Clostridium difficile Transmission: It's Complicated
Abstract & Commentary
By Richard R. Watkins, MD, MS, FACP
Division of Infectious Diseases, Akron General Medical Center, Akron, OH; Associate Professor of Internal Medicine, Northeast Ohio Medical
University, Rootstown, OH
Dr. Watkins reports no financial relationships relevant to this field of study. This article originally appeared in the November 2013 issue of Infectious Disease Alert.
Synopsis: This new study provides evidence that the widely
accepted paradigm of horizontal Clostridium difficile transmission
may not be valid.
Source: Eyre DW, et al. Diverse sources of C. difficile infection identified
on whole-genome sequencing. N Engl J Med 2013;369:1195-1205.
The prevailing theory of clostridium difficile transmission is that most cases occur after recent exposure to symptomatic patients in health care settings. Infected patients are known to shed large numbers of C. difficile spores and current infection control recommendations focus on preventing spore transmission from the environment through contact precautions and decontamination of surfaces and equipment. However, the rate of C. difficile infection (CDI) continues to increase, bringing into question the effectiveness of these methods.
A new study by Eyre and colleagues provides evidence that the widely accepted paradigm of horizontal C. difficile transmission may not be valid. Over a period of 3.6 years, the investigators performed whole genome sequencing on more than 1200 clinical isolates of C. difficile from the Oxford University Hospitals. Any two cases were considered to be linked if they met two conditions: no more than two single-nucleotide variants (SNVs), and the patients had direct or indirect contact while one was symptomatic or contact occurred within a 12-week incubation period.
The analysis determined only 35% of the isolates were genetically linked to a prior case, while 45% had more than 10 SNVs and therefore were not related. Patients with genetically linked isolates tended to be older than those with genetically distinct isolates (median age 81 years vs 76 years, respectively; P < 0.001). Moreover, 38% of the linked cases had ward contact with the previous genetically related case while 9% shared the same time in the hospital but were never on the same ward. Interestingly, no hospital-based contact could be established for 46% of the patients, implying diverse sources of acquisition. For instance, some of the patients were in the same medical practice (10%) or lived in the same postal-code district (11%) as previous cases, but no pairs of patients with two or fewer SNVs attended the same outpatient clinic on the same day. Among the 45% of cases that represented transmission from sources other than symptomatic cases, it is reasonable to conclude their CDI was acquired from either asymptomatic individuals or some other environmental reservoir (i.e. food, animals, or surfaces).
There are a few limitations to the study that require mention. First, the toxin testing used by Eyre and colleagues has been largely supplanted by more sensitive assays (e.g., polymerase chain reaction or nucleic acid tests), which probably led to an underestimation of C. difficile incidence in the local regions. Second, the institutions that participated in the study housed patients in four-bed bays, an arrangement that is unusual in the United States. Since the majority of patients in U.S. hospitals are in private rooms that afford less opportunity for patient-to-patient contact, the rate of C. difficile transmission is likely to be different between the two countries. Third, the study was conducted during a period when infection control precautions to limit the spread of C. difficile were widely practiced, which likely reduced the impact of symptomatic patients. Finally, it is possible that C. difficile has other still-unidentified reservoirs in the environment or hospitals that are important for transmission. In the study, the authors did not perform whole genome sequencing on any strains from extended-care facilities, which are known to be sources for outbreaks of C. difficile.
COMMENTARY
This study is significant because it provides evidence that in a majority of cases, CDI is not transmitted by symptomatic patients. Instead, asymptomatic individuals or other sources are likely to be the main perpetrators. This finding compels us to re-examine infection control protocols whose goal is to limit the transmission of multidrug-resistant pathogens like C. difficile. This is not to say we should abandon basic control methods since the infrequent transmission from symptomatic patients in the study hospitals may attest to their effectiveness. Rather, future studies that elucidate novel routes of transmission can identify currently unknown sources of CDI. For example, should all patients who are admitted to the hospital be screened for C. difficile? If so, what should happen to asymptomatic carriers especially since isolation in general is unpopular with patients, their families, and health care workers?
Another takeaway point from this study is that perhaps more emphasis should be placed on minimizing disruptions to the gut microbiome and its restoration (i.e., through probiotics) should be a priority. Although physicians and the public already have more than enough reasons to limit the use of unnecessary antibiotics, the threat of acquiring CDI is a valid concern and should be taken into consideration whenever these agents are prescribed. As Eyre et al note, the rate of fluoroquinolone and cephalosporin usage in the UK fell between 2006 and 2009, and during this period, active restriction of these agents in one Scottish hospital resulted in a relative reduction of 77% in the incidence of CDI. Thus, reducing the susceptibility of patients to CDI may be more effective than lowering transmission rates. Finally, this study demonstrated the usefulness of whole genome sequencing in determining the transmission of one major disease and this method holds great promise for uncovering the mechanisms of transmission for other serious pathogens.
