Scientific paper

Postinduction bacteremia in experimental acute pancreatitis

Intestinal barrier failure and subsequent bacterial translocation have been implicated in the development of organ dysfunction and septic complications associated with severe acute pancreatitis. Splanchnic hypoperfusion and ischemia/reperfusion injury have been postulated as a cause of increased intestinal permeability. The urinary concentration of intestinal fatty acid binding protein (IFABP) has been shown to be a sensitive marker of intestinal ischemia, with increased levels being associated with ischemia/reperfusion. The aim of the current study was to assess the relationship between excretion of IFABP in urine, gut mucosal barrier failure (intestinal hyperpermeability and systemic exposure to endotoxemia), and clinical severity. Patients with a clinical and biochemical diagnosis of acute pancreatitis were studied within 72 hours of onset of pain. Polyethylene glycol probes of 3350 kDa and 400 kDa were administered enterally, and the ratio of the percentage of retrieval of each probe after renal excretion was used as a measure of intestinal macromolecular permeability. Collected urine was also used to determine the IFABP concentration (IFABP-c) and total IFABP (IFABP-t) excreted over the 24-hour period, using an enzyme-linked immunosorbent assay technique. The systemic inflammatory response was estimated from peak 0 to 72-hour plasma C-reactive protein levels, and systemic exposure to endotoxins was measured using serum IgM endotoxin cytoplasmic antibody (EndoCAb) levels. The severity of the attack was assessed on the basis of the Atlanta criteria. Sixty-one patients with acute pancreatitis (severe in 19) and 12 healthy control subjects were studied. Compared to mild attacks, severe attacks were associated with significantly higher urinary IFABP-c (median 1092 pg/ml vs. 84 pg/ml; P < 0.001) and IFABP-t (median 1.14 μg vs. 0.21 μg; P = 0.003). Furthermore, the control group had significantly lower IFABP-c (median 37 pg/ml; P = 0.029) and IFABP-t (median 0.06 μg; P = 0.005) than patients with mild attacks. IFABP correlated positively with the polyethylene glycol 3350 percentage retrieval (r = 0.50; P < 0.001), CRP (r = 0.51; P < 0.001), and inversely with serum IgM EndoCAb levels (r = −0.32; P = 0.02). The results of this study support the hypothesis that splanchnic hypoperfusion contributes to the loss of intestinal mucosal integrity associated with a severe attack of pancreatitis. ( J Gastrointest Surg 2003;7:26–36.)

Background: Infectious complications in severe pancreatitis are the main factors determining clinical course and outcome. The taurocholate model for acute necrotizing pancreatitis was evaluated for frequency and time course of pancreatic and extrapancreatic bacterial infection. Methods: Sixty-five male Wistar rats were divided into 5 groups of 13 animals each. Specimens for bacteriologic examination were taken, and pancreatitis was induced by intraductal infusion of 3% taurocholate under sterile conditions. Animals were killed 8, 16, 24, or 32 hours thereafter, and bacteriologic examination was performed. A control group of animals with intraductal infusion of 0.9% saline solution were killed after 32 hours. Results: There was no significant pancreatic infection in the control group and in the 8-hour group (1 of 13 rats). Sixteen and 24 hours after induction of pancreatitis, infection and inflammation of the pancreas were found in 77% (10 of 13 rats), and after 32 hours pancreatic infection occurred in 69% (9 of 13 rats). Extrapancreatic bacterial infection after 16 hours occurred in the liver (62%), spleen (62%), and mesenteric lymph nodes (46%). Bacteria infecting the pancreas reflected the bacterial spectrum of the large bowel and terminal ileum before induction of pancreatitis (Escherichia coli [77%], Proteus [43%], Enterococcus [37%], and Staphylococcus [23%]). Conclusions: Pancreatic infection is an early and frequent finding in the taurocholate model of acute necrotizing pancreatitis. Infection occurs between 8 and 16 hours after induction of pancreatitis. The source of infecting bacteria seems to be the large bowel or the terminal ileum. We present a useful model of severe pancreatitis in which to study bacterial translocation, the further route of spread, and therapeutic approaches. (Surgery 2000;127:427-32.)

Sepsis accounts for 80% of deaths from acute pancreatitis. This study aimed to investigate early changes in intestinal permeability in patients with acute pancreatitis, and to correlate these changes with subsequent disease severity and endotoxemia. The renal excretion of enterally administered polyethylene glycol (PEG) 3350 and PEG 400 was measured within 72 hours of onset of acute pancreatitis to determine intestinal permeability. Severity was assessed on the basis of APACHE II scores and C-reactive protein measurements. Serum endotoxin and antiendotoxin antibodies were measured on admission. Eight-five patients with acute pancreatitis (mild in 56, severe in 29) and 25 healthy control subjects were studied. Urinary excretion of PEG 3350 (median) was significantly greater in patients who had severe attacks (0.61%) compared to those with mild disease (0.09%) and health control subjects (0.12%) (P <0.0001), as was the permeability index (PEG 3350/400 excretion) (P <0.00001). The permeability index was significantly greater in patients who subsequently developed multiple organ system failure and/or died compared with other severe cases (0.16 vs. 0.04) (P = 0.0005). The excretion of PEG 3350 correlated strongly with endotoxemia (r = 0.8; P = 0.002). Early increased intestinal permeability may play an important role in the pathophysiology of severe acute pancreatitis. Therapies that aim to restore intestinal barrier function may improve outcome.

