Assessment of intestinal barrier integrity and associations with innate immune activation and metabolic syndrome in acutely ill, antipsychotic-free schizophrenia patients

TL;DR

This study examines gut barrier function in schizophrenia patients without antipsychotics, finding that gut markers differ from controls, with some changes linked to smoking and others suggesting unique gut damage.

Contribution

The study identifies distinct gut-related processes in antipsychotic-free schizophrenia patients, highlighting the importance of accounting for smoking in interpreting immune and gut barrier markers.

Findings

LBP levels were higher in schizophrenia patients but linked to smoking rather than schizophrenia itself.

I-FABP levels were lower in schizophrenia patients and not influenced by smoking, suggesting gut epithelial damage.

LBP and I-FABP showed distinct immune and metabolic associations, indicating separate gut dysfunction mechanisms.

Abstract

Schizophrenia (Sz), once seen solely as a brain disorder, is now recognised as a systemic illness involving immune and metabolic dysregulation. The intestinal barrier has emerged as a key player in gut–brain–immune interactions. However, studies in early, antipsychotic free stages remain scarce and often neglect confounding factors such as smoking and metabolic syndrome.

We measured two complementary markers: lipopolysaccharide-binding protein (LBP), reflecting endotoxin exposure and systemic immune activation, and intestinal fatty acid-binding protein (I-FABP), indicating gut epithelial damage and permeability changes, in blood from 96 acutely ill, antipsychotic-free Sz patients (61 first-episode, 35 relapsed) and 96 matched controls. Associations with innate immunity, metabolic parameters, smoking, and clinical features were assessed using nonparametric statistics and random forest regression. Group differences were tested using covariate adjustment, as well as in a separate analysis of non-smokers (Sz: n = 42; controls: n = 84).

Median LBP was higher in Sz (21.96 µg/mL) vs. controls (18.10 µg/mL; FDR-adjusted p = 0.021, δ = 0.209) but became non-significant after adjusting for smoking (FDR-adjusted p = 0.199). In contrast, I-FABP was lower in Sz (218.2 pg/mL) than controls (315.0 pg/mL; FDR-adjusted p = 0.021, δ = –0.195) and remained robust across smoking-adjusted analyses. No differences were found between first-episode and relapsed patients for either marker.

LBP correlated strongly with CRP (r = 0.557, p < 0.001) and neutrophils (r = 0.468, p < 0.001) and was moderately predicted by immune models (pseudo-R2 = 0.354 overall; 0.273 Sz; 0.449 controls). Links to waist circumference and blood pressure were weaker (pseudo-R2: 0.048–0.104). I-FABP showed fewer immune associations and was not correlated with LBP (r = –0.017, FDR-adjusted p = 0.819), suggesting distinct mechanisms.

Our findings suggest separable gut‑related processes in antipsychotic-free Sz. The apparent LBP elevation was not schizophrenia‑specific; its strong correlations with CRP and neutrophils point to smoking related inflammation rather than a schizophrenia specific effect. Accordingly, prior findings of LBP elevations in Sz likely reflect unaccounted smoking. In contrast, reduced I-FABP, independent of smoking, may indicate epithelial injury. The absent correlation between LBP and I-FABP highlights distinct pathophysiological dimensions of gut dysfunction. Longitudinal studies, ideally spanning prodromal phases and integrating microbiome, dietary, smoking, and permeability assessments, are needed to clarify temporal dynamics and guide stratified treatments.

The online version contains supplementary material available at 10.1186/s12974-025-03584-3.

Figures