Objectives: The purpose of this investigation was to evaluate the practicality and short-term predictive value of acute myocardial perfusion imaging with technetium-99m sestamibi in emergency room patients with typical angina and a normal or nondiagnostic electrocardiogram (ECG).
Background: Accuracy of emergency room chest pain assessment may be improved when clinical and ECG variables are used in conjunction with acute thallium-201 myocardial perfusion imaging. Technetium-99m sestamibi is a new radioisotope that is taken up by the myocardium in proportion to blood flow, but unlike thallium-201, it redistributes minimally after injection. Technetium-99m sestamibi can thus be injected during chest pain, and images acquired 1 to 2 h later (when patients have been clinically stabilized) will confirm whether abnormalities of perfusion were present at the time of injection.
Methods: One hundred two emergency room patients with typical angina (on the basis of a standardized angina questionnaire) and a normal or nondiagnostic ECG had a technetium-99m sestamibi injection during symptoms and were followed up for occurrence of adverse cardiac events (cardiac death, nonfatal myocardial infarction, coronary angioplasty, coronary surgery or coronary thrombolysis).
Results: Univariate predictors of cardiac events included the presence of three or more coronary risk factors (p = 0.009, risk ratio 3.3) and an abnormal or equivocal acute technetium-99m sestamibi scan (p = 0.0001, risk ratio 13.9). Multivariate regression analysis identified an abnormal perfusion image as the only independent predictor of adverse cardiac events (p = 0.009). Of 70 patients with a normal perfusion scan, only 1 had a cardiac event compared with 15 patients with equivocal scans or 17 patients with abnormal scans, with a cardiac event rate of 13% and 71%, respectively (p = 0.0004).
Conclusions: Initial myocardial perfusion imaging with technetium-99m sestamibi when applied in emergency room patients with typical angina and a normal or nondiagnostic ECG appears to be highly accurate in distinguishing between low and high risk subjects.