Abstract
Aims
To evaluate the accuracy of abdominal fat distribution of subcutaneous (SAT) and visceral adipose tissue (VAT) assessed by ultrasound (US) compared to magnetic resonance imaging (MRI), as the gold standard technique. Additionally, to assess the association between abdominal fat distribution measured by US and metabolic conditions in adults with abdominal obesity.
Materials and methods
A cross‐sectional study (ClinicalTrials.gov: NCT05882149) included 113 individuals (65.5% women) with waist circumference ≥102 cm for men and ≥88 cm for women. VAT and SAT were measured by both US and MRI. Diagnostic performance was evaluated using ROC curve analysis.
Results
Participants were (mean ± SD) 52.8 ± 10.9 years old. VAT thickness measured by US showed a strong correlation with MRI (r = 0.82). For prediabetes, US‐measured VAT thickness showed fair diagnostic accuracy in women (AUC = 0.71) with a proposed cut‐off of 5.87 cm. For metabolic syndrome, US‐measured VAT thickness showed poor diagnostic accuracy in women (AUC = 0.69), with a proposed cut‐off of 4.83 cm. Also, VAT/SAT ratio (thickness) measured by US showed fair diagnostic accuracy for prediabetes in women (AUC = 0.72), with a proposed cut‐off of 3.04, and poor diagnostic accuracy for metabolic syndrome in women (AUC = 0.69), with a proposed cut‐off of 2.52.
Conclusions
Ultrasound represents a useful screening tool for VAT thickness evaluation with an easy translation to clinical practice. Particularly, VAT thickness and VAT/SAT ratio (thickness) are positively associated with prediabetes and metabolic syndrome, especially in women.
Keywords: abdominal obesity, metabolic syndrome, prediabetes, ultrasound, visceral adipose tissue
1. INTRODUCTION
Abdominal obesity is a major cardiometabolic disease (CMD) risk factor 1 and it is linked to metabolic syndrome. 2 , 3 The routinely applicable anthropometrical indicators of central obesity are waist circumference (WC) and waist‐to‐hip ratio (WHR). 4 However, these anthropometric indicators do not differentiate abdominal adipose tissue distribution. 5
Abdominal adipose tissue is differentiated into subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT). 6 SAT represents about 80% of total body adipose tissue. 6 Whereas VAT, or intra‐abdominal fat, including preperitoneal, omental, mesenteric, and perirenal fat, accounts for about 10–20% in men and 5–10% in women. 7 , 8
Individuals with central obesity but a healthy body mass index (BMI), often present disproportionately high VAT, associated to higher CMD risk and mortality than BMI‐defined obesity. 6
Excessive VAT accumulation is a predictor of increased cardiovascular disease (CVD) risk and metabolic conditions such as insulin resistance and hyperinsulinemia, 9 high blood pressure, 9 dyslipidaemia, and inflammation. 10 VAT is not a homogeneous tissue but comprises distinct compartments with unique anatomical locations and physiological functions. Preperitoneal, omental, mesenteric, and perirenal fat each play distinct roles in metabolic and inflammatory processes. These fat depots are linked respectively to metabolic dysfunction and liver disease, 11 systemic inflammation, 5 intestinal immune disorders, 12 and aldosterone‐related cardiometabolic and renal dysfunction. 13 In contrast, SAT, formerly regarded as a homogeneous compartment, is now recognized to be anatomically partitioned by Scarpa's fascia into superficial and deep SAT layers. 14 Superficial SAT has been associated with favourable metabolic characteristics, with respect to hepatic fat content, particularly among women with obesity, whereas deep SAT is linked to hepatic steatosis and adverse metabolic profiles, as HDL cholesterol levels in men with obesity. 14 Also, in women living with obesity, especially superficial SAT is associated with circulating leukocytes and inflammatory proteins. 15 Thus, sex differences should be considered when assessing abdominal fat distribution.