In the present study we demonstrated that a single injection of endotoxin (lipopolysaccharides, E. Coli 0111-B4) into the superior pancreaticoduodenal artery of rabbits induced a dose-dependent acute necrotizing pancreatitis. The lesions observed by light microscopy were significant for 10 μg lipopolysaccharides and were maximal for 20 μg. After 24 h the main findings were edema, acinar cell vacuolisation, polymorphonuclear neutrophil infiltration and tissue necrosis. The pancreatic lesions developed strictly in the area supplied by the artery injected with lipopolysaccharides, without significant intestinal involvement.

Since platelet-activating factor (1-O-hexadecyl-2-acetyl-sn-glycero-3-phosphocholine, PAF; 50–500 ng), a phospholipid mediator of endotoxin-induced inflammation and shock, was previously shown to cause an acute necrotizing pancreatitis in rabbits, the role of PAF in the development of acute pancreatitis induced by lipopolysaccharides was studied by evaluating: (1) the synergism between doses of lipopolysaccharides (5–10 μg), which produced a mild tissue injury, and doses of PAF (10 ng) not producing, per se, any significant injury, and (2) the effect of three structurally unrelated PAF receptor antagonists. The results obtained demonstrated that 10 ng of PAF significantly potentiated pancreatic tissue damage induced by 10 μg of lipopolysaccharides. Moreover, PAF receptor blockade resulted in complete prevention of the severe pancreatitis caused by 20 μg lipopolysaccharides with cis(+)-1-[2-hydroxy[tetrahydro-5-[(octadecyl-aminocarbonyl)oxy]methyl]-2-furanyl]methoxy-phosphinyl-oxy]ethyl]-quinolinium hydroxide, SRI 63675 and in a significant reduction of the incidence and the extension of tissue injury with (RS)-2-methoxy-3-(octadecylcarbamoyloxy)-propyl-2-(3-thiazolio)-ethylphosphate, CV 3988 and 6-(2-chloro-phenyl)-8,9-dihydro-1-methyl-8- (4-morpholinyl-carbonyl)-4H,7H-cyclopentyl[4,5]thienol[3,2-f][1,2,4]-triazolo-[4,3-α][1,4]diazepine, WEB 2170. These observations suggest a crucial role of PAF in the pathogenesis of pancreatic tissue damage induced by lipopolysaccharides.

Although pancreatic spesis is the most common cause of major morbidity and mortality associated with acute pancreatitis, the pathogenesis of such infections is unknown. Since intraperitoneal foci of inflammation are known to promote bacterial translocation, we hypothesized that acute pancreatitis promotes bacterial translocation that leads to infection of the inflamed pancreas and peripancreatic tissues. Non-lethal acute pancreatitis was induced in rats, and the translocation of live bacteria to the pancreas, mesenteric lymph nodes, liver, and spleen was determined. The presence of orally fed fluorescent beads, sensitive inert markers of translocation, was also determined in the pancreas and mesenteric lymph nodes. Live bacteria were recovered from 33% of the pancreata of rats with acute pancreatitis but from none of the control rats. Beads were visualized in 91% of the pancreata of rats with acute pancreatitis but in none of the pancreata from control rats. Beads were not visualized in the measenteric lymph nodes of rats with acute pancreatitis, suggesting a transperitoneal route of migration. We conclude that acute pancreatitis promotes bacterial translocation leading to transperitoneal infection of the pancreas. These results support the use of selective decontamination of the gut and peritoneal lavage for the prevention of pancreatic infections in acute pancreatitis.

In a prospective clinical study including 114 patients with acute necrotizing pancreatitis, but excluding patients with a pancreatic abscess, necrotic material obtained at surgery was tested bacteriologically. Intestinal microorganisms were cultured in 39.4% of the cases. The contamination rate was 23.8% in patients operated on during the first 7 days of the attack; it rose to 71.4% in the third week and decreased to 32.5% after the fourth week. Intra- and extrapancreatic necrosis was more widespread and pancreatitis-associated ascites was more frequent in patients with proven contamination. The number of objective signs was 4.5 (median) and postoperative mortality was 37.8% in bacteriologically positive subjects, whereas the number was 3.5 (median) and mortality was 8.7% in bacteriologically negative patients. Morphologic and clinical alterations were more severe, and the mortality rate was significantly elevated, in patients with a short history of disease and bacterial contamination of necrotic tissue. All 5 patients with pancreatic sepsis who were operated on in the first 7 days of the disease, as compared with 2 of 16 patients with sterile necrosis, died. Thus, it is demonstrated that bacterial contamination of pancreatic necrosis occurs early and frequently, causing a significant increase in morbidity and mortality, particularly when it develops in the initial stages of the attack.