Several years ago, as some US emergency departments (EDs) prepared to implement the recently approved high-sensitivity assay for cardiac troponins, there was still apprehension about broadly applying the exquisitely sensitive test to the mass of patients who would present with chest discomfort.
Would the assay’s prowess in detecting myocardial damage, whether or not from acute ischemia, unleash cascades of sometimes expensive, often unnecessary downstream tests and services?
Now, two new analyses, from regional US health systems that fully adopted the test for cardiac troponin T by high-sensitivity assay (hs-cTnT) soon after its regulatory approval, suggest such fears may have been unwarranted.
Together, they found that implementation of the new assays expanded the identification of high-risk patients with myocardial injury, including acute myocardial infarction (MI), in most cases without promoting more consults, testing, or other uses of hospital resources. On the contrary, some downstream testing and other big-ticket items like hospitalizations declined in most cases.
One of the studies compared patient ED visits at five centers, about 7500 for chest pain and more than 100,000 for other symptoms, both before and after full-scale implementation of the new assays on April 1, 2018. The rate of “cascade events” climbed overall for the chest-pain patients, but driven by demand for serial hs-TnT testing and electrocardiography while use of other tests, procedures, and services dropped off.
“When we drilled deeper into the different cascade events, we found lower rates of hospitalization and coronary intervention and less use of cardiac medications like beta blockers, statins, and antithrombotics. They were less likely to have cardiology get involved, whether a consultation in the ED or even admission to the cardiology service,” Ishani Ganguli MD, MPH, Brigham and Women’s Hospital, Boston, told theheart.org | Medscape Cardiology.
“And length of stay was shorter, both the length of the ED visit and the total length of stay including any hospitalization,” said Ganguli, who is lead author on the study, published May 3 in the Journal of the American College of Cardiology.
Downstream Tests Mostly Unchanged
A separate analysis, published in tandem and scheduled for presentation May 16 during the virtual American College of Cardiology (ACC) 2021 Scientific Sessions, was based on more than 4000 ED presentations before and after launch of the high-sensitivity assays at two hospitals in the same regional system.
In it, the proportion of patients with at least one positive hs-cTnT assay climbed sharply from 15% before implementation to 47% afterward (P < .0001). But use of the new assays led to fewer stress tests (4.9% vs 6.5%, P = .02) and less echocardiography (6.5% vs 12%, P < .0001). There were no other observed escalations in downstream resources or services except for an uptick in referrals to coronary angiography, from 2.3% before implementation to 3.5% afterward (P = .02).
Importantly, senior author Yader Sandoval, MD, said for the theheart.org | Medscape Cardiology, “there was a marked increase in patients sent directly home from the emergency department” among those without a positive assay. Their rate went from 60% to 74% (P < .0001).
Such patients “otherwise would have been admitted and might have had more troponin testing, maybe a stress test, and so on,” said Sandoval, of the Mayo Clinic, Rochester, Minnesota.
“A lot of the benefit from the high-sensitivity tests is from being very good at identifying patients at higher risk,” he said. More recently, they’ve been shown to be “very, very good tests for identifying patients at low risk.”
Also in the analysis, the rate of diagnosed acute MI overall went from 3.3% before implementation of the high-sensitivity assays to 8.1% afterward (P < .0001); nonischemic myocardial injury jumped from 11% to 38% (P < .0001).
Rates of both events went up for men and women separately, but disproportionately so in women. The rate of acute MI before implementation of the high-sensitivity assays was much higher in men than in women, 4.4% and 2.3%, respectively (P = .008). But the difference faded after the new assays took over, when the rates were 8.5% in men and 7.7% in women (P = .46).
The finding therefore corroborates earlier research suggesting that the high-sensitivity assays may partly reverse the problem of MI underdiagnosis in women, Sandoval said, and may expand “opportunities to evaluate them further.”
Economic Impact
Contrary to earlier European reports in which the advent of high-sensitivity assays led to substantial financial savings, the analysis from Ganguli and associates saw no overall cost benefits after their implementation despite the drop in downstream testing and other services.
The group believes that any cost savings in the study may have been countered by the relatively small increase in coronary artery bypass surgeries. “Because that surgery is so much more expensive than the other services that were examined, it offset any of the drops in spending,” Ganguli said when interviewed. “If you look at spending for individual services before and after, you would find that for many of them, the cost did go down overall.”
