The effects of adjusting for case mix on mortality and length of stay following radical cystectomy

Purpose: Prior studies evaluating quality of care following radical cystectomy have been constrained by the use of retrospective reviews of single institutional series and limited ability to examine risk factors in a comprehensive manner. Characterization of these factors could enhance preoperative patient counseling and facilitate perioperative management, thereby improving the quality of patient care. Materials and Methods: The National Surgical Quality Improvement Project is a prospective quality management initiative at 123 Veterans Affairs Medical Centers nationwide. The project collects preoperative clinical and intraoperative data, and outcomes on a wide variety of surgical procedures from multiple surgical disciplines. Since 1991, 2,538 radical cystectomies have been captured by the National Surgical Quality Improvement Project. Modeling using logistic regression was performed to identify preoperative risk factors associated with mortality and prolonged length of stay (greater than 90th percentile) after radical cystectomy. Results: The 30 and 90-day mortality rates following cystectomy were 2.9% and 6.8%, respectively, and median hospital stay was 11 days (90th percentile 30). Robust preoperative factors associated with mortality and prolonged length of stay that uniformly increased risk were older patient age (OR 1.2 to 1.4), American Society of Anesthesiologists class 3 or greater (OR 1.5 to 3.3), dependent functional status (OR 1.7 to 2.0) and low serum albumin (OR 2.1 to 12.0). Conclusions: A defined set of preoperative risk factors is independently associated with greater mortality and hospital stay following radical cystectomy. The breadth of these factors suggests that complex case mix adjustment is mandatory when comparing outcomes. Implementation of novel processes directed toward minimizing patient risk has the potential to improve outcomes following cystectomy.