POD obliteration using the 'sliding sign' after ∼40 examinations, performed in a referral clinic for pelvic pain. What is already known: With rectal DIE, the reasons for the obvious diagnostic problems are complex. Menstrual pain or cramps are still considered to be 'normal' and do not provide a reason for patients and even health-care providers to seek expert help. Furthermore, the performance of TVS for diagnosing pelvic endometriosis has been shown to be accurate only in experienced hands. Study design, size and duration: This prospective study included 121 selected patients with suspected endometriosis. Participants/materials, setting, methods: Symptomatic patients, referred to a pelvic pain clinic, were examined by an expert sonographer (E.S.) and consecutively by two trainees (T1/2). Main results and the role of chance: The learning curve using the cumulative sum shows that the trainees, listed as T1/T2, reached the predefined level of proficiency in detecting bowel nodules after examining 42 and 37 patients, for T1 and T2, respectively. The prevalence rate of bowel nodules demonstrated by the ES was 21%. The sensitivity, specificity, positive and negative predictive values (PPV, NPV) as well as the accuracy for TVS of T1 and T2 in comparison with the results of ES were 72 and 89, 96 and 95, 87 and 80, 90 and 98, and 89 and 94%, respectively. The prevalence rate of POD obliteration, as demonstrated by a negative sliding sign, was 27%. The trainees reached the predefined level of proficiency after examining 42 and 33 patients, for T1 and T2, respectively. The sensitivity, specificity, PPV, NPV as well as the accuracy of TVS for T1 and T2 in comparison with the results of the ES were 83 and 89, 95 and 95, 91 and 80, 90 and 98, and 91 and 94%, respectively. Limitations, reasons for caution: We performed this analysis in a tertiary referral centre with a high number of advanced cases of DIE, not reflecting a standard population. Wider implication of the findings: Integrated in TVS training courses, typical sonographic video clips for DIE of the rectum, including the use of disease-specific signs, could help to improve diagnostic accuracy in DIE and shorten diagnostic delays. Study funding/competing interests: No funding was received for this study. None of the authors has any competing interests.