Automatic calculation of myocardial external efficiency using a single 11C-acetate PET/CT scan: a dual-center validation

Objectives Myocardial oxygen consumption (MVO2) measured with 11C-acetate PET is used to calculate myocardial external efficiency (MEE), a powerful marker of cardiac performance. Calculation of MEE requires additional data on stroke volume and left ventricular (LV) mass from a second image modality. A dual scan protocol is challenging and differences in loading conditions between scans can cause errors in MEE estimates. Therefore, we sought to develop and validate in a dual-center setting an automatic method of calculating MEE using only 11C-acetate PET.

Methods 30 patients with aortic valve stenosis (AS) and 10 healthy controls (HC) were recruited in Aarhus, Denmark. In addition, 10 patients with mitral insufficiency (MI) were recruited in Uppsala, Sweden. All subjects underwent a dynamic 27 min 11C-acetate PET scan on a Siemens Biograph TruePoint 64 PET/CT (AS, HC) or on a GE Discovery ST PET/CT (MI). Using Cardiac VUer, the arterial input function was obtained automatically and utilized to calculate forward stroke volume (FSV) using the indicator dilution principle, and to calculate MVO2 by kinetic modeling. Using parametric images, the entire LV wall was segmented fully automatically and LV mass was derived. Gold standard LV mass and FSV were obtained with cardiac magnetic resonance (CMR) for all patients. MEE was calculated using MVO2 and either CMR derived (MEEMRI) or PET derived (MEEPET) FSV and LV mass .

Results Correlation between MEEMRI and MEEPET was high for both AS (r=0.84) and MI (r=0.83) with no difference between MEEMRI and MEEPET (p=0.15 and p=0.11 for AS and MI). MEEPET was 25.2±6.2% for HC, 17.3±4.6% for AS (p=0.003 vs HC) and 17.8±5.8% for MI (p=0.006 vs HC).

Conclusions Myocardial efficiency can be calculated automatically using only a single 11C-acetate PET scan. This facilitates clinical applicability and provides unbiased access to a sensitive marker of cardiac dysfunction.