==== Front Korean J RadiolKorean J RadiolKJRKorean Journal of Radiology1229-69292005-8330The Korean Society of Radiology 10.3348/kjr.2018.19.5.923Gastrointestinal ImagingOriginal ArticleCorrelation between Non-Alcoholic Fatty Liver Disease and Visceral Adipose Tissue in Non-Obese Chinese Adults: A CT Evaluation Yu Ai-Hong MD1*Duan-Mu Yang-Yang MS2*Zhang Yong MD1Wang Ling MS1Guo Zhe MD1Yu Yong-Qiang MD2Wang Yu-Sheng BS1Cheng Xiao-Guang MD11 Department of Radiology, Beijing Jishuitan Hospital, the 4th Medical College of Peking University, Beijing 100035, China.2 Department of Radiology, First Affiliated Hospital of Anhui Medical University, Hefei 230022, China.Corresponding author: Xiao-Guang Cheng, MD, Department of Radiology, Beijing Jishuitan Hospital, the 4th Medical College of Peking University, Beijing 100035, China. Tel: 86-10-58516688-6123, Fax: 86-21-64085875, xiao65@263.net*These authors contributed equally to this work. Sep-Oct 2018 06 8 2018 19 5 923 929 27 7 2017 17 2 2018 Copyright © 2018 The Korean Society of Radiology2018The Korean Society of RadiologyThis is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.Objective To investigate the correlation between non-alcoholic fatty liver disease and visceral adipose tissue in non-obese Chinese adults using computed tomography (CT). Materials and Methods The study included 454 subjects undergoing abdominal CT scan. Degree of CT attenuation in liver and spleen, and the degree of fat infiltration in liver were evaluated according to three indices: the attenuation value of liver parenchyma (CTLP), the attenuation ratio of liver and spleen (LSratio) and the attenuation difference between liver and spleen (LSdif). Visceral fat area (VFA) and total fat area (TFA) at L2/3 and L4/5 levels were measured, and the abdominal subcutaneous fat area (SFA) was calculated. Bivariate correlation analysis was carried out to determine the correlation among these factors. Results In men, VFA, SFA and TFA at L2/3 and L4/5 levels showed significant differences in terms of the three indices to distinguish fatty liver from non-fatty liver (all, p < 0.001). In men, all the three indices showed negative correlation with TFA, SFA and VFA (all, p < 0.001). The negative correlation between the three indices and VFA at the L2/3 level was higher than at L4/5 level (r = −0.476 vs. r = −0.340 for CTLP, r = −0.502 vs. r = −0.413 for LSratio, r = −0.543 vs. r = −0.422 for LSdif, p < 0.001, respectively). The negative correlation between LSratio, LSdif and VFA at L2/3 and L4/5 levels was higher than SFA at the corresponding level. In women, all the three indices showed negative correlation with VFA and TFA at L2/3 and L4/5 levels, and the negative correlation between CTLP and VFA was higher at L2/3 level than at L4/5 level (r = −0.294 vs. r = −0.254, p < 0.001). Conclusion In non-obese Chinese adults, the degree of hepatic fatty infiltration showed a strong correlation with abdominal fat on CT. VFA at L2/3 level was more closely related to fatty liver compared with VFA at L4/5 level. Fatty liverNASHVisceral fatComputed tomographyCT QuantificationQuantitative diagnosisBeijing Bureau of 215 program2013-3-0332009-02-03 ==== Body INTRODUCTION Fatty liver is defined as the accumulation of liver triglycerides exceeding more than 5% of the liver weight (1). Fatty liver without a history of alcohol intake is known as non-alcoholic fatty liver disease (NAFLD). NAFLD is a common chronic liver disease in the world (2). NAFLD is clinically defined as a transition from simple liver steatosis to non-alcoholic hepatitis, which may in turn develop into non-alcoholic hepatic cirrhosis and even hepatocellular carcinoma (3). Therefore, diagnostic screening for simple liver steatosis