Biochemical, Hematological, Inflammatory, and Gut Permeability Biomarkers in Patients with Alcohol Withdrawal Syndrome with and without Delirium Tremens

Abstract

Background: Delirium Tremens (DT) is known to be a serious complication of alcohol withdrawal syndrome (AWS). Neurotransmitter abnormalities, inflammation, and increased permeability are associated with the pathogenesis of AWS and DT. However, the biomarkers of these conditions are still poorly understood. Methods: In this work, biochemical, hematologic, inflammatory, and gut permeability biomarkers were investigated in the following three groups: healthy controls (n = 75), severe AWS patients with DT (n = 28), and mild/moderate AWS without DT (n = 97). Blood sampling was performed after resolution of the acute condition (on 5 ± 1 day after admission) to collect clinical information from patients and to investigate associations with clinical scales. Biomarker analysis was performed using automated analyzers and ELISA. Inflammatory biomarkers included the erythrocyte sedimentation rate (ESR), high-sensitivity C-reactive protein (hsCRP), and platelet-to-lymphocyte ratio (PLR). Results: Among the biochemical biomarkers, only glucose, total cholesterol, and alanine aminotransferase (ALT) changed significantly in the analyzed groups. A multiple regression analysis showed that age and ALT were independent predictors of the CIWA-Ar score. Hematologic biomarker analysis showed an increased white blood cell count, and the elevated size and greater size variability of red blood cells and platelets (MCV, RDWc, and PDWc) in two groups of patients. Gut permeability biomarkers (FABP2, LBP, and zonulin) did not change, but were associated with comorbid pathologies (alcohol liver disease and pancreatitis). The increase in inflammatory biomarkers (ESR and PLR) was more evident in AWS patients with DT. Cluster analysis confirmed the existence of a subgroup of patients with evidence of high inflammation, and such a subgroup was more frequent in DT patients. Conclusions: These findings contribute to the understanding of biomarker variability in AWS patients with and without DT and support the heterogeneity of patients by the level of inflammation.

Keywords: Delirium Tremens; FABP2; LBP; alcohol use disorder; alcohol withdrawal syndrome; biochemical test; blood test; gut permeability; inflammation; zonulin.

Figure 1

Plasma FABP2, LBP, and zonulin concentrations in patients with alcohol withdrawal syndrome (AWS) with and without Delirium Tremens (DT) compared with healthy controls (HCs). (A) Stacked histogram indicated the percentage of FABP2-positive and negative samples in the analyzed groups. The chi-square test was used to assess the significance of differences (p-value). (B–D) Plasma concentrations of FABP2 (B), LBP (C), and zonulin (D) in the analyzed groups. No significant differences were found using the Kruskal–Wallis test.

Figure 2

Dependence of biomarker levels on comorbid conditions in AWS. (A) FABP2 level in patients with and without ALD. (B) LBP level in patients with and without ALD. (C) Zonulin levels in patients with and without pancreatitis. (D) PLR in patients with and without hypertension. The differences were calculated using the Mann–Whitney test.

Figure 3

Correlation analysis of biochemical, hematological, inflammatory, and gut permeability biomarkers with clinical parameters and scales in a general group of patients. (A) Correlation heatmap (* p < 0.05, ** p < 0.01). (B–G) Scatterplots for the most significant correlations: zonulin and LBP (B), RDWc and age of patients (C), PDWc and AUDIT score (D), hsCRP and ESS score (E), uric acid and CIWA-Ar score (F), and PLR and CIWA-Ar score (G). Rs—Spearman’s correlation coefficient. For abbreviations, see Table 1, Table 2, Table 3 and Table 4.

Figure 4

Stratification of patients with AWS according to the inflammation level using cluster analysis. Stacked histogram indicated the percentage of participants assigned to “low-inflammation” and “high-inflammation” clusters. Clustering was performed using the K-mean algorithm based on the following inflammatory biomarkers: WBC count, PLT count, ESR, and PLR. Differences were calculated using the chi-square test.