It keeps mutating, acts fast—and it’s spreading. Meet the No. 1 superbug in hospitals today.

DAN HAWALESHKA August 14 2006


It keeps mutating, acts fast—and it’s spreading. Meet the No. 1 superbug in hospitals today.

DAN HAWALESHKA August 14 2006



It keeps mutating, acts fast—and it’s spreading. Meet the No. 1 superbug in hospitals today.


It was toward the end of a long day of scientific detective work— a sort of CSI for bacteria experts—when Glenn Songer, from the University of Arizona, stepped up to the audience microphone. The occasion, if not rare, was certainly unusual —an intensive one-day war council convened in May at the Centers for Disease Control and Prevention in Atlanta, Ga. The panellists were North America’s top experts on two potentially deadly bacteria: Clostridium difficile and Clostridium sordellii. C. difficile, in particular, is now behaving in alarmingly and inexplicably new ways. About 200 epidemiologists, microbiologists, bacteriologists, physicians and public health officials had gathered in the chill air of an overly air-conditioned CDC auditorium to weigh options for coping with what may be a silent epidemic in the making. And Songer was about to rattle them.

There are certainly dozens, perhaps hundreds of strains of C. difficile, most notably a superbug known as the epidemic strain. This hypervirulent variety has killed thousands of people—often elderly patients in hospital-across North America over roughly the past five years, particularly in Quebec. There, C. diff. killed an estimated 2,000 patients in 2003-2004 alone. Typically, the victims have been on antibiotics, which in addition to combatting infection, also wipe out the good bacteria normally found in the gastroin testinal tract. In their absence, C. diff. has no competition and thrives. The connection to recent exposure to antibiotics is so pronounced it is considered a virtual prerequisite for coming down with Clostridium difficile. Until now.

Now, CDC officials are concerned by evidence suggesting C. diff is spreading out of U.S. hospitals and into the community-atlarge. (We already know the bacterium has penetrated Quebec communities, and the


Public Health Agency of Canada is tracking C. diff in hospitals across the country and is cognizant of the potential threat to cities, towns and villages.) The CDC first raised the alarm last December, noting 33 cases of atypical C. diff infection had come to its attention. Twenty-three cases were communityacquired. The other 10 were women who were either pregnant or who had just given

birth. Both groups, the CDC noted, were “previously thought to be at low risk.” Uncharacteristically, one-quarter of these unusual cases had not recently taken any antibiotics; half of the community-associated patients were 18 years old or younger; almost 40 per cent of the victims suffered a


relapse; one woman died.

Based on these and other findings, the CDC concluded the disease could be changing, with new cases distinguished by traits previously unseen in the community, including young patient age, lack of antimicrobial exposure and a high recurrence rate.

The audience of scientific sleuths in Atlanta had heard much of this and more by the time Songer got up to speak. Songer said his lab had screened for the bacterium in meat bought in Arizona grocery stores. (In addition to striking humans, C. diff. also causes disease among commercial farm animals.) The samples included ground beef, pork and turkey, as well as a selection of beef, beef-pork and pork sausages. Incredibly, one-quarter of the samples were tainted with strains of Clostridium difficile. The workshop participants, who had been quietly listening, immediately broke into peals of nervous laughter intermixed with groans of disbelief.

In an interview outside the auditorium afterward, Songer cautioned there simply isn’t enough data to say much of anything right now. “No conclusions can be drawn at this point from what we’ve found,” Songer says. “It just encourages us to look further, and hopefully get a more definitive answer.” He said his fellow researchers are still trying to identify the strains of C. diff. implicated, and that “it would be very surprising if it was the epidemic type—I doubt that very much.” However, Scott Weese, a veterinarian at the

University of Guelph in Ontario, says his lab has already observed the epidemic strain in live cattle, pigs, and a dog. “You have to assume that if it’s in the cattle population,” Weese says, “it’ll end up in the meat.” Weese expects to soon publish a study that found C. diff. in commercial beef bought in Ontario and Quebec in percentages similar to Songer’s findings in Arizona. Because C. diff. spores have been known to survive the cooking process, people will want to know whether, for example, our hamburger meat could be making us sick. But Weese says we still don’t know if the C. diff. bacterium is transmissible between people and animals.

