HOSPITALISTS WHO ROUTINELY TEST for troponins or who call cardiologists in for every elevated level they find are putting the test at risk for overuse. That’s according to Jeffrey Michel, MD, interim chief of cardiology at Scott & White Memorial Hospital in Temple, Texas, and the lead author of a new study.
“Testing overuse can lead to more echocardiograms,” Dr. Michel points out. “It can also worsen outcomes if we subject people to treatments like blood thinners and antiplatelets or invasive procedures.”
The research, published in the May issue of the Journal of Hospital Medicine, found only a 3.5% rate of acute MI in more than 92,000 hospital visits associated with troponin testing. Further, 79% of elevated troponins were associated with other primary diagnoses, such as pneumonia and respiratory failure. The study was performed at 14 hospitals over one year.
“My advice would be to test only those patients likely to have had a heart attack.”
“Elevated troponins are probably a sign that systemically the whole patient is sick,” says Dr. Michel. “You can’t really separate organs out and say a disease involves only one organ. You can say the heart is involved in the illness, but it isn’t an MI.” That’s an important distinction, he says: While a suspected MI requires immediate cardiologist intervention, the other conditions do not.
“In our data, we’ve seen troponin tests done for patients having a UTI or for a patient in the ER who’s just getting a medication refill,” he points out. “It gets baked into the order sets.” And with an even more sensitive troponin test due out soon, the stakes are even higher. Dr. Michel talked to Today’s Hospitalist about the study results.
What concerns you most about the results?
The majority of elevated troponins were not from patients with heart attacks. That means the positive predictive value—the likelihood that you have the disease—is low.
Hunting expeditions to track elevated troponins because they may indicate a heart attack could lead to longer length of stay, unnecessary expense, time and worry, none of which change patient management. If you have pneumonia, you won’t be treated any differently if your troponin is elevated.
Why is there so much testing for non-MI patients?
It’s a tough job being in a busy ER or hospital service. And sometimes, people with heart attacks don’t show up with typical chest pain or chest pressure, especially women, so it makes sense for physicians to be on the alert. A negative cardiac troponin is a very good way to make sure patients are not having a heart attack. But some places take most patients with elevated troponin down to the cath lab to look for blocked arteries. They don’t find them, but they do a lot of procedures.
It’s similar in some ways to finding an elevated prostate specific antigen: Once it’s elevated, we worry that it’s prostate cancer, but it may not be.
Can guidelines help?
We have plenty of guidelines if the patient is admitted to the hospital with a likely MI. But we do not have guidelines for managing troponin elevation. We can’t say, ‘Don’t worry about it,’ because we know it’s associated with patients doing worse in their hospital stay. It’s the perfect storm of worry.
How will the new high-sensitivity troponin test affect testing trends?
The new test will be 1,000 times more sensitive than the current one, so it will be able to detect lower levels of troponin in the bloodstream and allow the ER doctor to diagnose an MI if the patient has had a heart attack. But the danger is that the positive predictive value will be even worse, meaning more sorting through even more patients with positive tests.
Hospitalists will see more elevated troponins for patients admitted from the ER with stroke, sepsis, pneumonia or kidney failure. If we don’t change how we use the test now, it will generate even more overuse when the new test is implemented.
When should hospitalists use the troponin test?
My advice would be to test only those patients likely to have had a heart attack. Specifically, look at patients with acute onset of chest pain and unexplained shortness of breath that isn’t accompanied by an obvious pneumonia or lung problem. A patient with new symptoms and an abnormal EKG would also raise my interest. In such patients, a negative troponin test is reassuring, and a positive one should be referred to a cardiologist.
What steps should hospitalists take before calling a cardiology consult?
Don’t be afraid to use your clinical judgment when the troponin is elevated. Look at the patient and ask yourself: Do I think this patient is having a heart attack? If the answer is “no” and you feel confident of that, document it.
For example, if the patient has sepsis, document it and talk to the patient. If he or she is not having chest pain and the EKG looks fine, you can say, “I don’t think there’s an acute MI. I think the elevation is from the sepsis and that connection is well known. There’s no reason to go further.”
But in a patient with unexplained troponin elevation who otherwise appears healthy—no stroke, sepsis or pneumonia—an echocardiogram can look for evidence of heart failure or pulmonary hypertension. Get one before calling a cardiologist.
You’ve developed a program at your hospital to help hospitalists better diagnose non-MI patients with elevated troponins. How does it work?
We did some education with hospitalists to explain that elevations are frequently seen in sepsis, pneumonia, stroke and kidney failure. We also review cases one-on-one to help doctors make decisions—without a consult—about whether there’s really a heart attack, and we have huddles with hospitalists once a week.
A nurse collects data on diagnoses related to elevated troponin. If a case is coded as an MI and the nurse feels that’s not the case, I’ll give the hospitalist feedback.
What are program results?
We’re all on the same page, with consistent coding practices and many fewer elevated troponins being equated with MIs. But we still see practice variations among our cardiologists and hospitalists. Some may do more stress tests on people with elevated troponin to be sure they’re not missing something. Others are more comfortable saying, “we don’t need this,” especially cardiologists.
We’re looking at 2016 data to drill down to which patient we should test or not, and to compile the true costs associated with testing.
Do you think it’s hard for hospitalists—and hospitals—to say no to the test?
I completely understand why hospitalists and the ER see a benefit to testing. Then they have something in the chart that says it’s negative in case the patient goes home and has a heart attack, and someone asks, “How did you miss this?”
The elephant in the room is the threat of liability. We live in an environment where physicians get sued for failure to diagnose, and MI is among the worst.
Paula S. Katz is a freelance health care writer based in Vernon Hills, Ill.Published in the September 2017 issue of Today’s Hospitalist