Advertisement

Assessment of Abdominal Fat Using High-field Magnetic Resonance Imaging and Anthropometric and Biochemical Parameters

Published:October 14, 2016DOI:https://doi.org/10.1016/j.amjms.2016.09.009

      Abstract

      Background

      To measure the abdominal subcutaneous fat (SF) and visceral fat (VF) volumes using high-field magnetic resonance imaging (MRI) and to investigate their association with selected anthropometric and biochemical parameters among obese and nonobese apparently healthy participants.

      Methods

      A cross-sectional study was conducted by recruiting 167 healthy participants. Abdominal scans were acquired at 3T MRI, and the SF and VF were segmented and their volumes were calculated. Selected anthropometric and biochemical measurements were also determined.

      Results

      A significant difference (P < 0.05) was observed between normal body weight and overweight and obese participants for SF and VF, total abdominal fat volumes, leptin, resistin, adiponectin and waist circumference. Waist circumferences were measured by tape and MRI. Findings revealed that MRI-measured fat volumes were different between males and females and had a significant (P < 0.01) strong positive correlation with body mass index, leptin, resistin and WC and had a negative correlation with adiponectin level. MRI-measured fat volumes were found to correlate moderately with interleukin-6 and weakly with cholesterol, serum triglyceride and low-density lipoprotein. Except for cholesterol, all measured biochemical variables and abdominal fat volumes in the current study were significantly associated with body mass index.

      Conclusions

      All anthropometric and biochemical parameters showed weak-to-strong associations with the MRI-measured fat volumes. Abdominal fat distribution was different between males and females and their correlations with some lipid profiles were found to be sex dependent. These findings revealed that MRI can be used as an alternative tool for obesity assessment.

      Key Indexing Terms

      Introduction

      Obesity is defined as an abnormal accumulation of body fat, usually 20% or more over an individual׳s ideal body weight.
      • World Health Organization
      Body fat is distributed into 2 main compartments—subcutaneous fat (SF) tissue and visceral fat (VF) tissue. The latter is known to be hormonally active component of total body fat and possesses unique biochemical characteristics that influence several normal and pathological processes in the human body.
      • Shuster A.
      • Patlas M.
      • Pinthus J.H.
      • et al.
      The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis.
      Studies on regional adipose tissue distribution and metabolism revealed that the proportion of abdominal adipose (either visceral or subcutaneous) tissue could be a driver of the health risks associated with overweight and obesity.
      • Tchernof A.
      • Després J.-P.
      Pathophysiology of human visceral obesity: an update.
      Abdominal or visceral adiposity is caused by many factors such as increase in the synthesis and secretion of very low-density lipoprotein by the liver; reduced clearance of triglyceride-rich lipoproteins; presence of low-density lipoprotein (LDL) particles and the reduced high-density lipoprotein (HDL) levels.
      • Tchernof A.
      • Després J.-P.
      Pathophysiology of human visceral obesity: an update.
      Adipose tissue, and specifically white adipose tissue, was found to be implicated in the synthesis and secretion of most proteins called adipokines.
      • Kershaw E.E.
      • Flier J.S.
      Adipose tissue as an endocrine organ.
      These adipokines include leptin, adiponectin, resistin, tumor necrosis factor-alpha, interleukin-6 (IL-6), plasminogen activator inhibitor-1, visfatin and chemerin.
      Different methods have been used to estimate the body fat distribution. Anthropometric measurements are among these techniques and they are considered as rudimentary and indirect markers for body fat distribution and require the use of many equations that have been validated for different age, sex and ethnic populations.
      • Deurenberg P.
      • Pieters J.J.
      • Hautvast J.G.
      The assessment of the body fat percentage by skinfold thickness measurements in childhood and young adolescence.
      • Deurenberg P.
      • Deurenberg-Yap M.
      Validation of skinfold thickness and hand-held impedance measurements for estimation of body fat percentage among Singaporean Chinese, Malay and Indian subjects.
      • Widhalm K.
      • Schönegger K.
      • Huemer C.
      • et al.
      Does the BMI reflect body fat in obese children and adolescents? A study using the TOBEC method.
      • Deurenberg P.
      • Deurenberg-Yap M.
      • Guricci S.
      Asians are different from Caucasians and from each other in their body mass index/body fat percent relationship.
      • De Lorenzo A.
      • Deurenberg P.
      • Pietrantuono M.
      • et al.
      How fat is obese?.
      Body mass index (BMI) is considered the most common anthropometric measure of adipose tissue.
      • Kvist H.
      • Chowdhury B.
      • Grangard U.
      • et al.
      Total and visceral adipose-tissue volumes derived from measurements with computed tomography in adult men and women: predictive equations.
      However, the validity of BMI as a measure of obesity has been questioned in different ethnic groups and age populations as it does not measure body fat directly and could not be used as a diagnostic tool for determining fat mass.
      • Widhalm K.
      • Schönegger K.
      • Huemer C.
      • et al.
      Does the BMI reflect body fat in obese children and adolescents? A study using the TOBEC method.
      • Deurenberg P.
      • Deurenberg-Yap M.
      • Guricci S.
      Asians are different from Caucasians and from each other in their body mass index/body fat percent relationship.
      • De Lorenzo A.
      • Deurenberg P.
      • Pietrantuono M.
      • et al.
      How fat is obese?.
      On the contrary, BMI can be used as a tool for monitoring and assessing weight status in populations and as a screening tool for identifying potential weight problems among individuals. Other anthropometric measures such as waist circumference (WC), umbilicus circumference and sagittal diameter are considered as easy techniques that could be used to estimate the body fat distribution, but results that come out upon using such techniques are still inconsistent because the body position during measurement may affect the findings.
      • Kvist H.
      • Chowdhury B.
      • Grangard U.
      • et al.
      Total and visceral adipose-tissue volumes derived from measurements with computed tomography in adult men and women: predictive equations.
      • Ball S.D.
      • Swan P.
      Accuracy of estimating intra-abdominal fat in obese women.
      • Seidell J.C.
      • Oosterlee A.
      • Deurenberg P.
      • et al.
      Abdominal fat depots measured with computed tomography: effects of degree of obesity, sex, and age.
      Therefore, the development of accurate, precise and reliable tools for the quantification of body fat and its distribution continues to evolve.
      • So R.
      • Sasai H.
      • Matsuo T.
      • et al.
      Multiple-slice magnetic resonance imaging can detect visceral adipose tissue reduction more accurately than single-slice imaging.
      Different imaging techniques have been proposed to measure body fat deposition and its distribution.
      • Mattsson S.
      • Thomas B.J.
      Development of methods for body composition studies.
      Among these imaging techniques are the dual energy X-ray absorptiometry, ultrasound, computed tomography
      • Seidell J.C.
      • Bakker C.J.
      • van der Kooy K.
      Imaging techniques for measuring adipose-tissue distribution—a comparison between computed tomography and 1.5-T magnetic resonance.
      • Lubura M.
      • Hesse D.
      • Neumann N.
      • et al.
      Non-invasive quantification of white and brown adipose tissues and liver fat content by computed tomography in mice.
      and magnetic resonance imaging (MRI).
      • Napolitano A.
      • Miller S.R.
      • Murgatroyd P.R.
      • et al.
      Validation of a quantitative magnetic resonance method for measuring human body composition.
      Several MRI studies have been conducted in an attempt to achieve different goals. Some of these studies were conducted to validate fat tissue segmentation method (manual, semiautomated or fully automated)
      • Poonawalla A.H.
      • Sjoberg B.P.
      • Rehm J.L.
      • et al.
      Adipose tissue MRI for quantitative measurement of central obesity.
      • Joshi A.A.
      • Hu H.H.
      • Leahy R.M.
      • et al.
      Automatic intra-subject registration-based segmentation of abdominal fat from water-fat MRI.
      • Mantatzis M.
      • Milousis T.
      • Katergari S.
      • et al.
      Abdominal adipose tissue distribution on MRI and diabetes.
      • Positano V.
      • Gastaldelli A.
      • maria Sironi A.
      • et al.
      An accurate and robust method for unsupervised assessment of abdominal fat by MRI.
      • Wang D.
      • Shi L.
      • Chu W.C.W.
      • et al.
      Fully automatic and nonparametric quantification of adipose tissue in fat-water separation MR imaging.
      • Addeman B.T.
      • Kutty S.
      • Perkins T.G.
      • et al.
      Validation of volumetric and single-slice MRI adipose analysis using a novel fully automated segmentation method.
      or approach (single, multislice or volumetric imaging)
      • Mantatzis M.
      • Milousis T.
      • Katergari S.
      • et al.
      Abdominal adipose tissue distribution on MRI and diabetes.
      • Addeman B.T.
      • Kutty S.
      • Perkins T.G.
      • et al.
      Validation of volumetric and single-slice MRI adipose analysis using a novel fully automated segmentation method.
      • Warren M.
      • Schreiner P.J.
      • Terry J.G.
      The relation between visceral fat measurement and torso level--is one level better than another? The Atherosclerosis Risk in Communities Study, 1990–1992.
      • Demerath E.W.
      • Shen W.
      • Lee M.
      • et al.
      Approximation of total visceral adipose tissue with a single magnetic resonance image.
      • Shen W.
      • Punyanitya M.
      • Wang Z.
      • et al.
      Visceral adipose tissue: relations between single-slice areas and total volume.
      • So R.
      • Matsuo T.
      • Sasai H.
      • et al.
      Best single-slice measurement site for estimating visceral adipose tissue volume after weight loss in obese, Japanese men.
      • Shen W.
      • Chen J.
      • Gantz M.
      • et al.
      A single MRI slice does not accurately predict visceral and subcutaneous adipose tissue changes during weight loss.
      accounting for variation in fat distribution between slices. Other studies were performed to evaluate the MR acquisition technique.
      • Poonawalla A.H.
      • Sjoberg B.P.
      • Rehm J.L.
      • et al.
      Adipose tissue MRI for quantitative measurement of central obesity.
      • Wang D.
      • Shi L.
      • Chu W.C.W.
      • et al.
      Fully automatic and nonparametric quantification of adipose tissue in fat-water separation MR imaging.
      • Lancaster J.L.
      • Ghiatas A.A.
      • Alyassin A.
      • et al.
      Measurement of abdominal fat with T1-weighted MR images.
      • Ludwig U.A.
      • Klausmann F.
      • Baumann S.
      • et al.
      Whole-body MRI-based fat quantification: a comparison to air displacement plethysmography.
      Further research attempts were undertaken to assess any possible association between MRI-derived fat volumes and some anthropometric or biochemical parameters,
      • Leenen R.
      • van der Kooy K.
      • Seidell J.C.
      • et al.
      Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
      or to evaluate some pathological conditions and health status.
      • Mantatzis M.
      • Milousis T.
      • Katergari S.
      • et al.
      Abdominal adipose tissue distribution on MRI and diabetes.
      • Leenen R.
      • van der Kooy K.
      • Seidell J.C.
      • et al.
      Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
      • Wander P.L.
      • Boyko E.J.
      • Leonetti D.L.
      • et al.
      Change in visceral adiposity independently predicts a greater risk of developing type 2 diabetes over 10 years in Japanese Americans.
      • Vogt L.J.
      • Steveling A.
      • Meffert P.J.
      • et al.
      Magnetic resonance imaging of changes in abdominal compartments in obese diabetics during a low-calorie weight-loss program.
      • Gray D.S.
      • Fujioka K.
      • Colletti P.M.
      • et al.
      Magnetic-resonance imaging used for determining fat distribution in obesity and diabetes.
      In these studies, MRI proved to be a valuable and reliable imaging technique to study and assess abdominal fat distribution. In addition, multislice segmentation approach provided a more accurate and consistent readings especially when measuring the VF that varies between slices
      • Kanaley J.A.
      • Giannopoulou I.
      • Ploutz-Snyder L.L.
      Regional differences in abdominal fat loss.
      and showed a higher level of correlation with total body fat volume as well as anthropometric and biochemical findings when compared to single slice segmentation approach.
      Therefore, the main aim of the present study was to measure the abdominal SF and VF volumes (from dual-echo T1-weighted images acquired using a 2-point chemical shift approach) across a range of healthy obese and nonobese individuals and assess any possible association between those volumes, anthropometric and biochemical parameters of those participants to provide a better assessment of obesity using such combined approach of evaluation.

