False-Positive Stress Echocardiograms: A Continuing Challenge

INTRODUCTION

Advancements in stress echocardiography during the past 2 decades have made this test an integral part of the cardiac diagnostic laboratory. In a large laboratory, 15-20 tests are performed per day.¹ Stress echocardiography is a cost-effective test with high sensitivity (88%) and specificity (83%) for detection of significant coronary artery stenosis, most often defined as >50% diameter stenosis on coronary angiography.²

A study of insurance billing data extracted from a national insurance provider from November 2004 through June 2007 identified 80,676 people aged 40-64 years (53% women, mean age 51.5 years) referred for outpatient stress testing. Fifty-four percent were referred for nuclear stress myocardial perfusion imaging, 21% underwent stress echocardiography, and 25% had exercise tolerance testing without imaging as their initial test.

Studies comparing the accuracy of nuclear perfusion imaging and stress echocardiography in the same patient population have shown that the tests have similar sensitivities for the detection of coronary artery disease, but stress echocardiography has higher specificity. In a pooled analysis of 18 studies with 1,304 patients who underwent exercise or pharmacologic stress echocardiography in conjunction with thallium- or technetium-labeled radioisotope imaging, sensitivity and specificity were 80% and 86%, respectively, for echocardiography. Corresponding sensitivity and specificity were 84% and 77% for myocardial perfusion imaging, respectively.²

Despite the excellent specificity of stress echocardiography, we continue to see a subset of patients with false-positive tests (ie, <50% diameter coronary artery stenosis on the subsequent angiogram) in the absence of left bundle branch morphology, right ventricular pacing, prior cardiac surgery, or abnormal wall tethering at baseline. These false-positive findings present a management challenge, as it remains unclear if and how to treat patients with a false-positive stress echocardiography.

CLINICAL STUDIES AND OUTCOMES

In a retrospective study of 1,477 patients who underwent stress echocardiography (either treadmill stress echocardiography or dobutamine stress echocardiography), From et al found that 480 patients (32.5%) had false-positive results.³ Furthermore, in patients with markedly abnormal stress echocardiographic findings (n=605), 28% had <50% stenosis or normal coronary arteries. Hypertensive response to exercise (defined as peak systolic blood pressure >210 mmHg in men and >190 mmHg in women) is considered to be one of the reasons for false-positive stress echocardiography. In the From et al study, 10% of the patients with abnormal stress echocardiography who were referred for angiography developed a hypertensive response to exercise.³ Subsequent multivariate analysis showed that those who were most likely to have a false-positive stress echocardiography were nondiabetic, nonhypertensive, younger females without a prior history of coronary artery disease and negative peak stress electrocardiogram (ECG). Eleven percent of the patients with abnormal stress echocardiography underwent additional testing. Four patients had positive acetylcholine studies (suggesting endothelial dysfunction), 3 patients had catheter-induced spasm, 3 patients had myocardial bridging, and 2 patients had anomalous coronary vessels. After 2.4 years of follow-up, no significant mortality difference was seen between patients with and without >50% luminal narrowing on angiography after abnormal stress echocardiography.

Another study with the same lead author evaluated 31 patients with normal left ventricular function at rest, extensive regional wall motion abnormalities at peak stress (with an increase in left ventricular end-systolic cavity size at peak stress), and angiographically normal coronary arteries, with a 2-year follow-up.⁴ The mid and apical ventricular segments were more likely to be abnormal on echocardiography, leading the authors to suggest that some false-positive stress echocardiograms may be a forme fruste variety of apical ballooning syndrome. Eighty-four percent of these patients were women, with a mean age of 61 years. Forty-five percent of the cohort were deemed low probability for obstructive coronary artery disease. Twenty-six percent of patients developed a hypertensive response to stress. Stress ECG was positive in only 6% of the patients, and no deaths had occurred at follow-up.

Yet another retrospective study from some of the same researchers examined abnormal treadmill stress echocardiography in 7,236 patients who achieved >10 metabolic equivalents.⁵ Twelve percent of the patients had exercise echocardiographic results that were positive for ischemia, and 4% of the patients had echocardiographic criteria consistent with extensive ischemia. Prior to stress testing, 30% of patients in the abnormal stress echocardiography group were thought to be at low probability for obstructive coronary artery disease. Subsequently, 3.2% of patients with abnormal stress echocardiography underwent coronary angiography within 30 days, and only one-fourth of those patients required coronary revascularization. A prognostic model did not add incremental value by adding stress echocardiography to the excellent short- and medium-term prognosis of these exertion-tolerant patients when followed for an average of 4.8 years.

DISCUSSION

These studies suggest the existence of a group of patients who develop hypertensive response to exercise and therefore may have false-positive stress echocardiography (10%-26% in the studies cited above). Hence, superior blood pressure control prior to stress echocardiography may prevent some false-positive tests.

In addition, a subset of patients has microvascular abnormalities, vasomotor changes, endothelial dysfunction, and/or small vessel coronary disease that can lead to false-positive stress echocardiography. In support of this hypothesis, perfusion imaging by single-photon emission computed tomography (SPECT) has shown clear defects in myocardial perfusion in the absence of angiographic stenosis in some cases.⁶ Abnormal coronary flow reserve has also been demonstrated in patients without significant coronary lesions by traditional coronary angiography despite an abnormal SPECT.⁷ Rodés-Cabau et al performed intravascular ultrasound on a group of patients with stable coronary disease with abnormal radionuclide stress results and normal coronary circulation on angiography and demonstrated a significantly increased plaque burden in this subset of patients.⁸ This group of patients may benefit from provocative testing as well as aggressive medical management.

Pooled accuracy studies of stress echocardiography have typically used >50% diameter stenosis by angiography as the definition of obstructive disease,² but we know from animal studies that >70% stenosis is required to cause enough of a decrease in coronary flow reserve to cause ischemia.⁹ Hence, we may need to challenge the current definition and redefine what we term false-positive.

CONCLUSION

The evidence is insufficient to state that false-positive stress echocardiography in the absence of obstructive coronary artery disease portends a poor outcome, but considerable evidence shows that some of these patients have microvascular abnormalities and endothelial dysfunction. Consequently, this group may benefit from aggressive medical management and further testing. Stress echocardiography remains useful for prognostication in select patients.