is important clinically. Studies indicated that NAFLD was more prevalent in obese population, and was related to type 2 diabetes, dyslipidemia and hypertension (4). A few studies considered NAFLD as a manifestation of metabolic syndrome in the liver (5). Meanwhile, a few studies reported a close relationship between abdominal obesity and risk factors for these metabolic diseases (6). In abdominal obesity, abdominal visceral adipose tissue was more closely related to metabolic risk factors compared with subcutaneous adipose tissue (6). However, recent studies explored the effects of these risk factors on liver in patients with morbid obesity or in those who had been diagnosed with metabolic syndrome (7). A few epidemiological studies have shown that NAFLD occurred even in non-obese patients with non-metabolic syndrome (89). In China, Xu et al. (10) investigated large samples of non-obese populations and showed that NAFLD was also prevalent in China's non-obese population, and the prevalence ratio was increased in the 5-year follow-up. However, there was no study related to body fat composition and NAFLD measured by CT in non-obese population. The purpose of this study was to investigate the correlation between NAFLD and abdominal fat in non-obese Chinese adults by computed tomography (CT). MATERIALS AND METHODS Subjects This study was approved by our hospital ethics committee, and all participants had signed informed consent before participating in the study. The subjects included in this study were participants of an ongoing study since June 2014 investigating degeneration of the spine in the Department of Orthopedics, Beijing Jishuitan Hospital. A routine CT scan of lumbar spine was performed. The present study analyzed existing data of the lumbar spine. The images of liver and spleen were reconstructed with a large field of view (FOV), and no radiation was involved. Participants' age, place of origin and past medical history were determined by a questionnaire survey. Indices such as height, weight, waist circumference and hip circumference with barefoot and underwear were measured in all the participants. Patients diagnosed with diabetes or Cushing's syndrome, patients who were treated with large doses of steroids or addicted to heavy drinking, diagnosed with hepatitis, history of cirrhosis or liver surgery, and obesity with a body mass index (BMI) ≥ 30 kg/m2 were further excluded. Finally, 454 (164 men and 290 women) participants were included, with an age range of 19–63 years and an average age of 38 years. CT Scan Images were obtained for all the subjects using an 80-slice CT scanner (Toshiba Medical Systems Corp., Tokyo, Japan) with a calibration phantom (Mindways Software Inc., Austin, TX, USA) beneath the waist. Routine calibration was performed before quantitative computed tomography (QCT) measurement. The scan parameters were as follows: 120 kV, 250 mA, 40 cm FOV, 120 cm bed height, 1 mm slice thickness, and 512 × 512 matrix. All subjects were scanned in the supine position with both the arms stretched above the head. The images of liver and spleen were reconstructed with large FOV. CT Attenuation Measurement Liver and spleen measurements were performed using a CT post-processing workstation, and only one image (level of portal vein into the liver) of each patient was selected to complete the attenuation measurement of liver and spleen. Three regions of interest in the liver were selected to avoid blood vessels, bile ducts and calcification; and 2 regions of interest in the spleen were selected at the same level. Finally, the respective means of the 3 attenuation values of the liver and 2 attenuation values of the spleen were calculated (Fig. 1). The following 3 indices were