      Materials and Methods

      Study Population

      The present study has been conducted in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki). All participants were asked to sign a consent form according to the approval of local research ethics committee and filled out the MRI safety questionnaire before participating in this study. A total of 167 apparently healthy participants (83 males, 84 females; mean age = 27.2 years; age range: 18-51 years) were recruited from the Royal Medical Services personnel (including the security manpower, hospital cleaners and employees in administrative positions) to participate in this cross-sectional study. In addition, other participants who met the inclusion criteria, which require that the participants were disease-free and at least 18 years old, were also recruited. All participants agreed to be enrolled in the present study except for 11 participants, who refused to give a blood sample and complete the anthropometric measurements. The exclusion criteria included being pregnant or lactating (for women) and suffering from eating disorders or having any disease. Sociodemographic and health data were collected by trained research assistant using interview-based questionnaires. The sociodemographic data included age, marital status, household income, education (illiterate, primary and secondary, diploma and BSc and postgraduate degrees), occupation and tobacco usage (current and previous smokers were categorized as smokers and those who never smoked were set as nonsmokers).
      Physical activity level was determined using 7-day Physical Activity Recall (PAR).
      • Sallis J.F.
      • Haskell W.L.
      • Wood P.D.
      • et al.
      Physical activity assessment methodology in the Five-City Project.
      The 7-Day PAR is a structured interview that depends on participant׳s recall of time spent engaging in physical activity over a 7-day period.
      • Sallis J.F.
      • Haskell W.L.
      • Wood P.D.
      • et al.
      Physical activity assessment methodology in the Five-City Project.
      Participants were asked to respond to a PAR question based on the way they used to behave in their usual days. The number of hours spent in different activity levels were obtained and converted into metabolic equivalents (METs). The total physical activity MET minutes per week was obtained by summing the METs and then performing categorical analysis (inactive, minimally active or health-enhancing physical activity active).
      • Sallis J.F.
      • Haskell W.L.
      • Wood P.D.
      • et al.
      Physical activity assessment methodology in the Five-City Project.

      Anthropometric Measurements

      Body weight was measured to the nearest 0.1 kg, with minimal clothing and without shoes, using a calibrated scale (Tanita, Model SC-331S, Japan).

      D Nieman, R Lee. Nutritional Assessment. 6th ed. McGraw-Hill Education; 2013.

