Does training of fellows affect peri-operative outcomes of robot-assisted partial nephrectomy?

Abstract

Objective: To evaluate the impact of fellows' involvement on the peri-operative outcomes of robot-assisted partial nephrectomy (RAPN).

Materials and methods: We analysed 216 patients who underwent RAPN for a small renal tumour. We stratified our cohort into two groups according to the involvement of a fellow surgeon during the procedure: expert surgeon operating alone (expert group) and fellow operating under the supervision of the expert surgeon (fellow group). Peri-operative data were compared between the two groups. Linear and logistic regression analyses were performed to assess the impact of fellows' involvement on peri-operative and postoperative outcomes. Trifecta and margins ischaemia complications (MIC) score achievement rates were used to assess the quality of surgery in both the expert and fellow groups. Trifecta was defined as a combination of warm ischaemia time <25 min, negative surgical margins and no peri-operative complications. MIC score was defined as negative surgical margins, ischaemia time <20 min, and absence of complications grade ≥3.

Results: Fellows were involved in a total of 89 procedures (41%). Patients' characteristics were similar in the two groups. Operating time and warm ischaemia time (WIT) were longer in the fellow group (180 vs 120 min, P < 0.001, and 18 vs 14 min, P = 0.002, respectively). Length of hospital stay (LOS) was longer in the fellow group (5 vs 4.3 days; P = 0.05) and patients in this group had higher estimated blood loss (EBL; 400 vs 300 mL; P = 0.01), but this had no impact on transfusion rate (14% vs 11%; P = 0.43). Positive surgical margin rates were similar in the fellow and expert groups (2.2% vs 3.1%; P = 0.70). Major complications were more frequent in the fellow group (12.3% vs 6.3%), but the difference was not significant (P = 0.10). In multivariable analysis, fellow involvement was predictive of longer WIT (β = 0.22; P = 0.003) and operating time (β = 0.49; P < 0.001), but was not associated with EBL (β = 0.12, P = 0.09) or LOS (β = 0.12, P = 0.11). Finally, fellow involvement was associated with a lower rate of trifecta and MIC score accomplishment (odds ratio [OR] 0.53, P = 0.05 and OR 0.46, P = 0.01, respectively).

Conclusion: Training fellows to perform RAPN is associated with longer operating time and WIT but does not appear to compromise other peri-operative outcomes.

Keywords: fellow; kidney neoplasms; learning curve; partial nephrectomy; robotics; surgical training.