Skip to main content
Premium Trial:

Request an Annual Quote

Q&A: UK Hospitals Find Cepheid's Xpert MRSA Test at Point of Care Can Shave Hours Off Time-to-Result


brenwald.jpgHospitals are looking for ever more rapid methods to screen new admissions for methicillin-resistant Staphylococcus aureus, since identifying MRSA-positive patients can rapidly reduce the chance that the infection will spread to other patients.

While real-time PCR-based MRSA tests promise to greatly reduce the amount of time required to get a result — requiring several hours as opposed to two to three days for culture-based methods — these systems have so far been relegated to hospital pathology laboratories with trained technicians. And because samples must be transported from hospitals to these labs, the actual time to result still sometimes takes a full day or longer.

Researchers in the UK, led by Nigel Brenwald in the department of microbiology at Birmingham's City Hospital, recently looked into the feasibility of bringing the PCR-based test into the point-of-care setting in order to further reduce the time-to-result.

In a paper published in the November issue of the Journal of Hospital Infection, Brenwald and colleagues describe how they installed Cepheid's GeneXpert system on four wards within the Sandwell and West Birmingham Hospitals NHS trust to see how well the hospital staff — after training — could perform the company's Xpert MRSA test compared to lab staff.

The study found 97.5 percent agreement between the results obtained on the ward and the results from the lab, and an average reduction in time of 10 hours in the wards as compared to the lab — results that indicate that the PCR-based test could be used routinely in the emergency-admissions setting, according to the authors.

"To our knowledge, this is one of the first studies looking at the potential for point-of-care [real-time PCR-based] testing for MRSA," Brenwald and colleagues write in the paper. "We found that all grades of healthcare staff involved in the study, once trained, could easily process samples and obtain results using the Xpert MRSA test."

PCR Insider recently spoke to Brenwald about the study and its implications for point-of-care MRSA testing. An edited version of the interview follows.

What has prevented PCR-based MRSA testing from being used at the point of care until now?

Really, there's nothing on the market that's applicable to the point-of-care setting. Most PCR tests would be done in the laboratory. The thing about the Cepheid test — the Xpert MRSA test — is that it's simple enough for anyone, really, to be able to perform the test.

Normally in the laboratory you'd have separate rooms preparing DNA from the sample, setting up the reaction, doing the actual thermal cycling, and then reading the test. With the Xpert MRSA test, you've basically got a laboratory inside a cartridge. The person performing the test would collect the nose swab, elute the swab into a buffer, do a vortex thing, and then simply transfer that material into the cartridge. They add two reagents, and that's essentially the test setup. It takes a few minutes to do that. And then the cartridge is put in the GeneXpert machine and everything — all the processing and reading — is done by the machine. Essentially an answer comes up in 70 minutes, so we felt that was sufficiently simple for someone who's been trained on the ward to be able to perform it, which is what we found.

[The amount of training that was required varied] depending on the unit. In the critical care unit they were obviously highly trained nurses, and in some of the admissions units they were below nurse level — they were healthcare assistants. We went out and trained them and they all managed to do the test

So right now in the UK there are no regulatory restrictions that would prevent you from rolling this out at the point of care?

In terms of who can do the test there aren't, but obviously at the laboratory we monitor how the test is being performed. It's still really in our control — we monitor the machine, make sure that maintenance is carried out, clean the machine, [and] do quality assurance on the machine. So we wouldn't just let the nursing staff have the machine and say, 'There you are, do the test.' I think that would be a recipe for disaster.

But there are guidelines that we have to follow in terms of using a point-of-care test, and we actually had to go through various committees to allow us to do that. So there are some local regulatory things to worry about, but not on a national scale.

[ pagebreak ]

We make the nurses responsible for the result they reduce, so that focuses their mind on getting the right answer.

I'm not sure when Cepheid will bring out the kits with the reagents in the cartridge. They've been talking about that for a little while now. Then it might be sufficiently simple to do to get [Food and Drug Administration] approval and allow the US to use it as a point-of-care test [Currently the test is approved for moderate-complexity CLIA labs in the US — Ed.].

In terms of the study, both the ward and the lab used the same GeneXpert platform, so any discrepancy was due to operator error?

That's right. It was just to see if [the hospital staff] could produce the same result as we would in the lab. Here in the lab, we're used to using this equipment — even basic things like pipettes we'd be very familiar with — but in the ward setting some of the nurses we were training had never seen a pipette, so we had to tailor our training to get them to use that sort of equipment.

But they did remarkably well. We didn't expect that. We thought they'd have a lot more trouble getting a result. So that was quite reassuring that they were able to do that.

You found a 97.5 percent agreement between the tests performed in the ward and the tests performed in the lab. It sounds like that was better than what you expected.

It was a lot better. I think even Cepheid were quite shocked that essentially non-scientific staff could do that, so it was quite reassuring.

The actual test itself is probably the easiest bit of the whole exercise, though. Taking the specimens from patients and entering the correct data on the machine has actually been more of a problem. All of the data is captured in the laboratory so we know what's going on in the ward. It all comes to a server in the laboratory. And the way [the staff] enters in the patient details has to be done in a very set way, and that's been the hardest bit — trying to get that side of it. The actual test itself has not proved very difficult.

It seems like a lot of vendors are trying to develop platforms that are targeted at the point-of-care setting.

