Hospitals struggle with a serious new gut microbe
About 3 years ago, physicians in Quebec noticed an alarming pattern in patients with diarrhea. "All of a sudden, we were having patients so sick that they needed the ICU [intensive-care unit]," says doctor and epidemiologist Sandra Dial. The same illness was also sending patients to the morgue as never before. They weren't succumbing to the ailment that brought them to the hospital. Instead, they seemed to have gotten sick from their antibiotic treatment.
As an antibiotic attacks its target, it can also kill harmless flora, the term for the billions of bacteria that live in healthy intestines. Invaders, mainly the pathogen Clostridium difficile, can flock to the colon's open real estate.
Doctors treat C. difficile by suspending the culpable antibiotic therapy and administering a different one. While usually curable, the diarrheal disease recurs repeatedly in some patients and occasionally causes life-threatening inflammation of the colon.
C-diff, as it's called, is a rising problem across North America and in parts of Europe. A virulent, once-rare strain has driven a string of recent outbreaks and is suspected to be responsible for much of the overall increase in the disease. New resistance to widely used antibiotics seems to have made the microbe more likely to cause diarrhea in sick and elderly people.
At the same time, scattered cases of C. difficile among people not normally susceptible to it have researchers concerned.
Before December 2002, Dial had seen only three cases of C. difficile in 8 years at the McGill University–affiliated ICUs where she works. Then, she says, "we had five patients in 1 month."
Parallel epidemics were soon under way elsewhere in the region. More than 7,000 C. difficile cases occurred in Quebec hospitals in 2003, affecting up to 2.5 percent of patients. By early 2005, 30 centers in the province were seeing at least five times as many cases as they had identified in previous years.
In the United States, "the rate of C-diff in hospital patients doubled between 2000 and 2003," says L. Clifford McDonald of the Centers for Disease Control and Prevention (CDC) in Atlanta.
In response, researchers have identified unique characteristics of the newly virulent strain—and have revealed a grimmer picture of C. difficile than the one familiar to most physicians.
"Ten years ago, we didn't believe people died of this," Dial says. "It was very unusual. Now, unfortunately, it's not unusual."
Previously, elderly people staying in health care facilities and nursing homes were virtually the only targets of C. difficile. Now, the bacterium is gaining notoriety for illnesses in seemingly vigorous people.
Last year, for instance, the pathogen felled a 31-year-old Pennsylvania woman who was pregnant with twins. Early in her second trimester, she went to the emergency room with worsening diarrhea. Despite treatment that night and during two subsequent hospital stays, within a month she delivered stillborn fetuses and died of complications of colon inflammation.
In the Dec. 2, 2005 Morbidity and Mortality Weekly Report, McDonald and his collaborators in four states describe this case and 34 other recent illnesses caused by C. difficile contracted outside health care facilities.
To tease out the changing characteristics of C-diff, Jacques Pépin and his colleagues at the University of Sherbrooke Hospital Center in Quebec combed through data on 5,619 hospital patients treated between January 2003 and June 2004. Nearly 300 of the patients had developed the disease.
Compared with the other patients, those who'd contracted C. difficile tended to be older and to have had longer hospital stays. Those infected were also more likely to have recently used an antibiotic of the cephalosporin family, the team found.
Those antibiotics had been previously linked to outbreaks. The diarrhea-causing bacterium is usually resistant to cephalosporins, but normal colon bacteria are not. Therefore, the drugs create a wide-open opportunity for C. difficile. Swallowing a spore of the bacterium during cephalosporin therapy could result in an infection that otherwise wouldn't have gained a foothold.
In the Nov. 1, 2005 Clinical Infectious Diseases, the Sherbrooke researchers also link C. difficile to a class of antibiotics that had only infrequently turned up in earlier work. Fluoroquinolones, the most widely used antimicrobials in the United States, fight infections of the respiratory and urinary tracts, the skin, and other tissues. Ciprofloxacin is the main drug in this class.
In early 2004, researchers at Sherbrooke and 11 other Quebec hospitals joined forces to investigate what they regarded as the province's plague. Led by microbiologist Vivian G. Loo of McGill, they ultimately studied 1,703 patients who acquired C. difficile during the first half of that year.
