Chased by an antibiotic-induced C difficile-shaped shadow

shadow-abx

A fascinating new JAMA Internal Medicine study suggests that being admitted to a room when the prior occupant had taken antibiotics increases the risk of the subsequent occupant of the same room developing C. difficile infection (CDI). Quite a few convincing epi studies have showed that admission to a room when the prior occupant was known to have a number of key pathogens (including C. difficile) increased the chance of acquisition for the subsequent occupant. But this study extends the ‘prior room occupancy’ concept into a new dimension!

The study looked retrospectively at a little over 100,000 pairs of patients who were sequential occupants of the same room in four New York hospitals between 2010 and 2014. 576 (0.6%) of the subsequent room occupants developed CDI. The subsequent room occupants who developed CDI had all of the risk factors for CDI that you’d expect (older age, receipt of antibiotics etc). Also, however, the receipt of antibiotics in the prior occupant significant increased the risk of CDI in the subsequent room occupant (0.43% of patients developed CDI when the prior occupant did not receive antibiotics vs. 0.72% when the prior room occupant did receive antibiotics). Receipt of antibiotics was the only factor in prior room occupants that was significantly associated with CDI in subsequent room occupants. In an important sensitivity analysis, prior room occupants with known CDI were excluded from the study. Still, the association with antibiotics in the prior room occupant remained significant.

So what’s going on here? It seems that the receipt of antibiotics in patients who were colonised by not infected with C. difficile resulted in more shedding of C. difficile spores into the environment, resulting in an increased risk of acquiring C. difficile in the subsequent room occupant. This idea fits in with data from other studies. My old research group in London published some work last year suggesting that the type of antibiotics given to a patient made a significant impact on the degree of environmental shedding of C. difficile spores, with fidaxomycin resulting in less contamination than vancomycin or metronidazole. This study was concerned with symptomatic C. difficile, but you can see how the same principle could apply to asymptomatic colonisation. There is an alternate (and not necessarily mutually exclusive) hypothesis that this association is driven by ‘collective microbiome’ alternations – that on a ward level, antibiotic exposures change the composition of everybody and everything’s microbiome on (patients, staff, visitors, environment), and this affects everybody’s risk of acquiring C. difficile!

A few key points to put this study into perspective though. The most important risk factors for developing CDI in subsequent room occupants were their own risk factors (e.g. receipt of antibiotics, age etc); the increased risk from the prior room occupant having received antibiotics was relatively small. Also, the prior room occupants’ receipt of antibiotics was coded as a binary variable – yes or no. There’s a lot more to learn here about how different antibiotics, multiple courses, and length of antibiotic exposure affect this risk.

The evidence that admission to a room previously occupied by a patient with MRSA, VRE, and A. baumannii is pretty compelling now (see this recent review). The most compelling evidence for me that the prior room occupant link is real and not some strange epi artefact (e.g. the sickest patients on the ward with the highest risk of acquisition are always placed in the same room), is that when you do a better job of cleaning and disinfection, the prior room occupant reduces or disappears. However, in the case of C. difficile, the prior room occupant link relies heavily on one study, which is rather small and can be challenged. Furthermore, there is some evidence from older studies including typing suggesting that an apparent prior room occupant link may be explained by something else. So, this is a welcome study to explore further the link between the prior room occupant and C. difficile. But a good follow-on study would be to see whether reducing the use of antibiotics in prior room occupants reduced the development of CDI in subsequent room occupants! In the reality though, this seems unlikely given the small effect size, and it’s probably better to focus on reducing CDI-inducing exposures in current inpatients (like unnecessary use of antibiotics), and improving cleaning and disinfection to deal with the prior room occupant link. These findings do, however, re-open the debate on whether we should be screening and isolating patients with asymptomatic C. difficile colonisation…

Image: Flickr.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s