| Popis: |
Abstract Background Although a high C-reactive protein-to-albumin ratio (CAR) is believed to increase mortality risk, the association between the physical activity (PA), CAR, and mortality among cancer survivors has not been investigated. This study aimed to examine this association among cancer survivors in the United States. Methods This cohort study used data from the National Health and Nutrition Examination Survey from 1999 to 2010. PA was self-reported using the Global Physical Activity Questionnaire, and C-reactive protein and albumin levels were obtained from laboratory data files. Mortality data were obtained by linkage of the cohort database to the National Death Index as of December 31, 2019. The analysis was conducted from November 1 to December 31, 2023. We used Cox proportional hazards multivariable regression to assess hazard ratios (HRs) and 95% confidence interval (CIs) for total and cancer-specific mortality risks attributable to PA and CAR. Results Among 2,232 cancer survivors, 325 (14.6%) reported no PA with a high CAR. During a follow-up of up to 20.75 years (median, 12.3 years; 27,453 person-years), 1,174 deaths occurred (cancer, 335; other, 839). A high CAR was observed to be consistently associated with the highest risks of total (HR, 1.59; 95% CI, 1.37–1.85) and cancer-specific (HR, 2.06; 95% CI, 1.55–2.73) mortality compared with a low CAR in a series of adjusted models. Multivariable models showed that PA was associated with a lower risk of all-cause (HR, 0.60; 95% CI, 0.52–0.69) and cancer-specific (HR, 0.64; 95% CI, 0.49–0.84) mortality compared with no PA. In the joint analyses, survivors with PA ≥ 600 metabolic equivalent min/wk and a low CAR were more likely to reduce the risk of total (HR, 0.41; 95% CI, 0.32–0.51) and cancer-specific (HR, 0.32; 95% CI, 0.20–0.50) mortality by 59% and 68% compared with those with no PA and a high CAR. Conclusion The pairing of adequate PA and a low CAR was significantly associated with reduced all-cause and cancer-related mortality risks. |
Full text from SpringerLink