Impact of hospital and surgeon volume on mortality and complications after prostatectomy.

Abstract

It remains controversial whether short-term surgical complications after radical prostatectomy can be decreased by increasing surgeon or hospital procedural volume. We determined whether hospital or surgeon volumes impacted various short-term surgical complications. We examined in-hospital mortality and complications following radical prostatectomy in all 25,404 men who underwent this surgery across 8 provinces in Canada between 1990 and 2001. Bayesian multilevel logistic regression models were used, adjusting for patient age, comorbidity, surgery year, and hospital and surgeon volume, while accounting for clustering by surgeon and hospital. Overall 50 men (0.2%) died and 5,087 (20.0%) had 1 or more in-hospital complications following surgery. In models adjusted for age, comorbidity and surgery year hospital volume was associated with in-hospital mortality (p = 0.037). In adjusted models doubling hospital volume was associated with a decreased risk of any, cardiac, respiratory, vascular, genitourinary, miscellaneous medical and miscellaneous surgical complications (each p <0.001), although not wound/bleeding complications (p = 0.40). Similarly doubling surgical volume was associated with a decreased risk of any, respiratory, wound/bleeding, genitourinary, miscellaneous medical and miscellaneous surgical complications (each p <0.01), although not cardiac and vascular complications (p = 0.58 and 0.17, respectively). Adjustment for clustering led to nonsignificant effects of hospital volume on miscellaneous surgical complications, and of surgeon volume on miscellaneous medical and miscellaneous surgical complications. However, this did not alter other findings. Increasing hospital and surgeon volume are associated with a decreased risk of most complications after radical prostatectomy even after adjusting for the effects of clustering.