Clinical Utility : Phase III TrialA total of 252 patients with ischemic heart disease or atypical chest pain were involved in an open-label, multi-center, Phase III clinical trial of MYOVIEWTM conducted in Europe and the US.15 Each patient received 185 to 296MBq (5 to 8mCi) Tc-99m-tetrofosmin at peak exercise and 555 to 888MBq (15 to 24mCi) at rest approximately 4 hours later. Patients also received TI-201 55.5 to 74MBq (1.5 to 2.0mCi) at peak exercise either 1 to 14 days before or 2 to 14 days after the MYOVIEW imaging study (Figure 8).7
In addition, 58 healthy subjects with low likelihood of CAD underwent exercise and rest MYOVIEW imaging. Planar and SPECT images were obtained, the former beginning 15 to 60 minutes following MYOVIEW injection. Planar images were read by four independent blinded readers having no clinical information about the patients. Each reader reported all MYOVIEW images prior to reporting the TI-201 data. A final arbitrated consensus (if disagreements were present) was derived from these individual readings. Only after the final interpretation had been recorded were comparisons made between TI-201 and MYOVIEW data sets. The low likelihood group images were interpreted by two readers and were read interspersed with a comparable number of abnormal studies. 15 The planar image data were evaluated with respect to the five standard anatomic regions: anterior / anterospectal, inferior / inferoposterior, lateral, apex, and septum. A separate interpretation was made for each region and for each patient based on one of four possible categories: normal, ischemia (reversible defect), infarct (fixed defect), or mixed (ischemia and infarct). For purposes of comparisons between imaging techniques, a mixed defect was counted as both ischemia and infarct.15 The agreement between MYOVIEW and TI-201 imaging data was expressed as percent concordance and also by kappa statistics which measure the amount of agreement over and above any agreement expected by chance alone. To help interpret kappa values, statistical literature suggests we regard values in the range 0.81 to 1.00 as very good, 0.61 to 0.80 as good, 0.41 to 0.60 as moderate, 0.21 to 0.40 as fair, and <0.20 as poor. If the observed number of agreements is less than the number expected by chance, kappa will be negative, effectively indicating that guessing would have resulted in better agreement.7 The physiologic parameters for the MYOVIEW and TI-201 exercise studies were comparable (Table 4).15 Maximum heart rate achieved was essentially identical. There were clinically nonsignificant but statistically significant differences between peak systolic blood pressure and rate pressure product. The incidence of chest pain and significant ST segment depression was comparable. Technically acceptable paired imaging data were available in 224 of 252 patients. These 224 patients form the basis of comparative study. Of these 224 patients, 111 had no previous myocardial infarction. 15 Table 4 - Exercise Data During MYOVIEW and Thallium Studies15
In general, MYOVIEW images were of superior technical quality and were somewhat easier to interpret than TI-201 images. Subdiaphragmatic activity did not present a problem with respect to image interpretation or creation of image artifacts.15 The concordance of MYOVIEW and TI-201 image data by patient is presented in Table 5. Precise concordance between the two agents for each of the four categories (ie, normal, ischemia, infarct, mixed) by patients was 59.4% (kappa statistic = 0.44, 95% confidence interval (CD: 0.35 to 0.53). When patients were diagnosed as normal or abnormal, concordance was 80.4% (kappa = 0.55,95% CI: 0.43 to 0.67). Similarly high values were obtained from patients with no previous history of infarction; precise concordance was 56.8% with a kappa statistic of 0.38 (95% CI: 0.25 to 0.51), and concordance for normal versus abnormal was 74.8% (kappa = 0.49, 95% CI: 0.32 to 0.65).15 Table 5 - Concordance of MYOVIEW and Tl-201 Image Data by Patient15 (%)
In addition, concordance by region ranged from 80% to 90% (normal vs. abnormal), 74% to 88% (ischemia), and 85% to 90% (infarct) in all patients (Table 6). When distinguishing normal from abnormal tissue, kappa values ranging from 0.56 (95% CI: 0.41 to 0.72) in the lateral wall to 0.73 (95% CI: 0.62 to 0.83) in the septum were observed.15 Table 6 - Concordance of MYOVIEW and Tl-201 by Region15 (All Patients) (%)
Results were comparable for patients without prior infarction (Table 7).15 Table 7 - Concordance of MYOVIEW and Tl-201 by Region15 (Patients Without Prior Myocardial Infarction) (%)
Zaret et. al. Circulation, 1995 (91) Catheterization data were available in 181 patients. One hundred sixty-seven of the patients had complete MYOVIEW studies available for comparative analysis and 162 had adequate TI-201 studies. Estimates of diameter stenoses were made at the investigational sites; there was no independent core laboratory review.15 Table 8 summarizes results of the comparisons between the blinded diagnoses derived from MYOVIEW and Tl-201 images and catheterization data.15 The results show that MYOVIEW and Tl-201 have comparable sensitivity, specificity, and both positive and negative predictive values in patients with >75% stenosis. Specificity was somewhat better for MYOVIEW compared with Tl-201. It was also observed that MYOVIEW sensitivity correlated with extent of coronary disease. Sensitivity increased from 74% to 80% when vessel involvement increased from single to double and to as high as 90% for triple-vessel involvement with >75% stenosis. Sensitivities of 82%,87%, and 83% were observed for >90% in single-, double-, and triple-vessel disease, respectively.15 Table 8 - Specificity and Sensitivity of MYOVIEW and Tl-201 (Any Defect) in Relation to Coronary Angiography15
Specificity is relatively low for both tracers, but it must be emphasized that this was not altogether unexpected.15 The patients entered into this study had a high pre-test probability of CAD and may, therefore, have been referred for catheterization because of an abnormal noninvasive study. In contrast, the normalcy rate of MYOVIEW imaging in the group of subjects with low likelihood of CAD was high, with 56 of 58 (97%) subjects demonstrating normal image patterns.15 In conclusion, the results derived from this study indicate that MYOVIEW provides high quality images and has a diagnostic utility for the detection of CAD comparable to that of Tl-201, the current gold standard.15 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||
GE Healthcare | Medical Diagnostics Web Site | Myoview Full Prescribing Information
Technology Overview | Product Profile | Clinical Practice | Quality | Resources
The information in this web site is intended for use only by customers, patients and health care professionals in the United States.
Copyright General Electric Company 1997-2009