selected to determine hepatic fat infiltration: 1) The attenuation value of liver parenchyma (CTLP); 2) The attenuation ratio of liver and spleen (LSratio); 3) The attenuation difference between liver and spleen (LSdif). Fatty liver was diagnosed by traditional standards: Attenuation value of liver parenchyma (CTLP) less than 48 HU, or attenuation ratio of liver and spleen (LSratio) less than 1.0, or attenuation difference of liver and spleen (LSdif) less than 5 HU, indicate a diagnosis of fatty liver (111213). CT Measurement of Abdominal Fat The acquired scan data were transferred to QCT bone density measurement analysis (QCT PRO) workstation (Mindways Software Inc.). Fat areas were measured using the body fat analysis software of the QCT bone density measurement. Using the intervertebral space as an anatomic mark, the following anatomical areas were used for CT measurement: 1) the horizontal level of L2/3; 2) the horizontal level of L4/5 (Fig. 2). Visceral fat area (VFA) and total fat area (TFA) at each level were obtained, respectively, and the subcutaneous fat area (SFA) was calculated using the following equation: SFA = TFA − VFA. All measurements were performed by an experienced radiologist. Previous studies have shown that this method showed good repeatability (14). Statistical Analysis SPSS 20.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Normal distribution data were expressed by mean ± standard deviation. Independent sample t test was used to compare the differences between VFA, TFA and SFA in the fatty and non-fatty liver groups and the differences were analyzed. Bivariate correlation analysis (Pearson's correlations analysis) was used to analyze the correlation between the three indices (CTLP, LSratio, LSdif) and VFA, TFA, SFA at L2/3 and L4/5 levels. P < 0.05 was considered statistically significant. RESULTS The mean CTLP, LSratio and LSdif were 56.0 ± 10.2 HU (ranging from 12.2 HU to 73.8 HU), 1.17 ± 0.20 (ranging from 0.31 to 1.88), 7.8 ± 8.6 HU (ranging from −28.8 HU to 29.9 HU), respectively. The mean VFA, SFA and TFA values at L2/3 levels were 105.8 ± 56.7 cm2 (ranging from 10.6 cm2 to 322.8 cm2), 107.5 ± 47.0 cm2 (ranging from 4.4 cm2 to 323.7 cm2), and 213.4 ± 86.2 cm2 (ranging from 31.3 cm2 to 492.8 cm2), respectively. The mean VFA, SFA and TFA values at L4/5 levels were 91.8 ± 36.2 cm2 (ranging from 20.7 cm2 to 260.3 cm2), 155.0 ± 60.0 cm2 (ranging from 19.2 cm2 to 362.1 cm2), and 246.8 ± 80.9 cm2 (ranging from 54.3 cm2 to 516.6 cm2), respectively. The mean BMI, waist circumference, waist-hip ratio, weight and age were 23.5 ± 2.9 kg/m2 (ranging from 16.1 kg/m2 to 29.9 kg/m2), 80.8 ± 10.0 cm (ranging from 60 cm to 181 cm), 0.84 ± 0.08 (ranging from 0.70 to 1.91), 64.2 ± 11.1 kg (ranging from 42 kg to 97 kg), and 38.3 ± 8.6 years (ranging from 19 years to 63 years), respectively (Table 1). The case numbers of fatty liver disease diagnosed in terms of the three indices in male and female groups are shown in Tables 2 and 3. The differences in VFA, SFA and TFA between fatty and non-fatty liver groups were evaluated according to the three indices, and the statistical results are presented in Tables 2 and 3. In men, significant differences were observed between VFA, SFA and TFA at L2/3 and L4/5 levels when using the three indices to distinguish between the fatty and non-fatty liver groups (all, p < 0.001) (Table 2). In women, significant differences between VFA, SFA and TFA at L2/3 and L4/5 levels were observed using CTLP to distinguish between the fatty and non-fatty liver groups (p = 0.011 in SFA at L2/3 and p < 0.001 in others). However, there was no significant difference in SFA between the two groups at different levels when using LSratio and LSdif to evaluate