Obesity, metabolic syndrome and diabetes are public health problems leading to increased morbidity and mortality. The accumulation of organ fat, particularly in the liver (known as non-alcoholic fatty liver disease; NAFLD), and the regional distribution and composition of abdominal adipose tissue, i.e. subcutaneous and visceral adipose tissue volumes (SAT and VAT) and their fatty acid unsaturation levels, play an important role to identify different obese phenotypes, e.g. metabolically benign and insulin-resistant obese, which has clinical implications on patient treatment. Great VAT volume and liver fat were reported to be the strongest determinant of insulin sensitivity [1]. Furthermore, VAT volume and liver fat have been shown to be closely related as increasing VAT volume was associated with higher incidence of NAFLD [2]. While fraction of unsaturated fatty acids in VAT inversely correlated with VAT volume [3], the relationship between liver fat and fatty acid unsaturation level in abdominal adipose tissue remains unclear. Therefore, the present study investigated the association between liver fat, abdominal adipose tissues distribution (all measured with chemical shift encoding-based water-fat MRI), and composition (i.e. fatty acid unsaturation levels in VAT and SAT obtained by using MRS).

MR measurements: Forty-six subjects (15 male, 31 and females) were recruited for this study (age range: 24 to 67 years, BMI range: 22 to 44,5 kg/m², BMI mean: 33±5 kg/m²). Two-point Dixon images were acquired covering the entire abdominal and pelvic region on a 3 T system (Ingenia, Phillips Healthcare) using anterior and posterior coils and two stacks with: TR = 4.0 s, TE1/TE2 = 1.32/2.6 ms, flip angle = 10°, bandwidth = 1004 Hz/pixel, 332x220 acquisition matrix size, FOV = 500x446 mm², acquisition voxel = 1.5x2.0x5.0 mm³, 44 slices, SENSE with R = 2.5. STEAM MRS was performed in regions within the VAT and deep SAT at the level of L5 to obtain the fatty acid composition in both adipose tissue compartments using a 20x20x20 mm3 voxel size, TR = 2 s, TE = 11/15/20/25 ms, 4096 samples, spectral bandwidth= 3.5 kHz and 16 averages. A six-echo gradient echo sequence (mDixon quant) was used to measure liver fat fraction with: TR = 7.8 ms, TE1/ΔTE= 1.3/1.1 ms, flip angle = 3°, bandwidth = 1523 Hz/pixel, 152x133 acquisition matrix size, FOV = 300x403 mm², acquisition voxel = 2.0x3.0x6.0 mm³, 25 slices, SENSE with R = 2.2x1.2 (in L/R and F/H respectively).

Data analysis: An automatic segmentation algorithm was used to determine SAT and VAT volumes based on the two-point Dixon water and fat-separated images, as described previously [4]. MR spectra at first TE were processed offline using custom-built frequency-based peak fitting routines. The area of the olefinic peak (peak A in Fig. 1) and the methylene peak (peak F in Fig. 1) were computed and the ratio of the olefinic peak to the methylene peak (ratio F to A) was determined as a measure of fat unsaturation in SAT and VAT. Liver proton density fat fraction (PDFF) maps were computed using the online fat quantification algorithm on the scanner (mDixon quant), which employs a water-fat signal model accounting for the presence of multiple fat peaks and a single T2*. Region of interests were drawn on the PDFF maps in segment VII of the liver to determine liver PDFF.

Table 1 lists the correlation coefficients between the measured parameters. SAT volume significantly correlated with VAT volume (r=0.34; p<0.05). Correlation of VAT and SAT unsaturation amounted (r=0.29; p<0.05). The liver PDFF was positively associated with VAT volume (r = 0.38; p<0.05) and SAT volume (r=0.47; p<0.05). Furthermore, liver PDFF was inversely correlated with SAT unsaturation level (r=-0.53; p<0.05). Figure 1 shows representative liver fat fraction maps and VAT spectra from two subjects.The present study allowed interesting insights into the obese phenotype. Liver fat fraction was not only associated with abdominal adipose tissue compartments (i.e. SAT and VAT volumes), but also with SAT unsaturation profiles. These findings suggest that abdominal adipose tissue dysfunction reflected by SAT unsaturation levels are closely related with the clinically important pathological fat accumulation in the liver. Thus, abdominal adipose tissue profiling may help to differentiate the multifaceted obese phenotype which is important for risk stratification and selection of appropriate treatment and lifestyle interventions options in clinical routine.