      Height was measured to the nearest 1 cm with participants in standing position without shoes using a calibrated portable measuring rod.

      D Nieman, R Lee. Nutritional Assessment. 6th ed. McGraw-Hill Education; 2013.

      BMI was calculated as the ratio of weight in kilograms to the square of height in meters and was categorized according to the classification system established by the National Institutes of Health: normal body weight, 18.5-24.9; overweight, 25.0-29.9 and obese, >30.0.
      Executive summary of the clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults..
      Waist circumference (WC) was measured by tape (WC_tape) at the narrowest level between the lowest rib and the iliac crest at the end of normal expiration in standing position.

      D Nieman, R Lee. Nutritional Assessment. 6th ed. McGraw-Hill Education; 2013.

      In addition, WC was measured from the MR images (WC_MRI) at a level just below the lower costal margin using image analysis software (Slice-O-Matic, Tomovision Inc., Montreal, Canada) in a semiautomated approach. All anthropometric measurements were performed by a trained dietitian.

      Biochemical Analysis

      Venous blood samples were withdrawn after 12-hour overnight fasting by specialized medical laboratory technicians. Serum samples were separated from the whole blood and stored at −80°C until subsequent analyses.

      Determination of Adipokines

      Serum inflammatory cytokine, IL-6 and obesity-related adipokines, or adipocytokines (adiponectin, resistin and leptin) concentrations were measured by ELX 800 TC models 96-well Elisa Microplate Readers, USA using commercially available enzyme-linked immunosorbent assay (ELISA) kits (RayBio Human IL-6 ELISA Kit, USA; Cat# ELH-IL-6, RayBio Human Acrp30 ELISA Kit USA; Cat# ELH-Adiponectin, RayBio Human Leptin ELISA Kit, USA; Cat# ELH-Leptin, RayBio Human Resistin ELISA Kit, USA and Cat# ELH-Resistin). The reproducibility of the intra-assay and interassay coefficients of variation were <10% and <12%, respectively for all RayBio ELISA kits.

      Determination of Lipid Profile

      Fasting blood lipid profile, including total cholesterol, LDL, HDL and triglyceride (TG) levels were measured by Jenway 6305 UV/visible spectrophotometer, USA using commercially available enzymatic kits from Teco diagnostics, USA.

      MRI Measurement

      MR scanning was performed on a 3-T Siemens Trio MR system (Siemens Healthcare, Erlangen, Germany), equipped with 4-channel phase-array body coil. Nonenhanced transverse T1-weighted in-phase and out-of-phase breath-hold spoiled gradient-echo MRI sequence provided by the manufacturer was performed with repetition time of 5 milliseconds, echo time of 1.225 milliseconds (OP imaging) and 2.45 milliseconds (IP imaging), flip angle of 10°, slice thickness of 5 mm, number of slices of 80, matrix size of 256 × 192, field of view of 380 × 285 mm2, number of averages of 1, acceleration factor of 2 and scan time of 18 seconds. The acquired slices covered the region between the level above the diaphragm and the head of the femur in supine position and full expiration. The MRI scanner calculated the “fat” and “water” images as follows:
      Fatimage=[IPOP]2=[(water+fat)(waterfat)]2
      (1)


      Waterimage=[IP+OP]2=[(water+fat)+(waterfat)]2
      (2)


      All abdominal MR images were checked by an MRI physicist for any artifact while the participant was still in the scanner. MR fat images of 14 participants were found to be affected by motion (8 cases), motion and phase (3 cases) or phase (3 cases) artifact. Fat-water signal swap artifact was attributed to a failure in phase correction owing to main magnetic field (Bo) inhomogeneity, which was not corrected by scanner׳s shimming at the beginning of scanning. Furthermore, motion artifact was caused by participant׳s breathing during scanning. So, a second scan was performed to re-do the shimming (to correct fat images for the phase error) and to give participants clear breathing instructions to avoid any motion artifact. As a result, all MR fat images of the 14 participants were free of any motion or phase artifacts or both on their second scan.

      Image After Processing and Analysis

      “Fat” MR images were imported into Slice-O-Matic in their standard formats (Digital Imaging and Communications in Medicine) to segment the SF and VF tissues. Slices covering the region between the top of the diaphragm and the top of the first sacral vertebra (S1) were used to segment the SF and VF tissues. Depending on the height of the participant, a total of 38-46 axial slices per participant were analyzed by a trained technician. Two different methods were used to segment both tissues. While the SF tissue was segmented using snake algorithm, VF tissue was segmented using a threshold-based growing region. Each segmented tissue was saved as a separate “tag” and the total volumes of SF and VF were calculated from all analyzed slices and saved into a separate file for further analysis. The model and method employed to segment the various tissues is fully described and illustrated elsewhere.
      • Shen W.
      • Chen J.
      Application of imaging and other noninvasive techniques in determining adipose tissue mass.
      Abdominal SF and VF volumes were calculated for all analyzed slices. The Figure shows an example of single axial “fat” MR image and its segmentation into SF and VF tissues in Slice-O-Matic software. The figure displays a 3-dimensional image rendering of both fat tissues (VF and SF) from the top of the diaphragm to the top of the first sacral vertebra.
      FIGURE
      FIGUREAn example of “fat” MR image in its gray scale (A), segmentation of subcutaneous and visceral fat tissue (B) and 3D rendering of fat tissues from all analyzed slices in different orientations (C and D). 3D, 3 dimensional; MR, magnetic resonance.

      Statistical Analyses

      All statistical analyses were performed using SPSS 20.0 (SPSS, Chicago, IL). The significance level was set at P < 0.05 (2-tailed). All the data were tested for normality and linearity. The differences in body fat composition (measured with MRI), anthropometric and biochemical parameters according to sex and BMI were evaluated using one-way analysis of variance and least significant difference tests for multiple comparisons. Pearson׳s correlation coefficients using bivariate and partial (adjusted for age, sex, physical activity and smoking) analyses were performed to evaluate all potential associations between body fat composition and BMI with other variables.