fatty liver (Table 3). In men, CTLP, LSratio and LSdif demonstrated negative correlation with TFA, SFA and VFA at L2/3 and L4/5 levels, respectively (all, p < 0.001) (Table 4). The negative correlation of LSratio and LSdif with VFA at L2/3 and L4/5 levels was higher than SFA at the corresponding level. The three indices were negatively correlated with VFA, and the negative correlations at L2/3 level were all higher than at L4/5 level (r = −0.476 vs. r = −0.340 for CTLP, r = −0.502 vs. r = −0.413 for LSratio, r = −0.543 vs. r = −0.422 for LSdif, p < 0.001, respectively) (Table 4). In women, CTLP, LSratio and LSdif showed negative correlation with VFA and TFA at L2/3 and L4/5 levels, respectively (p = 0.004 in TFA at L4/5 and p ≤ 0.001 in others). Neither LSratio nor LSdif demonstrated any correlation with SFA at L2/3 or L4/5 levels. The negative correlation between CTLP and VFA at L2/3 level was higher than at L4/5 level (r = −0.294 vs. r = −0.254, p < 0.001). Overall, the negative correlation between CTLP and VFA at L2/3 level was significant (r = −0.294, p < 0.001) (Table 5). DISCUSSION Non-alcoholic fatty liver disease is one of the most common chronic liver diseases worldwide, and obesity is an important cause underlying NAFLD. BMI has a direct correlation with NAFLD (15). Although BMI can be used as an independent predictor of NAFLD, a few studies showed that the body fat composition reflected health better than BMI and body weight (1516). Omagari et al. (17) investigated 3432 Japanese adults and found that 27.2% males and 59.2% females who had no history of alcohol intake were diagnosed with fatty liver, even though they were not overweight. This finding suggested unreliability of BMI as the sole indicator for determination of fatty liver. Further, the study indicated that central body fat distribution correlated with the development of fatty liver. Similar studies also showed that the central body fat distribution in obese individuals was more closely related to diabetes, dyslipidemia, hypertension and other cardiovascular diseases, compared with gluteo-femoral body fat distribution in obese patients (18). In obesity characterized by central body fat distribution, excessive accumulation of visceral fat was more closely related to metabolic syndrome, compared with subcutaneous fat (18). Although studies pointed to waist circumference as a predictor of central body fat distribution in obese patients (19), the index failed to distinguish between visceral and subcutaneous fat and resulted in errors in the evaluation of visceral fat. Currently, histological examination is considered as the gold standard for fatty liver diagnosis, even though in vivo biopsy is an invasive tool with obvious deficiencies (20). Alternative imaging modalities have been used for liver examination (2122). CT scan has the ability to diagnose and quantify the degree of liver fat infiltration (23). CT scan is an effective and practical modality that clearly distinguishes between visceral fat and subcutaneous fat, and also facilitates direct quantification with simple, highly repeatable and accurate results. Previous studies demonstrated that CT scan facilitated the determination of the degree of hepatic fat infiltration and obviated the need for unnecessary biopsy; it also has been used to quantify visceral fat adequately (2324). Hence, CT represents an important technique for clinical investigation of the correlation between fatty liver and visceral fat. In this study, three CT indices were used to evaluate fatty liver. Significant differences in VFA among males and females in the fatty and non-fatty liver groups were found. There was no significant difference in female SFA between the two groups. In both men