[In vivo proton MR spectroscopy of normal liver parenchyma: technique and results]

To analyze the proton magnetic resonance spectroscopic data ( (1)H MRS) of normal liver parenchyma with regard to age, sex, body mass index and location in the liver.

45 healthy volunteers age 24 to 65 years were examined with an optimized single-voxel (1)H MRS using a 1.5-T scanner. A spin echo sequence with a TR of 1500 ms and a TE of 135 ms was used, allowing in-phase detection of the choline signal. Weak water suppression was achieved using a chemical shift selective suppression (CHESS) technique. Each examination included the measurement of three voxels with a voxel size of 18 x 18 x 18 mm (3) in different areas of the liver. The volunteers were divided into different age-based groups (young: < or = 44 years; older: > or = 44 years), BMI (normal weighted: < 25 kg/m (2); obese: > 25 kg/m (2)) and sex.

In the acquired spectra different lipid (e. g. [CH (2)] (n)), choline, glutamine, glutamate and glycogen-glucose-complex resonances were detected. The analysis of the spectra, however, only focused on the concentrations of choline and (CH (2)) (n) and the relative concentrations of the choline-to-(CH (2)) (n)-ratios. In the older volunteers the relative concentration of the choline-to-(CH (2)) (n)-ratio was significantly decreased by 0.213 +/- 0.193 in comparison to the younger subjects (p = 0,031). Further statistical analysis confirmed a significant decrease of the choline-to-(CH (2)) (n)-ratio by 0.223 +/- 0.180 in obese volunteers compared to volunteers of a standard weight (p = 0,016). The significant difference between the choline-to-(CH (2)) (n)-ratio in female versus male volunteers was calculated with an increase of 0.483 +/- 0.172 (p = 0,000). The location of the voxel in the liver parenchyma did not yield a significant difference in the choline-to-(CH (2)) (n)-ratio.

The analysis of the proton liver MRS of healthy volunteers indicated a significant difference in the choline-to-(CH (2)) (n)-ratio depending on age, sex, and BMI with a confidence interval of 95 %. The different choline-to-(CH (2)) (n)-ratio could be the result of the body fat distribution depending on age and sex and also of the increased fat portion of the body in obese volunteers.