Pregnancy outcomes following surgical repair of female genital fistula in Uganda

Abstract

Background

Women developing genital fistula and undergoing fistula repair in sub-Saharan Africa and South Asia are largely of reproductive age, and fistula prevalence is highest in countries with high fertility and where social values are placed on childbearing. Optimizing women's health following female genital fistula and surgical repair requires further understanding of the risks to subsequent pregnancies and how to mitigate them, to enable women to achieve their desired family size without additional morbidity.

Objective

We sought to contribute to the postfistula repair evidence base through estimating rates of spontaneous abortion and stillbirth as well as the associated risk factors of these adverse outcomes in pregnancies following fistula repair.

Methods

We captured data on sociodemographic characteristics, obstetric and fistula history, and pregnancy and childbirth care and experiences, and outcomes for the first postrepair pregnancy from 302 women who became pregnant within 10 years following genital fistula repair, recruited from six fistula repair facilities in Uganda. We described sociodemographic characteristics, spontaneous abortion (pregnancy loss <20 weeks gestational age), and stillbirth (intrauterine fetal death at ≥20 weeks gestational age) and determined factors associated with these outcomes using logistic regression. We compared outcomes to two external data sources: a meta-analysis and propensity-score matched Ugandan women of reproductive-age.

Results

Overall, 14% (43/302) of the participants had spontaneous abortions and 5% (12/255) had stillbirths in postrepair pregnancies. The spontaneous abortion rate in our study was higher compared to a recent meta-analysis; however, the stillbirth rate was not. The stillbirth rate in our study was 2.5 percentage-points higher compared to the general population (95% CI 0.2–4.9, P=.036). Factors independently associated with increased risk of spontaneous abortion included fistula type, vaginal bleeding during pregnancy, any urine leakage, and educational attainment. Vesicovaginal fistula (VVF)-high (VVF types I and III) vs VVF-low (VVF type II Aa Ab Ba Bb) had significantly reduced odds of spontaneous abortion (adjusted odds ratios [aOR] 0.11, 95% CI 0.03–0.45, P=.002) and rectovaginal fistula and VVF-other (other or not indicated) had marginally reduced odds (aOR 0.38, 95% CI 0.012–1.14, P=.083 and aOR 0.26, 95% CI 0.05–1.25, P=.093, respectively). In bivariate analyses, any urine leakage, assisted vaginal delivery, and emergency cesarean section were highly correlated with stillbirth. Stillbirth risk was over-10-fold higher among individuals reporting urine leakage (OR 10.5, 95% CI 2.75–20.43, P=.001). Assisted vaginal birth and emergency cesarean birth were both associated with 17-fold increased odds of stillbirth (OR 16.93, 95% CI 1.45–198.08, P=.024 and 16.56, 95% CI 1.65–166.28, P=.017, respectively).

Conclusion

Our results demonstrate that in the study setting, greater attention to high-quality, comprehensive pregnancy care and birth planning are critical for improving outcomes among women who have undergone fistula repair, including facilitation of elective cesarean section which is recommended for postrepair births. Additional investments must be made to strengthen women's health access and knowledge that supports their postfistula repair reproductive goals.