Introduction: Posterior rhabdosphincter reconstruction (PRRS) following radical prostatectomy (RP) was designed to improve early urinary continence. The aim of our study was to determine anatomical features of the urethrovesical anastomosis using MRI to compare the differences depending on the reconstruction technique and continence outcomes.
Methods: We selected 40 patients from a randomized clinical trial (NCT03302169) with 158 men designed to compare early continence rates in patients undergoing urethrovesical anastomosis with or without PRRS. All patients had clinically localized prostate cancer and underwent MRI between 2-4m after RP. They were selected according to the reconstruction technique (PRRS vs non PRRS) and functional outcome (early vs late continence recovery) and 4 groups were made. Continence outcomes were assessed by time to have no leakage (day of dry pad), 24h PADtest 30d after RP and EPIC 26 questionnaires. Good early outcomes in continence recovery were defined as <=1pad/24h and PADtest <150 mL/24h 30d after RP. 2 independent blinded radiologists (1 and 2) assessed predetermined anatomical MRI benchmarks in order to determine the situation of the anastomosis in the pelvis: the vertical situation assessed as distance to the line coccyx-inferior pubic margin (ACPv) and antero-posterior situation as distance from the pubis to the line from coccyx-pubis through the anastomosis (AAP). Patients and data gatherers were blinded to treatment allocation.
Results: According to both radiologists, urethrovesical anastomosis was significantly superior in the pelvis situation (higher ACPv) in patients with PRRS when analyzing the group with good early outcome recovery, p=0.021 and p=0.01 reader 1 and 2, respectively. Reader 1 found that the anastomosis after PRRS was more anterior (lower AAP) in the same group, p=0.017; reader 2 didn’t find significant differences. Globally, PRRS group patients with better early recovery had a more anterior and superior position of the anastomosis (higher ACPv and lower AAP), according to both readers, p<0.05. The Pearson correlation between readers of ACPv and AAP were 0.975 (p<0.001) and 0.333 (p<0.046), respectively. No differences were found in the group using >1 pad/24h 30d after RP.
Conclusions: This is the first study that confirms positional differences according to the type of reconstruction in RP. It seems that the early continence improvement of PRRS is due to more superior and anterior situation of the urethrovesical anastomosis. Source of