NEW ORLEANS – A Montreal hospital grappling with high Clostridium difficile infections rates launched an intervention in October 2013 to screen patients at admission and detect asymptomatic carriers, and investigators found 4.8% of 7,599 people admitted through the ED over 15 months were carriers of C. difficile.
To protect Jewish General Hospital physicians, staff and other patients from potential transmission, these patients were placed in isolation. However, because they were fairly numerous – 1 in 20 admissions – and because infectious disease (ID) experts feared a substantial backlash, these patients were put in less restrictive isolation. They were permitted to share rooms as long as the dividing curtains remained drawn, for example. In addition, clinicians could skip wearing traditional isolation hats and gowns.
The ID team at the hospital considered the intervention a success. “It is estimated we prevented 64 cases over 15 months,” Dr. Longtin said during a packed session at the annual meeting of the American Society for Microbiology.
The hospital’s C. difficile rate dropped from 6.9 per 10,000 patient-days before the screening and isolation protocol to 3.0 per 10,000 during the intervention. The difference was statically significant (P less than .001).
“Compared to other hospitals in the province, we used to be in the middle of the pack [for C. difficile infection rates], and now we are the lowest,” Dr. Longtin said.
Asymptomatic carriers were detected using rectal sampling with sterile swab and polymerase chain reaction analysis. Testing was performed 7 days a week and analyzed once daily, with results generated within 24 hours and documented in the patient chart. Only patients admitted through the ED were screened, which prompted some questions from colleagues, Dr. Longtin said. However, he defends this approach because the 30% or so of patients admitted from the ED tend to spend more days on the ward. The risk of becoming colonized increases steadily with duration of hospitalization. This occurs despite isolating patients with C. difficile infection. Initial results of the study were published in JAMA Internal Medicine ( ).
Risk to health care workers
C. difficile carriers are contagious, but not as much as people with C. difficile infection, Dr. Longtin said. In one study, the microorganism was present on the skin of 61% of symptomatic carriers versus 78% of those infected (). In addition, C. difficile present on patient skin can be transferred to health care worker hands, even up to 6 weeks after resolution of associated diarrhea ( ).
Prior to the intervention, C. difficile prevention at Jewish General involved guidelines that “have not really changed in the last 20 years,” Dr. Longtin said. Contact precautions around infected patients, hand hygiene, environmental cleaning, and antibiotic stewardship were the main strategies.
“Despite all these measures, we were not completely blocking dissemination of C difficile in our hospital,” Dr. Longtin said. He added that soap and water are better than alcohol for C. difficile, “but honestly not very good. Even the best hand hygiene technique is poorly effective to remove C. difficile. On the other hand – get it? – gloves are very effective. We felt we had to combine hand washing with gloves.”
Hand hygiene compliance increased from 37% to 50% during the intervention, and Dr. Longtin expected further improvements over time.
Risk to other patients
“Transmission of C. difficile cannot only be explained by infected patients in a hospital, so likely carriers also play a role,” Dr. Longtin said.
Another set of investigators found that hospital patients exposed to a carrier of C. difficile had nearly twice the risk of acquiring the infection (odds ratio, 1.79) (
“For every patient with C. difficile infection, it’s estimated there are 5-7 C. difficile carriers, so they are numerous as well,” he said.
The bigger picture
During the study period, the C. difficile infection trends did not significantly change on the city level, further supporting the effectiveness of the carrier screen-and-isolate strategy.
There was slight increase in antibiotic use during the intervention period, Dr. Longtin said. “The only type of antibiotics that really decreased were vancomycin and metronidazole... which suggests in turn there were fewer cases of C. difficile infection.”
Long-term follow-up is ongoing, Dr. Longtin said. “We have more than 3 years of intervention. In the past year, our rate was 2.2 per 10,000 patient-days.”
Unanswered questions include the generalizability of the results “because we’re a very pro–infection control hospital,” he said. In addition, a formal cost-benefit analysis of this strategy would be worthwhile in the future.
Dr. Longtin is a consultant for AMG Medical and receives research support from Merck and BD Medical.