Measurement of skeletal muscle glucose utilization by dynamic 18F-FDG PET without arterial blood sampling

Abstract

Skeletal muscle glucose utilization (SMGU) can be measured by 18F-FDG PET to characterize insulin resistance. The aim of this study was to determine whether femoral muscle SMGU can be measured without arterial blood sampling by sequential PET imaging of the thoracic and femoral regions. Ten patients with possible insulin resistance underwent dynamic 18F-FDG PET of the femoral region during hyperinsulinaemic euglycaemic clamping (group A), and femoral muscle SMGU was calculated using PET data of various time periods and measured arterial input. SMGU was also calculated using venous plasma activity, instead of arterial activity, as input during the late phase. Another five patients underwent sequential PET of the thoracic and femoral regions after single tracer injection (group B). The input function was estimated from aorta activity on thoracic images during the early phase and from venous activity during the late phase, and SMGU with this estimated input was compared with that with measured arterial input. In group A, exclusion of early dynamic PET data from analysis had essentially no effect on the calculated SMGU, and partial substitution of venous activity for arterial activity only marginally changed the estimates. The difference between SMGUs with measured and estimated inputs was minimal in group B. Femoral muscle SMGU can be calculated without femoral imaging early after tracer injection, and the input function can be assessed using data of thoracic imaging and venous blood samples. These results support the validity of measuring femoral muscle SMGU without arterial sampling, simultaneously with measurement of myocardial glucose utilization.