and women, the correlation between the three indices of the degree of hepatic fatty infiltration and VFA was higher compared with SFA. Studies involving morbidly obese patients or those diagnosed with metabolic syndrome showed that abnormal visceral fat accumulation was a risk factor for fatty liver (7). These results were consistent with our study investigating non-obese Chinese adults, probably due to the endocrine function of adipose tissue, and the role of metabolites of abdominal subcutaneous tissue in body circulation, while visceral fat secretions were discharged into the portal system. Free fatty acids were produced by excessive metabolization of visceral fat, shunted into the liver via portal circulation, synthesized into fat, and deposited in the liver, resulting in the formation of fatty liver (25). Instead of total visceral fat volume, VFA in a single image was measured by CT in this study. This method has been widely accepted, but the level still remains disputed (2426272829). Researchers recommend the use of CT for VFA at a single image to replace the total visceral fat volume (24), and the level of the umbilicus (L4/5) was mostly selected (2627). Other studies indicated that VFA at L2/3 level better predicted the visceral total fat (2829). We selected L2/3 and L4/5 (umbilicus) levels and measured in this study. Results showed negative correlations between the three indices and VFA at L2/3 level, which were higher than at L4/5 level in males. By contrast, a negative correlation between CTLP and VFA at L2/3 level was higher than at L4/5 level in females, which was consistent with the study of Han et al. (28) and Demerath et al. (29). A few studies suggested that (30) visceral fat was more likely to accumulate in the upper abdominal region at the L1/2 to L3/4 levels, due to the presence of highly active visceral adipocytes in the omentum and mesentery than in the retroperitoneal adipocytes. This accumulation of visceral fat might increase the plasma adiponectin level and release additional inflammatory factors, triggering a series of metabolic diseases, such as insulin resistance, inflammation and NAFLD (30). Compared with the lower abdomen, the upper abdomen was more closely related to total visceral fat (30), and this interpretation was consistent with our study results. Waist circumference was measured at the level of umbilicus (L4/5) clinically. The study results showed that the measurement of waist circumference at L2/3 level facilitates clinical evaluation of visceral fat. Our study has several important strengths. The correlation between NAFLD and visceral adipose tissue in non-obese Chinese adults was analyzed using a large sample of CT data, and abdominal fat at two levels of L2/3 and L4/5 was measured concurrently. We found that abdominal fat strongly correlated with hepatic steatosis in non-obese adults and L2/3 was better than L4/5 for fat measurement. However, our study also has several limitations: First, CT was used to diagnose and evaluate fatty liver in this study, without including pathological diagnosis, making it difficult to distinguish between simple hepatic steatosis and NAFLD hepatitis. The three indices of CT were used to judge liver fat infiltration, and the best index among the three warrants additional study in the future. Further, this study failed to consider hemochromatosis or other diseases in patients, which were related to hepatic iron deposition. These diseases increase the CT attenuation value of the liver, resulting in missed diagnosis of fatty liver. In summary, in non-obese Chinese adults, the degree of liver fatty infiltration showed a strong correlation with abdominal fat, and the association of VFA with fatty liver was stronger than