Summary of Review Findings

The results of the evidence review are summarized in Table 18. As in previous USPSTF reviews, we found no studies that evaluated clinical outcomes associated with screening with resting or exercise ECG compared with no screening. We also found no studies on how screening affects use of therapies to reduce cardiovascular risk (lipid-lowering therapy or aspirin). Another critical research gap is that no studies directly evaluated the incremental value of screening ECG when added to traditional risk assessment for accurately classifying patients into different risk categories.

Table 18

Summary of Evidence.

The lack of information on reclassification is critical from a clinical perspective, since treatment decisions regarding therapies for reducing cardiovascular risk are often based on whether a patient is classified as low- (<10 percent risk over the next 10 years), intermediate- (10 to 20 percent risk), or high-risk (>20 percent risk) for future CHD events. From the information currently available, it is not possible to determine the degree to which performing resting or exercise ECG in an individual patient more accurately moves them from one risk category to another versus yielding a more precise estimate within the same risk category, which is less clinically useful. For example, in populations at very low risk (<5 percent) for CHD events, such as most young adults, even a doubling of risk is unlikely to move an individual from a low-risk to a higher-risk category. Similarly, in persons already at high risk based on traditional risk factor assessment, abnormalities on resting or exercise ECG are unlikely to change management decisions. The greatest potential benefits of screening with resting or exercise ECG are likely to be in intermediate-risk patients, since the presence of abnormalities could shift persons into the high-risk group, where additional interventions might be warranted, but no study reported how many persons classified as intermediate risk based on traditional risk factor assessment would be reclassified as high risk following screening. Two studies evaluated effects on the C statistic of adding resting or exercise ECG findings to traditional risk factor assessment compared with traditional risk factor assessment alone,^72,95 but this measure is of limited clinical usefulness because it does not provide information about the actual predicted risks in an individual patient, or the proportion of patients classified (or reclassified) as high, intermediate, or low risk.⁵⁷ Both showed slight improvements in the C statistic, though the difference did not appear statistically significant in one study,⁷² and confidence intervals for the estimates were not reported in the other.⁹⁵

The bulk of the available evidence came from over 60 studies of more than 250,000 subjects that evaluated whether abnormalities on resting or exercise ECG are associated with an increased risk of subsequent cardiovascular events. Unlike previous USPSTF reviews, we focused on studies that adjusted for at least five of the seven Framingham risk factors, in order to better understand the incremental value of adding resting or exercise ECG for predicting cardiovascular events. Based on pooled analyses, a number of abnormalities on resting (ST segment abnormalities, T wave abnormalities, LVH, left axis deviation, bundle branch block) or exercise (ST segment depression with exercise, failure to reach target heart rate) ECG were associated with an increased risk of subsequent cardiovascular events. The magnitude of increased risk ranged from an adjusted pooled hazard ratio of around 1.4 to around 2.1 for either resting or exercise ECG abnormalities (Table 1). An exception was variously defined “major” resting ECG abnormalities, which were associated with somewhat greater hazard ratios (range, 2.3 to 3.7) than those observed for “minor” resting ECG abnormalities (range, 1.1 to 2.1) (Table 5). Although statistical heterogeneity was present in most analyses, the point estimates from almost all studies favored an association, estimates were stable in stratified analyses (based on study quality, method of defining the ECG abnormality, and cardiovascular outcome assessment), and meta-regression analyses (based on differential duration of followup, proportion of male subjects, and number of Framingham risk factors adjusted for) did not explain the between-study variance. Low versus high exercise capacity or fitness during exercise ECG was also associated with increased risk of subsequent cardiovascular events or all-cause mortality, with hazard ratios ranging from 1.7 to 3.1, but data could not be pooled. Despite the strong evidence that abnormalities on resting or exercise ECG are associated with an increased risk of subsequent cardiovascular events beyond the risk accounted for by assessment of traditional risk factors, information on the diagnostic usefulness of these tests is incomplete, since understanding the usefulness of screening requires additional information on reclassification and whether reclassification leads to clinical actions that improve patient outcomes.¹⁶

