Patient and tumour-related factors for prediction of urinary incontinence after radical prostatectomy

Objective. The aim of this study was to identify preoperative patient and tumour-related factors associated with 12 months postoperative urinary incontinence. Material and methods. In total, 1529 men who had undergone radical prostatectomy for clinically localized prostate cancer between September 2008 and February 2010 at 15 Swedish hospitals completed a questionnaire before, 3 and 12 months after surgery. Urinary leakage, comorbidity and possible confounders were measured by self-administered validated questionnaires. Clinical data were collected preoperatively and postoperatively. The primary outcome, incontinence, was defined as the change of one pad or more per day. The ratio of proportions, estimated according to the log-binomial regression model, was analysed for 38 different factors and is presented as relative risks with 95% confidence intervals. Age-adjusted relative risk was calculated in the corresponding bivariate regression model. Results. Prospective data were available from 1360 men (response rate 89%). Results showed that age at surgery predicts long-term urinary incontinence exponentially. Patients reporting urinary leakage before prostate cancer diagnosis had an age-adjusted relative risk of 1.8 (95% confidence interval 1.3-2.4) for incontinence 12 months postoperatively. No statistically significant correlation was found between previous transurethral resection of the prostate, high body mass index or the other 34 evaluated factors and postoperative incontinence. Conclusions. Of 38 possible risk factors only age at surgery and preoperative urinary leakage were associated with 12 months postoperative incontinence in this study comprising 1360 men operated with radical prostatectomy. These findings may help the surgeon to have a targeted risk conversation with the patient before the treatment decision is made.