Noninvasive Tests for Chest Pain: New Insight Into Who Benefits Most
Because the advantage was mainly seen in high-risk patients, experts urge more thoughtful use of testing overall.
Among patients in Ontario discharged from the emergency department (ED) with chest pain, noninvasive testing was associated with a decrease in the composite outcome of cardiovascular death or MI over 1 year compared with no testing. Because this finding was driven by the subset of patients at high-risk, though, the researchers conclude that noninvasive testing may be overused in those at low or intermediate risk.
Neither of the two large contemporary trials in this space—PROMISE and SCOT-HEART—showed a robust overall benefit for noninvasive testing for chest pain evaluation and both had low rates of MI and death. According to Idan Roifman, MD (University of Toronto, Canada), the lead author of the latest study, “a general question that has arisen over the past couple of years is: is noninvasive testing warranted in certain situations and not others?”
This analysis “shows that right now we're probably doing too many of these tests in the low and intermediate groups, which is the vast majority of patients, and we may actually not be doing as many as we should in the high-risk group,” he told TCTMD. “These are patients that have had previous PCI, previous known coronary disease, previous MI, and some physicians may take a different approach on these patients and say, ‘We know that they have coronary disease, we're going to treat them medically, and we're not going to care about doing a test on them. It's not going to really change our management.’ Well, this study actually shows that maybe you should do a test on them because that's the only subgroup that actually showed benefit to testing.”
Commenting on the study for TCTMD, Michael Blaha, MD (Johns Hopkins Medical Institute, Baltimore, MD), said: “There's been a little bit of a movement among some people arguing that we don't need any testing, but of course that's not right. Testing still matters. But [this] makes the case that you really need to test the right people and if you test the right people you can improve outcomes.”
For the study, published online this week ahead of print in the Journal of the American Heart Association, Roifman and colleagues retrospectively looked at all 370,863 adults in Ontario, Canada, who were evaluated for chest pain in the ED and sent home between April 2010 and November 2015. They identified 96,457 propensity-matched pairs of patients who did and did not undergo noninvasive testing—graded exercise stress test, stress echocardiography, myocardial perfusion imaging, or coronary CTA—within 30 days of discharge.
Those who underwent testing were 11% more likely to see a family physician (HR 1.11; 95% CI 1.10-1.12) and 74% more likely to see a cardiologist (HR 1.74; 95% CI 1.72-1.77) within 1 year of discharge compared with the no testing arm. Additionally, at both 90 days and 1 year, the testing cohort was more likely to receive an invasive evaluation and revascularization, and those over 65 years old who received testing were also more likely to be prescribed medical therapy consisting of ACE inhibitors/angiotensin-receptor blockers, statins, or beta-blockers.
“We speculate that noninvasive testing may have initiated a cascade of events leading to improved downstream processes of care which ultimately may have led to the lower rates of MACE that we observed,” Roifman and colleagues write. “While this cascade of events is speculative, it is consistent with the overall rationale of evaluating outcomes related to diagnostic testing strategies. This rationale argues that evaluation of outcomes in this context is not simply a reflection of receipt or nonreceipt of the diagnostic test itself but rather is also indicative of a chain of real-world events and decisions triggered by the diagnostic test that ultimately lead to potential observed differences in cardiovascular outcomes.”
Testing still matters. But [this] makes the case that you really need to test the right people and if you test the right people you can improve outcomes. Michael Blaha
Rates of the primary composite endpoint—cardiovascular death or hospitalization for MI—were low and similar at about 0.3% for both groups at 90 days, although there was a trend favoring no testing (HR 1.17; 95% CI 1.00-1.37). At 1 year, the rates remained low but the testing arm showed a slight though significant advantage (0.68% vs 0.78%; HR 0.87; 95% CI 0.78-0.96). This was driven by the high-risk patient group with prior MI, unstable angina, or revascularization (HR 0.75; 95% CI 0.61-0.92), as there were no significant differences seen in the low- and intermediate-risk cohorts.
