Reproducibility of the revised American Fertility Society classification of endometriosis using laparoscopy or laparotomy

Objective: To assess the discrepancy between laparoscopic and laparotomic scoring methods using the revised American Fertility Society (AFS) classification of endometriosis. Method: In this prospective study, 84 patients with endometriosis were scored twice (laparoscopically and laparotomically) by the same subspecialty-certified reproductive endocrinologist. The magnitude of inter-method variability was reported quantitatively by the S.D. of the differences in scores between the pairs. The differences in the mean endometriosis scores between the two methods were assessed by the paired Student's t-test. P < 0.05 was considered as statistically significant. Discrepancy between the two methods in the staging of endometriosis patients was presented by kappa measure of agreement. Result: There was considerable variability in the scores between the two scoring methods by the same observer. Among individual components of the scoring system, the greatest variability occurred in the ovarian endometriosis and cul-de-sac obliteration subscores, with the least variability observed for peritoneum endometriosis. The inter-method variation in score was sufficient to alter the endometriosis staging in 34.5% of patients, including a difference of two stages in 3.6% of patients. The kappa coefficient was 0.49, indicating fair-to-good agreement between the two scoring methods. Conclusion: Inter-method variability between laparoscopic and laparotomic scoring methods was high for ovarian endometriosis subscore using the revised AFS classification of endometriosis. Agreement in endometriosis staging between the two methods was fair to good. (C) 1998 International Federation of Gynecology and Obstetrics.