“The general population shouldn’t be too concerned about it right now because we don’t have good evidence that it’s a foodborne pathogen,” Weese says. “That’s something we need to look at, but there’s no reason to panic at this point.”

That said, Songer’s lab results and the Atlanta audience’s nervousness underscore just how concerned health authorities have become with community-acquired Clostridium difficile. It now appears to be an “emerging infection,” says Dr. Paul Seligman, director of the U.S. Food and Drug Administration’s office of pharmacoepidemiology and statistical science, and one of the event organizers. “Clearly,” Seligman says, “we want to pay attention to anything that moves from an environment we’re familiar with, such as hospitals, to a community setting—particularly when you’re talking about severe, rapidly fatal illnesses.”

THERE ARE FEW GOOD ways to die, and lying in a pair of soiled diapers while waiting for the merciful end to arrive certainly isn’t one of them. But patients fighting a severe Clostridium difficile infection frequently suffer that indignity, enduring 10 to 20 bowel movements a day. In addition to chronic diarrhea, the epidemic strain can cause a ruthless inflammation of the colon called colitis that sometimes necessitates surgical removal of

that part of the large intestine. Sepsis, or blood poisoning, can also occur. Patients endure horrible abdominal pain, perilously low blood pressure and anemia that leaves them too weak to crawl out of bed.

In 2002, an epidemic of C difficile took hold of hospitals in the Sherbrooke region in Quebec’s Eastern Townships, and in Montreal, 140 km to the northwest, killing a staggering 2,000 Quebecers in the following year. “Even those who survived spent many days in a pitiful condition,” recalls Dr. Jacques Pépin, an epidemiologist at the University of Sherbrooke, who attended to many such patients. “I can tell you, it’s not a disease you would wish on anyone.”

C. difficile is spread through resilient spores found in feces—they can survive for months on a wide variety of surfaces. The spores are ingested when people put a hand down on a contaminated countertop or sink, then absentmindedly touch their mouths or eat food without first washing their hands with soap. Once ingested, C. difficile can lie dormant until a trigger, usually antibiotics, disrupts the gut’s normal bacterial flora. It is estimated that between three and 10 per cent of healthy North Americans are asymptomatic carriers of the bacterium, while up to 40 per cent of hospitalized patients are infected.

Physicians once considered C. diff. no more

than a clinical nuisance. It caused a brief bit of diarrhea and had no impact on mortality. But the epidemic strain is an accomplished pathogen, considered by many to be the No. 1 superbug in health care facilities. It is the most common cause of infectious diarrhea in hospitals in the developed world. The risk of contracting the disease is known to rise with increasing length of hospital stay, age, and if a person’s immune system has been compromised.

As with all bacteria, the various strains of Clostridium difficile mutate regularly. In the past decade, massive overuse by hospitals of a class of antibiotics known as fluoroquino-

lones favoured the genetic selection of a tougher C. diff. bacterium. It is this fluoroquinolone-resistant, hypervirulent variety that slammed Quebec. Dubbed the epidemic strain, the bacterium has been detected in British Columbia, Alberta, Saskatchewan, Ontario, Nova Scotia and Newfoundland, as well as in 16 states, the United Kingdom and the Netherlands.

Hospitals are not supposed to make people sick, much less the communities that surround them. But creaking health care facilities with crowded wards and inadequate access to sinks for handwashing—particularly in Quebec—have turned these sites into microbial incubators-cum-distribution hubs for Clostridium difficile. The superbug can spread among patients, who then unwittingly carry the disease with them into the community once they’ve been discharged. In Sherbrooke, from 1991 to 2003, the rate of C. diff. infection in the general population went from 35 per 100,000 inhabitants to 156 per 100,000.

