Value of Ischemia and Coronary Anatomy in Prognosis and Guiding Revascularization Among Patients With Stable Ischemic Heart Disease

BACKGROUND: The value of physiological ischemia versus anatomic severity of disease for prognosis and management of patients with stable coronary artery disease (CAD) is widely debated. METHODS: A total of 1764 patients who had rest-stress cadmium-zinc-telluride single-photon emission computed tomography myocardial perfusion imaging and angiography (invasive or computed tomography) were prospectively enrolled and followed for cardiac death/nonfatal myocardial infarction. The CAD prognostic index (CADPI) was used to quantify the extent and severity of angiographic disease. Prognostic value was assessed using Cox models, adjusted for pretest risk, known CAD, stressor, left ventricular ejection fraction, %ischemia and infarct, CADPI, and early (90-day) revascularization. Incremental prognostic value was evaluated using net reclassification index. RESULTS: The mean age was 69.7±9.5 years, 24.4% were women, and 29.3% had known CAD. Significant ischemia (>10%) was present in 28.4%. Nonobstructive, single, and multivessel disease was present in 256 (14.5%), 772 (43.8%), and 736 (41.7%), respectively. Early revascularization occurred in 579 (32.8%). Cardiac death/myocardial infarction occurred in 148 (8.4%) over a 4.6-year median follow-up. Both %ischemia and CADPI provided independent and incremental prognostic value over pretest clinical risk ( P <0.001). In a model containing both ischemia and anatomy, ischemia was prognostic (hazard ratio per 5% ↑, 1.35 [95% CI, 1.11–1.63]; P =0.002) but CADPI was not (hazard ratio per 10-unit ↑, 1.09 [95% CI, 0.99–1.20]; P =0.07). Early revascularization modified the risk associated with %ischemia (interaction P =0.003) but not with CADPI (interaction P =0.6). %Ischemia and single-photon emission computed tomography variables added incremental prognostic value over clinical risk and CADPI (net reclassification index, 20.3% [95% CI, 9%–32%]; P <0.05); however, CADPI was not incrementally prognostic beyond pretest risk, %ischemia, and single-photon emission computed tomography variables (net reclassification index, 3.1% [95% CI, −5% to 15%]; P =0.21). CONCLUSIONS: Ischemic burden provides independent and incremental prognostic value beyond CAD anatomy and identifies patients who benefit from early revascularization. The anatomic extent of disease has independent prognostic value over clinical risk factors but offers limited incremental benefit for prognosis and guiding revascularization beyond physiological severity (ischemia).