Can nitrogen-13 ammonia kinetic modeling define myocardial viability independent of fluorine-18 fluorodeoxyglucose?

OBJECTIVES: The hypothesis of this study was that evaluation of myocardial flow and metabolism using nitrogen-13 (N-13) ammonia kinetic modeling with dynamic positron emission tomographic (PET) imaging could identify regions of myocardial scar and viable myocardium as defined by fluorine-18 fluorodeoxyglucose (F-18 FDG) PET. BACKGROUND: Uptake of most perfusion tracers depends on both perfusion and metabolic retention in tissue. This characteristic has limited their ability to differentiate myocardial scar from viable tissue. The kinetic modeling of N-13 ammonia permits quantification of blood flow and separation of the metabolic component of its uptake, which may permit differentiation of scar from viable tissue. METHODS: Sixteen patients, > 3 months after myocardial infarction, underwent dynamic N-13 ammonia and F-18 FDG PET imaging. Regions of reduced and normal perfusion were defined on static N-13 ammonia images. Patients were classified into two groups (group I [ischemic viable], n = 6; group II [scar], n = 10) on the basis of percent of maximal F-18 FDG uptake in hypoperfused segments. Nitrogen-13 ammonia kinetic modeling was applied to dynamic PET data, and rate constants were determined. Flow was defined by K1; volume of distribution (VD = K1/k2) of N-13 ammonia was used as an indirect indication of metabolic retention. RESULTS: Fluorine-18 FDG uptake was reduced in patients with scar compared with normal patients with ischemic viable zones (ischemic viable 93 +/- 27% [mean +/- SD]; scar 37 +/- 16%, p < or = 0.01). Using N-13 ammonia kinetic modeling, flow and VD were reduced in the hypoperfused regions of patients with scar (ischemic viable flow: 0.65 +/- 0.20 ml/min per g, scar: 0.36 +/- 0.16 ml/min per g, p < or = 0.01; VD: 3.9 +/- 1.3 and 2.0 +/- 1.07 ml/g, respectively, p < or = 0.01). For detection of viable myocardium in these patients, the sensitivity and specificity were 100% and 80% for N-13 ammonia PET flow > 0.45 ml/min per g; 100% and 70% for VD > 2.0 ml/g; and 100% and 90% for both flow > 0.45 ml/min per g and VD > 2.0 ml/g, respectively. The positive and negative predictive values for the latter approach were 86% and 100%, respectively. CONCLUSIONS: In this cohort, patients having regions with flow < or = 0.45 ml/min per g or VD < or = 2.0 ml/g had scar. Viable myocardium had both flow > 0.45 ml/min per g and VD > 2.0 ml/g. Nitrogen-13 ammonia kinetic modeling permits determination of blood flow and metabolic integrity in patients with previous myocardial infarction and can help differentiate between scar and ischemic but viable myocardium.