SFA. VFA at L2/3 level was more closely related to fatty liver compared with VFA at L4/5 level. This study was supported by Grants from the Beijing Bureau of 215 program (no. 2013-3-033; 2009-02-03). Fig. 1 Attenuation measurements of liver and spleen. Three regions of interest in liver and 2 regions of interest in spleen at level of hepatic portal vein were selected (avoiding blood vessels, bile ducts and calcification). Fig. 2 CT measurement of abdominal fat. Sagittal and coronal CT images showed that L2/3 and L4/5 were horizontal levels of L2 and L4 vertebral body lower edges, respectively. Cross-sectional image showed that blue region represented adipose tissues, yellow region denoted muscles and soft tissues, and green region indicated vertebral bodies and accessories; outside green dots suggested subcutaneous fat, and inside green dots pointed to visceral fat. SFA = subcutaneous fat area, VFA = visceral fat area Table 1 Demographic Data of Participants Mean ± SD Range CTLP (HU) 56.0 ± 10.2 12.2–73.8 LSratio 1.17 ± 0.20 0.31–1.88 LSdif (HU) 7.8 ± 8.6 −28.8–29.9 L2/3 VFA (cm2) 105.8 ± 56.7 10.6–322.8 L2/3 SFA (cm2) 107.5 ± 47.0 4.4–323.7 L2/3 TFA (cm2) 213.4 ± 86.2 31.3–492.8 L4/5 VFA (cm2) 91.8 ± 36.2 20.7–260.3 L4/5 SFA (cm2) 155.0 ± 60.0 19.2–362.1 L4/5 TFA (cm2) 246.8 ± 80.9 54.3–516.6 WC (cm) 80.8 ± 10.0 60–181 WHR 0.84 ± 0.08 0.70–1.91 BMI (kg/m2) 23.5 ± 2.9 16.1–29.9 Weight (kg) 64.2 ± 11.1 42–97 Age (years) 38.3 ± 8.6 19–63 BMI = body mass index, CTLP = attenuation value of liver parenchyma, LSdif = attenuation difference between liver and spleen, LSratio = attenuation ratio of liver and spleen, SD = standard deviation, SFA = subcutaneous fat area, TFA = total fat area, VFA = visceral fat area, WC = waist circumference, WHR = waist-hip ratio Table 2 VFA, SFA, TFA, BMI, WHR, WC, and Weight between Fatty Liver Group and Non-Fatty Liver Group in Men according to Three Indexes CTLP Lsratio Lsdif Fatty Liver Non-Fatty Liver P* Fatty Liver Non-Fatty Liver P* Fatty Liver Non-Fatty Liver P* Number of patients 46 118 39 125 57 107 L2/3 VF (cm2) 190.7 ± 54.6 126.7 ± 56.2 < 0.001 200.2 ± 46.7 127.3 ± 56.7 < 0.001 186.0 ± 56.0 122.3 ± 54.4 < 0.001 L2/3 SF (cm2) 121.3 ± 44.6 83.7 ± 40.5 < 0.001 122.3 ± 41.6 85.5 ± 42.4 < 0.001 115.9 ± 39.9 82.7 ± 43.3 < 0.001 L2/3 TF (cm2) 312.0 ± 78.8 210.4 ± 88.1 < 0.001 322.5 ± 69.6 212.8 ± 89.2 < 0.001 301.9 ± 81.9 205.3 ± 87.2 < 0.001 L4/5 VF (cm2) 131.2 ± 37.9 102.0 ± 39.2 < 0.001 138.2 ± 35.1 101.4 ± 38.7 < 0.001 131.8 ± 40.6 98.7 ± 36.3 < 0.001 L4/5 SF (cm2) 176.5 ± 52.6 128.5 ± 54.9 < 0.001 183.2 ± 54.1 129.2 ± 53.6 < 0.001 173.9 ± 52.4 125.0 ± 54.1 < 0.001 L4/5 TF (cm2) 307.8 ± 70.4 230.5 ± 83.6 < 0.001 321.4 ± 64.0 230.6 ± 82.2 < 0.001 305.7 ± 73.1 223.7 ± 80.6 < 0.001 BMI (kg/m2) 26.5 ± 2.1 24.1 ± 2.7 < 0.001 26.7 ± 1.9 24.1 ± 2.7 < 0.001 26.2 ± 2.1 24.0 ± 2.8 < 0.001 WHR 0.91 ± 0.05 0.87 ± 0.05 < 0.001 0.92 ± 0.05 0.87 ± 0.05 < 0.001 0.90 ± 0.05 0.87 ± 0.05 < 0.001 WC (cm) 92.6 ± 6.3 85.2 ± 7.2 < 0.001 93.4 ± 5.5 85.4 ± 7.3 < 0.001 91.6 ± 6.4 85.0 ± 7.4 < 0.001 Weight (kg) 80.0 ± 8.8 71.8 ± 9.3 < 0.001 80.8 ± 7.1 72.0 ± 9.6 < 0.001 79.7 ± 7.5 71.1 ± 9.6 < 0.001 *Student's t test. Table 3 VFA, SFA, TFA, BMI, WHR, WC, and Weight between Fatty Liver Group and Non-Fatty Liver Group in Women according to Three Indexes CTLP LSratio LSdif Fatty Liver Non-Fatty Liver P* Fatty Liver Non-Fatty Liver P* Fatty Liver Non-Fatty Liver P* Number of patients 37 253 19 271 49 241 L2/3 VF (cm2) 118.5 ± 54.0 78.8 ± 32.9 < 0.001 130.5 ± 47.9 80.6 ± 35.6 < 0.001 103.7 ± 56.4 79.9 ± 32.4 < 0.001 L2/3 SF (cm2) 133.1 ± 50.8 112.4 ± 45.4 0.011 146.5 ± 64.7 112.8 ± 44.3 0.068 123.1 ± 54.9 113.4 ± 44.6 0.253 L2/3 TF (cm2) 251.7 ± 91.1 191.2 ± 70.5 < 0.001 276.9 ± 93.1 