Also, the study included spending estimates only for documented tests and services and did not account for the cost of hospitalizations, she said. “If we had been able to do that, we for sure would have seen a cost reduction.”
In other findings, the patients who presented with chest pain were 10.5% more likely than the patients with other symptoms to have multiple troponin tests performed after compared to before implementation of the high-sensitivity assays. Their net increase in electrocardiographic examinations was 7.1 per 100 patient-visits.
Those early increases in testing were offset by net declines in CT angiography (CTA) scans (by 1.5 such tests per 100 patient-visits), stress tests (by 5.9 per 100 patient-visits), percutaneous coronary interventions (by 0.65 procedures per 100 patient-visits), hospitalizations (by 5.8%), and mean length of stay (by 0.24 days), Ganguli and colleagues reported.
Transatlantic Differences
Both studies used the Roche Gen-5 cTnT assay, the first of several high-sensitivity troponin assays to gain approval in the United States. And they both interpreted biomarker concentrations via the Fourth Universal Definition of Myocardial Infarction (UDMI-4), using sex-specific cutoff values for myocardial injury, which was defined as at least one assay with levels >99th percentile of normal.
However, there is an “ongoing controversy whether uniform cut-offs or sex-specific cut-offs,” the first preferred in a European Society of Cardiology (ESC) statement and latter by the writers of UDMI-4, “should be used in the evaluation of patients with acute chest discomfort,” says an editorial accompanying the two reports.
“Given the broad differential diagnosis in women presenting with acute chest discomfort,” sex-specific cutoffs in the two current reports “resulted in a disproportional increase in the rule-in of conditions for which currently no evidenced-based therapy is available, such as type 2 MI and myocardial injury,” write the authors, led by Christian Mueller, MD, University Hospital of Basel, Basel, Switzerland.
Sex-specific cutoffs “will also invariably lead to an under-diagnosis of type 1 MI in men (due to the higher sex-specific cut-off), a condition for which evidence-based therapy is available,” continue the authors, who recommended the ESC-backed criteria over those based on UDMI-4.
Guidance for US clinicians based on the high-sensitivity assay experience in Europe, for example, where the tests had been widely used for at least a decade, has been helpful but in some ways limited by differences from the United States in hospital processes and the economics and culture of healthcare delivery.
Those differences help explain the frustrating delay before the high-sensitivity assays reached US clinicians, who — compared to their European counterparts — were considered more likely to use them in patients with less specific symptoms for whom a diagnosis involving the myocardium may not have been as likely.
Acute MI “is a don’t-miss diagnosis. So there are reasons why we act on that possibility,” Ganguli said. “There is the malpractice landscape and our interest in pursuing technology and doing more diagnostic services that may make a different story than in Europe.”
Or as she and her colleagues bluntly say in their report, there was apprehension in the United States that the high-sensitivity assays, “which are more likely to detect cardiac injury caused by conditions other than acute MI, might instead increase utilization driven by clinician concern about missing an acute MI in a healthcare ecosystem that places greater emphasis on diagnostic completeness and malpractice risk management.”
But, “having a test that gives you earlier reassurance that it’s not a misdiagnosis can reduce some of the need for downstream testing,” Ganguli noted. “That’s what we’re learning.”
Ganguli has received consulting fees from Blue Cross Blue Shield Massachusetts and grant funding from International Business Machines; disclosures for the other authors are in their report. Sandoval has previously served on the Advisory Boards for Roche Diagnostics and Abbott Diagnostics without personal compensation, and as a speaker without personal financial compensation for Abbott Diagnostics; disclosures for the other authors are in the report. Mueller reports receiving research support from Abbott, Beckman Coulter, Brahms, Novartis, Ortho Diagnostics, Quidel, Roche, Siemens, Singulex, Sphingotec, and speaker honoraria/consulting honoraria from Amgen, Astra Zeneca, Boehringer Ingelheim, and Bristol-Meyers Squibb; disclosures for the other editorialists are in their report.
American College of Cardiology (ACC) 2021 Scientific Sessions: Session 2249. Presented May 16, 2021.
J Am Coll Cardiol. Published online May 3, 2021. Sandoval et al, Abstract; Ganguli et al, Abstract; Editorial
Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.