Asymptomatic C. difficile Triples Diarrhea Risk

Major Finding: After adjustment for age and antibiotic use, the relative risk of diarrhea in C. difficile–colonized patients was more than 3-fold greater (odds ratio 3.3) than in noncolonized patients; the risk for clinical diagnosis and treatment was 10-fold greater (OR 10.1).

Data Source: A follow-up study of 320 hospitalized patients.

Disclosures: Dr. Leekha reported having nothing to disclose.

ATLANTA — Hospitalized patients who were asymptomatically colonized with toxigenic Clostridium difficile had a significantly greater risk of developing clinically significant C. difficile infection and diarrhea, compared with noncolonized patients, in a study of 320 adults admitted to one hospital.

The finding is contrary to previous reports and should be explored further, Dr. Surbhi Leekha said at the Decennial International Conference on Healthcare-Associated Infections.

“The take-home is that there may be an association between initial colonization at the time of hospitalization and the subsequent development of diarrhea, but we cannot determine based on these results whether the increased risk is a direct consequence of that colonization or whether colonization is a marker for other [factors] with that patient, such as severity of illness, immune status, or recurrent hospitalization, that in turn predispose to CDI,” she concluded.

Asymptomatic colonization with C. difficile occurs in approximately 8%–20% of hospitalized patients, who then can serve as “reservoirs” contributing to nosocomial transmission. At least three prior studies have suggested that these patients are not at risk for symptomatic disease and may even be at lower risk, provided a “critical period” of about 1–2 weeks has passed following acquisition of the organism (Clin. Infect. Dis. 1994;18:181–7; Lancet 1990;336:97–100; Lancet 1998;351:633–6).

A previous part of the current study had enrolled 320 adults admitted to Saint Marys Hospital, Rochester, Minn., who had stool specimens tested for toxigenic C. difficile using polymerase chain reaction assay within 5 days of admission between March 1 and April 30, 2009. Of these patients, 30 (9.4%) were found to be colonized without symptoms. Factors associated with C. difficile colonization included recent hospitalization (relative risk 2.3) and chronic dialysis (RR 7.6). Dr. Leekha, of the Mayo Clinic, Rochester, reported those findings at the 2009 meeting of the Infectious Diseases Society of America.

The current follow-up was done 3–4 months after determination of C. difficile colonization status. Of the 320 asymptomatic patients who had a history of C. difficile infection (CDI), 12 were excluded. Of the remaining 308, follow-up information was obtained via telephone calls and chart reviews for 272 patients. Of those, 25 were colonized and 247 were not.

Those who were colonized were significantly more likely to have been hospitalized recently (64% vs. 36%), to be on chronic hemodialysis (12% vs. 2%), to be on proton pump inhibitors (52% vs. 37%), and to have recent corticosteroid use (32% vs. 15%). Antibiotic use and subsequent hospitalization during the follow-up period did not differ significantly between the two groups, Dr. Leekha reported.

Diarrhea developed in 32 (12%) of the 272 patients, including 7 of the 25 who were colonized (28%) and 25 of the 247 who were not (10%).

Clinical diagnosis and treatment for CDI occurred in 8 of the 272 patients (3%), including 4 of the 25 C. difficile–colonized patients (16%) and 4 of the 247 noncolonized patients (2%). After adjustment for age and antibiotic use, the relative risk of diarrhea for colonized patients was more than 3-fold greater (odds ratio 3.3), compared with noncolonized patients; the risk for clinical diagnosis and treatment was 10-fold greater (OR 10.1), she said.

Of the eight patients who were treated for CDI, two—one colonized, one not—did not have diarrhea and therefore would probably not have been tested for C. difficile in a usual clinical setting. All four of the treated noncolonized patients had used systemic antibiotics within 2 weeks of symptom onset. But interestingly, of the four treated colonized patients, one had not used antibiotics within 2 weeks and one had undergone outpatient ocular surgery and had used only ocular antibiotics. Whether those play a role in CDI is unknown, Dr. Leekha commented.

It also is not known why the findings of this study differ from those of previous studies, which had suggested a “protective effect” of colonization. Carriers were found to have had higher levels of toxin A IgG, compared with those with symptoms, which was postulated to play a role. Further elucidation of host factors is needed, she said.