摘要
Abstract Background Prostate cancer (PCa) is a common malignancy in males and obesity may play a role in its development and progression. Associations between visceral obesity measured by a body shape index (ABSI) and PCa mortality have not been thoroughly investigated. This study assessed the associations between ABSI, body mass index (BMI), and long‐term PCa‐specific mortality using a nationally representative US database. Methods This population‐based longitudinal study collected data of males aged ≥40 years diagnosed with PCa and who underwent surgery and/or radiation from the National Health and Nutrition Examination Survey database 2001–2010. All included participants were followed through the end of 2019 using the National Center for Health Statistics Linked Mortality File. Associations between PCa‐specific mortality, BMI, and ABSI were determined using Cox proportional hazards regression and receiver operating characteristic (ROC) curve analysis. Results Data of 294 men (representing 1,393,857 US nationals) were analyzed. After adjusting for confounders, no significant associations were found between BMI (adjusted hazard ratio [aHR] = 1.06, 95% confidence interval [CI]: 0.97–1.16, p = 0.222), continuous ABSI (aHR = 1.29, 95% CI: 0.83–2.02, p = 0.253), or ABSI in category (Q4 vs. Q1–Q3: aHR = 1.52, 95% CI: 0.72–3.24, p = 0.265), and greater risk of PCa‐specific mortality. However, among participants who had been diagnosed within 4 years, the highest ABSI quartile but not in BMI was significantly associated with greater risk for PCa‐specific mortality (Q4 vs. Q1–Q3: aHR = 5.34, 95% CI: 2.26–12.62, p = 0.001). In ROC analysis for this subgroup, the area under the curve of ABSI alone for predicting PCa‐specific mortality was 0.638 (95% CI: 0.448–0.828), reaching 0.729 (95% CI: 0.490–0.968 when combined with other covariates. Conclusions In US males with PCa diagnosed within 4 years, high ABSI but not BMI is independently associated with increased PCa‐specific mortality.