Echo for Acute MI Ups Hospital Stays and Costs, Without Boosting Outcomes

Imaging has become so routine that one author advises clinicians to define their diagnostic question before ordering a test.

Echo for Acute MI Ups Hospital Stays and Costs, Without Boosting Outcomes

Greater use of transthoracic echocardiography in patients with acute myocardial infarction doesn’t seem to improve clinical outcomes but is associated with greater hospital costs and longer lengths of stay, according to a new retrospective cohort study.

“Echo is a wonderful, useful test,” lead author Quinn Pack, MD (University of Massachusetts Medical School-Baystate, Springfield), told TCTMD, adding that “in the right patient” it can give important diagnostic and prognostic information. “But if it's just ordered for a routine indication because somebody said [they] should and because the box was checked and because that's just what we always do—it's our routine, it's our culture—that kind of thinking is, I think, leading to waste.”

This approach applies to most of medicine, agreed Christine Jellis, MD (Cleveland Clinic, OH), who was not involved with the study. “We have to target the tests that we use—be cognizant of their utility, how it will change management—and while cost has to come into consideration, I think the main driving factor is patient outcome and whether the test is going to be clinically appropriate and clinically useful,” she told TCTMD.

For the study, published online this week in JAMA Internal Medicine, Pack and colleagues looked at close to 100,000 patients hospitalized for acute MI at 397 US hospitals in 2014, 70.4% of whom received at least one transthoracic echocardiogram. Nuclear cardiac imaging (4.8%), ventriculography (1.7%), and cardiac MRI (0.3%) were uncommon. Compared with hospitalized patients who did not have an echocardiogram, those who did were more likely to have heart failure, receive ICU care, and receive noninvasive ventilation, invasive ventilation, inotropes, vasopressors, balloon pumps, and nuclear imaging studies.

Observed echocardiography rates ranged from 2% to 95.3% across hospitals, with a median of 73.9%. After adjustment, there were no differences observed between the hospitals in the highest and lowest quartiles of echocardiography use with regard to inpatient mortality (OR 1.02; 95% CI 0.88-1.19) or 3-month readmission (OR 1.01; 95% CI 0.93-1.10). Notably, patients treated at hospitals in the highest quartile of echocardiography use stayed 0.23 days longer in the hospital (P = 0.01) and their mean admission costs were $3,164 higher (P < 0.001) than those in the lowest quartile.

Ritual vs Thought

In his experience, Pack said that the ordering of echocardiograms has become “so ritualistic that nobody stops and actually thinks: ‘Are we going to get something from this?’” While it’s an easy test to order, it’s noninvasive, and “there’s almost no risk to it,” all of this “makes it so pervasively easy to order that we have become thoughtless in our approach to using this study,” he said.

Echocardiography is a diagnostic test, he stressed. “You use diagnostic tests in people in whom you have a diagnostic question. But if you don't have a diagnostic question, then you shouldn't be using the test. And if you can deduce the information you might get from a test from careful clinical assessment, then you can save yourself the time and the money that the test would waste.”

Think before you order an echo, and if you know you're not going to get anything from the test, then don't order it. Quinn Pack

While certain guidelines recommend that all acute MI patients should undergo echocardiography, Pack said he disagrees. “There are going to always be some circumstances where you're not going to gain anything from doing the echo, and so really, if there was just one takeaway, it's just: think before you order an echo, and if you know you're not going to get anything from the test, then don't order it.”

Instead, he recommends first ordering an N-terminal pro-B-type natriuretic peptide (NT-proBNP) to help guide whether an echocardiogram is necessary, and additionally a good physical exam. “[These] are much less expensive than an echocardiogram, particularly the physical exam, which is essentially free,” Pack said, adding that more research still needs to be done focusing on who would benefit most from extra diagnostic screening.

“A test all by itself will not gain you anything for patient care unless the test results teach you something that changes the way you manage the patient,” he said.

What About ‘Bedside Echos’?

Commenting on the study for TCTMD, R. Sacha Bhatia, MD (Women’s College Hospital, Toronto, Canada), said he was initially surprised to see that it was done, given how routine echocardiography is after acute MI. However, “we often tend to do things as part of a routine but haven't necessarily looked to see if they add significant value,” he said. “It was certainly interesting to me to look to see that maybe what we need to be looking at is what are the impacts both in a positive and negative way for even routine things that we're doing to make sure that we're very evidence-based in our decisions around things that we just take for granted.”

It's hard to know from the data, however, just how sick the studied patients were on an individual level, Bhatia said, adding that more research is needed before modifying any echocardiography practice in acute MI. “The other thing we just also don't know is how the echos . . . themselves change the way that a patient was treated once they left hospital,” he said. “One of the things that we don't know is if you had a patient who had a big infarct, for example, and we found that out in hospital or they had valvular disease or they had some other finding, how did that change management? Were medication prescriptions changed? Did they go see a heart failure specialist? Did it change their management at all? And that's really important because the whole idea of diagnostic testing is to influence the way we treat patients.”

Jellis called the study results “thought-generating” and also agreed that more information is needed before changing practice. “In a study like this, there's obviously the potential for a lot of confounding, but I think I wonder if there were still perhaps other factors that were not easily recognizable that meant that the patients who had echos were perhaps a sicker group or more complex group perhaps in sort of higher quaternary-type hospitals,” she said. “That may have influenced the fact that some of these patients ended up with echos compared to patients who were less sick and perhaps in more smaller or regional centers where access to echocardiography was less readily available.”

Additionally, Jellis mentioned the increased availability today of “bedside echos,” or point-of-care ultrasound; these may not be billed for directly and thus might not be captured in a database. “Maybe there was a proportion of patients who had bedside echos that were not able to be accounted for and that that may have influenced management in a way that we don't know,” he said.

  • Pack reports receiving grants from the National Heart, Lung, and Blood Institute during the conduct of the study.