“Importantly, neither the PROMISE nor SCOT-HEART trials had ‘no-testing’ arms,” note the authors. “In this study, we gained insights into the utility of noninvasive testing for coronary artery disease by evaluating the subset of patients who presented to the ED with chest pain and who were subsequently discharged home.”
They acknowledge that their study did not use a validated risk score to stratify patients into high-, intermediate-, and low-risk groups, and also the fact that the observational dataset is subject to selection bias beyond what might be overcome with propensity matching.
Ultimately, they conclude, “these results suggest that noninvasive testing for CAD after ED discharge may currently be overutilized and that risk-based testing should be considered for patients discharged from the ED for chest pain.”
‘Too Many’ Tests
As to the reason behind the discrepancy between outcomes seen at 90 days and 1 year, Roifman said “we don't know exactly what's causing this” and added that the time frames were set a priori. “The overall message of this study is the rates of MI and cardiovascular death in patients being discharged from the ED after evaluation for chest pain is super, super low. And so it doesn't matter whether you get tested or not, the rates are super, super low. Testing overall confers some benefit, which was apparent after a year. It wasn't apparent in the first 90 days, but when you dig deeper in the numbers, it's really only the subgroup of patients who are at high risk . . . that drove the entire relationship.”
The high-risk patients made up only 6.4% of the testing cohort, he explained. This “highlights that we could do a lot less of these tests,” Roifman said, encouraging physicians to use their risk-stratification tool when patients leave the emergency department to guide care instead of jumping right to a test. “It gives an idea about how much your patient can potentially benefit from a noninvasive test or [can help determine if] it is worthwhile to forgo the noninvasive test for now, manage them medically, titrate their medications. And then if they have ongoing symptoms, that's when you can do a test.”
“If the results are credible and reliable, it would imply that the vast majority of noninvasive post-chest pain admission to the ED are not warranted,” Sanjay Kaul, MD (Cedars-Sinai Medical Center, Los Angeles, CA), told TCTMD in an email. “That would translate into huge cost savings!”
The overall message of this study is the rates of MI and cardiovascular death in patients being discharged from the ED after evaluation for chest pain is super, super low. Idan Roifman
Blaha said these findings are in line with subanalyses of PROMISE and SCOT-HEART in that testing can pinpoint higher-risk patients with the potential for benefit. While this study only follows patients for a year, “if you take higher-risk patients and you find out who needs either downstream preventative therapy or I guess in some cases will need a revascularization, you can improve their outcomes.”
One thing that muddies the results a bit, he said, is that the study doesn’t stratify results by individual noninvasive tests, which can communicate different information, nor by type of chest pain. “You're higher risk if you have more risk factors, [and] you're also higher risk if your more-typical chest pain,” Blaha said.
The message here is “less testing in low-risk [and a] more thoughtful approach to intermediate risk where you can see some benefit, but you probably shouldn't test everyone,” he continued. “My approach to that is to consider the risk factors and their symptom quality. If their symptoms are atypical, we're less likely of course to test. If their symptoms are typical, we're much more likely to test. And I think we can dispense with the notion that testing is pointless [as] some people might argue. Testing definitely has value, but we need to be very thoughtful about who is likely to have that value.”
The reason Roifman said many low-risk patients end up undergoing noninvasive testing is because even though there might be no objective evidence of MI in the ED and the patient is discharged, “you still feel the obligation to do some kind of noninvasive test on them after they go home to make sure you’re not missing anything. . . . That’s the natural tendency.”
This is leading to “too many” tests, Roifman concluded. “That's a lesson to be gleaned from this study is that patients really can be discharged home safely, and, yeah, they probably do need some sort of outpatient assessment by a cardiologist, but they don't necessarily need a test.”
Roifman I, Han L, Koh M, et al. Clinical effectiveness of cardiac non-invasive diagnostic testing in patients discharged from the emergency department for chest pain. J Am Heart Assoc. 2019;Epub ahead of print.
- This study was supported by ICES, which is funded by an annual grand from the Ontario Ministry of Health and Long-Term Care.
- Roifman and Blaha report no relevant conflicts of interest.