Chest Pain in the ED: More Referrals for Cardiac Tests Not Tied to Better Outcomes

For lower-risk patients with acute myocardial injury already ruled out for their chest pain, an increase in referrals for noninvasive cardiac testing (NICT) was not associated with improved outcomes, a retrospective cohort study suggested.

Compared with the low NICT referral group, there was no significant decrease in the adjusted hazard ratio of major adverse cardiac events (MACE) — a composite of myocardial infarction, cardiac arrest, cardiogenic shock, or cardiac death — for the intermediate referral group (adjusted HR 1.08, 95% CI 1.02-1.14) and high referral group (adjusted HR 1.05, 95% CI 0.99-1.11) within 2 years, reported Dustin Mark, MD, of Kaiser Permanente Medical Center in Oakland, California, and colleagues.

Results were similar across HEART (history, ECG changes, age, risk factors, and troponin) risk scores, and when the investigators also considered all-cause mortality and coronary revascularization as clinical outcomes, they reported in Circulation: Cardiovascular Quality and Outcomes.

The findings therefore support the current guideline recommendation against urgent use of NICT in low-risk emergency department (ED) patients with chest pain, given the lack of evidence of improved clinical outcomes. NICT is instead reserved for patients at high coronary risk or those with elevated serum cardiac troponin levels.

“To be clear, these findings do not indicate that patients without evidence of acute myocardial injury do not benefit from NICT but rather that accepted determinants of clinical risk (such as the HEART score and its components) may not be ideally suited to identifying which patients might benefit,” Mark’s group wrote.

Additionally, only 43.2% of those referred actually ended up with a record of cardiac testing within 30 days — by far most commonly with exercise treadmill tests, which are known to have relatively low sensitivity for coronary artery disease. Only about a third of tests were imaging-based.

“Thus, while the utility of routine and urgent functional NICT in low- and intermediate-risk ED patients has rightly been questioned, in large part due to low testing yields, there may still be a role for anatomic NICT in mitigating long-term risks in selected patients,” the authors wrote.

Indeed, the ongoing TARGET-CTCA study is expected to provide more answers regarding the benefit of CT coronary angiography when selected for patients with mildly elevated, but nondiagnostic, high-sensitivity troponin.

Mark and colleagues had previously reported no association between higher NICT referrals and MACE at 60 days, beyond a positive association between NICT and coronary revascularization.

In an accompanying editorial, Florentina Simader, MD, and Rasha Al-Lamee, MBBS, PhD, both of Imperial College London, stressed that “longer follow-up periods are essential for a comprehensive evaluation of effects about major adverse cardiac events, especially among patients categorized as lower risk.”

In this study, patients had a median HEART score of 4; 48.2% were low risk, 49.2% were moderate risk, and 2.7% were high risk.

“Consequently, the main findings heavily rely on a patient cohort characterized by a lower risk profile, for whom existing literature indicates no discernible benefit from NICT,” Simader and Al-Lamee noted. “Thus, it is not particularly surprising that an aggressive referral approach did not yield benefits compared with a more restricted one.”

“[T]o answer the fundamental question of whether ‘To test or not to test?’, the unfortunate answer is that we still do not know,” they wrote. “The only reliable method to address this question definitively is through a well-executed randomized controlled trial. Until such research is available, we must rely on conflicting data from observational studies.”

This study included 144,577 eligible patient encounters at 21 EDs within an integrated healthcare system in Northern California. Patients presented with chest pain, with an accompanying serum troponin level measurement, from 2013 to 2019.

Patients with ST-segment-elevation myocardial infarction or myocardial injury by serum troponin testing were excluded, as were patients under 30 or over 80.

Median age was 58, and 57% were women. Of these patients, 23% had diabetes, and 13% had coronary artery disease.

Thirty-day NICT referral was 13%, 19.9%, and 27.8% in the low, intermediate, and high NICT referral groups, respectively. ED physicians were separated into tertiles of NICT referral after adjustment for facility-specific practice. Patient baseline variables were comparable across the three groups.

Mark and colleagues said their provider-level-stratified analysis was designed to account for any confounding by indication. Even so, they cautioned, the observational nature of the study precluded any causal associations from being made between NICT referral and patient outcomes.

Please enable JavaScript to view the

comments powered by Disqus.