Comprehensive prediction model of urinary incontinence one year following robot-assisted radical prostatectomy.

OBJECTIVES: Urinary incontinence (UI) still remains one of the major functional complications after robot-assisted radical prostatectomy (RARP). As the cause for UI is multifactorial, it is quite difficult to make a prediction preoperatively. Considering intraoperative and postoperative risk factors, besides the preoperative ones, we designed an incontinence prediction model, administered 1 month after the surgery, in order to identify incontinent patients at 1 year. PATIENTS AND METHODS: We retrospectively reviewed 244 patients who underwent RARP at our institution. Only 209 patients had sufficient data, a 1-year follow-up and were continent preoperatively. The association of UI with the risk factors was assessed by univariable and multivariable regression models.
RESULTS: There was a 17.2% global UI rate at 1 year after RARP. Only age-adjusted Charlson comorbidity index, erectile function assessed by International Index of Erectile Function-5, prostate volume, nerve-sparing status and 24-hour urine loss at 1 month correlated with UI (p = 0.032, 0.009, 0.031, 0.018 and <0.001, respectively). The accuracy of the prediction model of UI was 92.8% (c-index), with an area under the curve of 91.9%.
CONCLUSION: Age-adjusted Charlson comorbidity index, International Index of Erectile Function-5, prostate volume, nerve-sparing status and 24-hour urine loss at 1 month after RARP can predict an individual's risk of UI at 1 year after RARP with good accuracy. Further external validation is required in order to generalize the use of this model.