201Tl-redistribution analysis in early and delayed myocardial scintigrams of patients with coronary heart disease (CHD)

Diagnostic accuracy of myocardial perfusion imaging with single photon emission computed tomography and positron emission tomography: a comparison with coronary angiography

OBJECTIVE

The aim of this study was to compare the diagnostic accuracy of myocardial perfusion imaging (MPI) by positron emission tomography (PET) with the diagnostic accuracy of MPI by single photon emission computed tomography (SPECT) in two comparable patient cohorts, using coronary angiography (CA) as the standard of reference.

METHODS

A "SPECT-group" of 80 patients (15 female, 65 male; mean age 60+/-9 years) and a "PET-group" of 70 patients (14 female, 56 male; mean age 57+/-10 years) underwent a one day stress/rest examination either with attenuation-corrected (13)N-ammonia PET or attenuation-corrected (201)TlCl SPECT or (99m)Tc-hexakis-methoxy-isobutyl-isonitril (MIBI) SPECT. PET and SPECT results were semiquantitatively graded using a 6-segment heart model. All patients underwent CA, and stenoses were graded as a diameter reduction >or=50%.

RESULTS

Coronary findings between both groups did not significantly differ at CA. For the SPECT-group overall sensitivity and specificity for localisation of stenoses was 77% and 84%. Respective values for the PET-group were 97% and 84%. The specificity of MPI by SPECT in the detection of ischemia was 74% and 91% for MPI by PET. The diagnostic accuracy of MPI improves when the individual coronary dominance and previous coronary revascularisations are taken into account.

CONCLUSION

MPI by (13)N-ammonia PET is more sensitive in the detection and localisation of coronary stenoses, and more specific in the detection of ischemia than MPI by (201)TlCl/(99m)MIBI SPECT.

Processing of 201T1 serial images for the assessment of myocardial clearance and redistribution

Regional activity changes over the myocardium after IV injection of 201T1 were assessed in 90 patients by applying a superposition algorithm for paired images. The kinetics were expressed as quotients C1/C2 of the initial divided by the delayed count rate within any myocardial segment. Regarding the total left ventricular myocardium, normal subjects showed count rate ratios (C30 min/C4 h) of 1.75 +/- 0.075 (SD, n = 19) when the nuclide was injected under submaximal (80% of maximum) exercise. In patients with triple-vessel disease (n=21) the corresponding values were 1.46 +/- 0.05, in single-vessel (n=14) disease the quotients averaged 1.62 +/- 0.12. Average quotients in poststenotic myocardial segments increased from 1.46 before to 1.73 following bypass surgery in cases of graft patency (n=8), while a decrease was observed in a patient with occlusion of the bypass graft. In small-vessel disease and in congestive cardiomyopathy, global values of 1.50 +/- 0.07 (n=7) and 1.53 +/- 0.12 (n=19) were obtained. The display of C1/C2 matrices provided typical patterns in uptake defects caused by ischemic heart disease and by congestive cardiomyopathy. The procedure described may provide diagnostic information in addition to conventional myocardial scintigraphy in triple-vessel disease, in patients with angina who fail to demonstrate hemodynamically significant coronary artery stenoses, and in patients with inhomogeneous T1-scintigrams.