      Results

      The anthropometric, biochemical and lifestyle characteristics of males and females as well as for the whole study samples are shown in Table 1. For participant׳s height, weigh, and WC_tape and WC_MRI, a significant statistical difference (P < 0.05) was detected between males and females with males showing higher values than females. Although the SF tissue volume in males was found to be significantly (P = 0.001) lower (3,779.4 cm3) than in females (5,222.7 cm3), VF tissue volume was significantly (P = 0.001) higher (2,967.1 cm3) in males than in females (1,881.7 cm3). This was also reflected in the ratio of the VF to SF volumes that was found to be significantly (P = 0.000) higher (0.8243) in males than in females (0.3743). No significant difference was found between males and females for adipokines level and lipid profile, except for TG that was significantly (P = 0.001) higher in males (121.4 mg/dL) than in females (84 mg/dL) although it was still within the reference range for both the sexes.
      TABLE 1Anthropometric, biochemical and lifestyle differences within the study sample based on sex.
      ParameterMaleFemale
      Significant at P < 0.05; BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; SEM, standard error of the mean; TG, triglycerides.
      P Value
      Total
      nMeanSEMnMeanSEMnMeanSEM
      Height (cm)77172.40.8181158.60.660.001157165.30.75
      Weight (kg)7779.11.78168.71.360.00115773.81.52
      BMI (kg/m2)7726.70.598127.50.630.25515727.10.43
      Physical activity (MET)5223,080.962,449.166320,463.811,476.720.34411521,647.221,370.41
      Waist circumference (cm) by:
       Tape6593.81.47586.81.30.01714090.00.99
       MRI8390.51.38483.81.40.00116787.10.97
      Subcutaneous fat (cm3)833,779.4220.4845,222.7273.60.0011674,505.4184.1
      Visceral fat (cm3)832,967.1171.3841,881.7100.20.0011672,421.1107.3
      Total abdominal fat (cm3)836,746.5356.4847,104.4358.40.5911676,926.5252.4
      Visceral to subcutaneous ratio830.82430.0317840.37430.11250.0001670.59530.2416
      Leptin (ng/mL)767.50.45807.30.370.8041567.40.29
      Resistin (ng/mL)768.60.44808.50.50.9561568.50.34
      IL-6 (pg/mL)769.20.5809.50.50.6761569.30.35
      Adiponectin (pg/mL)7611.10.58011.30.580.78415611.210.38
      Cholesterol (mg/mL)76163.04.380153.33.20.071156158.02.7
      TG (mg/mL)76121.48.08084.04.70.001156102.24.8
      HDL (mg/mL)7641.01.28042.21.00.41815641.60.77
      LDL (mg/mL)76112.93.780106.22.90.150156109.52.3
      BMI categories, N (%)
       Normal35 (45.5)35 (43.8)0.45870 (44.6)
       Overweight23 (29.9)20 (25.0)43 (27.4)
       Obese19 (24.6)25 (31.2)44 (28.0)
      Marital status, N (%)
       Married31 (40.8)24 (29.6)0.05355 (35.0)
       Single45 (59.2)53 (65.4)98 (62.4)
       Divorced4 (4.9)4 (2.5)
      Education, N (%)
       Illiterate1 (1.5)0.7611 (0.6)
       Primary and secondary education14 (20.6)22 (28.2)36 (21.4)
       Diploma12 (17.6)13 (16.7)25 (14.9)
       Bachelor39 (57.4)42 (53.8)101 (61.3)
       Master and PhD2 (2.9)1 (1.3)3 (1.8)
      Occupation, N (%)
       Yes43 (64.2)45 (57.7)0.06388 (60.7)
       No24 (35.8)33 (42.3)57 (39.3)
      Current smoking, N (%)
       Yes34 (50.9)10 (13)0.00444 (30.6)
       No33 (49.1)67 (87)100 (69.4)
      low asterisk Significant at P < 0.05; BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; SEM, standard error of the mean; TG, triglycerides.
      Table 2 shows significant (P < 0.05) differences between normal body weight, overweight and obese participants when they were compared for the SF, VF, total abdominal fat volume, leptin, resistin, adiponectin and WC_tape and WC_MRI. Serum levels of IL-6, cholesterol, LDL and HDL were significantly (P < 0.05) greater in obese participants than those reported for overweight and normal body weight participants. TG serum concentration was the only biochemical variable which was significantly higher (P < 0.05) in overweight and obese participants when compared with normal body weight participants.
      TABLE 2Anthropometric and biochemical parameters of the study sample based on BMI.
      ParameterNormalOverweightObese
      Significant at P < 0.05 with different letters indicating significant differences between groups based on the least significant difference (LSD) test such that, for the same parameter, similar letters indicate no significant difference between the 2 groups of the participants.
      P Value
      nMeanSEMnMeanSEMnMeanSEM
      Waist circumference by:
       Tape6381.8a0.823890.8b1.26238103.3c1.30.001
       MRI7077.5a0.854288.7b1.244100.2c1.20.001
      Subcutaneous fat (cm3)702,793.9a1,19.2424,526.8b187.6447,104.4c315.00.001
      Visceral fat (cm3)701,535.7a86.4422,674.8b203.9443,543.9c197.60.001
      Total abdominal fat (cm3)704,329.6a163.1427,201.6b257.84410,648.3c349.70.001
      Leptin (ng/mL)684.9a0.18407.9b0.354211.4c0.530.001
      Resistin (ng/mL)686.5a0.33408.3b0.374212.5c0.740.001
      IL-6 (pg/mL)687.9a0.46409.2a0.524211.4b0.820.001
      Adiponectin (pg/mL)6814.7a0.44010.4b0.51426.25c0.460.001
      Cholesterol (mg/mL)68149.5a3.540158.8a5.142171.5b5.90.005
      TG (mg/mL)6884.2a5.140109.7b11.442125.9b10.20.008
      HDL (mg/mL)6844.5a1.34040.5a1.24238.8b1.30.022
      LDL (mg/mL)68101.0a2.940109.9a3.742123.1b5.50.001
      BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; SEM, standard error of mean; TG, triglycerides.
      BMI was categorized according to NIH (21).
      low asterisk Significant at P < 0.05 with different letters indicating significant differences between groups based on the least significant difference (LSD) test such that, for the same parameter, similar letters indicate no significant difference between the 2 groups of the participants.
      Findings illustrated in Table 3 revealed that SF, VF and total abdominal fat volumes had significant (P < 0.01) positive correlation with BMI (r = 0.889, 0.640 and 0.900, respectively), leptin (r = 0.863, 0.728 and 0.870, respectively), resistin (r = 0.698, 0.659 and 0.726, respectively), WC_tape (r = 0.778, 0.713 and 0.803, respectively) and WC_MRI (r = 0.950, 0.816 and 0.963, respectively). In contrast, SF, VF and total abdominal fat volume showed a significant (P < 0.01) strong negative correlation with adiponectin level (r = −0.792, −0.727 and −0.817, respectively). Serum IL-6 level was found to have a moderate correlation with the SF (r = 0.533, P < 0.01), VF (r = 0.462, P < 0.01) and total abdominal fat (r = 0.541, P < 0.01). Cholesterol, TG and LDL showed a significant weak positive correlation with SF (r = 0.290, 0.306 and 0.307, respectively, with P < 0.01), VF (r = 0.238, 0.349 and 0.273, respectively, with P < 0.01) and total abdominal fat (r = 0.291, 0.336 and 0.314, respectively, with P < 0.01). An inverse weak correlation was also detected between HDL and VF as well as total abdominal fat (r = −0.335 and −0.274, respectively, with P < 0.05).
