Accuracy of PET in predicting functional recovery after revascularisation in patients with chronic ischaemic dysfunction: head-to-head comparison between blood flow, glucose utilisation and water-perfusable tissue fraction

Accurate prediction of improvement in left ventricular ejection fraction (LVEF) after revascularisation is critical in the therapeutic decision-making process in patients with chronic ischaemic dysfunction. Positron emission tomography (PET) allows non-invasive evaluation of myocardial blood flow (MBF), metabolic rate of glucose (MRG, absolute and relative) and the water-perfusable tissue fraction (PTF). Each of these indices has been proposed for the prediction of functional recovery. Their relative merits, however, are unknown, because a direct head-to-head comparison of their predictive accuracy in the same patients has not been performed. In this study, MBF, MRG (absolute and relative) and PTF were evaluated in 34 patients with severe ischaemic LV dysfunction (LVEF 32% +/- 9%). MBF and PTF were determined by oxygen-15 labelled water PET, and MRG (absolute and relative) was determined by fluorine-18 fluorodeoxyglucose (FDG) PET during hyperinsulinaemic euglycaemic clamp. LVEF was measured by radionuclide ventriculography before and 4-6 months after surgery. Sensitivities of MBF, PTF, absolute MRG and relative MRG in predicting improvement in LVEF were 80%, 80%, 90% and 100%, respectively. Their specificities were 54%, 67%, 71% and 71%, respectively. The lowest specificity was obtained for MBF, an intermediate specificity was obtained for PTF and the highest specificities were obtained with FDG PET using absolute and relative MRG. It is concluded that metabolic imaging is superior to perfusion-based indexes for assessment of the potential for functional recovery after revascularisation.