Surgical checklist adherence across urology expertise levels impacts transurethral resection of bladder tumour quality indicators.

Objectives To address the association of perioperative surgical checklist across variable surgical expertise with transurethral resection of bladder tumour (TURBT) accuracy and oncological outcomes in non-muscle-invasive bladder cancer. Patients and Methods We relied on our prospective collaborative database of patients treated with TURBT between 2012 and 2017. Surgical experience was stratified into three groups: resident vs young vs expert consultants. The association of surgical experience with detrusor muscle (DM) presence and adherence to the standardised peri-procedural nine-items TURBT checklist was evaluated with logistic regression models. A Cox regression model was used to investigate the association of surgical experience with recurrence-free survival (RFS). Results A total of 503 patients were available for analysis. TURBT was performed by expert consultants in 265 (52.7%) patients, by young consultants in 149 (29.6%) and by residents in 89 (17.7%). Residents were more likely to have DM in the TURBT specimen than expert consultants (odds ratio [OR] 1.75, 95% confidence interval [CI] 1.03-2.99, P = 0.04). Conversely, no differences in DM presence were seen between young vs expert consultants (OR 1.09, 95% CI 0.71-1.70, P = 0.69). The median checklist completion rate was higher for both residents and young consultants when compared to experts' counterparts (56% and 56% vs 44%, P = 0.009). When focusing on patients receiving a second-look TURBT, the persistent disease was associated with resident status (OR 4.24, 95% CI 1.14-17.70, P = 0.037) at initial TURBT. Surgical experience was not associated with 5-years RFS. Conclusion Surgeon's experience in the case of adequate perioperative surgical checklist implementation was inversely associated with the presence of DM in the specimen but directly linked to higher probability of persistent disease at re-TURBT, although no 5-year RFS differences were noted. [ABSTRACT FROM AUTHOR]