Evidence suggests that reductions in specific adipose depots, including deep SAT, are associated with improvements in metabolic syndrome parameters, such as enhanced insulin sensitivity and favourable alterations in lipid profiles. 16
The visceral‐to‐subcutaneous fat ratio (VAT/SAT ratio) is associated with CMD risk factors, independently of BMI, WC, and VAT. 17 The VAT/SAT ratio may be a unique risk factor independent of absolute fat volumes. 17
SAT and VAT have been assessed traditionally with magnetic resonance imaging (MRI) or computed tomography (CT), both considered gold standard techniques. 18 However, their clinical application is limited by cost, radiation in CT, and technical difficulties. 18 Ultrasound (US) offers a low‐cost, non‐invasive alternative for SAT and VAT thickness real‐time assessment. 18 , 19 , 20 Although abdominal US effectively measures both compartments, it is not yet considered a gold standard technique. 18 , 19 In this context, the hypothesis is that greater VAT thickness and higher VAT/SAT ratio, assessed by US, are associated with altered metabolic conditions. Confirming these associations would support the clinical utility of abdominal US.
Therefore, this study aims to evaluate the accuracy of US‐derived SAT and VAT measurements against MRI and to assess their associations with metabolic conditions in adults with abdominal obesity.
2. METHODS AND PROCEDURES
2.1. Study design and participants
A cross‐sectional study was conducted between May 2023 and January 2025 in Reus (Spain), and visits held at Centre Mèdic Quirúrgic (CMQ). The study followed the Helsinki Declaration and Good Clinical Practice Guidelines of the International Conference of Harmonization (ICH GCP), and was approved on 27/04/2023 by the Ethics Committee of the Pere Virgili Health Research Institute (IISPV) (074/2023) and registered on ClinicalTrials.gov (NCT05882149). All participants provided written informed consent.
2.2. Eligibility criteria
The inclusion criteria were as follows: (a) male and female subjects ≥18 years old; (b) abdominal obesity (WC ≥102 cm for men; ≥88 cm for women); (c) signed informed consent; (d) willingness to follow the protocol; and (e) females of child‐bearing potential required a negative pregnancy test.
The exclusion criteria were the following: (a) WC <102 cm for men, <88 cm for women, or >150 cm; (b) BMI ≥40 kg/m2; (c) diagnosed and pharmacologically treated type 1 or type 2 diabetes (fasting blood glucose ≥7 mmol/L); (d) severe chronic diseases (e.g., autoimmune, cardiovascular, liver, kidney, dementia, pancreatic disease, history of cancer within past 5 years, anaemia, or any other disease or condition considered by the researcher that could interfere); (e) immunosuppression or related treatments; (f) pharmacologically treated (medication/supplements) dyslipidaemia; (g) recent (during the last month before inclusion) antibiotic, probiotic, prebiotic, weight‐loss drug, laxative, or fibre use; (h) chronic active inflammatory disorders; (i) bariatric surgery; (j) any chronic gastrointestinal disease (e.g., IBD); (k) regular use of systemic or inhaled corticosteroids, or immunomodulatory drugs; (l) significant change in tobacco, snuff, nicotine, and e‐cigarette use habits in the past 3 months or planned cessation of the use of these products during the trial; (m) active or recent (last 3 months) participation in a weight loss program; (n) weight change (increase or loss) of 3 kg during the past 3 months; (o) pregnancy, planning pregnancy, or breastfeeding; (p) substance abuse; (q) hypersensitivity to study products; (r) participation in another clinical trial within 60 days prior to inclusion; and (s) inability to follow study guidelines.
2.3. Assessment of abdominal fat distribution by ultrasound
US measurements of SAT and VAT were taken in the supine position with the transducer placed about 2 cm above the umbilical scar, the position corresponding to the third and fourth lumbar vertebra (L3‐L4), where the aorta bifurcates into the iliac arteries, in the axial plane, with minimal transducer pressure, identified when lateral imaging of the US becomes blur. 18 , 21 Scans were done during exhalation using a VINNO 5 device (Vinno (Suzhou) Co., Ltd., Suzhou, China) in the HAR‐mode with the Abdominal Resolution preset at a frequency of 5 MHz with a convex transducer (F2‐5CE). All ultrasound measures were carried out three times by a single researcher, and the mean was used for analysis. SAT (cm), including superficial and profound SAT, was measured as the distance between the skin and the linea alba, and VAT (cm), including preperitoneal, omental, and mesenteric fat, but not perirenal fat, as the distance from the posterior face of the linea alba to the anterior wall of the aorta. 18 , 21 VAT/SAT ratio was calculated and is expressed as VAT/SAT ratio (thickness).
