C-Reactive Protein-to-Albumin Ratio as a Prognostic Marker in ICU Patients with Pre-Existing Hypertension and Diabetes

Background/Objective: The C-reactive protein-to-albumin ratio (CAR) reflects both systemic inflammation and nutritional status and has been proposed as a prognostic marker in critical illness, yet its value in intensive care unit (ICU) patients with pre-existing hypertension is not well defined. Methods: This was a retrospective single-center study of 341 critically ill adults with pre-existing hypertension (June 2001–October 2012) from the Medical Information Mart for Intensive Care III database. CAR was calculated from the first C-reactive protein (CRP) and albumin measurements after ICU admission. Using adjusted Cox proportional hazard models, we examined the association of CAR with 30-day mortality in the overall hypertensive cohort, across hypertension groups, and in patients with coexisting diabetes. Results: Non-survivors had higher CAR than survivors (35.5 vs. 18.1, p = 0.008). CAR showed moderate discriminative ability in the overall hypertensive cohort (AUC = 0.637, 95% CI: 0.543–0.732), with better discrimination in patients with normal/elevated blood pressure (BP) (AUC = 0.748, 95% CI: 0.637–0.858) and a relatively higher AUC in the subgroup with diabetes and normal/elevated BP (0.833, 95% CI: 0.671–0.995). In univariable Cox analysis, high CAR was associated with increased 30-day mortality in the overall hypertensive cohort (HR: 3.02, 95% CI: 1.48–6.17, p = 0.0024), in patients with normal/elevated BP (HR: 8.90, 95% CI: 2.00–39.17, p = 0.0038), and in patients with diabetes and normal/elevated BP (HR: 10.00, 95% CI: 1.20–83.10, p = 0.0331). These associations remained significant after multivariable adjustment in the overall hypertensive cohort (adjusted HR: 3.01, 95% CI: 1.45–6.21, p = 0.0030), in patients with normal/elevated BP (adjusted HR: 10.12, 95% CI: 2.20–46.59, p = 0.0030), and in patients with diabetes and normal/elevated BP (adjusted: HR: 19.41, 95% CI: 1.37–275.28, p = 0.0284). Conclusions: These results suggest that CAR measured early after ICU admission may serve as a practical tool for mortality risk stratification in ICU patients with pre-existing hypertension, particularly those with diabetes.