Are endometrial nerve fibres unique to endometriosis? A prospective case–control study of endometrial biopsy as a diagnostic test for endometriosis in women with pelvic pain

• Published in: Human reproduction (Oxford) 2015-12, Vol.30 (12), p.2808-2815
• Main Authors: Ellett, Lenore, Readman, Emma, Newman, Marsali, McIlwaine, Kate, Villegas, Rocio, Jagasia, Nisha, Maher, Peter
• Format: Article
• Language: English
• Subjects: Adult; Biomarkers; Biopsy; Case-Control Studies; Endometriosis - pathology; Endometrium - innervation; Endometrium - pathology; Female; Humans; Nerve Fibers - pathology; Pelvic Pain - pathology; Prospective Studies; Sensitivity and Specificity; Young Adult
• ISSN: 0268-1161; 1460-2350
• DOI: 10.1093/humrep/dev259

STUDY QUESTION Can the presence of endometrial nerve fibres be used as a diagnostic test for endometriosis in women with pelvic pain? SUMMARY ANSWER Endometrial fine nerve fibres were seen in the endometrium of women both with and without endometriosis, making their detection a poor diagnostic tool for endometriosis. WHAT IS KNOWN ALREADY Laparoscopy and biopsy are currently the gold standard for making a diagnosis of endometriosis. It has been reported that small density nerve fibres in the functional layer of the endometrium are unique to women with endometriosis and hence nerve fibre detection could function as a less invasive diagnostic test of endometriosis. However, it may be that other painful conditions of the pelvis are also associated with these nerve fibres. We therefore focused this prospective study on women with pelvic pain to examine the efficacy of endometrial nerve fibre detection as a diagnostic test for endometriosis. STUDY DESIGN, SIZE, DURATION This prospective case–control study conducted between July 2009 and July 2013 included 44 women with pelvic pain undergoing laparoscopic examination for the diagnosis of endometriosis. Immunohistochemical nerve fibre detection in endometrial curettings and biopsies using anti-protein gene product 9.5 was compared with surgical diagnosis. PARTICIPANTS/MATERIALS, SETTINGS, METHODS Paired endometrial biopsies and curettings were taken from patients with (n = 22, study group) and without (n = 22, control group) endometriosis. Tissue was analysed by immunohistochemistry and nerve fibres were counted whenever they were present in the functional layer of the endometrium. MAIN RESULTS AND THE ROLE OF CHANCE Fine nerve fibres were present in the eutopic endometrium of patients both with and without endometriosis. The presence of nerve fibres in curettings was not effective for either diagnosing or excluding endometriosis; sensitivity and specificity were 31.8 and 45.5% respectively, positive predictive value was 36.8% and negative predictive value was 40.0%. Few endometrial biopsy specimens were found to have nerve fibres present; sensitivity and specificity for endometrial biopsy were 13.6 and 68.2% respectively, positive predictive value was 30.0% and negative predictive value was 44.1%. LIMITATIONS, REASONS FOR CAUTION This was a relatively small sample size and studies like this are subject to the heterogeneous nature of the patient population and tissue samples, despite our best efforts to regulate