Certainly in the UK there are people desperate to get simpler point-of-care tests for things like MRSA and influenza even. So I think we'll hopefully see in the next few years quite a few point-of-care tests. The benefits, as we outlined in our paper, are that you just get a much faster result. You don't have to wait for samples to be transported to laboratories. Certainly in our trust we have hospitals that may be a 20 minute drive away from the laboratory, so to get a swab from the patient there — it's got to be transported and then ported to the lab. Whereas if you can do it next to the patient, that's obviously saving a lot of hours, which can be used to decolonize a patient, if you're talking about MRSA..

So that's the potential benefit of point-of-care testing.

The paper mentions that you're looking at the impact of this on patients. How far along are you in that work?

We're doing a major audit at the moment on patients, which is an ongoing thing that hopefully we'll have finished around March.

We have seen some benefits in our critical care units. They're fairly small units, but we've seen a big drop in the acquisition of MRSA by patients, which is obviously what we hoped would happen because if you can pick up a patient before they go into the unit, if you know they're MRSA positive they can be isolated and decolonized immediately before they go onto the unit, and I think we've seen the effect of the isolation and treatment on the infections that people acquire when they're on the unit. We're certainly seeing drops in that.

[ pagebreak ]

It's a little bit early days to say whether we've seen a complete drop in MRSA, but we've certainly seen a drop in MRSA bacteremia since we've started the study.

So that's good news, but in terms of the actual nuts and bolts of whether these patients are actually getting decolonized earlier, we're still auditing that part of it.

Would you say that the performance of the point-of-care test was good enough to replace lab-based testing?

It probably wouldn't replace lab testing totally. We focused on emergency admission patients, and obviously there is a whole other group of elective patients who know when they are going to come in, so we tend to use a culture method for that because we have a few more days to get an answer from those patients.

The PCR test is relatively quite expensive. Culture is a lot less expensive. So we tend to split patients into two groups: the elective ones get the culture, unless if they turn up for an operation and they haven’t been screened, and then we can use the PCR, and we would do that in the laboratory. Then the emergency patients get the PCR. So we're not using PCR totally on everyone as a point-of-care test. I think the lab probably still has a role in a lot of these things.

And obviously, doing PCR you're not isolating the organism, so any positives we get we then culture so that we have the organisms for susceptibility testing, and also if there's an outbreak we can then look at the strains and fingerprint the strains.

Are there any disadvantages or areas for further improvement that you'd like to see for the PCR-based MRSA assay in order to roll this out more broadly?

I think some of the MRSA tests we used were 70-minute tests. From a microbiology point of view that seems quite quick, but when you have a patient coming in on the ward, they need a lot of procedures done — MRSA testing is just one of them — so if we had a faster test that would help. They're more likely to be able to get the results before the patient leaves the admissions unit to go out on the ward. So speed is quite an important issue.

I know that Cepheid is looking at reducing the time down to about 50 minutes, which doesn't seem like a big difference but every little bit helps, really. And just making the test a bit simpler — not having to add reagents, for example, would make things a lot easier for people setting the test up.

With PCR, again, depending on the targets that are being looked for, certainly there is the potential for false positive results, which might occur with some strains. And then, again, if there are very small numbers there, you might miss a case, possibly. There are some false negative samples with PCR, but essentially you'd get that with any test that you use.

So probably speed and the ease of setting up would be the critical things for a point-of-care test.

The paper describes the training as being a bit of an arduous process. Would you have any recommendations for other hospitals, at least in the UK, who might be interested in adopting this technology at the point of care?

Yes, the training was quite arduous — particularly on somewhere like a critical care unit where all the nurses wanted to be trained. So that could be 70 staff. Just the way they nurse patients, they tend to allocate a member of staff to a patient, and there didn't seem to be anyone around who could do the test while another nurse did the nursing, so we did have to train a lot of nurses.

What we found is that training a trainer on the ward was the best way forward. So that's what we're instituting at the moment — maybe a handful of nurses who are trained up as the core trainers for each unit, and then they can handle the 70-plus staff and do the training. We would then be monitoring the trainers to make sure they're doing that correctly. They have a set procedure, a training manual that they had to follow before we let them loose on the other members of the staff.

[Also,] lab people tend to not work at night whereas critical care is working 24 hours per day, so patients could be entering at any time of day. So having core trainers on the ward makes life a lot easier. It means we don't have to get in at midnight to train the nurses.

Is there anything else that you've learned from this project that is worth noting?

I think it would be nice to see machines that are a lot faster, and maybe hand-held devices that you can use to do a quick test.

We've had the machines in the admissions units, where there are a lot of patients coming through. I think maybe as the machines get faster and smaller and applicable to sort of hand-held type devices, we may see them on more units. So each ward might have a machine for testing.

There's another hospital in the UK that's looking at having GeneXpert systems in more units — rather than just having them in the admissions unit where all patients have to come through, they would just have machines on the different wards so they'd have a lot more machines. That may be a better way forward, but it means more training, obviously.

The Scan

Study Links Genetic Risk for ADHD With Alzheimer's Disease

A higher polygenic risk score for attention-deficit/hyperactivity disorder is also linked to cognitive decline and Alzheimer's disease, a new study in Molecular Psychiatry finds.

Study Offers Insights Into Role of Structural Variants in Cancer

A new study in Nature using cell lines shows that structural variants can enable oncogene activation.

Computer Model Uses Genetics, Health Data to Predict Mental Disorders

A new model in JAMA Psychiatry finds combining genetic and health record data can predict a mental disorder diagnosis before one is made clinically.

Study Tracks Off-Target Gene Edits Linked to Epigenetic Features

Using machine learning, researchers characterize in BMC Genomics the potential off-target effects of 19 computed or experimentally determined epigenetic features during CRISPR-Cas9 editing.