An easily applicable single condition-based mortality index for patients undergoing radical prostatectomy or radical cystectomy

Purpose: There is no generally accepted instrument to measure comorbidity in patients with cancer. We determined which single comorbid conditions are independently associated with competing mortality after radical prostatectomy or radical cystectomy in order to develop a mortality index. Methods: The study samples consisted of 2,961 consecutive patients who underwent radical prostatectomy between 1992 and 2007 for clinically localized prostate cancer and 932 consecutive patients who underwent radical cystectomy between 1993 and 2012 for high-risk non muscle-invasive or muscle-invasive urothelial or undifferentiated bladder cancer. Competing mortality was the study endpoint. Proportional hazard models for the subdistribution of competing risks were used for analysis. Results: Age, angina pectoris, peripheral vascular disease, cerebrovascular disease, chronic lung disease, diabetes mellitus, moderate or severe renal disease, current smoking, and American Society of Anesthesiologists (ASA) physical status class 3 to 4 were independent predictors of competing mortality after radical prostatectomy. After identifying radical cystectomy, age, angina pectoris, chronic lung disease, diabetes mellitus, current smoking, ASA class 3 to 4, and male sex as independent predictors of competing mortality, a combined mortality index using the conditions independently associated with competing mortality in both samples stratified the patients into risk groups with 0% 10-year competing mortality in the lowest and approximately 50% in the highest-risk classes. Conclusions: This simple and plausible combined mortality index based on age, ASA class, smoking status, and the presence of the conditions such as angina pectoris, chronic lung disease, and diabetes mellitus may be used to predict competing mortality in candidates for radical prostatectomy or radical cystectomy. (C) 2017 Elsevier Inc. All rights reserved.