Home Q&A Hospitalists miss the mark on handheld ultrasound

Hospitalists miss the mark on handheld ultrasound

June 2008

Published in the June 2008 issue of Today’s Hospitalist

SHOULD HOSPITALISTS be in the business of using hand-carried ultrasound to supplement the information they gather via the physical exam? And even if they want to use hand-held ultrasound on the wards, can they learn what it takes to acquire and interpret images?

While previous research has offered limited evidence that the answer to both questions might be “yes,” no study had ever directly examined the use of hand-carried ultrasound by hospitalists. But a recent study did exactly that “and came away with less-than-stellar conclusions about the ability of internists to use the technology.

The article, which appeared in the November 2007 issue of The American Journal of Medicine, had 10 hospitalists perform an average of 35 hand-carried echocardiograms. When the hospitalists’ performance was compared to that of certified echo technicians, they fared considerably worse on image acquisition and interpretation.

When the hospitalists were compared to senior cardiology fellows, they fared a little better. But while their skills approached the level of the fellows’, the hospitalists still didn’t match up.

Today’s Hospitalist talked to Eric E. Howell, MD, one of the investigators and director of the collaborative inpatient medicine service at Johns Hopkins Bayview Medical Center in Baltimore. Dr. Howell was one of the hospitalists there who underwent training in hand-carried ultrasound.

How did you hope that giving hospitalists hand-carried echocardiography would be helpful?
While it wouldn’t be helpful for every patient, we thought this might be one of those tests that a lot of patients could use. There are many data showing that the physical exam isn’t as good as it used to be and that hand-carried ultrasound gives you a lot of information that the physical exam can’t.

So one question was whether hospitalists should be screening everyone with hand-carried ultrasound. For this study, we decided to ask whether hospitalists can actually learn the skills.

Why was it so difficult for hospitalists to learn?
It’s a skill that requires training. Some of the images, like the short axis and long axis views, were relatively easy to acquire. But doing the four-chamber view was fairly difficult.

Every patient is different, everyone has a different body habitus and there is no one place you can put the probe to find the image of the heart. It’s like using a stethoscope, and we all know that you don’t learn to use the stethoscope after placing it on 35 patients. Looking for mitral regurg and things of that nature is probably going to require 100 studies.

Was it more difficult to learn the procedures than you expected?
We knew that some things would be difficult, but we thought that others would be fairly easy. We thought LV function, for example, might be fairly straightforward because there had been previous studies suggesting that medical students and nurses can discover LV dysfunction. It turns out that they do that with a high number of false positives, and we were trying to narrow down the false positives and false negatives.

While we could learn the skills, we weren’t as good as cardiologists or even the echo techs. It’s very interesting that echo techs are excellent at what they do.

Did echo techs feel threatened by the experiment?
Not at all. Our purpose was never to replace the platform echocardiogram. Our thought was that as a hospitalist, you may be able to do a bedside echo and if there is an abnormality or even a normality, you would back it up with a platform echo.

We don’t think this will supplant echo technicians at all, and definitely not cardiologists. We thought that hand-held echocardiography might be a bridge, but we were also concerned that if we produced too many false positives, we might be actually ordering more echocardiograms.

So you were thinking of this as a rule-out process, similar to stress testing?
Yes, if we had fairly low suspicion and everything looked normal, we would document that and do no further testing. But high-risk patients would certainly get a follow-up echocardiogram.

Were you surprised by any of the results?
One surprise was the time that it takes to set up the machine and to acquire the images. The other is that I didn’t know any of the patients, because they were study subjects, not my own patients. Patients really liked looking at their hearts, so it provided an opportunity to develop a physician-patient relationship in a very unique way. Besides giving hospitalists more data, it could be a big patient satisfier.

Are there any financial upsides of having hospitalists do ultrasound?
We never actually billed for any of these procedures, and we don’t have any plans to. So the question really is whether it improves patient care, and if you actually make patient care more efficient. If the answer to those questions is ” yes,” then at some point in the future somebody may actually pay us to do these procedures, like they do in the ED now. For now, however, we’re assuming that we would not be paid.

How about other benefits of having hospitalists use ultrasound?
One of the goals is to make hospitalists’ jobs more fun. Could having hand-carried ultrasound make hospitalists feel special because they have something that really identifies them as a specialist? ED doctors do portable, hand-carried ultrasound, and it’s fun.

Do you have plans for more research on the topic?
There are a couple of ideas on the table. One is to focus on the things that worked well, like the fact that you can find the E/A ratio fairly easily, and then tailor hospitalist training to learn those skills and apply them clinically.

For example, cardiologists say that in a subset of patients, the E/A ratio is a good predictor for patients who currently have heart failure. Maybe we can use that as a way to measure whether or not the patient is in heart failure.

Edward Doyle is Editor of Today’s Hospitalist.