2.4. Assessment of abdominal fat distribution by magnetic resonance imaging
SAT (thickness, expressed in cm and area, expressed in cm2) and VAT (thickness, expressed in cm and area, expressed in cm2) were measured by an MRI transverse body scan in one axial slice 5 cm over intervertebral space of L4‐L5. 22 The MRI study was performed with a General Electric 3 Tesla HDXT MRI after 6 h of fasting. VAT/SAT ratio was calculated for thickness (VAT/SAT ratio [thickness]) and area (VAT/SAT ratio [area]).
2.5. Assessment of anthropometric parameters
Anthropometric parameters were obtained while the subjects were wearing lightweight clothing and no shoes. Trained dietitians measured the body weight and body composition of the subjects using a body composition analyser (Tanita MC 780‐MA; Tanita Corp., Barcelona, Spain) and the height of the subjects using a wall‐mounted stadiometer (Tanita Leicester Portable; Tanita Corp., Barcelona, Spain).
The WC was measured at 2 cm above the umbilicus scar using a 150‐cm anthropometric steel measuring tape. 23 The waist (WC, cm) to height (cm) ratio (WHR) and the conicity index (CI), defined as WC (cm)/[0.109 × square root of weight (kg)/height (m)], were calculated.
2.6. Assessments of cardiovascular disease risk factors and metabolic conditions
Systolic and diastolic blood pressure (SBP and DBP) were measured twice after 2–5 min of patient respite, seated, with a 1‐min interval in between, using an automatic sphygmomanometer (OMRON M6 Comfort, HEM‐7360‐E; Peroxfarma, Barcelona, Spain). The pulse pressure (PP) was determined by the difference between the SBP and DBP readings. 24
The lipid profiles, such as total cholesterol, high‐density lipoprotein cholesterol (HDL‐c), low‐density lipoprotein cholesterol (LDL‐c), triglycerides (TG), very low density lipoprotein (VLDL), and fasting blood glucose (FBP) were measured in serum using standardized enzymatic automated methods in an autoanalyzer (Beckman Coulter‐Synchron, Galway, Ireland).
Moreover, participants were classified based on whether they met the diagnostic criteria for metabolic syndrome, according to the definitions established by the 2009 Harmonized Criteria, which include elevated WC (≥102 cm in men and ≥88 cm in women), elevated blood pressure (systolic ≥130 and/or diastolic ≥85 mmHg) or drug treatment, elevated triglycerides (>150 mg/dL) or drug treatment, reduced HDL‐c (<40 mg/dL in men and <50 mg/dL in women), and elevated fasting glucose (≥100 mg/dL) or drug treatment. 25
2.7. Statistical analysis
Categorical variables were presented as percentages (%). The Normality was tested by Kolmogorov–Smirnov test. Continuous variables were presented as mean ± standard deviation (SD) (normal distribution) or median and interquartile range (IQR) (non‐normal distribution), separated by sex, and compared by t test or Mann–Whitney U test, depending on the nature of the variable.
Spearman correlation coefficients (r) were calculated to determine significant correlations between VAT, SAT, and VAT/SAT ratio, by US and MRI, and CVD risk factors, interpreted as ≤0.20, none; 0.21–0.40, weak; 0.41–0.60, moderate; 0.61–0.80, high; and ≥0.81, very high. 26 Linear regression models adjusted for sex and age were fitted to assess the association between VAT, SAT, and VAT/SAT ratio, assessed by US and MRI. The level of agreement in VAT and SAT between the US and MRI was assessed using Bland–Altman plots. The mean difference/bias between the two methods was calculated and tested against zero using a paired t test.
Finally, ROC analysis was performed to assess the diagnostic accuracy of VAT, assessed by US and MRI, for metabolic syndrome and for high fasting glucose using the area under the curve (AUC) and 95% confidence interval (95% CI).
