Martin De Smet

Intestinal barrier dysfunction in a randomized trial of a specific probiotic composition in acute pancreatitis

Objectives:To determine the relation between intestinal barrier dysfunction, bacterial translocation, and clinical outcome in patients with predicted severe acute pancreatitis and the influence of probiotics on these processes. Summary of Background data:Randomized, placebo-controlled, multicenter trial on probiotic prophylaxis (Ecologic 641) in patients with predicted severe acute pancreatitis (PROPATRIA). Methods:Excretion of intestinal fatty acid binding protein (IFABP, a parameter for enterocyte damage), recovery of polyethylene glycols (PEGs, a parameter for intestinal permeability), and excretion of nitric oxide (NOx, a parameter for bacterial translocation) were assessed in urine of 141 patients collected 24 to 48 h after start of probiotic or placebo treatment and 7 days thereafter. Results:IFABP concentrations in the first 72 hours were higher in patients who developed bacteremia (P = 0.03), infected necrosis (P = 0.01), and organ failure (P = 0.008). PEG recovery was higher in patients who developed bacteremia (PEG 4000, P = 0.001), organ failure (PEG 4000, P < 0.0001), or died (PEG 4000, P = 0.009). Probiotic prophylaxis was associated with an increase in IFABP (median 362 vs. 199 pg/mL; P = 0.02), most evidently in patients with organ failure (P = 0.001), and did not influence intestinal permeability. Overall, probiotics decreased NOx (P = 0.05) but, in patients with organ failure, increased NOx (P = 0.001). Conclusions:Bacteremia, infected necrosis, organ failure, and mortality were all associated with intestinal barrier dysfunction early in the course of acute pancreatitis. Overall, prophylaxis with this specific combination of probiotic strains reduced bacterial translocation, but was associated with increased bacterial translocation and enterocyte damage in patients with organ failure.

Systemic inflammation increases intestinal permeability during experimental human endotoxemia.

Although the gut is often considered the motor of sepsis, the relation between systemic inflammation and intestinal permeability in humans is not clear. We analyzed intestinal permeability during experimental endotoxemia in humans. Before and during experimental endotoxemia (Escherichia coli LPS, 2 ng/kg), using polyethylene glycol (PEG) as a permeability marker, intestinal permeability was analyzed in 14 healthy subjects. Enterocyte damage was determined by intestinal fatty acid binding protein. Endotoxemia induced an inflammatory response. Urinary PEGs 1,500 and 4,000 recovery increased from 38.8 ± 6.3 to 63.1 ± 12.5 and from 0.58 ± 0.31 to 3.11 ± 0.93 mg, respectively (P < 0.05). Intestinal fatty acid binding protein excretion was not affected by endotoxemia. The peak serum IL-10 concentrations correlated with the increase in PEG 1,500 recovery (r = 0.48, P = 0.027). Systemic inflammation results in an increased intestinal permeability. The increase in intestinal permeability is most likely caused by inflammation-induced paracellular permeability, rather than ischemia-mediated enterocyte damage.

Upper gastrointestinal responses to intraduodenal nutrient in type 1 diabetes mellitus.

Background Abnormal nutrient-related small-intestinal feedback may contribute to disordered gastric motility and upper gastrointestinal symptoms in patients with diabetes. Aim To evaluate the motor, sensory and incretin responses to intraduodenal nutrients in patients with type 1 diabetes and in controls. Methods Eight type 1 diabetes patients (two with autonomic neuropathy) and nine controls were studied during euglycaemia. A manometric catheter was positioned across the pylorus, and nutrient was infused intraduodenally (90 kcal over 30 min). Blood glucose and plasma glucagon-like peptide 1 and gastric inhibitory polypeptide were measured, and sensations were assessed with visual analogue questionnaires. Results During nutrient infusion, neither the number of antral waves nor the stimulation of phasic or basal pyloric pressures differed between patients and controls. Upper gut sensations and areas under the plasma incretin peptide curves did not differ between the groups. Conclusions During euglycaemia, the upper gastrointestinal motor, sensory and incretin peptide responses to small-intestinal nutrient are comparable in patients with relatively uncomplicated type 1 diabetes and in healthy subjects.

Lipid solubilization in human gallbladder versus hepatic biles

BACKGROUND/AIMS Cholesterol crystallizes more rapidly in gallbladder than in hepatic biles, supposedly due to formation of cholesterol-supersaturated vesicles in concentrated gallbladder biles because of preferential micellization of phospholipids compared to cholesterol. We therefore aimed to compare lipid solubilization in hepatic and gallbladder biles. METHODS Mixed micellar and vesicular phases were separated from hepatic and associated gallbladder biles of seven cholesterol gallstone patients by using state-of-the-art gel filtration with bile salts at intermixed micellar/intervesicular compositions and concentrations in the eluant. RESULTS Vesicles were found in 6 out of 7 hepatic biles, but only in 2 of the corresponding gallbladder biles. Both percentage (7.8+/-5.1 vs. 36.3+/-7.6%; p = 0.01) and amount (0.9+/-0.2 vs. 1.7+/-0.3 mM; p = 0.06) of vesicular cholesterol were lower in gallbladder biles. Similar results were found for vesicular phospholipids (1.3+/-0.8 vs. 11.6+/-6.0%; p = 0.05; and 0.3+/-0.1 vs. 1.1+/-0.5 mM; p = 0.07). The vesicular cholesterol/ phospholipid ratio was 1.7+/-0.5 in hepatic bile but 4.3 and 1.8 in the 2 gallbladder biles which contained vesicles. Mixed micelles in gallbladder biles had a higher cholesterol saturation index than mixed micelles in hepatic biles (1.43+/-0.11 vs. 1.15+/-0.07; p = 0.02). CONCLUSIONS Concentration of bile in the gallbladder leads to decreased vesicular lipid contents. The finding of supersaturated mixed micelles in the absence of vesicles in a significant number of patients points to the possibility of non-vesicular modes of crystallization.