Predicting superficial peritoneal endometriosis in symptomatic women with negative examination and imaging
Elisabeth Reiser1, Lena Tschiderer2, Olaf Buchweitz3
1Department of Gynecological Endocrinology and Reproductive Medicine, Medical University of Innsbruck, Innsbruck, Austria.
Objective:
To develop a predictive risk score for superficial peritoneal endometriosis (SPE) in patients with negative examination and imaging, based on symptoms and quality of life measures.
Design:
Multicenter nested case-control study conducted from September 2022 to January 2024 at 18 certified endometriosis centers in Austria, Germany, and Switzerland. Women undergoing first-time surgery for suspected endometriosis had surgical findings documented using the #ENZIAN classification. Only women with isolated peritoneal endometriosis or no endometriosis were included in this analysis. Subjects completed standardized preoperative questionnaires, which assessed pain characteristics, symptom duration, analgesic intake, sexual function (FSFI), and health-related quality of life (EHP-30). Multiple imputation, logistic regression, and receiver operating characteristic curve analyses were performed.
Subjects:
Of 845 women in the full study cohort, 229 met inclusion criteria, completed all questionnaires, and had complete surgical data available: 121 (53%) had SPE only, and 108 (47%) had no endometriosis. A validation cohort of 268 subjects (206 [77%] SPE and 62 [23%] no endometriosis) was analyzed.
Exposure:
Pain-related variables, internally validated by bootstrapping, and externally validated in an independent cohort.
Main Outcome Measures:
Surgically diagnosed peritoneal endometriosis RESULTS: Severe dysmenorrhea (odds ratio [OR] 2.01; 95% CI, 1.14-3.53) and pain duration >3 years (OR 2.40; 95% CI, 1.40-4.11) were significantly associated with SPE. Dyspareunia, dysuria, dyschezia, pain only during menses, analgesic intake, family history, health-related quality of life, and sexual function were not. A 5-item predictive score (maximum six points) showed moderate discrimination in the development cohort (area under the curve [AUC] 0.64; 95% CI, 0.57-0.72), with similar performance on internal validation (AUC 0.64; 95% CI, 0.57-0.71). External validation demonstrated no predictive value (AUC 0.49; 95% CI, 0.41-0.58).
Conclusion:
Although severe dysmenorrhea and long duration of symptoms were associated with SPE, symptom profiles and functional questionnaires were insufficient for accurate prediction. The calculated score performed modestly internally but failed external validation. Given that symptoms common to endometriosis are also reported in women with other etiologies of pelvic pain, symptom-based models appear inadequate for reliable identification of SPE. Integrating clinical factors with biomarkers or emerging diagnostic technologies may be necessary to improve noninvasive diagnostic accuracy. Clinicians and patients should be aware of the high rate of negative surgical findings when making the decision to proceed with laparoscopy versus considering conservative medical therapies instead.