Evidence on harms associated with screening using resting or exercise ECG is limited. Nonetheless, serious direct harms appear to be minimal with resting ECG (other than possibly anxiety or labeling) and small or rare with exercise ECG (ischemia associated with exercise, injuries related to exercise), assuming appropriate attention to absolute and relative contraindications to exercise testing and adherence to standard safety precautions for terminating a test. Perhaps of greater concern than the direct harms associated with the test itself are the downstream harms that could result from additional testing or interventions as a result of screening. For example, some patients undergo angiography following screening ECG, and are therefore exposed to the potential harms related to that procedure, including bleeding, radiation exposure, and contrast allergy or nephropathy. Similarly, patients who are placed on lipid-lowering therapy or aspirin as a result of ECG screening are exposed to the harms related to those interventions. Evidence on downstream harms associated with screening is not available, with data primarily limited to rates of patients who subsequently undergo angiography (range, 0.6 to 1.7 percent). A small proportion (<1 percent) of patients undergo revascularization with coronary artery bypass graft surgery or a percutaneous coronary intervention following screening with exercise ECG, despite the lack of evidence on benefits associated with these interventions in asymptomatic persons and the known risks associated with those procedures.^95,99

Limitations

We only included English-language studies, which could result in language bias. Studies that evaluated the risk associated with various resting or exercise ECG abnormalities varied in quality and duration of followup, assessed different cardiovascular outcomes, and used different methods to define the abnormalities. We therefore used a random effects model to perform meta-analysis. Although statistical heterogeneity was present in several of the meta-analyses, stratified analyses and meta-regression had little impact on estimates and conclusions. Referral bias could have resulted in underestimates of the risk associated with ECG abnormalities if their identification led to increased use of treatments effective at reducing cardiovascular risk.

Emerging Issues/Next Steps

Resting and exercise ECG are technologies that have been available for many years. Most of the studies included in this review evaluated the usefulness of long-established and widely recognized abnormalities on resting or exercise ECG for predicting future cardiovascular events. However, newer abnormalities (or refinements of established abnormalities) on resting (such as the QRS/T angle, high QRS nondipolar voltage, and decreased heart rate variability)^90,91 or exercise ECG (such as heart rate adjustment of ST depression, QT interval and T wave subintervals, and heart rate recovery)¹⁴⁵ have been proposed as potentially better predictors of cardiovascular events, and may warrant further study.

Future Research

Studies that directly evaluate how screening with resting or exercise ECG affects clinical outcomes compared with not screening, or how screening affects use of interventions to reduce cardiovascular risk, are needed. Any study of screening should also evaluate harms associated with screening, as well as downstream harms related to additional testing and therapies. Although randomized trials would be desirable, well-conducted prospective studies with adequate sample sizes and sufficient duration of followup could also be informative.

In lieu of direct evidence on the clinical effects of screening, future studies on risk prediction should provide data to enable estimates of reclassification, from which potential benefits of screening might be extrapolated, based on the known efficacy of interventions in high-risk populations. Decisions to allocate resources to update this or similar reviews on the usefulness of ECG screening might be predicated on the availability of such evidence that can be identified using literature scans or other methods. Many of the studies included in this review evaluated large sample sizes over long periods of time, and the information needed to assess reclassification rates in these databases likely already exists. Therefore, a more efficient method than initiating new studies for obtaining information on reclassification would be to reanalyze preexisting databases.

Some studies suggest that the association between abnormalities on resting or exercise ECG and subsequent cardiovascular events might vary in subpopulations defined by race⁸² or sex.^76,116 Research is needed to better understand whether and how the usefulness of different ECG abnormalities as predictors varies in different subpopulations, in order to inform optimal screening strategies. From a comparative effectiveness perspective, studies that evaluate newer compared with more traditional abnormalities on resting or exercise ECG would be valuable, as would be studies that evaluate the usefulness of combinations of ECG abnormalities compared with single findings, and studies that compare screening with resting or exercise ECG versus cardiac CT, carotid artery intima-media thickness ultrasonography, or other imaging modalities.

Conclusions

There is no direct evidence on benefits of screening with resting or exercise ECG on clinical outcomes, and no evidence on how screening affects use of therapies to reduce cardiovascular risk. There is strong evidence that abnormalities on resting or exercise ECG are associated with mildly increased risk of subsequent cardiovascular events after adjusting for traditional cardiovascular risk factors. Estimates of increased risk were similar for resting and exercise ECG abnormalities. The clinical implications of these findings are unknown, as pooled risk estimates do not necessarily indicate the degree to which resting or exercise ECG results in accurate reclassification of persons into CHD risk categories or has an impact on clinical decisions and subsequent patient outcomes. Evidence on harms associated with screening is limited. Although direct harms associated with screening appear to be small, downstream harms related to subsequent testing and interventions are likely to be an important factor in assessing the balance of benefits and harms associated with screening.