RESUMO
Various regions of interest (ROI) are used to semiquantify lung-to-heart (LHR) and liver-to-heart uptake ratios (LH) in myocardial perfusion imaging (MPI) with technetium-99m analogues. However, in bibliography, these ratios are correlated to a moderate number of clinicolaboratory indices for coronary artery disease (CAD). The aim of this study was: a) to find the best ROI positioning among one hepatic and multiple pulmonary ROI used to semiquantify technetium-99m tetrofosmin, single photon emission tomography myocardial perfusion scan (99mTc-TF SPET), LHR and LH ratios, b) to compare the diagnostic potential of the better selected ROI in correlation with: SPET scores, rest left verticular ejection fraction, five risk factors (RF) of CAD, (smoking, diabetes mellitus, hypertension, hyperlipemia and abnormal exercise-ECG) and coronary angiography (CAR), regarding the latter as the "gold standard" for the severity of CAD and c) to study the multivariate correlations among all the aforementioned diagnostic parameters, so that to rank them according their ability to screen and stratify the existence and the severity of CAD. We have studied 73 patients (54 men and 19 women, mean aged 58.5+/-10.3 and 59.8+/-8.7 years respectively), who underwent stress-rest 99mTc-TF SPET scan. All patients were classified according the following criteria: (a) CAR criteria: Group I: 34 patients with low likelihood of CAD, Group II: 19 patients with moderate CAD, Group III: 7 patients with severe CAD, Group IV: 13 patients with myocardial infarction. (b) RF criteria: Group A: 24 patients with 0-1 RF and Group B: 49 patients with >or=2 RF. (c) LVEF criteria: Group 0: 25 patients with LVEF>50, Group 1: 31 patients with LVEF=40-50, and Group 2: 4 patients with LVEF<40. (d) Regarding the existence (yes/no) of each RF (10 more subgroups). All patients underwent anterior planar imaging before the SPET scan acquisition. ROIs were placed on the following regions: Over the whole myocardium, peripherally, around the upper part of the previous ROI, on the lower right lung field, encompassing the whole right lung, on the middle mediastinum and around the upper part of the liver. Our results by multivariate regression analysis, showed that in 99mTc-TF SPET scan a modified technique of a peripheral ROI, drawn around the upper part of the heart, should be proposed as the optimal method for the calculation of LHR. This index, with normal cutoff