Supplementary Material for: Dose-response relationship between C-reactive protein/albumin ratio and in-hospital mortality in elderly patients with acute ischemic stroke

Introduction: The C-reactive protein/albumin ratio is a reliable indicator of outcome risk in several diseases. This study aims to evaluate prognostic power of the C-reactive protein/albumin ratio for in-hospital mortality and the dose-response relationship between the two in the oldest-old patients with acute ischemic stroke. Methods: A longitudinal observational study was conducted including patients with acute ischemic stroke (aged ≥ 80 years) from two tertiary hospitals between January 1, 2014, and January 31, 2020. Based on the tertiles of the C-reactive protein/albumin ratio, the patients were divided into three groups. Restrictive cubic spline and robust locally weighted regression analysis were performed on continuous variables to examine the dose–response relationship between the C-reactive protein/albumin ratio and in-hospital mortality risk. All-cause mortality during hospitalization was the outcome for this study. Results: The study included 584 patients (mean age = 84.6 ± 3.1 years; 59.6% men). The C-reactive protein/albumin ratio was divided into three groups, namely, T1 of < 0.73, T2 of 0.73 – 2.03, and T3: > 2.03. After adjusting for demographic and clinical characteristics, a higher C-reactive protein/albumin ratio was independently associated with in-hospital mortality. The hazard ratio for this association was 2.01 (95% confidence interval: 1.12 – 3.60, P = 0.019). A dose–response relationship between the C-reactive protein/albumin ratio and in-hospital mortality risk was observed. Sensitivity analysis found no attenuation in the hazard ratio in uninfected individuals, whereas no difference in the hazard ratio was noted in individuals with infections. Conclusions: When predicting in–hospital mortality in the oldest-old patients with ischemic stroke, the C-reactive protein/albumin ratio might be a helpful and convenient metric.