90-DAY MORTALITY AFTER RADICAL CYSTECTOMY FOR BLADDER CANCER

Introduction/Aim: Radical cystectomy represents the gold standard treatment in muscle-invasive bladder cancer and is one of the most challenging procedures in urologic surgery. This procedure is associated with significant perioperative morbidity and mortality, ranging from 20 to 64% and from 0.3 to 5.7%, respectively. The considerable inter-individual variability of perioperative mortality has led to the development of several models of individual perioperative mortality prediction for patients undergoing radical cystectomy. The aim of our study was to evaluate the predictive accuracy of the nomograms of Isbarn and Aziz and the identification of perioperative mortality risk factors in a series of patients undergoing radical cystectomy for muscle invasive bladder cancer at our institution. Patients and Methods: We retrospectively reviewed data regarding 145 consecutive patients who underwent radical cystectomy and urinary diversion for urothelial bladder cancer at our Institute between 2002 and 2012. The following pre-operative variables, such as age at intervention, gender, body mass index (BMI), operative volume, Charlson comorbidity index, presence of carcinoma in situ (CIS) to endoscopic resection bladder (TURV), American society of anesthesiologists (ASA) score, clinical stage according to the TNM and 90-day mortality, were collected and analyzed. The Isbarn and Aziz nomograms were, moreover, applied to our cohort. Results: Median age at radical cystectomy was 68 years and 85% of patients were male, with a median BMI of 26 (IQR=25-27). The most represented ASA score was 2, whereas the most frequent Charlson score (62.76%) was 0. Median in-hospital stay was 15 days, with a range between 7 and 35 days. Median follow-up was 26 months (IQR=11-45); five deaths were registered within 90 days (3.4%). Applying the nomograms of Aziz and Isbarn to our patients, we obtained an average mortality risk <10% and of 2.4%, respectively. At multivariate analysis, no variable was independently related to perioperative mortality risk. Evaluating the receiver operating characteristic (ROC) curves, the Aziz nomogram showed the highest predictive accuracy, while ASA score was found to be the single variable with the highest accuracy in predicting 90 days mortality. Results: In our series, at the multivariate analysis, none of the variables resulted as an independent risk factor for 90-day mortality; however, only ASA score seemed to have a trend in this sense. This retrospective study has a small number of participants with few events, thus making the multivariate analysis unreliable. Conclusion: In our series, 90- day mortality after radical cystectomy was 3.4% (5/145 patients). On univariate analysis, only Charlson comorbidity index (ref. 0-2; p= 0.019; 0.013), ASA score ( p= 0.004) and the adjusted ICC age (0.022) were indipendent risk factors of 90- day perioperative mortality, whereas at multivariate analysis, no variable was independently related to mortality risk. The Aziz nomogram presents the highest accuracy in predicting a 90-day mortality of patients undergoing radical cystectomy.