Preoperative comparison of different noninvasive strategies for predicting improvement in left ventricular function after coronary artery bypass grafting

Abstract

Thallium-201 (Tl-201) imaging and dobutamine stress echocardiography (DSE) are the most frequently used tests in the clinical setting for assessing viability. However, Tl-201 has a suboptimal specificity and DSE a suboptimal sensitivity to predict functional improvement after revascularization. F18-fluorodeoxyglucose (FDG) imaging is considered highly accurate, but availability is limited. Sequential imaging of Tl-201 and DSE may improve accuracy for assessing viability and may be comparable to FDG. Forty-seven patients with ischemic cardiomyopathy underwent Tl-201 single-photon emission computed tomography (SPECT) at rest (4-hour delayed imaging), DSE, and FDG SPECT before bypass surgery. Sensitivity, specificity, and accuracy of 2 sequential strategies were compared with those of FDG SPECT. Strategy 1 considered Tl-201 imaging as the first step, followed by DSE in patients with an intermediate likelihood of viability on Tl-201. Strategy 2 considered DSE as the first step, followed by Tl-201 imaging. Left ventricular ejection fraction was assessed before and 6 months after revascularization, and improvement of >/=5% was considered significant. Tl-201 had a high sensitivity (95%, p <0.05 vs DSE) with a low specificity (57%, p <0.05 vs DSE). DSE had a low sensitivity (63%, p <0.05 vs Tl-201) with a high specificity (89%, p <0.05 vs Tl-201). Both strategies 1 and 2 resulted in high sensitivities (89% and 89%, respectively) and high specificities (89% and 86%, respectively), compared with FDG SPECT (sensitivity 89%, specificity 86%). Sequential testing by Tl-201 SPECT and DSE has a comparable accuracy to FDG SPECT to predict improvement in left ventricular ejection fraction after revascularization. In centers without access to FDG, sequential imaging with Tl-201 and DSE offers an accurate alternative for assessing myocardial viability.