Mortality after urologic cancer surgery: Impact of non-index case volume

OBJECTIVES To quantify the degree to which overall urologic oncology volume either reduces or enhances the effect of single procedure volume on short-term outcomes after urologic oncology surgery. METHODS Urologic oncology procedures for prostate, kidney, and bladder cancer performed between 1988 and 2003 were identified in the Nationwide Inpatient Sample. Procedure-specific volume and urologic oncology volume (excluding the procedure of interest) were determined for each cancer and each hospital. Multivariable logistic regression models were constructed to measure the independent effect of urologic oncology volume (non-index procedures) on operative mortality after prostatectomy, cystectomy, and nephrectomy (index procedures) after adjusting for patient and hospital factors. RESULTS Unadjusted operative mortality for prostatectomy, cystectomy, and nephrectomy was 0.2%, 2.8%, and 1.4%, respectively. For prostatectomy and cystectomy, the magnitude of the volume-mortality association was reduced after adjusting for non-index urologic oncology case volume. For example, the relationship between surgical volume and mortality was reduced by 20% for radical prostatectomy and 60% for radical cystectomy. CONCLUSIONS The volume-outcome effect for index urologic oncology procedures is modified by experience with other non-index specialty-related procedures. Efforts to identify transferable, effective processes of care should focus on a subset of high-volume centers.