      TABLE 3Correlation coefficients (r) of subcutaneous, visceral and total fat volumes with anthropometric and biochemical parameters for the study sample.
      ParameterSubcutaneous fatVisceral fatTotal abdominal fat
      Age (years)
      r0.9410.7740.943
      P0.0010.0010.001
      BMI
      r0.8890.6400.900
      P0.0010.0010.001
      Waist circumference (cm) by:
       Tape
        r0.7780.7130.803
        P0.0010.0010.001
       MRI
        r0.9500.8160.963
        P0.0010.0010.001
      Leptin (ng/mL)
      r0.8630.7280.870
      P0.0010.0010.001
      Resistin (ng/mL)
      r0.6980.6590.726
      P0.0010.0010.001
      IL-6 (pg/mL)
      r0.5330.4620.541
      P0.0010.0010.001
      Adiponectin (pg/mL)
      r−0.792−0.727−0.817
      P0.0010.0010.001
      Cholesterol (mg/mL)
      r0.2900.2380.291
      P0.0130.0440.013
      TG (mg/mL)
      r0.3060.3490.336
      P0.0090.0030.004
      HDL (mg/mL)
      r−0.229−0.335−0.274
      P0.0530.0040.020
      LDL (mg/mL)
      r0.3070.2730.314
      P0.0090.0200.007
      BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; TG, triglycerides.
      Table 4 shows a significant positive relationship between the 3 fat volumes (SF, VF and total abdominal fat volumes) and BMI, WC_tape, WC_MRI, leptin and resistin (P < 0.001) in both males and females. Adiponectin showed a significant (P < 0.001) strong negative correlation with SF, VF and total abdominal fat volumes in both males (r = −0.885, −0.819, −0.904, respectively) and females (r = −0.762, −0.732, −0.794, respectively). Additionally, data showed that IL-6 had a significant strong positive correlation with SF (r = 0.627, P = 0.009) and moderate positive correlation with total abdominal fat volumes (r = 0.591, P = 0.016) in males, whereas it had a significant moderate correlation with SF, VF and total abdominal fat volumes (r = 0.477, 0.509 and 0.511, respectively) in females (P < 0.001). Regarding the lipid profile, cholesterol, TG and LDL showed a significant (P < 0.05) positive but weak association with the 3 fat volumes in females. In males, LDL showed a significant moderate association only with SF (r = 0.512, P = 0.042), whereas HDL had an inverse moderate association with the 3 fat volumes (P < 0.05).
      TABLE 4Correlation coefficients (r) of subcutaneous, visceral and total fat volumes with anthropometric and biochemical parameters based on sex.
      MaleFemale
      ParameterSubcutaneous fatVisceral fatTotal abdominal fatSubcutaneous fatVisceral fatTotal abdominal fat
      Age (years)
      r0.1890.5610.3870.2840.4470.342
      P0.0870.0010.0010.0090.0010.001
      BMI
      r0.9670.8400.9680.9330.7780.939
      P0.0010.0010.0010.0010.0010.001
      Waist circumference (cm) by:
       Tape
        r0.9200.8420.9360.7220.6730.747
        P0.0010.0010.0010.0010.0010.001
       MRI
        r0.9720.8410.9720.9430.8280.961
        P0.0010.0010.0010.0010.0010.001
      Leptin (ng/mL)
      r0.8860.8260.9070.8700.6620.858
      P0.0010.0010.0010.0010.0010.001
      Resistin (ng/mL)
      r0.8690.7200.8570.6310.6570.672
      P0.0010.0020.0010.0010.0010.001
      IL-6 (pg/mL)
      r0.6270.4450.5910.4770.5090.511
      P0.0090.0840.0160.0010.0010.001
      Adiponectin (pg/mL)
      r−0.885−0.819−0.904−0.762−0.732−0.794
      P0.0010.0010.0010.0010.0010.001
      Cholesterol (mg/mL)
      r0.3260.2030.2970.3490.3120.357
      P0.2180.4520.2640.0100.0220.008
      TG (mg/mL)
      r0.4640.3720.4540.3210.3600.349
      P0.0700.1560.0780.0180.0070.010
      HDL (mg/mL)
      r−0.551−0.555−0.579−0.050−0.186−0.090
      P0.0270.0260.0190.7220.1790.517
      LDL (mg/mL)
      r0.5120.3930.4940.2930.2800.305
      P0.0420.1330.0520.0320.0400.025
      Pearson׳s linear correlation coefficients are given at P < 0.05 after adjustment for age, physical activity and smoking.
      r, correlation coefficients; BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; TG, triglycerides.
      Table 5 shows that BMI had a significant strong positive correlation with WC_tape (r = 0.829, P = 0.001) and WC_MRI (r = 0.934, P = 0.001), leptin (r = 0.867, P = 0.001), resistin (r = 0.748, P = 0.001) and IL-6 (r = 0.569, P = 0.001). However, BMI showed a strong inverse correlation with adiponectin (r = −0.837, P = 0.001). In addition, BMI showed a weak positive correlation with TG (r = 0.257, P = 0.03), LDL (r = 0.259, P = 0.028) and weak negative correlation with HDL (r = −0.327, P = 0.005). Similar results were obtained for almost all biochemical variables when males were separated from females. However, serum cholesterol (in males), TG (in males), HDL (in females) and LDL (in females), showed no significant association with BMI.
      TABLE 5Correlation coefficients (r) of BMI with subcutaneous, visceral and total fat volumes, and anthropometric and biochemical parameters based on sex.
      ParameterMaleFemaleTotal
      rP ValuerP ValuerP Value
      Waist circumference (cm) by:
       Tape0.9420.0010.7860.0010.8290.001
       MRI0.9720.0010.9260.0010.9340.001
      Subcutaneous fat (cm3)0.9670.0010.9330.0010.9410.001
      Visceral fat (cm3)0.8400.0010.7780.0010.7740.001
      Total abdominal fat volume (cm3)0.9680.0010.9390.0010.9430.001
      Leptin (ng/mL)0.9060.0010.8640.0010.8670.001
      Resistin (ng/mL)0.8970.0010.6980.0010.7480.001
      IL-6 (pg/mL)0.5800.0180.5340.0030.5690.001
      Adiponectin (pg/mL)−0.9260.001−0.8070.001−0.8370.001
      Cholesterol (mg/mL)0.2830.2880.2760.0430.2030.088
      TG (mg/mL)0.4730.0640.2770.0430.2570.030
      HDL (mg/mL)−0.6630.005−0.1250.366−0.3270.005
      LDL (mg/mL)0.5190.0390.2530.0650.2590.028
      Pearson׳s linear correlation coefficients are given at P < 0.05 after adjustment for age, sex, physical activity and smoking.
      r, correlation coefficients; BMI, body mass index; HDL, high-density lipoprotein cholesterol; IL-6, interleukin-6; LDL, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; TG, triglycerides.

