Whole-Body Imaging of Tissue-specific Insulin Sensitivity and Body Composition by Using an Integrated PET/MR System: A Feasibility Study.

Abstract

Purpose To develop, evaluate, and demonstrate the feasibility of a whole-body protocol for simultaneous assessment of tissue-specific insulin-mediated fluorine 18 (18F) fluorodeoxyglucose (FDG) influx rates, tissue depots, and whole-body insulin sensitivity (referred to as the M value). Materials and Methods An integrated positron emission tomography (PET)/magnetic resonance (MR) imaging system combined with hyperinsulinemic euglycemic clamp (HEC) was used. Dynamic whole-body PET imaging was used to determine the insulin-mediated 18F-FDG tissue influx rate (Ki) in the whole-body region by using the Patlak method. M value was determined with the HEC method at PET imaging. Tissue depots were quantified by using water-fat separated MR imaging and manual segmentations. Feasibility of the imaging protocol was demonstrated by using five healthy control participants and five patients with type 2 diabetes. Associations between M value and Ki were studied in multiple tissues by using the Pearson correlation. Results Positive correlations were found between M value and Ki in multiple tissues: the gluteus muscle (r = 0.875; P = .001), thigh muscle (r = 0.903; P , .001), calf muscle (r = 0.825; P = .003), and abdominal visceral adipose tissue (r = 0.820; P = .004). A negative correlation was found in the brain (r = 20.798; P = .006). The MR imaging-based method for quantification of tissue depots was feasible for determining adipose tissue volumes and fat fractions. Conclusion This PET/MR imaging protocol may be feasible for simultaneous assessment of tissue-specific insulin-mediated 18F-FDG influx rates, tissue depots, and M value. © RSNA, 2017 Online supplemental material is available for this article.