PET measurements of myocardial glucose metabolism with 1-11C-glucose and kinetic modeling.

Abstract

The aim of this study was to investigate whether compartmental modeling of 1-(11)C-glucose PET kinetics can be used for noninvasive measurements of myocardial glucose metabolism beyond its initial extraction. 1-(11)C-Glucose and U-(13)C-glucose were injected simultaneously into 22 mongrel dogs under a wide range of metabolic states; this was followed by 1 h of PET data acquisition. Heart tissue samples were analyzed for (13)C-glycogen content (nmol/g). Arterial and coronary sinus blood samples (ART/CS) were analyzed for glucose (mumol/mL), (11)C-glucose, (11)CO(2), and (11)C-total acidic metabolites ((11)C-lactate [LA] + (11)CO(2)) (counts/min/mL) and were used to calculate myocardial fractions of (a) glucose and 1-(11)C-glucose extractions, EF(GLU) and EF((11)C-GLU); (b) (11)C-GLU and (11)C-LA oxidation, OF((11)C-GLU) and OF((11)C-LA); (c) (11)C-glycolsysis, GCF((11)C-GLU); and (d) (11)C-glycogen content, GNF((11)C-GLU). On the basis of these measurements, a compartmental model (M) that accounts for the contribution of exogenous (11)C-LA to myocardial (11)C activity was implemented to measure M-EF(GLU), M-GCF(GLU), M-OF(GLU), M-GNF(GLU), and the fraction of myocardial glucose stored as glycogen M-GNF(GLU)/M-EF(GLU)). ART/CS data showed the following: (a) A strong correlation was found between EF((11)C-GLU) and EF(GLU) (r = 0.92, P < 0.0001; slope = 0.95, P = not significantly different from 1). (b) In interventions with high glucose extraction and oxidation, the contribution of OF((11)C-GLU) to total oxidation was higher than that of OF((11)C-LA) (P < 0.01). In contrast, in interventions in which glucose uptake and oxidation were inhibited, OF((11)C-LA) was higher than OF((11)C-GLU) (P < 0.05). (c) A strong correlation was found between GNF((11)C-GLU)/EF(GLU) and direct measurements of fractional (13)C-glycogen content, (r = 0.96, P < 0.0001). Model-derived PET measurements of M-EF(GLU), M-GCF(GLU), and M-OF(GLU) strongly correlated with EF(GLU) (slope = 0.92, r = 0.95, P < 0.0001), GCF((11)C-GLU) (slope = 0.79, r = 0.97, P < 0.0001), and OF((11)C-GLU) (slope = 0.70, r = 0.96, P < 0.0001), respectively. M-GNF(GLU)/M-EF(GLU) strongly correlated with fractional (13)C-content (r = 0.92, P < 0.0001). Under nonischemic conditions, it is feasible to measure myocardial glucose metabolism noninvasively beyond its initial extraction with PET using 1-(11)C-glucose and a compartmental modeling approach that takes into account uptake and oxidation of secondarily labeled exogenous (11)C-lactate.