      Discussion

      Our findings showed that there was a sex difference in abdominal fat distribution with females having a higher proportion of SF and lower VF compared to males although the difference in total abdominal fat in both males and females was insignificant. This could be attributed to the differences in basal fatty acid oxidation, in the regulation of lipolysis and in postprandial fatty acid storage between males and females.
      • Lemieux S.
      • Bouchard C.
      • Tremblay A.
      • et al.
      Sex differences in the relation of visceral adipose tissue accumulation to total body fatness.
      • Blaak E.
      Gender differences in fat metabolism.
      All these findings are in agreement with many studies.
      • Bidulescu A.
      • Liu J.
      • Hickson D.A.
      • et al.
      Gender differences in the association of visceral and subcutaneous adiposity with adiponectin in African Americans: the Jackson Heart Study.
      • Liu J.
      • Fox C.S.
      • Hickson D.A.
      • et al.
      Impact of abdominal visceral and subcutaneous adipose tissue on cardiometabolic risk factors: the Jackson Heart Study.
      • Claussen C.
      • Eschweiler G.
      • Ludescher B.
      • et al.
      Gender specific correlations of adrenal gland size and body fat distribution: a whole body MRI study.
      • Karcher H.-S.
      • Holzwarth R.
      • Mueller H.-P.
      • et al.
      Body fat distribution as a risk factor for cerebrovascular disease: an MRI-based body fat quantification study.
      The results of the current study showed that SF, VF and total abdominal fat measured by MRI were strongly and positively correlated with WC_tape and WC_MRI. Some previous studies have reported that WC was correlated better with fat mass and SF than with VF.
      • Browning L.M.
      • Mugridge O.
      • Dixon A.K.
      • et al.
      Measuring abdominal adipose tissue: comparison of simpler methods with MRI.
      • Camhi S.M.
      • Bray G.A.
      • Bouchard C.
      • et al.
      The relationship of waist circumference and BMI to visceral, subcutaneous, and total body fat: sex and race differences.
      • Wajchenberg B.L.
      Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome.
      Perry et al
      • Perry A.C.
      • Applegate E.B.
      • Jackson M.L.
      • et al.
      Racial differences in visceral adipose tissue but not anthropometric markers of health-related variables.
      attributed that to the amount of SF that was 3.5 times higher than that of VF.
      The significant strong correlations that was detected between the 3 fat volumes (SF, VF and total fat) and leptin were in agreement with those found in a study by Lubkowska et al
      • Lubkowska A.
      • Radecka A.
      • Bryczkowska I.
      • et al.
      Serum adiponectin and leptin concentrations in relation to body fat distribution, hematological indices and lipid profile in humans.
      in which a direct association was found between leptin and all body fat components expressed in absolute values and percentages including VF and SF. In another study by Park et al,
      • Park K.-G.
      • Park K.-S.
      • Kim M.-J.
      • et al.
      Relationship between serum adiponectin and leptin concentrations and body fat distribution.
      the level of leptin was found to be strongly influenced by SF, whereas the level of adiponectin was significantly affected by VF. In all cases, serum adiponectin level was found to be inversely correlated with SF and VF.
      • Park K.-G.
      • Park K.-S.
      • Kim M.-J.
      • et al.
      Relationship between serum adiponectin and leptin concentrations and body fat distribution.
      Many studies have produced inconsistent results about the levels of resistin in obesity and especially its etiopathogenetic relationship to insulin resistance.
      • Conneely K.N.
      • Silander K.
      • Scott L.J.
      • et al.
      Variation in the resistin gene is associated with obesity and insulin-related phenotypes in Finnish subjects.
      • Anderlová K.
      • Kremen J.
      • Dolezalová R.
      • et al.
      The influence of very-low-calorie-diet on serum leptin, soluble leptin receptor, adiponectin and resistin levels in obese women.
      • Dostalova I.
      • Smitka K.
      • Papezova H.
      • et al.
      Increased insulin sensitivity in patients with anorexia nervosa: the role of adipocytokines.
      To our knowledge, this was the first study to show a positive strong correlation between resistin and VF, SF and total abdominal fat.
      The positive correlations that were detected between the 3 fat volumes (SF, VF and total fat volume) and IL-6 were in agreement with some other related studies
      • Mohamed-Ali V.
      • Goodrick S.
      • Rawesh A.
      • et al.
      Subcutaneous adipose tissue releases interleukin-6, but not tumor necrosis factor-alpha, in vivo.
      • Fain J.N.
      • Madan A.K.
      • Hiler M.L.
      • et al.
      Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans.
      • Fried S.K.
      • Bunkin D.A.
      • Greenberg A.S.
      Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: depot difference and regulation by glucocorticoid.
      • Fontana L.
      • Eagon J.C.
      • Trujillo M.E.
      • et al.
      Visceral fat adipokine secretion is associated with systemic inflammation in obese humans.
      in which large portion of circulating IL-6 were found to be produced from both VF and SF depots in obese participants during basal conditions.
      The present study showed that cholesterol, TG and LDL were significantly but weakly associated with VF, SF and total abdominal fat. Moreover, a significant weak negative correlation was detected between HDL and the 3 fat volumes. Hoenig et al highlighted the presence of significant (P < 0.05) negative correlation between VF and HDL (r = −0.361) and positive with TG (r = 0.499). However, an inverse association between LDL and VF (r = −0.348) was detected, which indicates a potential discordance between cardiovascular risk and LDL.
      • Hoenig M.
      • Cowin G.
      • Buckley R.
      • et al.
      Low density lipoprotein cholesterol is inversely correlated with abdominal visceral fat area: a magnetic resonance imaging study.
      Leenen et al
      • Leenen R.
      • van der Kooy K.
      • Seidell J.C.
      • et al.
      Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
      conducted a study on 91 apparently healthy obese subjects to assess the associations between serum lipids and specific fat depots at abdominal and hip levels using MRI. In women, an accumulation of VF was associated with a deteriorated lipid profile, even after adjustment for age and body fat percentage; higher TG (P < 0.001), lower levels of HDL cholesterol (P < 0.01) and a diminished HDL cholesterol- to-LDL cholesterol ratio (P < 0.01) compared to those in men.
      • Leenen R.
      • van der Kooy K.
      • Seidell J.C.
      • et al.
      Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
      In men, there was a significant inverse relationship between the abundance of VF and the HDL-to-LDL ratio where significant positive correlations with total LDL cholesterol and TG disappeared after adjustment for age and fat percentage. Within each sex, SF was not significantly related to serum lipids. The above mentioned study concluded that there was a sex difference in the associations between VF accumulation and serum lipids.
      • Leenen R.
      • van der Kooy K.
      • Seidell J.C.
      • et al.
      Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
      In our study, BMI was significantly correlated with SF, VF, total abdominal fat and WC_tape and WC_MRI. Similar results were documented by Janssen et al (2002), in which BMI and WC were found to independently contribute to the prediction of nonabdominal, abdominal SF and VF in white men and women.
      Except for cholesterol, all measured biochemical variables in the current study were significantly associated with BMI. These include leptin, resistin, lL-6, adiponectin, TG, HDL and LDL. Arnardottir et al
      • Arnardottir E.S.
      • Maislin G.
      • Jackson N.
      • et al.
      The role of obesity, different fat compartments and sleep apnea severity in circulating leptin levels: the Icelandic Sleep Apnea Cohort study.
      demonstrated that leptin levels were highly associated with BMI, total abdominal fat and SF than VF volume per se. Many studies reported that adiponectin was negatively correlated with BMI and body fat mass.
      • Bidulescu A.
      • Liu J.
      • Hickson D.A.
      • et al.
      Gender differences in the association of visceral and subcutaneous adiposity with adiponectin in African Americans: the Jackson Heart Study.
      • Vikram N.K.
      • Misra A.
      • Pandey R.M.
      • et al.
      Adiponectin, insulin resistance, and C-reactive protein in postpubertal Asian Indian adolescents.
      • Matsubara M.
      • Maruoka S.
      • Katayose S.
      Inverse relationship between plasma adiponectin and leptin concentrations in normal-weight and obese women.
      A cross-sectional analysis of Azuma et al
      • Azuma K.
      • Katsukawa F.
      • Oguchi S.
      • et al.
      Correlation between serum resistin level and adiposity in obese individuals.
      showed that serum resistin was significantly higher in obese than in lean volunteers (24.58 ± 12.93 ng/mL; n = 64 versus 12.83 ± 8.30 ng/mL; n = 15; P < 0.01).
      Although the 2-point chemical shift MRI approach has been used and validated in many studies, this is the first study, to the best of our knowledge, to use MRI (multislice, semiautomated segmentation), anthropometric and biochemical parameters, collectively, to study the deposition of VF and SF tissues and investigate their effect on obesity in our country. A main limitation of the current study was the relatively small sample size owing to the limited time available to use the MRI scanner at the Royal Medical Services as it was dedicated for clinical imaging only. So, additional studies with a larger number of participants are required to validate these findings.

