Recurrence and Pregnancy Post-repair of Obstetric Fistula in Guinea

Overview

Obstetric fistula, also called vaginal fistula, is a serious medical condition which affects women in low income countries. Despite the strengthening of research on fistula, there is little data on the follow-up after fistula repair. The objective of this study is to analyse the factors associated with the recurrence of fistula and the outcomes of pregnancy following fistula repair in Guinea. It will target women who got a closed fistula at discharge after repair in 2012/2015 at three fistula repair sites supported by the Fistula Care Project in Guinea (Kissidougou Prefectoral Hospital, Labé Regional Hospital and Jean Paul II Hospital of Conakry). The outcomes of interest are fistula recurrence and pregnancy. The predictors of interest will include patient characteristics, fistula characteristics, the context of repair and the context of reintegration. Participants giving an informed consent after a home visit by the Fistula Counsellors who managed women during surgery will be interview at enrolment and every six month from inclusion. The study duration is estimated at 48 months (January 2012 to March 2016) including the retrospective part. A sample size of 364 women will estimate the recurrence of fistula with a plus/minus 2% margin of error (width of confidence interval is 4%) and 95% confidence interval and is sufficient to estimate the rate of pregnancy with a two-sided 95% confidence interval and 10% precision. The cumulative incidence rate of fistula recurrence will be calculated using Kaplan-Meier methods and the risk factor analysis will be performed using adjusted cox regression. For the outcomes of pregnancy, Pearson's Chi Square (χ2) will be used to compare proportions of pregnancy outcomes between potential predictors and logistic regression models will be used and associations will be reported as risk ratios with 95% confidence intervals. Analysis will be done using STATA version 13 (STATA Corporation, College Station, TX, USA) with a level of significance set at P\<0.05.

Conditions