Absolute Resting N-Ammonia PET Myocardial Blood Flow for Predicting Myocardial Viability and Recovery of Ventricular Function after Coronary Artery Bypass Grafting.

OBJECTIVE

We aimed to evaluate the feasibility of resting myocardial blood flow (rMBF), quantified with dynamic 13 N-Ammonia (NH) PET, for identifying myocardial viability and predicting improvement of left ventricular ejection fraction (LVEF) after coronary artery bypass grafting (CABG).

METHODS

Ninety-three patients with coronary artery disease (CAD) and chronic LVEF < 45%, scheduled for CABG, had dynamic NH PET and F-FDG PET imaging. The perfusion/metabolism polar maps were categorized in four patterns: normal (N), mismatch (M1), match (M2) and reverse mismatch (RM). The value of rMBF for identifying viable myocardium (M1, RM) and post CABG improvement of LVEF≥8% was analyzed by receiver operating characteristic (ROC) curves. Correlations of rMBF in segments to ΔLVEF post CABG were verified.

RESULTS

Mean rMBFs were significantly different (N=0.60±0.14; M1=0.44±0.07, M2=0.34±0.08, RM=0.53±0.09 ml/min/g, P<0.001). The optimal rMBF cutoff to identify viable myocardium was 0.42 ml/min/g (sensitivity=88.3%, specificity=82.0%) and 0.43 ml/min/g for predicting improvement of LVEF ≥8% (74.6%, 80.0%). The extent and rMBF of combined M1/RM demonstrated a moderate to high correlation to improved LVEF (r=0.78, 0.71, P<0.001).

CONCLUSION

Resting MBF, derived by dynamic NH PET, may be positioned as a supplement to F-FDG PET imaging for assessing the presence of viable myocardium and predicting potential improvement of LVEF after CABG.