Britain has also undergone an exponential rise in the rates of C. diff. cases diagnosed in the community, from less than one case per 100,000 persons in 1994 to 22 per 100,000 inhabitants in 2004. Confounding scientists’ ability to explain the formidable

hike is the fact antibiotic use within the community actually decreased during the same period, which suggests something else is going on to trigger the disease. But what? “Who knows?” says Michelle Alfa, a C. difficile researcher at St. Boniface General Hospital in Winnipeg. “Maybe there are other things out there that cause an upset in people’s gastrointestinal tract that then allows C. diffi to do its merry thing.”

Dr. Sandra Dial, a researcher at McGill University in Montreal, offers one possible explanation, one gaining momentum and widely discussed at the CDC gathering in Atlanta. Dial studied 1,233 patients drawn from a British database who came down with C. diffi in their communities between 1994 and 2004. Of these, only 37 per cent had taken antibiotics in the 90 days prior to being diagnosed with Clostridium difficile. Dial’s study, published last December in the Journal of the American Medical Association, found that heartburn drugs, and medication used to treat gastroesophageal reflux disease, increased a person’s chances of coming down with C. ¿(//^-associated diarrhea.

Dial reported that drugs known as proton pump inhibitors, or PPIs—including Nexium, Prevacid, Prilosec and Protonix—increased the risk almost threefold. So-called H2 receptor antagonists—Pepcid and Zantac are twodoubled the risk. People taking painkillers known as non-steroidal anti-inflammatory drugs, or NSAIDs (but not Aspirin), had a 30 per cent higher rate of C. difficile illness. Dial speculates that C. diffi may have mutated to take advantage of the decrease in stomach acid in people who take these drugs. “It has tremendous repercussions, as you can imagine,” says Dr. Clifford McDonald, a CDC epidemiologist, and one of the Atlanta meeting’s panellists. “PPIs are used widely.” Predicting exactly what C. diffi is up to, says Dr. Allison McGeer, director of infection control at Toronto’s Mount Sinai Hospital, is a little like predicting when the next avian flu pandemic will strike. “We honestly don’t know the extent to which this strain is going to stay concentrated in hospitals, or whether it’s going to spread and cause similar problems in the community and in younger people,” says McGeer, adding that the public-health implications are nevertheless plain to see. “This is not a disease that is just in hospitals and just affecting people who are going to die anyway,” McGeer warns. “This is already a disease that has caused a huge amount of public-health damage. This disease poses significant risks to all of us, now.”

WHEN THE CDC first raised the alarm last December, it was unable to determine with any certainty what strain of C. diffi. it was dealing with. All the agency knows is the bug did things previously unheard of in 33 people who ordinarily had no business coming down with the disease. Of the atypical cases of C. diffi, the CDC singled out two as signposts:

• One was a 31-year-old woman who was 14 weeks pregnant with twins. She sought medical attention at her local emergency department after three weeks of intermittent diarrhea, followed by three days of cramping and watery, black stools four to five times daily. She was admitted after her stool tested positive for C. difficile toxin. She had taken trimethoprim-sulfamethoxazole for a urinary tract infection about three months before she was admitted. Doctors prescribed the antibiotic metronidazole, one of only two standard treatments for C. diffi, and discharged her the same day, but had to readmit her less than 24 hours later. She stayed for almost three weeks. Treatment with the antibiotic vancomycin eased the severe inflammation of her colon. She was discharged a second time, but four days later was readmitted with diarrhea and low blood pressure. She spontaneously aborted her fetuses, and later died despite aggressive treatment that included the surgical removal of her colon. “It’s very disturbing when a pregnant woman in her 30s dies,” says Dr. André Weltman, a public-health physician at the Pennsylvania Department of Health. “That’s not supposed to happen.”

