Staff screening for MDROs: closing Pandora’s Box

A brave study from the Palmore/Frank group at NIH has opened the Pandora’s Box that is screening staff for MDROs, and, I’m delighted to say, firmly closed it with their findings! Only 3% of staff carried ESBLs, one carried a CPE, and none carried VRE, and this despite extensive contact with MDRO patients for many of the staff sampled!

There is no doubt that staff can contribute to the transmission of MDROs occasionally. But are staff ‘surfaces’ that most often become transiently contaminated, or do staff regularly acquire medium/long term colonisastion? And is staff carriage a regular contributor to cross-transmission? The gnarly question is whether routine sampling of staff should be performed. Would this identify a lot of MDRO staff carriers? And what would you do with these carriers if you found them? With MRSA, there is the option to suppress / decolonise using topical agents. But with intestinal MDROs (like VRE and MDR-GNB), there is no effective suppression / decolonisation therapy. So, would you remove them from clinical duties?

The study was performed in the 200 bed NIH Clinical Center, and recruited 400 healthcare personnel (HCP) with patient contact / lab staff with clinical specimen contact, and 400 staff without recent direct patient contact as a comparator. All participants were asked to submit two perirectal swabs separated by a few days, and complete a questionnaire. Interestingly, the staff received payment to incentivise participation! (I hope that the staff who only half did the study e.g. by only submitting one of the requested two swabs, or not completing the questionnaire only received half the payment!) Also, the dataset was anonomysed, so that carriage status could not be traced to individual staff.

4.0% of HCPs and 3.2% of the control participants grew ESBLs, none grew VRE, and one of the control participants grew a CPE. No significant risk factors were associated with ESBL carriage. The HCP group was made up of clinical and laboratory staff. It would have been good to see a breakdown of carriage rate between these two groups. I would expect acquisition of ESBLs from agar plates in a lab to be considerably rarer than acquisition of ESBLs in the far less controlled clinical environment, involving direct contact with colonised patients!

This is a really low ESBL carriage rate – the sort of rate you’d expect to see from healthy volunteers in the community, rather than staff working in a clinical setting where ESBLs are commonplace. Which brings me to one of the key limitations of the study: what exactly is a ‘perirectal swab’? Ask ten people to collect two ‘perirectal’ swabs and you would probably get 20 slightly different areas sampled! In a recent study, we found that sampling the perineum only detected half as many ESBLs as sampling the rectum, and that staff-collected swabs detected more ESBL-E than patient-collected swabs (especially for perineal swabs); these issues may well have influenced the ESBL detection rate. The methods suggest that specific guidelines were provided in exactly how to collect the samples, but I suspect there would still have been considerable variability in sample quality. This is reinforced by the finding that 20% of the samples failed to grow faecal flora. The study also doesn’t specify when the staff were asked to collect the specimens. Was this during a shift for clinical staff, or at the start of the shift? Or was this not specified?

Quibbles, questions, and limitations aside, this is a great study, that reinforces the status quo that routine sampling of clinical staff would not be useful in preventing the transmission of MDROs, aside from protracted outbreak settings where the epidemiology indicates the involvement of staff carriers.

ps Thanks to Tara Palmore for sending me the article PDF within about 10s of me asking!

Image: Flikr.


One thought on “Staff screening for MDROs: closing Pandora’s Box

  1. Some time ago we investigated (using math modelling) under what conditions HCW would contribute importantly to transmission of bugs in health crae settings (see, BMC ID). Nothing unexpected: “The potential role of persistently colonized health care workers in … transmission depends on the proportion of persistently colonized health care workers and the likelihood per colonized health care worker to transmit.” So, for gut-colonizing MDRGNB that proportion is low, and – if they keep their trousers on – the per person likelihood to transmit is also low. That should be enough to convince those that always come up with this suggestion….


Leave a Reply

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

You are commenting using your 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