Intestinal T lymphocyte homing is associated with gastric emptying and epithelial barrier function in critically ill: a prospective observational study

Abstract

Background: Impaired gastric emptying is common in critically ill patients. Intestinal dysmotility, a major cause of feed intolerance, may foster infectious complications due to mucosal barrier disruption. However, little is known about gut-directed immune activation, intestinal barrier function and its association with impaired gastric emptying in critically ill patients at ICU admission.

Methods: We conducted a prospective observational study at two tertiary care medical ICUs. Fifty consecutive patients needing invasive mechanical ventilation were recruited within 24 h of ICU admission, prior to any nutritional support. The acute physiology and chronic health evaluation (APACHE) II score, the sequential organ failure assessment (SOFA) score and the multiple organ dysfunction score (MODS) were used to assess illness severity and multiple organ dysfunction. Gastric emptying was assessed by paracetamol absorption test. Peripheral blood mononuclear cells were freshly isolated and cultured for 24 h, and TNF-α, IL-1β and IL-10 measured in cell culture supernatants and in serum by ELISA. The intestinal epithelial barrier was assessed, quantifying serum concentrations of intestinal fatty acid binding protein (I-FABP), ileal bile-acid binding protein (I-BABP) and zonulin-1 by ELISA. Small bowel homing T lymphocytes (CD4+ α4β7 + CCR9+) were analyzed by flow cytometry. The Mann-Whitney test and Spearman correlation were used in statistical evaluation.

Results: CD4 + α4β7 + CCR9+ T lymphocytes were inversely correlated with gastric emptying. Patients with delayed gastric emptying at ICU admission (n = 35) had significantly higher serum and PBMC-induced TNF-α and IL-1β and increased intestinal barrier disruption reflected by higher I-FABP, I-BABP and zonulin-1. Patients who died in the ICU had significantly impaired gastric empting at admission compared to ICU survivors. No differences were observed in APACHE II, SOFA or MODS in patients with delayed gastric emptying compared to patients with normal gastric emptying.

Conclusions: Exaggerated CD4 + α4β7 + CCR9+ T lymphocyte homing with increased pro-inflammatory cytokine release and intestinal epithelial barrier disruption are associated with delayed gastric emptying. This is not simply due to differences in overall severity of illness at ICU admission and may represent a pathophysiological mechanism of gut-directed immune activation leading to impaired barrier function in the critically ill.

Keywords: Critically ill; Gastric emptying; Gut homing; Intensive care; Mucosal barrier; T lymphocytes.

Figures

Fig. 1

Paracetamol absorption in ICU survivors compared to non-survivors. Patients who died in the ICU had significantly lower paracetamol absorption (p = 0.004). AUC area under the curve

Fig. 2

Severity of illness reflected by the acute physiology and chronic health evaluation (APACHE) II score, severity of multiple organ dysfunction assessed by the sequential organ failure assessment (SOFA) or the multiple organ dysfunction score (MODS) in patients with delayed (area under the curve (AUC) <600 min*mg/l) and normal gastric emptying (AUC ≥600 min*mg/l)

Fig. 3

Percentage of CD4 + α4β7 + CCR9+ T lymphocytes was inversely correlated (r = -0.5; p = 0.001) with gastric emptying as reflected by paracetamol absorption (area under the curve (AUC) min*mg/l)

Fig. 4

TNF-α and IL-1β (pg/ml) in peripheral blood mononuclear cell culture supernatants and serum of patients with delayed gastric emptying (area under the curve (AUC) <600 min*mg/l) and normal gastric emptying (AUC ≥600 min*mg/l). Error bars range, bold line median, box 5th–95th centiles)

Fig. 5

Serum concentration of zonulin-1 (ZO-1), intestinal fatty acid binding protein (I-FABP) and ileal bile acid binding protein (I-BABP) in patients with delayed gastric emptying (area under the curve (AUC) <600 min*mg/l) and normal gastric emptying (AUC ≥600 min*mg/l). Error bars, bold line and box respectively represent range, median and 5th-95th centiles Error bars range, bold line median, box 5th–95th centiles)