In a nation-wide study into the effects of probiotics in patients with severe acute pancreatitis, more people died in the group that was given probiotics than in the group that was not. The investigators were taken by surprise by the outcome. Previous smaller studies performed abroad showed that the treatment would reduce the rate of infection. This study demonstrates that this is not the case. The results of the study will shortly be published in a leading scientific journal.
Two hundred and ninety-six patients with acute pancreatitis, a rare but very serious disease, took part in the study. Twenty-four patients (sixteen percent) in the study group died completely unexpectedly, and nine patients (six percent) in the control group.
The study investigators do not yet know what has caused the higher death rate. Three factors are probably at play here:
- the administration of probiotics in the acute phase of this serious disease,
- the administration of probiotics to intensive-care patients, and
- the administration of probiotics through a feeding tube (directly into the intestine).
The investigators advise colleagues not to use probiotic bacteria in situations that meet these three criteria.
The patients were treated in 15 Dutch hospitals, among which all the eight university medical centers. The study was funded by Senter Novem, a government body, and conducted under the guidance of University Medical Center Utrecht. All the patients involved and the relatives of the deceased patients have been notified.
The results from two earlier small-scale studies on probiotics in the treatment of patients with pancreatitis were hopeful, but the studies were too small to convince doctors of the need to use probiotics in all patients with serious acute pancreatitis. This is why the scale of the current study, undertaken to prove the capacity of probiotics to reduce the infection rate, is much larger.
The study’s finding did not emerge until after the study was completed. This is because it was conducted ‘double blind’, as is required in medical research. Halfway through the study, a specially created monitoring commission established that there were no significant differences, neither in the infection rate nor in the risk of death. The total death rate during the study period was 11 percent, which is not significantly higher than the 10 percent that is known from medical literature. This is why during the study period the strong difference between patient groups went unnoticed. The differences between the two patient groups did not become visible until a code was broken after completion of the study.
All the participating patients gave prior consent to take part in the study and all the guidelines of the Central Committee on Research Involving Human Subjects (CCMO) were followed.
For more detailed information, please contact UMC Utrecht, Ilse van Wijk, 0031 88 755 7483.