Does hospital setting post robotic fellowship training affect outcomes? A multi-institutional comparison of initial outcomes between academic and community settings.

The effect of practice setting on skill development post robotic fellowship training is currently unknown. We sought to compare learning curves between a high-volume academic center and a similar volume community hospital, in the setting of building a new robotic prostatectomy program. In addition, we sought to characterize benchmarks for learning curve development for post-fellowship training in robotic surgery. At two institutions, one academic (AC) and the other in the community (CO), the first 150 patients who underwent robotic laparoscopic prostatectomy over a period of 1 year were evaluated. We compared the following outcomes, operative time (OT), estimated blood loss (EBL), and positive surgical margin (PSM) rates, by two surgeons. Both surgeons completed the same surgical robotic fellowship in the same year. Cases were divided by tertile and primary outcomes measures were compared. Demographic data were similar between the two groups. Statistical differences were seen in age, preoperative Sexual Health Inventory for Men score, clinical and pathologic stage, and bladder neck reconstruction rate (p < 0.05). Overall, there was no significant difference in OT between AC (174 min) and CO (181 min) (p = 0.1099). Both EBL and PSM were lower in the AC (155 vs. 197 ml, p < 0.001 and 10 vs. 26 %, p < 0.05). The difference in OT was significant only in the first tertile of cases (AC 168 min vs. CO 193 min, p = 0.002). However, OT increased by 13 min in AC and decreased by 22 min in CO, when comparing the first and last tertile. EBL was different between AC (161ml) and CO (212 ml) only in the first tertile of cases (p = 0.002). Both AC and CO had increased EBL over the last tertile of cases (16.2 vs. 26.5 ml, respectively). These results demonstrate minor differences in outcomes between the two practice settings. Fellowship training in robotic surgery demonstrates a shorter learning curve towards achieving proficiency. Larger and longer term series will be required to assess functional outcomes and time to proficiency.