Higher diagnostic performance of 82-Rubidium-PET in comparison to SPECT myocardial perfusion scintigraphy using CZT cameras for the detection of three-vessel coronary artery disease

Objectives: The use of CZT cameras for SPECT-myocardial perfusion scintigraphy (MPS) and Rubidium(Rb)-PET-MPS is continuously increasing, but very few data exist on the criteria that should be used to identify balanced myocardial ischemia in patients with three-vessel coronary artery disease (3V-CAD) with each imaging technique. The aim of this study was to identify retrospectively in a cohort of 276 patients with low to intermediate prevalence of CAD the optimal threshold values for balanced myocardial ischemia in patients with 3V-CAD using ^99mTechnetium-MIBI (Tc) SPECT-MPS with CZT gamma cameras and Rb-PET-MPS, and to compare the diagnostic performance of each of these criteria for the detection of patients with 3V-CAD.

Methods: 308 patients were recruited prospectively in the clinical study and underwent gated MPS with ^99mTc-MIBI using CZT cameras in association to the best feasible stress (exercise was prioritized) and gated MPS with ⁸²Rb-PET in association with pharmacological stress (dipyridamole). Patients with at least one MPS considered as positive were referred for coronary angiography (CA). Patients were classified as positive for myocardial ischemia in case of significant stenosis (≥ 70%, or ≥ 50 % and FFR ≤ 0.8) on CA or, in absence of CA, occurrence of cardiovascular event during the following year. Patients with an abnormal SPECT or PET who did not undergo CA or with cardiomyopathy (n = 32) were excluded from this analysis. Optimal threshold values for the identification of patients with 3V-CAD were identified for Tc-SPECT-MPS and Rb-PET-MPS using ROC analysis. The diagnostic performance of these thresholds were then evaluated in the population of the study.

Results: Among 276 patients, 12 patients had 3V-CAD, 23 patients 1- or 2-vessel CAD disease and 241 no significant coronary stenosis, or normal SPECT- and PET-MPS with no cardiovascular event after one-year follow-up. Patients with 3V-CAD presented with both imaging techniques moderate stress LVEF dysfunction in comparison to patients without 3V-CAD, but the magnitude of the difference between stress and rest LVEF was higher with Rb-PET-MPS than with Tc-SPECT-MPS in patients with 3V-CAD vs. without 3V-CAD ΔLVEF stress / rest = - 3.6 ± 6.0 % vs. - 0.1 ± 7.1 % for Tc-SPECT-MPS, p = 0.04; ΔLVEF stress / rest = - 4.6 ± 9.6 % vs. 6.0 ± 6.3 % for Rb-PET-MPS, p < 0.001). Global stress MBF and myocardial flow reserve (MFR) quantified with Rb-PET-MPS were significantly lower in patients with vs. without 3V-CAD (stress MBF: 1.1 ± 0.3 vs. 2.0 ± 0.7 ml/mn/g; MFR: 1.6 ± 0.4 vs. 2.6 ± 0.9, respectively; p < 0.001 for both). The AUC for the detection of patients with 3V-CAD were measured at 0.71 for positive stress ECG (> 1-mm ST-depression) and 0.75 for ΔLVEF stress / rest &#8804; - 2 % with Tc-SPECT-MPS; and at 0.76 for positive stress ECG, 0.85 for ΔLVEF stress / rest &#8804; - 0 %, 0.86 for stress MBF &#8804; 1.57 ml/mn/g and 0.85 for myocardial flow reserve (MFR) &#8804; 2.02 for Rb-PET-MPS. The AUC for the detection of patients with 3V-CAD based on MFR quantified with Rb-PET was significantly higher than the AUC for ΔLVEF stress / rest on Tc-SPECT-MPS (p = 0.04), but similar to the AUC of ΔLVEF stress / rest measured on Rb-PET (p = 0.83).

Conclusions: Rb-PET-MPS allows for a more accurate identification of patients with 3V-CAD than SPECT-MPS with CZT camera thanks to improved detection of moderate LV dysfunction and quantification of stress MBF.