PD222 - C-REACTIVE PROTEIN TO ALBUMIN RATIO, NUTRITIONAL RISK, AND OUTCOMES IN ICU PATIENTS WITH PULMONARY SEPSIS

PD222

C-REACTIVE PROTEIN TO ALBUMIN RATIO, NUTRITIONAL RISK, AND OUTCOMES IN ICU PATIENTS WITH PULMONARY SEPSIS

A. Flevari¹, P. Vasileiou¹, E. Ntikoudi^1,*, O. Apostolopoulou¹, M. Karagianni¹, V. Koutsoukou¹, E. Bourgani¹, C. Merkouri¹, G. Georgakopoulou¹, M. Bourikou¹, M. Ntaganou¹

¹Polyvalent ICU Department, Sotiria General Hospital Thoracic Diseases, Athens, Greece

Rationale: The C-reactive protein (CRP) to albumin ratio (CAR) is a relatively novel biomarker that combines a positive (CRP) and a negative (albumin) acute-phase reactant, reflecting systemic inflammation and metabolic stress. We aimed to evaluate the prognostic performance of the CAR ratio in ICU patients with pulmonary sepsis and compare it with established nutritional risk assessment score modified mNUTRIC.

Methods: This retrospective observational study included ICU patients with pulmonary sepsis and acute respiratory failure (ARF). On ICU admission, demographic data, CRP (mg/L), albumin (g/dL), CAR, APACHE II, SOFA and mNUTRIC scores were recorded. Outcomes were ICU length of stay (LOS) and mortlity at Day 28 and ICU discharge. Group comparisons used the Mann–Whitney U test. Discriminatory performance of CAR and mNUTRIC was assessed using ROC curves and AUCs with comparison by DeLong’s test. A p-value < .05 was considered significant.

Results: Of 202 ICU admissions in 2025, 74 patients (41.9% female) of median age 71.5 years (IQR 58-78) had pulmonary sepsis with ARF. Median APACHE_II was 21(IQR 18-27), mean SOFA score 8.35±3 and median mNUTRIC 6 (IQR 5-7). Day-28 mortality was 24.4%, and ICU mortality 36.5%. Median CAR was significantly higher in non-survivors than survivors [51.3 (IQR 24.8–74.6) vs 12.4 (IQR 6.5–36.8), p = 0.0004], while age and mNUTRIC scores were similar between groups. CAR demonstrated significant discrimination for Day-28 mortality (AUC 0.70, 95% CI 0.54–0.85, p=0.013), whereas mNUTRIC did not (AUC 0.64, p = 0.106). AUCs did not differ significantly (DeLong p = 0.62). The optimal CAR cut-off was 18, with 82% sensitivity and 47% specificity.

Image:

Conclusion: CAR was higher in non-survivors and showed acceptable discrimination for Day-28 mortality (AUC 0.70), suggesting inflammatory burden rather than nutritional status. Larger studies are needed to confirm utility.

Disclosure of Interest: None declared