LBA02-08 THE IMPROVE TRIAL: SURGICAL TECHNIQUE REMAINS THE MOST IMPORTANT FACTOR ASSOCIATED WITH RECOVERY OF URINARY CONTINENCE AFTER RADICAL PROSTATECTOMY

Abstract

INTRODUCTION AND OBJECTIVE: Post‐radical prostatectomy (RP) urinary incontinence (UI) is not uncommon. Both pelvic floor muscle training (PFMT) and duloxetine have been shown to be effective in improving post‐RP UI in retrospective series. We aim to assess the efficacy of PFMT and duloxetine in urinary continence recovery (UCR) after robot‐assisted RP (RARP). METHODS: We conducted a prospective, randomized controlled trial (NCT02367404) involving patients who experienced UI after RARP. Patients were randomized to 1 of 4 arms: (1) PMFTbiofeedback (BFB) only, (2) duloxetine only, (3) combined PMFT‐BFB and duloxetine, and (4) control group. The PMFT program consisted of pelvic muscle contraction exercises conducted with electromyographic feedback weekly, for 3 months. Patients in duloxetine arms were instructed to take oral duloxetine 60 mg at bedtime for 3 months. The primary end point was continence rate at 6 months, defined as no leakage of urine during 3 consecutive days on the 24‐hour pad test. Urinary symptoms and quality of life (QoL) were assessed by using a visual analog scale, the International Prostate Symptom Score questionnaire, and the King's Health Questionnaire. RESULTS: A total of 240 patients (60 in each arm) were included in the trial between 2015 and 2018. Overall, 89% of patients completed 1 year of follow‐up. From the patients allocated to receive treatment, 58% (69 of 120) had properly taken duloxetine and 38% (46 of 120) performed at least 10 sessions of PMFT. In the control group 53% of patients have achieved continence at 6 months, compared with 35% (p=0.07) in the PMFT‐BFB arm, 39% (p=0.2) in the duloxetine arm, and 27% (p=0.009) in the combined treatment arm, without difference in time to UCR. Moderate to severe urinary symptoms were less frequent in the control group (11%) compared with treatment arms: 27% (p=0.03), 30% (p=0.01), 24% (p=0.07), respectively. At 6 months, QoL was qualified as uncomfortable or worse in 17% of patients in the control group, compared with 44% (p=0.02), 45% (p=0.01), and 38% (p=0.06) in treatment arms, respectively. After accounting for other variables, neurovascular bundle preservation was the only factor found to be associated with UCR (OR 3.5 [IQR 1.2‐10.3]; p=0.02). CONCLUSIONS: PMFT‐BFB and duloxetine do not improve UCR after RP, and may have negative impact on QOL; therefore, they should not be routinely recommended to patients who have undergone RP. Neurovascular bundle preservation was the only factor found to be associated with RUC.