The infected patients represented 2.25 percent of all people hospitalized at the 12 centers during the study. That reflects an infection rate four times as high as that in earlier data.
Almost 7 percent of the 2004 infections were fatal. That rate, too, is at least four times as high as that found in the past, Loo says.
As in the Sherbrooke study, patients who developed C-diff were more likely than other patients to have used fluoroquinolones while hospitalized, the researchers report in the Dec. 8, 2005 New England Journal of Medicine (NEJM).
More than 80 percent of the infected patients had acquired a single, fluoroquinolone-resistant form of the pathogen, making that strain primarily responsible for the Quebec outbreaks, Loo's team concludes.
A novel mutant
To characterize the guilty strain, Pépin, McDonald, and six collaborators analyzed 124 C. difficile samples taken from Sherbrooke patients in 2004 and early 2005. They also looked at 30 recent samples from other patients in Quebec, the United States, and the United Kingdom.
Disease-causing C. difficile produce at least two colon-inflaming substances, called toxin A and toxin B. Rare strains that make neither toxin don't result in illness. Scientists have occasionally identified strains that produce a third toxin, called binary toxin.
The new outbreak strain always produces binary toxin, the researchers determined. Furthermore, they found, it produces 16 times as much toxin A and 23 times as much toxin B as typical hospital strains of C. difficile do.
Genetic traits also set apart the dangerous strain. For example, the researchers identified a standard C. difficile gene that was uniformly mutated in the samples. The mutation may account for the excess production of toxins A and B, the team suggested in the Sept. 24, 2005 Lancet. Extra toxin may lead to more severe diarrhea, and, therefore, greater spread of the pathogen in hospitals.
The mutant strain, which the researchers labeled NAP1/027, had infected patients in all three countries in the study.
It has also caused recent outbreaks in the Netherlands and Belgium, says Dale N. Gerding of Hines Veterans Affairs Medical Center in Illinois.
Until recently, NAP1/027 was rare, he says. "It was definitely present in other hospitals 20 years ago, but it wasn't causing epidemics," he says.
Gerding, McDonald, and their collaborators made that assessment after comparing 187 recent samples of C. difficile, collected between 2001 and 2003, with more than 6,000 older samples. About half of the recent samples were NAP1/027.
Genetic similarities indicated that 14 older samples dating to 1984 were of the same, newly troublesome strain. NAP1/027 made binary toxin in the 1980s, but it was not resistant to most fluoroquinolones before 2001, the researchers report in the Dec. 8, 2005 NEJM.
"Fluoroquinolone resistance is unusual and new for this strain," Gerding says. Use of those antibiotics in hospitals has increased tremendously over the past 2 decades, which could explain the success of the newly resistant variant, he says.
But Dial suggests that antibiotics aren't the only medications contributing to the epidemic. She and three McGill colleagues have found possible links between C. difficile infection and some commonly used drugs, particularly acid reducers that treat heartburn, indigestion, and stomach ulcers in millions of people.
In studying a U.K. database that doesn't track individual strains, Dial's team found that people who acquired C. difficile outside the hospital were three times as likely as uninfected people to use proton pump inhibitors (PPIs), such as Prilosec, and twice as likely to use H2 blockers, an older class of acid reducers.
Dormant C. difficile spores are resistant to acid, but replicating cells are susceptible. So, normal concentrations of stomach acid could suppress the organism's activity—until a person takes an acid reducer, Dial says.
Use of nonsteroidal anti-inflammatory drugs is also slightly elevated among the people with C. difficile, her team reports in the Dec. 21, 2005 Journal of the American Medical Association.
Gerding, along with doctors from a Portland, Maine, hospital, separately found an association between PPIs and C. difficile. Robert C. Owens Jr. of the Maine Medical Center presented that finding at the Interscience Conference on Antimicrobial Agents and Chemotherapy in Washington, D.C., last December.
But Gerding says that it's "an open question" whether the acid reducers contribute to infection. Data sets analyzed by Loo and by Pépin don't suggest a link.
