The risk for cardiovascular events conferred by decreased renal function is curvilinear with exponentially greater increases in risk as estimated glomerular filtration rate (eGFR) declines. In 13 non-diabetic pre-dialysis chronic kidney disease (CKD) patients, we employed quantitative F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) as a means to measure myocardial metabolic changes. Methods: Dynamic cardiac FDG PET images were acquired after 6 h fasting and glucose loading. Corrections for attenuation, scatter, randoms, dead time and decay were applied to the PET data and myocardial glucose utilization (MGU) was calculated using the Patlak method in conjunction with standardized myocardial regions of interest and an image-derived input function (left atrium). MGU was compared with eGFR based on a serum creatinine drawn within 2 weeks of the study date. Results: MGU was relatively uniform between the myocardial sectors (coefficient of variation = 16.2 ± 6.8%) within each patient. Between patients, whole myocardium MGU varied considerably with a range of 37.3–156.2 µmol/min/100 g and a mean of 68.9 ± 38.3 µmol/min/ 100 g. eGFR ranged from 11–89 ml/min/1.73 m2 with a mean of 42.8 ± 26.9 ml/min/1.73 m2. There was an inverse correlation between whole myocardium MGU and eGFR (Spearman’s rho correlation = –0.615, p = 0.025). In multivariate analysis, the relationship between MGU and eGFR was sustained with adjustment for age, race and gender (adjusted β = –1.56 ± 0.48, p = 0.01). There was no correlation between cardiac workload and eGFR (p = NS). Conclusions: A significant inverse correlation between MGU and eGFR is supportive of the hypothesis that CKD is associated with myocardial metabolic changes, which could not be attributed to demographic factors or cardiac workload. Dynamic FDG PET could provide a sensitive, non-invasive, quantitative tool for investigating pre-clinical myocardial abnormalities in patients with CKD.