Serum albumin demonstrates comparable or superior prognostic value compared to albumin-based ratios in sepsis.

Abstract

Background: Serum albumin is a well-known biomarker in sepsis, and several albumin-based ratios have been proposed to enhance its prognostic performance. However, it remains unclear whether these composite indices outperform serum albumin alone in predicting outcomes in septic patients.

Methods: We conducted a prospective observational study including 413 adult patients with sepsis admitted to the Intermediate Care Unit of Santorso Hospital, Italy, between January 2023 and June 2024. Clinical data, laboratory parameters, and six albumin-based ratios were collected at admission. The primary outcome was 30-day mortality. Prognostic performance was assessed using ROC curves, Brier Scores, and Decision Curve Analysis (DCA). Correlation with clinical severity was evaluated using Spearman's correlation with SOFA and APACHE II scores.

Results: Among the 413 patients enrolled, 30-day mortality was 16.9%. Serum albumin demonstrated the highest predictive value compared to all albumin-based ratios. DCA confirmed the superior net clinical benefit of serum albumin across a wide range of threshold probabilities, especially in early risk stratification. LAR showed comparable performance in AUC but lower clinical utility in DCA. Albumin also showed the strongest inverse correlation with SOFA (ρ = -0.276, p < 0.001) and APACHE II (ρ = -0.391, p < 0.001) scores.

Conclusions: Serum albumin is a strong, independent predictor of 30-day mortality in patients with sepsis and outperforms multiple albumin-based ratios in both statistical and clinical utility. Its use in early risk stratification should be encouraged in clinical practice. Further multicenter studies are warranted to confirm these findings and explore their generalizability.