(1496) Assessing Biomarkers for Predicting Severity in the Second Infectious Hit Following Sepsis

Introduction: Considerable knowledge exists regarding the correlation between various biomarkers and the severity of sepsis. However, alterations in innate and adaptive immunity following an initial infectious insult may modify the utility of these biomarkers during subsequent infectious events. Therefore, this study aims to assess which biomarker best predicts severity in a second infectious event.

Methods: A retrospective cohort study was conducted at two high-complexity hospitals, including ICU patients admitted for sepsis or septic shock between January 2016, and October 2019. Patients who developed a new infectious event between day seven and day thirty after the initial septic event were included. We first assessed the correlation between biomarkers (procalcitonin [PCT], neutrophils, neutrophil-to-lymphocyte ratio, lactate, PCT-to-albumin ratio, and lactate-to-albumin ratio) and sepsis severity (highest SOFA score achieved during each infectious hit). Correlation was assessed using Spearman or Pearson analysis according to distribution. Secondly, we evaluated the Area Under the ROC Curve (AUROC) of each biomarker in the second hit to predict in-hospital mortality. Finally, we selected the best cutoff point with the Youden index and compared mortality in patients with values above this cutoff point.

Results: A total of 49 patients were included. During the first septic event, the PCT-to-albumin ratio was the biomarker with higher correlation with severity (Spearman's Rho = 0.40, p< 0.01). In the second infectious episode, the lactate-to-albumin ratio emerged as the biomarker with highest correlation with severity (Rho = 0.52, p< 0.01). The AUROC of the lactate-to-albumin ratio to predict in-hospital mortality was 0.72 (95% CI 0.57, 0.87). Patients with a lactate-to-albumin ratio ≥1.5 experienced a fivefold increase in in-hospital mortality (57% [n=12] vs. 11% [n=3], p< 0.01) compared to those with a ratio < 1.5.

Conclusions: PCT-to-albumin ratio was the biomarker with higher correlation with severity in the first septic hit. However, during a second hit, possibly due to changes in the cytokine and immunological response to infection, the correlation of PCT with severity decreases, making the lactate-to-albumin ratio the best biomarker for predicting severity.