User Tools (skip):
In 2003, Quebec hospitals were blindsided by an outbreak of Clostridium difficile - a bacterial microbe that can cause an infection of the bowel. Fairly common to hospital settings, C. difficile can result in diarrhea, stomach pain and, in severe cases, could lead to death. The Public Health Agency of Canada notes that over that year, the reported cases of C. difficile in Quebec more than doubled from just a few years earlier.
On top of the dramatic increase in frequency, the infection was more virulent, with fatality rates also rising. Clearly, Quebec hospitals were facing a crisis.
Working at the Jewish General Hospital, McGill assistant professor of medicine, Sandra Dial, noticed something out of the ordinary. It had been a long-held belief in medical circles that patients using specific types of antibiotics were at an increased risk of getting the bug. However, in July 2004, Dial published a paper demonstrating how patients receiving antibiotics along with antacids (also called proton pump inhibitors, or PPIs) were twice as likely as to contract the infection as those receiving only antibiotics. In simple terms, Dial claimed that by lowering the level of gastric acid, PPIs create a hospitable environment for the colonization of the C. difficile bacteria.
The study, which was rejected by six journals before finally being published by the Canadian Medical Association Journal, wasn't well received by everyone. "The issue of linking PPIs to the infection was fairly new," Dial says, "so many people doubted the results. Also, it was commonly believed that the way C. difficile spores are transmitted make them resistant to gastric acid, so PPIs shouldn't matter."
Undaunted by the nay-sayers, Dial looked to verify her results with another, more extensive study. Enter Samy Suissa, director of the Division of Clinical Epidemiology at the MUHC and James McGill Professor of Epidemiology, Biostatistics and Medicine.
Thanks to a grant from the Canada Foundation for Innovation, Suissa had access to the General Practice Research Database (GPRD), a massive computerized database of the medical records of more than 3 million active outpatients. Suissa suggested that they use this powerful tool to try and replicate Dial's original results. As the methodological mentor, Suissa was responsible for the epidemiological design that would be used for the study.
Suissa recruited Chris Delaney, a PhD student in Epidemiology who also happened to be familiar with the GPRD. As the primary statistician for the project, Delaney spent the bulk of his time accessing data and crunching numbers. "We went as far away from Dr. Dial's original research as we could," he says. "This was a different country and they were community based medical records as opposed to hospital-based. If the original effect was true, it should show up in different populations, under different circumstances."
The group was rounded out by Dr. Alan Barkun a gastroenterologist who offered critical clinical insight and a healthy dose of cynicism. "What better way to make sure your data is good then by convincing a skeptic?" asks Dial.
At study's end nine month's later, Dial was vindicated. Just like their Canadian counterparts, UK patients using antibiotics in conjunction with PPIs were more likely to contract C. difficile than those on antibiotics alone. In fact, the study showed that the stronger the antacid, the greater the risk of infection - further strengthening her case that the bacteria is not resistant to gastric acidity.
The study also yielded some surprises. More than 70 percent of the UK patients who developed C. difficile had not been admitted to a hospital in the past year, and fewer than 50 percent had taken antibiotics in the three months prior to developing the infection, which disproved two long-held assumptions about the bacteria, says Suissa. "C. difficile has left the hospital setting to reach the community and antibiotics are not the only culprit."
Although the product of a strong collaborative effort, the study's heart and soul may very well belong to Dial. "It all started with the sharp eye of a very bright clinician who saw things she was not taught at medical school," says Suissa. "Sandra went against the common beliefs of her peers and said 'wait a minute, I'm going to question this.'"