External Validation of the SF-36 Quality-of-Life Questionnaire in Italian and Brazilian Populations to Select Patients With Colorectal Endometriosis for Surgery

Study Objective To evaluate the external validity of the validated French model of the quality-of-life questionnaire (QOL) SF-36 in predicting improvement after colorectal resection for endometriosis. Design Italian and Brazilian cohort studies (Canadian Task Force classification II-3). Setting Tertiary referral university hospital in Brazil and expert center in endometriosis in Italy. Patients Patients with colorectal endometriosis from an Italian population (n = 63) and a Brazilian population (n = 151). Intervention Laparoscopic colorectal resection for treatment of endometriosis. Measurements and Main Results Preoperative and postoperative evaluations of the Physical Component Summary (PCS) and the Mental Component Summary (MCS) of the SF-36 were performed. Substantial improvement in PCS and MCS was observed after colorectal resection in both populations. In the Brazilian population, the receiver operating curve (ROC) (area under the curve [AUC]) was 0.83 (95% confidence interval [CI], 0.77–0.89) for MCS and 0.78 (95% CI, 0.71–0.83) for PCS, demonstrating good discrimination performance. The mean difference between the predicted and calibrated probabilities was 19.6% for MCS and 32.8% for PCS. In the Italian population, the ROC curve (AUC) was 0.65 (95% CI, 0.52–0.78) for PCS and 0.67 (95% CI, 0.55–0.78) for MCS. The model demonstrated poor discrimination and calibration performance for PCS (p < .001) and MCS (p = .003). The mean difference between the predicted and calibrated probabilities was 17.5% for MCS and 21.8% for PCS. Conclusion Despite the use of validated translations of the SF-36, our results underline the limits of this tool in selection of patients for colorectal resection due to underestimation of predicted quality of life, possibly because of variations in epidemiologic characteristics of the populations.