      Conclusions

      All anthropometric and biochemical parameters showed weak-to-strong associations with the MRI-measured fat volumes (SF, VF and their total volumes). Abdominal fat distribution was different between males and females and their correlations with some lipid profiles were found to be sex dependent. Based on the above observations and the agreement of these results with those reported in many studies, the 2-point chemical shift MRI approach, used in this study, proved to be a valuable and reliable imaging technique for studying fat deposition and can be used to assess abdominal fat distribution and their relation to anthropometric and biochemical measurements among obese and nonobese healthy participants.

      Acknowledgments

      The authors would like to thank the Royal Medical Services for their great help and support and for allowing us to use their 3T MRI to scan our volunteers.

      References

        • World Health Organization
        Obesity: preventing and managing the global epidemic, Report of a WHO Consultation (WHO Technical Report Series 894). 2000;
        • Shuster A.
        • Patlas M.
        • Pinthus J.H.
        • et al.
        The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis.
        Br J Radiol. 2012; 85: 1-10
        • Tchernof A.
        • Després J.-P.
        Pathophysiology of human visceral obesity: an update.
        Physiol Rev. 2013; 93: 359-404
        • Kershaw E.E.
        • Flier J.S.
        Adipose tissue as an endocrine organ.
        J Clin Endocrinol Metab. 2004; 89: 2548-2556
        • Deurenberg P.
        • Pieters J.J.
        • Hautvast J.G.
        The assessment of the body fat percentage by skinfold thickness measurements in childhood and young adolescence.
        Br J Nutr. 1990; 63: 293-303
        • Deurenberg P.
        • Deurenberg-Yap M.
        Validation of skinfold thickness and hand-held impedance measurements for estimation of body fat percentage among Singaporean Chinese, Malay and Indian subjects.
        Asia Pac J Clin Nutr. 2002; 11: 1-7
        • Widhalm K.
        • Schönegger K.
        • Huemer C.
        • et al.
        Does the BMI reflect body fat in obese children and adolescents? A study using the TOBEC method.
        Int J Obes Relat Metab Disord. 2001; 25: 279-285
        • Deurenberg P.
        • Deurenberg-Yap M.
        • Guricci S.
        Asians are different from Caucasians and from each other in their body mass index/body fat percent relationship.
        Obes Rev. 2002; 3: 141-146
        • De Lorenzo A.
        • Deurenberg P.
        • Pietrantuono M.
        • et al.
        How fat is obese?.
        Acta Diabetol. 2003; 40: s254-s257
        • Kvist H.
        • Chowdhury B.
        • Grangard U.
        • et al.
        Total and visceral adipose-tissue volumes derived from measurements with computed tomography in adult men and women: predictive equations.
        Am J Clin Nutr. 1988; 48: 1351-1361
        • Ball S.D.
        • Swan P.
        Accuracy of estimating intra-abdominal fat in obese women.
        J Exerc Physiol Online. 2003; 6: 1-7
        • Seidell J.C.
        • Oosterlee A.
        • Deurenberg P.
        • et al.
        Abdominal fat depots measured with computed tomography: effects of degree of obesity, sex, and age.
        Eur J Clin Nutr. 1988; 42: 805-815
        • So R.
        • Sasai H.
        • Matsuo T.
        • et al.
        Multiple-slice magnetic resonance imaging can detect visceral adipose tissue reduction more accurately than single-slice imaging.
        Eur J Clin Nutr. 2012; 66: 1351-1355
        • Mattsson S.
        • Thomas B.J.
        Development of methods for body composition studies.
        Phys Med Biol. 2006; 51: R203-R228
        • Seidell J.C.
        • Bakker C.J.
        • van der Kooy K.
        Imaging techniques for measuring adipose-tissue distribution—a comparison between computed tomography and 1.5-T magnetic resonance.
        Am J Clin Nutr. 1990; 51: 953-957
        • Lubura M.
        • Hesse D.
        • Neumann N.
        • et al.
        Non-invasive quantification of white and brown adipose tissues and liver fat content by computed tomography in mice.
        PLoS One. 2012; 7: e37026
        • Napolitano A.
        • Miller S.R.
        • Murgatroyd P.R.
        • et al.
        Validation of a quantitative magnetic resonance method for measuring human body composition.
        Obesity. 2008; 16: 191-198
        • Poonawalla A.H.
        • Sjoberg B.P.
        • Rehm J.L.
        • et al.
        Adipose tissue MRI for quantitative measurement of central obesity.
        J Magn Reson Imaging. 2013; 37: 707-716
        • Joshi A.A.
        • Hu H.H.
        • Leahy R.M.
        • et al.
        Automatic intra-subject registration-based segmentation of abdominal fat from water-fat MRI.
        J Magn Reson Imaging. 2013; 37: 423-430
        • Mantatzis M.
        • Milousis T.
        • Katergari S.
        • et al.
        Abdominal adipose tissue distribution on MRI and diabetes.
        Acad Radiol. 2014; 21: 667-674
        • Positano V.
        • Gastaldelli A.
        • maria Sironi A.
        • et al.
        An accurate and robust method for unsupervised assessment of abdominal fat by MRI.
        J Magn Reson Imaging. 2004; 20: 684-689
        • Wang D.
        • Shi L.
        • Chu W.C.W.
        • et al.
        Fully automatic and nonparametric quantification of adipose tissue in fat-water separation MR imaging.
        Med Biol Eng Comput. 2015; 53: 1247-1254
        • Addeman B.T.
        • Kutty S.
        • Perkins T.G.
        • et al.
        Validation of volumetric and single-slice MRI adipose analysis using a novel fully automated segmentation method.
        J Magn Reson Imaging. 2015; 41: 233-241
        • Warren M.
        • Schreiner P.J.
        • Terry J.G.
        The relation between visceral fat measurement and torso level--is one level better than another? The Atherosclerosis Risk in Communities Study, 1990–1992.
        Am J Epidemiol. 2006; 163: 352-358
        • Demerath E.W.
        • Shen W.
        • Lee M.
        • et al.
        Approximation of total visceral adipose tissue with a single magnetic resonance image.
        Am J Clin Nutr. 2007; 85: 362-368
        • Shen W.
        • Punyanitya M.
        • Wang Z.
        • et al.
        Visceral adipose tissue: relations between single-slice areas and total volume.
        Am J Clin Nutr. 2004; 80: 271-278
        • So R.
        • Matsuo T.
        • Sasai H.
        • et al.
        Best single-slice measurement site for estimating visceral adipose tissue volume after weight loss in obese, Japanese men.
        Nutr Metab (Lond). 2012; 9: 56
        • Shen W.
        • Chen J.
        • Gantz M.
        • et al.
        A single MRI slice does not accurately predict visceral and subcutaneous adipose tissue changes during weight loss.
        Obesity (Silver Spring). 2012; 20: 2458-2463
        • Lancaster J.L.
        • Ghiatas A.A.
        • Alyassin A.
        • et al.
        Measurement of abdominal fat with T1-weighted MR images.
        J Magn Reson Imaging. 1991; 1: 363-369
        • Ludwig U.A.
        • Klausmann F.
        • Baumann S.
        • et al.
        Whole-body MRI-based fat quantification: a comparison to air displacement plethysmography.
        J Magn Reson Imaging. 2014; 40: 1437-1444
        • Leenen R.
        • van der Kooy K.
        • Seidell J.C.
        • et al.
        Visceral fat accumulation measured by magnetic resonance imaging in relation to serum lipids in obese men and women.
        Atherosclerosis. 1992; 94: 171-181
        • Wander P.L.
        • Boyko E.J.
        • Leonetti D.L.
        • et al.
        Change in visceral adiposity independently predicts a greater risk of developing type 2 diabetes over 10 years in Japanese Americans.
        Diabetes Care. 2013; 36: 289-293
        • Vogt L.J.
        • Steveling A.
        • Meffert P.J.
        • et al.
        Magnetic resonance imaging of changes in abdominal compartments in obese diabetics during a low-calorie weight-loss program.
        PLoS One. 2016; 11: e0153595
        • Gray D.S.
        • Fujioka K.
        • Colletti P.M.
        • et al.
        Magnetic-resonance imaging used for determining fat distribution in obesity and diabetes.
        Am J Clin Nutr. 1991; 54: 623-627
        • Kanaley J.A.
        • Giannopoulou I.
        • Ploutz-Snyder L.L.
        Regional differences in abdominal fat loss.
        Int J Obes (Lond). 2007; 31: 147-152
        • Sallis J.F.
        • Haskell W.L.
        • Wood P.D.
        • et al.
        Physical activity assessment methodology in the Five-City Project.
        Am J Epidemiol. 1985; 121: 91-106
      1. D Nieman, R Lee. Nutritional Assessment. 6th ed. McGraw-Hill Education; 2013.

