IMPACT OF BODY MASS INDEX ON RADICAL CYSTECTOMY

Abstract

Radical cystectomy has a significant rate of morbidity and it is important to elucidate the factors that contribute to this risk. Obesity is a major problem in the United States and is associated with increased health hazard. Morbid obesity may even preclude definitive surgical treatment. This study examines the impact of body mass index (BMI) on radical cystectomy. Retrospective analysis was performed on 498 patients who underwent radical cystectomy primarily for bladder cancer from July 1, 1990 to May 10, 2002. Patient BMI was defined as normal (less than 25 in 151), overweight (25 to 29.9 in 198), obese (30 to 34.9 in 98) and morbidly obese (35 or greater in 51). The bivariate relationships among BMI categories and clinical parameters were assessed using the chi-square test, the analysis of variance and the log rank test. Multivariate analyses were performed using Cox regression models. Median followup for the cohort was 3.3 years. Mean BMI was 28 and 70 of the study group was above normal weight. Compared to normal BMI, upper weight BMI groups were younger (p <0.0001), and had increased estimated blood loss (p = 0.01) and operative time (p = 0.001). Complication number (p = 0.0004) was increased in these groups but complication severity was similar (p = 0.54). Morbidly obese patients underwent incontinent diversion more often (p = 0.03). In multivariate models increased BMI was independently associated with increased estimated blood loss (p = 0.004), prolonged operative time (p = 0.006) and increased complication rate (p = 0.01). Increased BMI independently poses a greater perioperative risk to the patient and contributes to the technical challenge of the cystectomy. This is most appreciated in the morbidly obese population and likely contributes to a greater use of incontinent diversion in this group. The increased perioperative risk associated with elevated BMI is significant but not prohibitive and should not preclude cystectomy as definitive treatment.