193.4 ± 71.8 < 0.001 226.7 ± 99.8 193.3 ± 68.8 0.030 L4/5 VF (cm2) 104.4 ± 38.0 78.1 ± 25.3 < 0.001 117.7 ± 33.9 78.9 ± 26.4 < 0.001 94.9 ± 39.7 78.7 ± 24.9 < 0.001 L4/5 SF (cm2) 192.5 ± 79.4 157.9 ± 54.8 < 0.001 206.1 ± 101.4 159.3 ± 54.3 0.061 172.8 ± 79.7 160.2 ± 54.4 0.296 L4/5 TF (cm2) 296.9 ± 99.5 236.0 ± 70.3 < 0.001 323.8 ± 120.5 238.2 ± 70.2 0.006 267.7 ± 96.1 238.9 ± 69.1 0.074 BMI (kg/m2) 25.0 ± 3.1 22.4 ± 2.5 0.000 26.4 ± 3.5 22.5 ± 2.5 0.000 23.9 ± 3.7 22.5 ± 2.4 0.01 WHR 0.82 ± 0.06 0.82 ± 0.09 0.542 0.85 ± 0.05 0.81 ± 0.08 0.067 0.82 ± 0.06 0.82 ± 0.09 0.706 WC (cm) 81.0 ± 8.0 76.6 ± 9.4 0.007 84.1 ± 8.5 76.7 ± 9.2 0.001 78.8 ± 9.2 76.8 ± 9.3 0.178 Weight (kg) 63.8 ± 8.5 57.9 ± 6.8 0.000 66.6 ± 8.8 58.1 ± 6.9 0.000 61.4 ± 9.1 58.1 ± 6.8 0.018 *Student's t test. Table 4 Correlations between Three Indexes and VFA, SFA, and TFA at L2/3 and L4/5 Levels in Men* L2/3 VFA L2/3 SFA L2/3 TFA L4/5 VFA L4/5 SFA L4/5 TFA CTLP −0.476 −0.367 −0.478 −0.340 −0.385 −0.417 P* < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 LSratio −0.502 −0.314 −0.470 −0.413 −0.346 −0.425 P* < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 LSdif −0.534 0.362 −0.513 −0.422 −0.386 −0.456 P* < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 *Pearson's correlations analysis. Table 5 Correlations between Three Indexes and VFA, SFA, and TFA at L2/3 and L4/5 Levels in Women* L2/3 VFA L2/3 SFA L2/3 TFA L4/5 VFA L4/5 SFA L4/5 TFA CTLP −0.294 −0.196 −0.268 −0.254 −0.205 −0.252 P* < 0.001 0.001 < 0.001 < 0.001 < 0.001 < 0.001 LSratio −0.211 −0.033 −0.127 −0.225 −0.071 −0.137 P* < 0.001 0.573 0.031 < 0.001 0.231 0.019 LSdif −0.247 −0.066 −0.165 −0.258 −0.096 −0.169 P* < 0.001 0.265 0.005 < 0.001 0.103 0.004 *Pearson's correlations analysis. ==== Refs 1 Adams LA Angulo P Lindor KD Nonalcoholic fatty liver disease CMAJ 2005 172 899 905 15795412 2 Vernon G Baranova A Younossi ZM Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults Aliment Pharmacol Ther 2011 34 274 285 21623852 3 Musso G Gambino R Cassader M Pagano G Meta-analysis: natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity Ann Med 2011 43 617 649 21039302 4 Hamaguchi M Kojima T Takeda N Nakagawa T Taniguchi H Fujii K The metabolic syndrome as a predictor of nonalcoholic fatty liver disease Ann Intern Med 2005 143 722 728 16287793 5 Kwon YM Oh SW Hwang SS Lee C Kwon H Chung GE Association of nonalcoholic fatty liver disease with components of metabolic syndrome according to body mass index in Korean adults Am J Gastroenterol 2012 107 1852 1858 23032980 6 Fox CS Massaro JM Hoffmann U Pou KM Maurovich-Horvat P Liu CY Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study Circulation 2007 116 39 48 17576866 7 Sogabe M Okahisa T Tsujigami K Fukuno H Hibino S Yamanoi A Visceral fat predominance is associated with non-alcoholic fatty liver disease in Japanese women with metabolic syndrome Hepatol Res 2014 44 515 522 23617326 8 Chen CH Huang MH Yang JC Nien CK Yang CC Yeh YH Prevalence and risk factors of nonalcoholic fatty liver disease in an adult population of taiwan: metabolic significance of nonalcoholic fatty liver disease in nonobese adults J Clin Gastroenterol 2006 40 745 752 16940890 9 Das K Das K Mukherjee PS Ghosh A Ghosh S Mridha AR Nonobese population in a developing country has a high prevalence of nonalcoholic fatty liver and significant liver disease Hepatology 2010 51 1593 1602 20222092 10 Xu C Yu C Ma H Xu L Miao M Li Y Prevalence and risk factors for the development of nonalcoholic fatty liver disease in a nonobese Chinese population: the Zhejiang Zhenhai Study Am J Gastroenterol 2013 108 1299 1304 23567356 11 Limanond P Raman SS Lassman C Sayre J Ghobrial RM Busuttil RW Macrovesicular hepatic steatosis in living related