Multiply Recurrent C difficile Infection Rates Sharply Rise

Ricki Lewis, PhD

July 03, 2017

The incidence of multiply recurrent Clostridium difficile infection (CDI) is increasing at more than four times the rate of CDI in general, according to results of a retrospective cohort study published online today in the Annals of Internal Medicine.

CDI affects nearly 500,000 individuals annually in the United States, at a cost of more than $5 billion. Recurrence within 8 weeks of initial infection is common, and some patients suffer more than one event, termed multiply recurrent CDI (mrCDI). According to a 2008 meta-analysis, 13% to 50% of patients with CDI had one or more recurrences (Garey et al. J Hosp Infect. 2008;70:298-304). At the same time, drug-resistant strains of C difficile are arising and may be related to the increase in mrCDI, the authors suggest.

New drugs, such as fidaxomicin and bezlotoxumab, are used to treat mrCDI, as is fecal microbiota transplantation (FMT). However, little is known about the epidemiology of mrCDI and risk factors — knowledge that might better inform initial treatment, including whether a patient is a good candidate for FMT.

FDA guidance on the use of FMT to treat CDI has been shifting, as reported by Medscape Medical News, but clearly requires obtaining informed consent.

In the new study, Gene K. Ma, MD, from the University of Pennsylvania Perelman School of Medicine in Philadelphia, and colleagues analyzed records for 38,911,718 commercially insured patients, for the years 2001 to 2012, in the OptumInsight Clinformatics Database, which includes medical and pharmacy data. They sought to evaluate mrCDI incidence and its relationship to overall CDI increase, and to identify risk factors.

The researchers used the standard definition of mrCDI as a patient who received at least 3 courses of antibiotics for the infection (oral or intravenous metronidazole, oral vancomycin, or oral fidaxomicin), with each course lasting at least 14 days but no longer than 56 days after the start of the previous antibiotic course. At least one antibiotic must have been vancomycin or fidaxomicin.

Patient characteristics considered included age, sex, calendar year of diagnosis, geographic region or state, antibiotic use other than for CDI, use of proton pump inhibitors or corticosteroids within 90 days before CDI diagnosis, and history of inflammatory bowel disease, chronic kidney disease, or diabetes mellitus.

Of the 38,911,718 patients, 45,341 had at least one episode of CDI, and 1669 had mrCDI. During the study period, the annual incidence of CDI increased by 42.7%, from 0.4408 to 0.6289 cases per 1000 person-years (P = .004). However, the annual incidence of mrCDI increased by 188.8% during the same period, from 0.0107 to 0.0309 cases per 1000 person-years (P < .001). Known risk factors for CDI did not affect the observed increase in mrCDI, according to the authors.

Patients with mrCDI were older (median age, 56.0 vs 49.0 years for CDI; adjusted odds ratio [aOR] per 10-year increase in age, 1.25; 95% confidence interval [CI] 1.21 - 1.29) and were more likely to be female (63.8% vs 58.7%; aOR 1.24; 95% CI, 1.11 - 1.38). They were also more likely to have used antibiotics (72.3% vs 58.8%; aOR, 1.79; 95% CI, 1.59 - 2.01), proton pump inhibitors (24.6% vs 18.2%; aOR, 1.14; 95% CI, 1.01 - 1.29), or corticosteroids (18.3% vs 13.7%; aOR, 1.15; 95% CI, 1.00 - 1.32) within 90 days of CDI diagnosis.

Increased risk for mrCDI was associated with chronic kidney disease (10.4% vs 5.6%; aOR, 1.49; 95% CI, 1.24 - 1.80) and diagnosis in a nursing home (2.1% vs 0.6%; aOR, 1.99; 95% CI, 1.34 - 2.93).

The researchers suggest that first-line treatment might be reexamined for patients with initial CDI who have the risk factors of proton pump inhibitor or non-CDI antibiotic use and/or nursing home setting. Treatment options might include different antibiotics or FMT.

No other examined factors were associated with the increase in incidence of mrCDI. The investigators hypothesize that because even after adjusting for use of proton pump inhibitors and antibiotics, calendar year of diagnosis remained strongly associated with mrCDI, perhaps the rise in incidence suggests a change in the pathogen, such as the emergence of the North American pulsed-field gel electrophoresis type 1 strain.

A limitation of the investigation is the inability to identify a cause or causes for the increase in incidence of mrCDI, as well as the use of administrative data. Nevertheless, the findings indicate "an increasing demand for therapies to treat mrCDI," the authors write.

In an accompanying editorial, Sameer D. Saini, MD, and Akbar K. Waljee, MD, from the VA Ann Arbor Center for Clinical Management Research and the University of Michigan, caution that use of standardized case definitions, considering only patients with insurance, and not including those on Medicare, might have underestimated the incidence of CDI, especially by omitting older individuals. The study also did not include some specific risk factors associated with recurrent CDI, such as prior fluoroquinolone use.

Dr Saini and Dr Waljee write that better understanding of the host-response relationships of specific strains of C difficile and the influences of risk factors could enable prediction models that could lead to a more individualized approach to care. "But to do so, we must first have a better understanding of mrCDI, its scope and epidemiology, and its associated risk factors. The study by Ma and colleagues begins this important work."

The researchers and editorial writers have disclosed no relevant financial relationships.

Ann Int Med. 2017;167:152-158. Article, Editorial

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.