MYOCARDIAL PERFUSION IMAGING WITH TC-99M TETROFOSMIN - COMPARISON TO (TL)-T-201 IMAGING AND CORONARY ANGIOGRAPHY IN A PHASE-III MULTICENTER

Background Our objective was to compare the sensitivity and specificity of tetrofosmin, a new Tc-99m-labeled myocardial perfusion imaging agent for the detection of myocardial perfusion abnormalities, with those of Tl-201 and coronary angiography. Our hypothesis was that same-day stress/rest tetrofosmin imaging could provide data comparable to those of Tl-201 imaging. Myocardial perfusion imaging plays an important role for the evaluation of coronary artery disease. Newer Tc-99m-labeled agents offer several advantages over Tl-201, the conventional myocardial perfusion imaging agent. Tetrofosmin is a new Tc-99m-labeled agent with promising results in preliminary studies. Methods and Results Two hundred fifty-two patients with suspected coronary artery disease were enrolled in 10 centers in the United States and Europe. All patients underwent exercise and rest myocardial perfusion imaging with Tc-99m-tetrofosmin using two separate injections of the radiotracer 4 hours apart on the same day. Planar images were obtained in three standard views 15 to 60 minutes after radiotracer injection. Patients also underwent standard exercise and redistribution planar Tl-201 imaging within 2 weeks of tetrofosmin imaging. In addition, 58 healthy subjects with low likelihood of coronary artery disease underwent exercise and rest tetrofosmin imaging. Coronary angiograms were available in 181 patients with suspected coronary artery disease. All radionuclide images were processed in the central core laboratory and interpreted blindly by a panel of four experienced readers. Tl-201 images and tetrofosmin images were read separately. Discrepancies were resolved by consensus. The workload, peak heart rate, and double products were comparable during exercise for both imaging agents. Technically acceptable paired Tl-201 and tetrofosmin images were available in 224 of 252 patients. Tetrofosmin images were generally of good quality, with low extracardiac activity, and easy to interpret. Patients were categorized as showing normal, ischemia, infarction, or mixture with each imaging modality. Precise concordance for each of these categories was 59.4% (kappa=0.44; 95% CI, 0.35 to 0.53). When patients were categorized as normal or abnormal, the concordance was 80.4% (kappa=0.55; 95% CI, 0.43 to 0.67). When each of five anatomic territories (septal, anterior, inferior, lateral, and apical) was categorized as normal versus abnormal, the concordance Varied from 81% to 90%. When similar comparison was made for the specific category of abnormality, the concordance was 64% to 84%. When coronary angiography was used as the criterion, the sensitivity and positive and negative predictive accuracy of tetrofosmin and Tl-201 were comparable. The normalcy rate of tetrofosmin images in the healthy subjects with low likelihood of coronary artery disease was 97%. Conclusions Tc-99m tetrofosmin is a new myocardial imaging agent with favorable imaging characteristics with results comparable to those of Tl-201.