The C-reactive protein (CRP)/albumin ratio has recently emerged as a marker for poor prognosis or mortality across various patient groups. This study aimed to identify the association between CRP/albumin ratio and 28-day mortality and predict the accuracy of CRP/albumin ratio for 28-day mortality in medical intensive care unit (ICU) patients. This was a retrospective cohort study of 875 patients. We evaluated the prognostic value of CRP/albumin ratio to predict mortality at 28 days after ICU admission, using Cox proportional hazard model and Kaplan-Meier survival analysis. The 28-day mortality was 28.0%. In the univariate analysis, the Acute Physiology and Chronic Health Evaluation II (APACHE II) score (p < 0.001), CRP level (p = 0.045), albumin level (p < 0.001), and CRP/albumin ratio (p = 0.032) were related to 28-day mortality. The area under the receiver operating characteristic (ROC) curve (the area under the ROC curves (AUC)) of CRP/albumin ratio was higher than that of CRP for mortality (0.594 vs. 0.567, p < 0.001). The cut-off point for CRP/albumin ratio for mortality was 34.3. On Cox proportional-hazard regression analysis, APACHE II score (hazards ratio (HR) = 1.05, 95% confidence interval (CI) = 1.04⁻1.07, p < 0.001) and CRP/albumin ratio (HR = 1.68, 95% CI = 1.27⁻2.21, p < 0.001 for high CRP/albumin ratio) were independent predictors of 28-day mortality. Higher CRP/albumin ratio was associated with increased mortality in critically ill patients.
Keywords: C-reactive protein; albumin; intensive care unit; mortality.