Biomarkers Facilitate the Assessment of Prognosis in Critically Ill Patients with Primary Brain Injury: A Cohort Study

Abstract

Primary injuries to the brain are common causes of hospitalization of patients in intensive care units (ICU). The Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system is widely used for prognostication among critically ill subjects. Biomarkers help to monitor the severity of neurological status. This study aimed to identify the best biomarker, along with APACHE II score, in mortality prediction among patients admitted to the ICU with the primary brain injury. This cohort study covered 58 patients. APACHE II scores were assessed 24 h post ICU admission. The concentrations of six biomarkers were determined, including the C-reactive protein (CRP), the S100 calcium-binding protein B (S100B), neuron-specific enolase (NSE), neutrophil gelatinase-associated lipocalin (NGAL), matrix metalloproteinase 9 (MMP-9), and tissue inhibitor of metalloproteinase 1 (TIMP-1), using commercially available ELISA kits. The biomarkers were specifically chosen for this study due to their established connection to the pathophysiology of brain injury. In-hospital mortality was the outcome. Median APACHE II was 18 (IQR 13–22). Mortality reached 40%. Median concentrations of the CRP, NGAL, S100B, and NSE were significantly higher in deceased patients. S100B (AUC = 0.854), NGAL (AUC = 0.833), NSE (AUC = 0.777), and APACHE II (AUC = 0.766) were the best independent predictors of mortality. Combination of APACHE II with S100B, NSE, NGAL, and CRP increased the diagnostic accuracy of mortality prediction. MMP and TIMP-1 were impractical in prognostication, even after adjustment for APACHE II score. S100B protein and NSE seem to be the best predictors of compromised outcome among critically ill patients with primary brain injuries and should be assessed along with the APACHE II calculation after ICU admission.