RESUMO
INTRODUCTION AND HYPOTHESIS: This study aims to compare the prognostic value of two obstetric fistula classification systems. METHODS: Prospective analysis of 202 patients evaluated for obstetric fistula (OF) at the General Referral Hospital of Panzi, Bukavu, DRC, from April through December 2009. Fistula classification using both Goh's and Waaldijk's systems, as well as preoperative, surgical, and follow-up assessment were included. Receiver operating characteristics (ROC) curves were used to compare the accuracy of the two systems to discriminate successful closure from persistent fistula. RESULTS: Two hundred two women underwent fistula repair. Ten were lost to follow-up. At longest follow-up, 181 patients (88.3%) had successful fistula closure. On multivariate analysis, the independent variables of multiparity and a primary or secondary repair were more likely to have a successful closure. In Waaldijk's system, no single component was more predictive of successful closure than another. In Goh's system, type 4 fistulae were more likely to have failed closure compared to those with type 1 or 2 (p = 0.0144). When comparing ROC curves, Goh's system had significantly better ability to predict successful closure than the Waaldijk's system, p = 0.0421. CONCLUSIONS: Waaldijk and Goh are the two most commonly used obstetric fistula classification systems. In this series of OF patients at Panzi Hospital in the Democratic Republic of Congo, Goh's classification system demonstrated a significantly better prediction of OF closure than the Waaldijk's system.