      2. Executive summary of the clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults..
        Arch Intern Med. 1998; 158: 1855-1867
        • Shen W.
        • Chen J.
        Application of imaging and other noninvasive techniques in determining adipose tissue mass.
        Methods Mol Biol. 2008; 456: 39-54
        • Lemieux S.
        • Bouchard C.
        • Tremblay A.
        • et al.
        Sex differences in the relation of visceral adipose tissue accumulation to total body fatness.
        Am J Clin Nutr. 1993; 58: 463-467
        • Blaak E.
        Gender differences in fat metabolism.
        Curr Opin Clin Nutr Metab Care. 2001; 4: 499-502
        • Bidulescu A.
        • Liu J.
        • Hickson D.A.
        • et al.
        Gender differences in the association of visceral and subcutaneous adiposity with adiponectin in African Americans: the Jackson Heart Study.
        BMC Cardiovasc Disord. 2013; 13: 9
        • Liu J.
        • Fox C.S.
        • Hickson D.A.
        • et al.
        Impact of abdominal visceral and subcutaneous adipose tissue on cardiometabolic risk factors: the Jackson Heart Study.
        J Clin Endocrinol Metab. 2010; 95: 5419-5426
        • Claussen C.
        • Eschweiler G.
        • Ludescher B.
        • et al.
        Gender specific correlations of adrenal gland size and body fat distribution: a whole body MRI study.
        Horm Metab Res. 2007; 39: 515-518
        • Karcher H.-S.
        • Holzwarth R.
        • Mueller H.-P.
        • et al.
        Body fat distribution as a risk factor for cerebrovascular disease: an MRI-based body fat quantification study.
        Cerebrovasc Dis. 2013; 35: 341-348
        • Browning L.M.
        • Mugridge O.
        • Dixon A.K.
        • et al.
        Measuring abdominal adipose tissue: comparison of simpler methods with MRI.
        Obes Facts. 2011; 4: 9-15
        • Camhi S.M.
        • Bray G.A.
        • Bouchard C.
        • et al.
        The relationship of waist circumference and BMI to visceral, subcutaneous, and total body fat: sex and race differences.
        Obesity (Silver Spring). 2011; 19: 402-408
        • Wajchenberg B.L.
        Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome.
        Endocr Rev. 2000; 21: 697-738
        • Perry A.C.
        • Applegate E.B.
        • Jackson M.L.
        • et al.
        Racial differences in visceral adipose tissue but not anthropometric markers of health-related variables.
        J Appl Physiol. 2000; 89: 636-643
        • Lubkowska A.
        • Radecka A.
        • Bryczkowska I.
        • et al.
        Serum adiponectin and leptin concentrations in relation to body fat distribution, hematological indices and lipid profile in humans.
        Int J Environ Res Public Health. 2015; 12: 11528-11548
        • Park K.-G.
        • Park K.-S.
        • Kim M.-J.
        • et al.
        Relationship between serum adiponectin and leptin concentrations and body fat distribution.
        Diabetes Res Clin Pract. 2004; 63: 135-142
        • Conneely K.N.
        • Silander K.
        • Scott L.J.
        • et al.
        Variation in the resistin gene is associated with obesity and insulin-related phenotypes in Finnish subjects.
        Diabetologia. 2004; 47: 1782-1788
        • Anderlová K.
        • Kremen J.
        • Dolezalová R.
        • et al.
        The influence of very-low-calorie-diet on serum leptin, soluble leptin receptor, adiponectin and resistin levels in obese women.
        Physiol Res. 2006; 55: 277-283
        • Dostalova I.
        • Smitka K.
        • Papezova H.
        • et al.
        Increased insulin sensitivity in patients with anorexia nervosa: the role of adipocytokines.
        Physiol Res. 2007; 56: 587-594
        • Mohamed-Ali V.
        • Goodrick S.
        • Rawesh A.
        • et al.
        Subcutaneous adipose tissue releases interleukin-6, but not tumor necrosis factor-alpha, in vivo.
        J Clin Endocrinol Metab. 1997; 82: 4196-4200
        • Fain J.N.
        • Madan A.K.
        • Hiler M.L.
        • et al.
        Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans.
        Endocrinology. 2004; 145: 2273-2282
        • Fried S.K.
        • Bunkin D.A.
        • Greenberg A.S.
        Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: depot difference and regulation by glucocorticoid.
        J Clin Endocrinol Metab. 1998; 83: 847-850
        • Fontana L.
        • Eagon J.C.
        • Trujillo M.E.
        • et al.
        Visceral fat adipokine secretion is associated with systemic inflammation in obese humans.
        Diabetes. 2007; 56: 1010-1013
        • Hoenig M.
        • Cowin G.
        • Buckley R.
        • et al.
        Low density lipoprotein cholesterol is inversely correlated with abdominal visceral fat area: a magnetic resonance imaging study.
        Lipids Health Dis. 2011; 10: 12
        • Arnardottir E.S.
        • Maislin G.
        • Jackson N.
        • et al.
        The role of obesity, different fat compartments and sleep apnea severity in circulating leptin levels: the Icelandic Sleep Apnea Cohort study.
        Int J Obes (Lond). 2013; 37: 835-842
        • Vikram N.K.
        • Misra A.
        • Pandey R.M.
        • et al.
        Adiponectin, insulin resistance, and C-reactive protein in postpubertal Asian Indian adolescents.
        Metabolism. 2004; 53: 1336-1341
        • Matsubara M.
        • Maruoka S.
        • Katayose S.
        Inverse relationship between plasma adiponectin and leptin concentrations in normal-weight and obese women.
        Eur J Endocrinol. 2002; 147: 173-180
        • Azuma K.
        • Katsukawa F.
        • Oguchi S.
        • et al.
        Correlation between serum resistin level and adiposity in obese individuals.
        Obes Res. 2003; 11: 997-1001