Introduction:

Long term stress urinary incontinence affects 4-8% of men who undergo open or robotic assisted radical prostatectomy (RARP). Urologists often refer patients with SUI to pelvic floor physical therapy (PFPT), but little is known regarding patient factors that make PFPT more effective in continence recovery. We sought to identify predictors of response to PFPT in treating post-prostatectomy incontinence (PPI).

Methods:

A retrospective chart review was performed on patients who underwent RARP and pre- or post-op PFPT at the University of Miami in 2019. Multivariate analysis was performed to identify predictors of success of PFPT in managing PPI. We performed a separate analysis in the subset of patients who underwent preoperative multiparametric magnetic resonance imaging (mp-MRI) to identify possible anatomic factors associated with PFPT response including type of prostatic apex (using the Lee type classification), membranous urethra length, and presence of median lobe. Degree of PPI was measured by number of pads used a day. Patients were categorized in the following groups: no improvement, some improvement, and complete resolution.

Results:

Our cohort included 76 men: 80% were white, mean age was 63.2 years (SD 6.6), and average prostate volume was 49cc (SD 32.6). In this cohort, 25% had complete resolution of PPI with PFPT, 30% showed some improvement, and 25% had no improvement. Patients with complete PPI resolution started PFPT sooner than patients who saw no improvement (40.9 vs 47.6 days, p=0.02). Patients with early return of continence (no pad usage at 3 months) had the shortest time between surgery and onset of PFPT (25.5days). We found no significant predictors of PFPT response on multivariable analysis. In our sub-cohort with pre-op mp-MRI (N=52), average membranous urethral length was 1.3cm (SD 0.3), 50.9% had presence of a median lobe, and type of prostatic apex was almost evenly distributed between B,C, & D (27.3%, 27.3%, & 29.5%, respectively). 

Conclusion:

Following RARP, patients with complete PPI resolution started PFPT sooner than patients with no response. There were no significant predictors of PFPT response on multivariable analysis, and no anatomic feature could be linked to response of PFPT in treating PPI, likely due to small sample size.  Further research and larger sample size is needed to better define patient characteristics and anatomic features that make PFPT more effective in the treatment of PPI.

Funding: n/a