Dr. Thomas Sandora shares recent data on screening for asymptomatic carriage of Clostridium difficile and provides insight into what the preferred first-line treatment for the infection in children.
With close to half a million individuals becoming infected each year, Clostridium difficile (C. diff) continues to plague health care facilities everywhere. Despite many advances made in the fight against the infection over the past few years, and new guidelines released by the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) recently, many questions remain when it comes to transmission and treatment.
In a session at the SHEA Spring 2018 Conference, Thomas J. Sandora, MD, MPH, associate professor of pediatrics at Harvard Medical School and the hospital epidemiologist and medical director of Infection Prevention and Control at Boston Children’s Hospital, addressed challenging questions that remain when dealing with C. diff in pediatric patients.
He began his presentation by saying that although historically, hospitalization used to be considered a risk factor for C. diff, recent data has shown that most hospital-onset cases are actually not acquired from other C. diff patients.
He provided a snapshot of 1 study published in 2013 in The New England Journal of Medicine which used whole genome sequencing (WGS) to look at genetic relatedness between symptomatic patients in the United Kingdom. The study found that only 35% of cases could be linked with another case, whereas the majority of cases (65%) were not connected to any known prior case.
Another study using WGS to analyze isolates collected from children with C. diff at Lurie Children’s found that out of 104 genetically distinct isolates, “only 12% of incident C. diff cases were caused by a strain isogenic to another one in the cohort,” he said. Furthermore, there was only a small number of potential transmission events (10) and the researchers could only identify shared epidemiologic exposures in 7 of those cases. Therefore, the authors concluded that C. diff transmission among symptomatic children is actually very uncommon.
“The question becomes: If these cases are not coming from exposure to other cases, then where could they be coming from?” Dr. Sandora asked conference goers.
The answer might very well be asymptomatic carriers, at least that’s what 2 larger notable studies seem to suggest. The first study, published in 2017, was a population-based prospective cohort study conducted from 2012 to 2013 in 8 medical wards within 2 hospitals in Denmark. Those patients were housed either in single rooms or rooms with 2 to 4 other patients. At admission, the investigators attempted to screen for C. diff by polymerase chain reaction (PCR); if patients were positive, they collected a culture and then genotyped it. The results were blinded to both the clinicians and the investigators. Patient location was documented (what ward and room) until discharge. Bowel habits were also documented and if patients went on to develop diarrhea (which was defined as 3 or more unformed stools per day with a clinical suspicion of C. diff), the stool was tested by PCR and the patient was then isolated in a single room.
The patients were classified into 7 categories:
“They also calculated something they called the infectious pressure index which was the cumulative number of contacts with other carriers—the numerator—divided by the cumulative number of patient contacts in general—the denominator,” Dr. Sandora explained. “This is analogous to something we often call colonization pressure and if you developed C. diff infection (CDI) then you were excluded from this infectious pressure index because you were immediately placed into strict isolation.” This number ranges from 0 to 1 which provides a sense of how much exposure there is to asymptomatic C. diff carriers.
“For the analysis, they defined transmission as a non-carrier who later developed C. diff infection either during or after exposure to an asymptomatic carrier at the ward or room level and then they used multivariable logistic regression to evaluate the association between carrier status and C. diff infection,” Dr. Sandora continued. “They then used the typing and linkage analyses to determine how many carriers could be linked to direct transmission to non-carriers either at the room or the ward level.”
Their findings? The risk of a patient who was not exposed to an asymptomatic carrier was 2.6%, whereas a patient who was exposed to an asymptomatic carrier at the room level had a 4.6% risk of developing infection, which was a statistically significant increase.
The trend carried over when looking at risk on the ward level; for those patients not exposed to an asymptomatic carrier, their risk was 1.5%, whereas those who were exposed had a 3% risk of developing infection, “but that didn’t quite make statistical significance,” Dr. Sandora admitted.
However, the authors concluded that there was a potential link between asymptomatically colonized individuals and C. diff infection, and, as such, they suggested screening asymptomatic carriers with intervention (perhaps contact precautions) could be considered if there is a “persistently high CDI rate despite the usual isolation precautions and antimicrobial stewardship efforts that you’re using in your hospital,” he said.
He pointed out that there are several downsides to screening and isolating asymptomatic carriers, such as labor and cost and the impact of isolation precautions on families. Furthermore, this step only addresses 1 of the many different potential sources for C. diff.
Although the authors of the new C. diff guidelines did not find sufficient data to recommend screening for asymptomatic carriers and for placing these individuals on contact precautions, Dr. Sandora warned that as this is in an area of particular interest, we can probably expect changes in the future.
Dr. Sandora then switched gears to talk about treatment options for a first episode of a mild C. diff infection in an otherwise healthy child. A quick poll of the room showed that the overwhelming majority of conference goers felt that they would recommend the use of metronidazole for this case, what Dr. Sandora referred to as the “historically suggested approach.” Published 4 years ago, a survey of pediatric infectious disease physicians found that 100% of them would choose metronidazole as well. However, that survey also found that when the underlying patient started to vary, physicians were increasingly turning to vancomycin.
He then proceeded to share information about 2 studies “that are really changing the paradigm about treatment for C. diff infection.” Conducted between 2005 and 2007 at several sites in the United States, Canada, Europe, and Australia, a set of phase 3 multi-centered double-blind randomized control trials of adult patients with primary C. diff infection looked at the results seen in patients receiving either tolevamer, oral vancomycin, or oral metronidazole.
“We followed these patients for 4 weeks and the primary endpoint was clinical success which was defined as resolution of diarrhea and absence of severe abdominal pain for more than 2 days, including day 10,” Dr. Sandora explained. “We also looked at some other endpoints including time to resolution of diarrhea and recurrence of C. diff infection.”
The study yielded several interesting findings. First, tolevamer was inferior to both metronidazole and vancomycin, leading the team to conclude that it should not be used to treat C. diff infection. Conversely, a statistically increased clinical success rate was found in patients receiving vancomycin compared with those receiving metronidazole. In addition, those patients receiving metronidazole experienced more adverse events related to the study medication than those who received vancomycin. Despite some limitations to the study, the findings “were very powerful and influential,” according to Dr. Sandora. So much so that now, the revised C. diff guidelines state that either vancomycin or fidaxomicin is recommended over metronidazole for the treatment of C. diff in adults.
“So, the question is should we also do this for kids?” he asked.
Currently, there are no clinical trials being conducted that are comparing the 2 treatments in a pediatric population. “We have these data showing that every infectious disease physician who responded to that survey would choose metronidazole; most of you in this room voted for metronidazole. We have years of accumulated good experience with giving patients metronidazole for C. diff infection,” he said. Therefore, although this is a debatable matter, it was decided that for an initial episode of mild C. diff in children it can be treated with vancomycin or metronidazole; it’s up to the physician.
“However, we desperately need better and more data about the comparative efficacy of these 2 agents in kids to know really the best way to go when we’re managing initial C. diff infection in children,” he concluded.