Patient related risk factors for recurrent stress urinary incontinence surgery in women treated at a tertiary care center

Purpose: We examined patient related risk factors for recurrent stress urinary incontinence in women treated at a tertiary referral center. Materials and Methods: A case-control study was done in 18 to 75-year-old women with signs and symptoms of genuine or mixed stress urinary incontinence and no prior surgical treatment who underwent an open anti-incontinence procedure between 1990 and 2002 at our institution. Cases were defined as patients who underwent more than 1 anti-incontinence surgery and controls were defined as patients who underwent only 1 anti-incontinence procedure with followup during that period. Cases and controls were matched for surgery type, surgeon and date of surgery within 1 year. A total of 47 variables were examined, including patient age, parity, incontinence type, urodynamic findings, medical history (peripheral vascular, pulmonary and cardiac disease), past and concomitant pelvic surgery, social history (alcohol and tobacco use) and body mass index. Univariate conditional logistic regression was done first to determine which variables were potential protective or risk factors. Multivariate conditional logistic regression analysis was then used to determine which factors were statistically significant. Results: The records of 2,550 women with stress or mixed urinary incontinence who underwent an open surgical procedure between 1990 and 2002 were reviewed. A total of 53 cases and 146 controls were identified. Each case was matched with 1 to 4 controls. Data on cases and controls were collected using a standardized form. At a significance level of 0.05 the possible risk factors for recurrent stress urinary incontinence based on univariate analysis were diabetes mellitus (OR 3.579, p = 0.026), pelvic organ prolapse (OR 5.635, p = 0.03) and concomitant rectocele repair (OR 5.353, p = 0.04). Smoking was marginally protective (OR 0.497, p = 0.068). After multivariate conditional logistic regression analysis diabetes mellitus (adjusted OR 3.413, p = 0.045), pelvic organ prolapse (adjusted OR 8.195, p = 0.021) and concomitant rectocele repair (adjusted OR 17.079, p = 0.012) remained significant risk factors, while smoking remained a protective factor (adjusted OR 0.264, p = 0.012). Body mass index, age, race, parity and estrogen status were not identified as risk factors for recurrent stress urinary incontinence requiring a second anti-incontinence procedure. Conclusions: In a cohort of women with stress or mixed urinary incontinence treated at our institution between 1990 and 2002 women with diabetes mellitus, pelvic organ prolapse or concomitant rectocele repair were at increased risk for repeat anti-incontinence surgery, while women who smoked were at slightly decreased risk.