[Early jejunal nutrition with combined pre- and probiotics in acute pancreatitis--prospective, randomized, double-blind investigations].

BACKGROUND

Development of infection of pancreatic tissue in patients with severe acute pancreatitis dramatically increases morbidity and mortality. Colonisation of the lower gastrointestinal tract and oropharynx, mostly with gram-negative but sometimes also gram-positive bacteria is known to precede the contamination of the pancreatic tissue by a few days. A few specific lactic acid bacteria such as Lactobacillus plantarum 299 were effective in preventing colonisation of the gut by potential pathogens, to reduce endotoxemia and to stimulate the gut-associated lymphatic system (GALT) and the immune system.

METHODS

Patients with acute pancreatitis, arriving within 48 hours after onset of disease showing typical clinical picture and laboratory signs of pancreatitis (plasma amylase > 200 U/l, CRP > 150 mg/l and an Imrie-score = or > 3) were randomised into two groups During the first week the treatment group received a freeze-dried preparation containing 10(9) live Lactobacillus plantarum 299 together with an oat fibre substrate. The control group received a similar preparation, but the Lactobacillus plantarum 299 had been inactivated by heat. For seven days the treatment was repeated twice every day. The preparations were delivered to the hospital in sachets so the content was unknown to the investigators, staff and patients.

RESULTS

Forty five patients completed the study before there was indication that one group differed from the other in a statistically significant way, at which time the study was discontinued and the code broken. At this time 22 patients had received treatment with live and 23 patients with heat-killed Lactobacillus plantarum 299. Infected necrosis and abscesses occurred in 1/22 (4.5%) in the treatment group vs. 7/23 (30%) (p = 0.023) in the control group. The length of stay was 13.7 days in the treatment group vs. 21.4 days in the control group (not statistically significant).

CONCLUSIONS

Supplementing Lactobacillus plantarum 299 is an effective tool to prevent pancreatic sepsis, to reduce the number of operations and length of stay. The only patient who developed sepsis in the treatment group did so eight days after the treatment had been discontinued. One week treatment, as in the present study, is too short. It should be provided for at least 2 weeks, or more appropriately, as long as the patients are treated with antibiotics or have signs of GI colonisation.