THE REPRODUCIBILITY OF THE REVISED AMERICAN-FERTILITY-SOCIETY CLASSIFICATION OF ENDOMETRIOSIS

Objective: To assess the degree of intraobserver and interobserver var iability in endometriosis staging using the revised American Fertility Society (AFS) classification of endometriosis. Design: Videotapes of laparoscopies of 20 patients with endometriosis were each scored twice by five observers. Setting: The reproductive endocrine unit of a tert iary care, university-affiliated hospital. Subjects: Five subspecialty -certified reproductive endocrinologists. Interventions: None. Main Ou tcome Measures: Variability in assigned score was measured for each of the five components of the AFS classification, as well as total score s and stage of endometriosis. Results: There was considerable variabil ity in the scores assigned to each videotape, both by the same observe r and by different observers. The grand total score, which ranged from 0 to 90, varied with an SD of 13.44 when a single patient was rated t wice by the same observer and varied with an SD of 17.12 when a single patient was rated by two different observers. Among individual compon ents of the score, the greatest variability occurred in endometriosis of the ovary and cul-de-sac obliteration, with less variability observ ed for peritoneum endometriosis and for ovarian and tubal adhesions. C omparison of intraobserver and interobserver scores resulted in a chan ge in endometriosis stage in 38% and 52% of patients, respectively. Th ere were statistically significant differences in mean endometriosis s cores among the observers in four of the five anatomic categories exam ined. Conclusions: Intraobserver and interobserver variability was hig h for ovarian endometriosis and cul-de-sac subscores using the revised AFS classification of endometriosis.