liver donors: correlation between CT and histologic findings Radiology 2004 230 276 280 14695401 12 Fatty Liver and Alcoholic Liver Disease Study Group of Chinese Liver Disease Association [Diagnostic criteria of nonalcoholic fatty liver disease] Zhonghua Gan Zang Bing Za Zhi 2003 11 71 12648394 13 Yajima Y Narui T Ishii M Abe R Ohtsuki M Goto Y Computed tomography in the diagnosis of fatty liver: total lipid content and computed tomography number Tohoku J Exp Med 1982 136 337 342 7071850 14 Wang L Wang W Deng W Jiang F Xu J Zhao H [The reproducibility of abdominal fat area and distribution measured by QCT] Chin J Osteoporos 2012 11 999 1003 15 Ha Y Seo N Shim JH Kim SY Park JA Han S Intimate association of visceral obesity with non-alcoholic fatty liver disease in healthy Asians: a case-control study J Gastroenterol Hepatol 2015 30 1666 1672 25974139 16 Lee JY Kim KM Lee SG Yu E Lim YS Lee HC Prevalence and risk factors of non-alcoholic fatty liver disease in potential living liver donors in Korea: a review of 589 consecutive liver biopsies in a single center J Hepatol 2007 47 239 244 17400323 17 Omagari K Kadokawa Y Masuda J Egawa I Sawa T Hazama H Fatty liver in non-alcoholic non-overweight Japanese adults: incidence and clinical characteristics J Gastroenterol Hepatol 2002 17 1098 1105 12201871 18 Wajchenberg BL Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome Endocr Rev 2000 21 697 738 11133069 19 Lear SA Humphries KH Kohli S Birmingham CL The use of BMI and waist circumference as surrogates of body fat differs by ethnicity Obesity (Silver Spring) 2007 15 2817 2824 18070773 20 Oliva MR Mortele KJ Segatto E Glickman JN Erturk SM Ros PR Computed tomography features of nonalcoholic steatohepatitis with histopathologic correlation J Comput Assist Tomogr 2006 30 37 43 16365570 21 Xie S Li Q Cheng Y Zhang Y Zhuo Z Zhao G Impact of liver fibrosis and fatty liver on T1rho measurements: a prospective study Korean J Radiol 2017 18 898 905 29089822 22 Yoo H Lee JM Yoon JH Kang HJ Lee SM Yang HK T2* mapping from multi-echo dixon sequence on gadoxetic acid-enhanced magnetic resonance imaging for the hepatic fat quantification: can it be used for hepatic function assessment? Korean J Radiol 2017 18 682 690 28670163 23 Joy D Thava VR Scott BB Diagnosis of fatty liver disease: is biopsy necessary? Eur J Gastroenterol Hepatol 2003 15 539 543 12702913 24 Kuk JL Church TS Blair SN Ross R Does measurement site for visceral and abdominal subcutaneous adipose tissue alter associations with the metabolic syndrome? Diabetes Care 2006 29 679 684 16505526 25 Jakobsen MU Berentzen T Sørensen TI Overvad K Abdominal obesity and fatty liver Epidemiol Rev 2007 29 77 87 17478441 26 Kobayashi J Tadokoro N Watanabe M Shinomiya M A novel method of measuring intra-abdominal fat volume using helical computed tomography Int J Obes Relat Metab Disord 2002 26 398 402 11896496 27 Demerath EW Sun SS Rogers N Lee M Reed D Choh AC Anatomical patterning of visceral adipose tissue: race, sex, and age variation Obesity (Silver Spring) 2007 15 2984 2993 18198307 28 Han TS Kelly IE Walsh K Greene RM Lean ME Relationship between volumes and areas from single transverse scans of intra-abdominal fat measured by magnetic resonance imaging Int J Obes Relat Metab Disord 1997 21 1161 1166 9426384 29 Demerath EW Shen W Lee M Choh AC Czerwinski SA Siervogel RM Approximation of total visceral adipose tissue with a single magnetic resonance image Am J Clin Nutr 2007 85 362 368 17284730 30 Motoshima H Wu X Sinha MK Hardy VE Rosato EL Barbot DJ Differential regulation of adiponectin secretion from cultured human omental and subcutaneous adipocytes: effects of insulin and rosiglitazone J Clin Endocrinol Metab 2002 87 5662 5667 12466369