Glucose uptake and metabolism in the human brain during hypoglycaemia assessed by in vivo magnetic resonance

Introduction

In type-1 diabetes, iatrogenic hypoglycaemia is a serious complication of intensive insulin therapy that primarily endangers brain function. Little is known about the effect of hypoglycaemia on brain glucose uptake and metabolism. Magnetic Resonance (MR) provides several non-invasive methods to investigate these factors in the human brain in vivo, i.e. 13C MR spectroscopy (MRS) to follow uptake and metabolism of 13C-labeled glucose, 1 H MRS to determine local glucose concentrations, 31P MRS to assess high energy phosphate compounds and pH, and arterial spin labeling (ASL) to measure blood flow.

Aim of the study

To test the feasibility of the application of these MR methods on the human brain under euglycaemic and hypoglycaemic conditions for the assessment of local changes in glucose uptake and (energy) metabolism.

Methods

Eleven healthy volunteers (9F/2 M, 21.9 ± 1.6 years) underwent a hyperinsulinaemic (60–120 mU×m-2×min-1) euglycaemic (∼ 5.3 mmol/L; 60 min.) – hypoglycaemic (∼ 2.9 mmol/L; 60 min.) glucose clamp ¹ .

Measurements included

13C MRS using 13C-1-glucose isotopic enrichment to determine cerebral glucose uptake and subsequent metabolism (n=6), 1 H MRS to determine cerebral glucose content (n=5), 31P MRS to measure global cerebral energy demand (n=2), and ASL to measure cerebral perfusion (n=3). In one subject undergoing 13C MRS, a 100% 13C-enriched glucose solution was used during the hypoglycaemic phase of the clamp. All experiments were performed on a sophisticated 3 T Siemens Trio MR spectrometer using extended tubing.

Results

Cerebral glucose content, measured by 1 H MRS, averaged 1 mmol/L under euglycaemic conditions, but was undetectable during hypoglycaemia. Plasma 13C isotopic enrichment (determined by hf 1H-NMR) was 29±1% during euglycaemia and 25±1% during hypoglycaemia, except in the subject receiving 100% 13C-enriched glucose where it increased to 68%. 13C MRS resulted in spectra of good quality allowing monitoring of brain glucose uptake and time-dependent conversion to metabolites under both glycaemic conditions. The experiment using 100% 13C-enriched glucose revealed ongoing cerebral glucose uptake and subsequent metabolism under hypoglycaemic conditions. Overall brain perfusion measured by ASL was 47.5±5.8 mL×min-1×100 mL-1 brain tissue during euglycaemia and 46.9±4.6 mL×min-1×100 mL during hypoglycaemia (P=NS). The 31P MRS experiments resulted in spectra with clearly resolved resonances of phosphomonoesters, phosphodiesters, intracellular inorganic phosphate (Pi), phosphocreatine (PCr), and alpha, beta, gamma-ATP. Upon changing the glycaemic condition to hypoglycaemia, PCr/ATP remained unaltered, but Pi resonance shifted slightly equaling an intracellular pH increase in the order of ∼0.01–0.02.

Conclusion

Hyperinsulinaemic euglycaemic-hypoglycaemic clamp conditions can be safely applied in a 3 T MR magnet. Acute hypoglycaemia results in a lower glucose content in the human brain, but does not affect cerebral blood flow or global cerebral energy demand and does not abolish cerebral glucose metabolism. All MR methods worked sufficiently to allow implementation of these techniques for future studies on hypoglycaemia in diabetic patients.