• The other U.S. case involved a 10-yearold girl who went to a hospital suffering from intractable diarrhea, projectile vomiting and abdominal pain. It had been a year since she’d taken any antibiotics. Her stool tested positive for C. difficile toxin. The girl had fallen ill about two weeks earlier, a few days after her younger brother had come down

with a fever and diarrhea. He got better on his own, but she developed a fever as high as 102° F. Her symptoms worsened until she was having liquid stools up to 14 times a day. She was admitted to hospital and recovered after doctors administered intravenous fluids, electrolytes and metronidazole.

Thirty-three cases and a single death may not sound like much, says John Dyke, a clinical microbiologist with the Michigan Department of Community Health, but to ignore them would be a grave mistake—we could easily be seeing only the tip of the epidemiological iceberg. In most states, Dyke notes, C. diffi is not a reportable disease, and so we likely don’t know the true incidence in the population. Dyke says the Atlanta meeting he attended will help draw attention to the fact that we may have a disease, as he put it, “smouldering in the population.” Only two samples of the microbes were isolated from the 33 patients. “Neither of them were the epidemic strain,” says the CDC’s McDonald. “That doesn’t mean the epidemic strain isn’t playing a role, but we don’t know it yet.” What is known, however, is both the epidemic strain and the U.S. samples were eerily similar. “The two isolates both produced a previously uncommon, extra toxin—a binary toxin—that has been found in the epidemic strain,” McDonald explains. “In addition, each in its own way shared at least one other significant genetic characteristic with the epidemic strain.”

Pépin and his fellow researchers have looked for the epidemic strain in C. diffi patients in

the Sherbrooke region who had no connection to a hospital, and have found it in about one-third of all cases. That is a rough, preliminary estimate, he cautions, based on a small sample size, but there is no disputing the fact that the nasty strain of C. diffi has spread within the community. “The opposite would have been very surprising because in Quebec we’ve had literally, over two years, something like 14,000

cases in hospitals,” says Pépin, “and obviously these patients go back home, where they’re still shedding the bacteria in their stools, and then it enters the chain of transmission in the community.” However, unlike a significant portion of the communityassociated cases singled out by the CDC, Pépin has yet to observe the disease in individuals with no antibiotic exposure.

Currently, only Quebec and Manitoba require health care providers to report cases of C. difficile. Ontario is considering doing likewise, while the Public Health Agency of Canada will collect data from 41 so-called sentinel hospitals from across the country, starting this year. The agency will look for the hypervirulent strain and whether first-line antibiotics remain effective against the bacterium. PHAC is also working to develop a surveillance system to track cases of communityacquired C. diff., says Denise Gravel, acting manager of the PHAC’s division of blood safety surveillance and health care-acquired infections. “We haven’t been able to come up with a valid strategy for the moment,” Gravel says. “But even if they are acquired in the community, we are going to see them in these hospitals, so we’re not that concerned.”

It is entirely too easy for a physician to mistakenly dismiss diarrhea as being caused by either a virus, or food-borne bacteria, such as salmonella, shigella and campylobacter. That’s why the CDC says clinicians need to consider Clostridium difficile as a possible source of a person’s diarrhea, “even if the patients do not necessarily have traditional risk factors such as recent hospitalization or antimicrobial use.” The CDC recommends that patients should seek medical attention for diarrhea lasting longer than three days, or if it’s accompanied by blood or high fever.

Even with C. diff.’s apparent changing pathology, the disease remains closely associated with antibiotic use, says Dr. Thomas Louie, medical director of infection prevention and control for the Calgary Health Region. “We have seen patients in Calgary that have not been on antibiotics and have C. diff, but I’d have to review 100 to 500 cases to find one that is like that,” Louie says. “The big picture is that C. diff. is still, largely, antibiotic induced—we’re talking 99 per cent. But I don’t know about that one per cent,” Louie concedes, “and why that happened to that lady.” He’s referring to the woman, pregnant with twins, who died. Her death, coupled with the other unusual cases flagged by the CDC, factor in Louie calling for more proactive surveillance of C. difficile. Louie hopes her death is an anomaly. “And if it isn’t, we have to take a good, hard look at ourselves, at what we’re doing in health care—both in the hospital, and in the community.” M