In any case, says Gerding, "antibiotics are still the overwhelming risk factor."
In the United States, the newly troublesome strain continues to spread. "CDC has documented the strain in 16 states," McDonald says.
In Quebec, the worst-hit Canadian province, diligent efforts in hospitals have brought the new bug under control—to a degree. Compared with the epidemic's early&150;2003 peak, says Loo, "our rates are halved." But they're still significantly elevated from pre-epidemic rates, she notes.
Hospitals employed a multifaceted approach to rein in the epidemic. First, they increased the frequency of cleanings. Rooms with C. difficile–infected patients now get scrubbed down twice a day, Loo says. And to fight C. difficile's hardy spores, bleach has been substituted for gentler agents.
Second, hospitals spent new funds from the Canadian government to buy more of their most frequently used medical items, such as blood pressure cuffs, so that equipment that might pick up spores doesn't get shuttled from room to room. "We also eliminated rectal thermometers," Loo says.
Furthermore, hospitals installed more sinks to encourage hand washing and created new rooms to reduce patient crowding in some older hospitals.
"We also changed our hand-hygiene practice," Loo says. "Alcohol hand rinses decrease the concentration of the bacteria, but washing at the sink was still better."
Finally, she says, the hospitals made an effort to educate physicians and medical residents on appropriate antibiotic use. That effort has somewhat reduced unnecessary administration of fluoroquinolones and other drugs, but more progress is needed, says Loo.
Fire with fire
At the December conference on antimicrobial agents, Gerding described new tests conducted on a strain of C. difficile that might prevent, rather than cause, disease.
Scientists had previously observed that hospitalized people who already carry a harmless form of the bacterium are unlikely to pick up a disease-causing strain. So, Gerding and his team took a non–toxin-making strain from a healthy person and introduced it into laboratory hamsters. Then, they exposed the animals to a strain that normally kills hamsters within 48 hours. The hamsters stayed healthy.
Gerding says that he is seeking the Food and Drug Administration's permission to test the protective strain in people.
While the approach is "counterintuitive," says McDonald, "there's a couple of lines of evidence to suggest it holds promise."
Other experimental approaches aim to strengthen the gut's defenses. The closest to fruition, according to McDonald, is a strategy in which a polymer binds to and inactivates the bacterium's toxins. Genzyme Corp. of Cambridge, Mass., developed the polymer, which it calls tolevamer. After promising preliminary results, the company launched a treatment trial that it projects will include more than 1,000 C. difficile–infected volunteers by the end of this year.
A positive outcome could lead to the first treatment for the antibiotic-associated illness that does not employ an antibiotic.
3650 St. Urbain, Room K1 14
Montreal, QC H2X 2P4
Dale N. Gerding
Research Service (151)
P.O. Box 1490
Veterans Affairs Hospital
Hines, IL 60141
Vivian G. Loo
Department of Microbiology
McGill University Health Center
1650 Cedar Avenue, Room D16.168
Montreal, QC H3G 1A4
L. Clifford McDonald
Epidemiology and Laboratory Branch
Division of Healthcare Quality Promotion
Centers for Disease Control and Prevention
1600 Clifton Road, MS A35
Atlanta, GA 30333
Centre Hospitalier Universitaire de Sherbrooke
3001 12th Avenue N.
Sherbrooke, QC J1H 5N4
Bartlett, J.G., and T.M. Perl. 2005. The new Clostridium difficile—What does it mean? New England Journal of Medicine 353(Dec. 8):2503-2505. Extract available at [Go to].
Davidson, D., J. Peppa, and T. Louie. 2004. A phase 2 study of the toxin binding polymer tolevamer in patients with C. difficile associated diarrhea. 1st International Clostridium difficile Symposium. May 5-7. Kranjska Gora, Slovenia.
Harder, B. 2002. Germs that do a body good. Science News 161(Feb. 2):72-74. Available at [Go to].
Norén, T. 2005. Outbreak from a high-toxin intruder: Clostridium difficile. Lancet 366(Sept. 24):1053-1054. Abstract available at [Go to].