M. Schmid

436 Predictors of wound dehiscence in a prospective dataset of 2,586 cystectomies

INTRODUCTION AND OBJECTIVES: Wound dehiscence is major complication following radical cystectomy. It is a significant cause of readmission, reopearation and potentially delays lifesaving adjuvant therapies. We sought to investigate the incidence and predictors of wound dehiscence in patients undergoing radical cystectomy. METHODS: 2556 patient records with Current Procedural Terminology (CPT) codes for cystectomy between 2005 and 2012 were extracted from the American College of Surgeons National Quality Improvement Program (ACS-NSQIP). Stratification was made on the basis of the occurrence of postoperative wound dehiscence. This was defined as full disintegrity of the skin and fascial layer. Outcomes of interest included overall complication, mortality, prolonged length of stay and prolonged operative time, the latter two defined as measures above the 75th percentile. Descriptive and logistic regression models were performed to identify predictors of postoperative wound dehiscence. RESULTS: Of 2556 patients analyzed, 74 (2.9%) had a documented wound dehiscence. In multivariable analyses, smoking (OR 2.2 p1⁄4.002), prolonged operative time (OR 1.6;p1⁄40.05) and BMI were associated with increased odds of postoperative wound dehiscences. Female gender was associated with decreased odds of dehiscence (OR 0.4; p1⁄40.022). Elevated preoperative creatinine (>1.2 mg/dl), chronic steroid use and diabetes were not independent predictors. CONCLUSIONS: Our study is the first to identify predictors of wound dehiscence following radical cystectomy in a large multi-institutional prospective cohort. Identifying patients at risk for postoperative wound complications may guide the use preventative measures at the time of surgery. Source of Funding: none

641 Racial disparities in the surgical care of localized prostate cancer

INTRODUCTION AND OBJECTIVES: The Cancer of the Prostate Risk Assessment Postsurgical (CAPRA-S) score predicts prostate cancer (PCa) recurrence based on pathologic information from radical prostatectomy (RP). We independently validated the CAPRA-S in a European, single-institution database. METHODS: The study cohort comprised of 14532 patients treated with radical prostatectomy between January 1992 and August 2012. Prediction of biochemical recurrence (BCR), metastasis and cancer-specific mortality (CSM) by CAPRA-S was assessed by Kaplan-Meier analysis and the c-index. Performance of CAPRA-S in predicting BCR was assessed by calibration plot and decision curve analysis. RESULTS: Median follow up was 48.8 months (IQR 24.6 e 95.9). Biochemical recurrence occurred in 20.3% of men at a median time of 21.2 mo (interquartile range [IQR]: 7.7e44.9). When stratifying patients by CAPRA-S risk groups, biochemical recurrence-free survival estimates at 5 years were 91.4%, 70.4%, and 29.3% for the low, intermediate and high-risk group, respectively. The c-index for CAPRA-S in predicting BCR was 0.80. The cindices for CAPRA-S in predicting metastasis and CSM were 0.85 and 0.88, respectively. 374 men developed metastasis, and 184 men died from PCa. CONCLUSIONS: The postoperative CAPRA-S score was accurate when applied in a European study cohort and predicted biochemical recurrence, metastasis and CSM after RP with c-indices> 0.80. The score can be valuable for decision-making for adjuvant therapy.