Associations of C-reactive protein/albumin ratio with frailty and the risk of mortality: An observational study.

Patients with frailty still have a high risk of postoperative death even after undergoing medium and low stress surgeries. Early and effective identification of frailty helps improve the poor prognosis of those who need surgery. This study aims to explore the relationship between C-reactive protein/albumin ratio (CAR) and frailty and the risk of mortality. The clinical data of 14,743 participants in National Health and Nutrition Examination Survey were analyzed. The weighted logistic regression model was used to estimate the odds ratios of CAR and frailty. The weighted Cox regression model was used to estimate the hazard ratios of CAR and all-cause mortality, cardiovascular disease (CVD)-specific mortality, and cancer-specific mortality of frail participants. In addition, the nonlinearity of the above associations was evaluated and subgroup analysis was performed. The fully adjusted weighted logistic regression model showed a positive correlation between CAR and frailty [odds ratio (95% confidence interval [CI]): 1.23 (1.15-1.31), P < .0001]; restricted cubic spline regression indicated that this association was linear (nonlinear P = .059). Subgroup analysis suggested that the association between CAR and frailty was stronger in hypertension and CVD. In survival analysis, CAR significantly predicted all-cause mortality [hazard ratio [HR] (95% CI): 1.12 (1.05-1.20), P = .001], CVD-specific mortality [HR (95% CI): 1.18 (1.06-1.32), P = .003], and cancer-specific mortality [HR (95% CI): 1.12 (1.01-1.24), P = .03] in frail participants. CAR is independently and linearly positively correlated with frailty. In addition, an increase in CAR in frail participants also indicates a higher risk of death.