Login to Access Video or Poster Abstract: MP54-08
Sources of Funding: MedDrive Grant of the Medical Faculty, TU Dresden

Introduction

Radical cystectomy (RCE) shows the highest mortality and morbidity among urologic routine surgery. We analysed in-hospital outcomes of all RCE in Germany from 2006 to 2013 with a focus on the institutions’ annual caseload and surgical approach.

Methods

By using remote data processing we analysed the nationwide German hospital billing data from 2006 to 2013. All cases with a bladder cancer diagnosis combined with RCE were eligible for evaluation. We calculated mortality and transfusion rates during the hospital stay and the length of stay. The results were stratified for hospital characteristics, caseload, and the surgical approach.

Results

Total annual RCE numbers increased from 5,627 in 2006 to 7,399 in 2013. The share of open surgery declined from 99.3% to 96.6%, conventional laparoscopy increased from 0.7% to 1.6%, and the robot-assisted approach from 0% to 1.8%. The patients’ mean age was 68.2 ± 9.9 years. The average in-hospital mortality rate was 4.5% for open RCE; in comparison it was lower with 3.8% for laparoscopic (p=0.35) and 2.5% for robotic RCE (p=0.002). Hospitals with high annual caseloads >50 RCE showed lower mortality rates with 3.3% vs. 4.1% (26-50 RCE), 5.0% (11-25 RCE), 5.2% (4-10 RCE), and 7.0% (<3 RCE) (p<0.001). The need for blood transfusion during the hospital stay was higher for open with 60.0% vs. laparoscopic 50.6% vs. robotic RCE with 35.9% (p<0.0001). The mean length of stay was longer for open with 25.3 days and laparoscopic RCE with 26.0 days vs. robotic RCE with 21.4 days (p<0.0001). Hospitals with high annual caseload >50 RCE showed a shorter hospital stay with 23.3 days vs. 24.7 days (26-50 RCE), 26.1 days (11-25 RCE), 26.3 days (4-10 RCE), and 24.5 days (<3 RCE) (p<0.0001). _x000D_ Multivariate models indicated that the patient’s age and the type of urinary diversion were the most important factors for mortality and the need for blood transfusion (p<0.0001). Also on multivariate analysis hospitals with very low annual caseload (<3 RCE) had higher mortality (p=0.02), blood transfusion (p=0.0004), and a longer hospital stay (p<0.0001)._x000D_

Conclusions

Hospitals with high annual caseload volumes show an improved outcome with lower rates of in-hospital mortality, blood transfusion, and a shorter hospital stay. Compared to the open approach robotic RCE showed lower rates of in-hospital mortality, blood transfusion and a shorter hospital stay. Whether this result is due to selection bias warrants further examinations.

Funding

MedDrive Grant of the Medical Faculty, TU Dresden