An Evaluation of Peer-Rated Surgical Skill and its Relationship With Detrusor Muscle Sampling in Transurethral Resection of Bladder Tumor

Abstract

ObjectiveTo assess the reliability of peer-review of TURBT videos as a means to evaluate surgeon skill and its relationship to detrusor sampling. MethodsUrologists from an academic health system submitted TURBT videos in 2019. Ten blinded peers evaluated each surgeon's performance using a 10-item scoring instrument to quantify surgeon skill. Normalized composite skill scores for each surgeon were calculated using peer ratings. For surgeons submitting videos, we retrospectively reviewed all TURBT pathology results (2018-2019) to assess surgeon-specific detrusor sampling. A hierarchical logistic regression model was fit to evaluate the association between skill and detrusor sampling, adjusting for patient and surgeon factors. ResultsSurgeon skill scores and detrusor sampling rates were determined for 13 surgeons performing 245 TURBTs. Skill scores varied from -6.0 to 5.1 [mean: 0; standard deviation (SD): 2.40]. Muscle was sampled in 72% of cases, varying considerably across surgeons (mean: 64.5%; SD: 30.7%). Among 8 surgeons performing >5 TURBTs during the study period, adjusted detrusor sampling rate was associated with sending separate deep specimens (odds ratio [OR]: 1.97; 95% confidence interval [CI]: 1.02-3.81, P = .045) but not skill (OR: 0.81; 95% CI: 0.57-1.17, P = .191). ConclusionSurgeon skill was not associated with detrusor sampling, suggesting there may be other drivers of variability of detrusor sampling in TURBT.