Lean Chinese males are more insulin sensitive than lean Asian-Indian males, matched for age and BMI [1,2]. The mechanisms underlying these ethnic differences in insulin sensitivity are not clear. Mitochondrial function may play a key role in skeletal muscle insulin sensitivity [3,4]. In this study we used ³¹P MRS to examine skeletal muscle mitochondrial function at rest and exercise state between adult Chinese and Asian-Indian males.

10 Chinese (age 28±6 years, BMI 23 ±1 kgm^-2) and 7 Asian-Indian males (age = 26±3 years, BMI = 23 ±3 kgm^-2) were subjected to MRS at their right calf muscle using a Siemens Prisma 3T system using a double tuned flexible ³¹P-¹H coil . A 18ch body coil was used for acquisition of high resolution ¹H images. All subjects had normal glucose tolerance test and all underwent hyperinsulinemic euglycemic clamp for assessment of insulin sensitivity.

T1-weighted TFE images: High resolution MR images were acquired with the following parameters for assessment of muscle cross-sectional area and subsequent calculation of contractile power. Muscle cross-sectional area was analysed by region drawing in ImageJ.

³¹P MRS: ³¹P MR spectra were acquired using a pulse-acquire sequence and the following parameters: BW = 3000, Number of data-points = 2048. For assessment of PCr recovery rates, individual spectra were acquired with TR=2000ms during an exercise paradigm. Spectra were averaged in groups of 4 (8s) to improve SNR prior to peak fitting. Data were analysed using jMRUI. 2048 data-points were truncated and 12Hz Lorentzian line-broadening was added before phase correction. Peaks were fitted using the AMARES algorithm in jMRUI.

Exercise: Prior to scanning, maximum voluntary contraction (MVC) was first calculated. Ergometer pressure was set to maximum and was decreased until the subject could perform one full compression of the pedal. This was repeated twice more and the average value was taken. Following positioning in the scanner and acquisition of resting ³¹P data, the ³¹P exercise data was acquired continuously while subjects performed the exercise protocol which sequentially involved 5 minutes resting, 3 minutes of exercise and then 8 minutes of rest. Exercise was performed at 44% of calculated MVC.

NIRS: A near infra-red spectroscopy probe (OxiplexTS) was positioned on the muscle to allow measurement of muscle oxy-haemoglobin concentration [HbO₂], deoxy-haemoglobin concentration [Hb] and oxygen saturation [OS]. Mean values during exercise (taken from 1.5-2.5 minutes) and the recovery rate (from 1 minute after cessation of exercise) were calculated.

Data from four subjects were excluded from the analysis due to poor spectral quality from external noise entering the scanner. A typical ³¹P spectrum is shown in Figure 1.

Group differences: Fasting blood glucose levels and insulin sensitivity (M-value) were similar between Chinese and Asian-Indians. No ethnic differences were seen for any of the parameters measured (τPCr, THC, HbO_­2­, Hb, OS).

Correlations: τPCr versus HbA1c and τPCr versus fasting blood glucose levels showed a trend for longer τPCr with increasing HbA1c values and fasting blood glucose levels; however, these did not reach significance. No correlation was seen between τPCr and M-value.

Although correlation between τPCr and both HbA1c and fasting blood glucose levels showed a trend for longer τPCr with increasing HbA1c values and fasting blood glucose levels, indicating poorer mitochondrial function with poor blood glucose control, this did not reach significance. No correlation was observed with M-value, another marker of blood glucose control. However, the low sampling sizes in this study and non-significant difference between all group measures mean we are unable to assess inter-ethnic differences in this study.In lean Asians, mitochondrial function, as measured by τPCr, is not significantly associated with fasting blood glucose levels or insulin sensitivity. There is no ethnic difference in mitochondrial function.