Factors associated with obstetric fistula repair failure among women admitted at Gynocare women’s and Fistula Hospital in Kenya, 2012-2016: a case control study

Abstract

Background: Obstetric fistula is an abnormal hole that connects a woman’s vagina and bladder or rectum or both leading to continuous leakage of urine and/or stool. It is estimated to affect about 2 million women in the world, with Kenya contributing about 3,000 incidences each year. Currently surgery is the main treatment option. However, surgical failure rate remains a major challenge. Objective: The study aimed at determining factors associated with obstetric fistula repair failure at Gynocare Women’s and Fistula Hospital in Kenya. Methodology: A case control study design was adopted to investigate factors influencing the probability of fistula repair failure. Study population comprised of obstetric fistula patients who underwent fistula repair at Gynocare Women’s and Fistula Hospital and repair outcomes at discharge known between January, 2012 to December, 2016 and a sample size of 357 (119 cases and 238 controls) was used. Simple random sampling was used to select the cases and controls using a computer generated random numbers. STATA 13 SE was used to code, clean and analyze data. For categorical variables frequencies and proportions were reported and presented in tables. Continuous variables were summarized using measures of central tendency (mean/median) and dispersion (standard deviation/inter-quantile range); summaries were presented in tables and distribution in histograms/box plots. Bivariate analysis was done to check for association between variables; Chi-square/Fisher’s Exact test and Independent samples t-test/Mann-Whitney U test were used respectively for categorical versus categorical and categorical versus numerical variables. Logistic regression model was used to evaluate the adjusted odds ratio at alpha significance level of 0.05. Results: On average age at development of fistula was 21 years (IQR 17, 28) but 30 years (IQR 21, 40) at the time of repair with an average fistula time period of 4.3years(IQR 06, 16.1). Study participants were mostly (62.2%) married with low or no formal education (90.1%) women. Delivery that led to fistula development occurred in the hospital for 85.2% of the study participants and 66.7% resulted into C/S. Only 1/3 (n = 120) had previous repair(s), while patients classified to have vesicovaginal fistula class IIA were 20.5%, class IIB were 35.0% and class III were 9.2%, the rest (35.3%) were classified as vesicovaginal fistula class I. The odds of failure were 2.9 times more among those with previous repair attempts compared to those with no previous repair attempts. Women with vesicovaginal fistula class IIB were 4 times more likely to develop failure. While women who attained at least secondary education level were 77% less likely to have fistula repair failure. Conclusion: After controlling the effects of age, marital status, comorbidities, parity, time to repair, and post-operative complications: having not attained at least secondary education level, having previous repair attempts, and vesicovaginal fistula class IIB were found to be independent predictors of closure failure. Further studies required to investigate other factors not included in this study and probably extend the follow-up period beyond hospital discharge.