ROC analysis was significant when AUC is >0.5 and the lower 95% CI value is >0.5 (reference line = 0.5). 27 The diagnostic accuracy based on AUC was the following: 0.50–0.59, fail; 0.60–0.69, poor; 0.70–0.79, fair; 0.80–0.89, good; and ≥0.9, excellent. 27 The Youden's index was used to determine the cut‐off point for each variable based on sensitivity and specificity to confirm the metabolic syndrome diagnosis; the higher Youden's index determined the best cut‐off point. 27
All statistical analyses were performed using SPSS IBM (Corp. Released 2023. IBM SPSS Statistics for Windows, Version 29.0.1.0, IBM Corp., Armonk, NY). A p‐value of <0.05 was considered significant.
3. RESULTS
3.1. Characteristics of the study participants
Baseline characteristics of all participants are presented in Table 1. A total of 113 participants with abdominal obesity were included, 34.5% were men (n = 39/113), and 65.5% were women (n = 74/113). The mean ± SD age of participants was 52 ± 10.9 years, and the median (IQR) BMI was 30(5.15) kg/m2. There were significant differences between sexes for anthropometric measures, US measures, MRI measures, and cardiovascular parameters (p < 0.05). Frequencies of CVD risk factors and metabolic conditions of the participants are presented in Table 2. There were no significant differences in BMI when compared by sex.
TABLE 1.
Descriptive characteristics of the participants.
| Total (n = 113) | Women (n = 74) | Men (n = 39) | p‐value a | |
|---|---|---|---|---|
| Age (year) | 52 ± 10.93 | 53 ± 11.16 | 52 ± 10.60 | 0.610 |
| Anthropometric measures | ||||
| Weight (kg) | 82.00 (20.58) | 78.65 (12.80) | 96.15 (22.18) | <0.001 |
| Height (m) | 1.67 (0.13) | 1.63 (0.10) | 1.76 (0.13) | <0.001 |
| BMI (kg/m2) | 30.00 (5.15) | 29.10 (5.23) | 31.25 (5.98) | 0.086 |
| Waist circumference (cm) | 104.2 ± 9.5 | 100.6 ± 8.4 | 111.1 ± 7.7 | <0.001 |
| Hip circumference (cm) | 111.00 (8.38) | 111.00 (10) | 110.75 (6) | 0.847 |
| Waist‐to‐hip ratio | 0.94 ± 0.07 | 0.90 ± 0.06 | 1.00 ± 0.05 | <0.001 |
| Conicity index | 1.43 ± 1.08 | 1.47 ± 1.33 | 1.37 ± 0.06 | <0.001 |
| Fat mass (kg, %) |
30.38 ± 6.57 35.47 ± 5.59 |
30.89 ± 6.91 38.37 ± 3.95 |
29.62 ± 6.44 30.13 ± 3.97 |
<0.001 <0.001 |
| Fat free mass (kg, %) |
52.40 (15.95) 64.36 ± 5.59 |
48.90 (5.88) 61.63 ± 3.91 |
68.25 (13) 66.59 ± 3.69 |
<0.001 <0.001 |
| Lean body mass (kg, %) |
49.75 (15.18) 61.28 ± 5.34 |
46.40 (5.60) 57.70 (5.10) |
64.85 (12.40) 66.70 (5.15) |
<0.001 <0.001 |
| MRI measures | ||||
| Waist circumference (cm) | 102.53 (12.21) | 98.89 (12.43) | 106.97 (13.11) | <0.001 |
| VAT (cm2) | 183.94 ± 77.41 | 149.87 ± 55.32 | 248.57 ± 72.35 | <0.001 |
| SAT (cm2) | 320.72 ± 102.35 | 340.65 ± 90.61 | 282.91 ± 113.41 | <0.001 |
| VAT/SAT ratio (area) | 0.50 (0.46) | 0.42 (0.20) | 0.87 (0.83) | <0.001 |
| VAT thickness (cm) | 6.56 ± 2.44 | 5.66 ± 2.01 | 8.28 ± 2.28 | <0.001 |
| SAT thickness (cm) | 2.89 ± 1.12 | 3.00 ± 1.06 | 2.68 ± 1.21 | 0.007 |
| VAT/SAT ratio (thickness) | 2.01 (2.13) | 1.73 (1.66) | 2.72 (3.31) | 0.01 |
| US measures | ||||
| VAT thickness (cm) | 6.23 (4.03) | 5.44 (3.14) | 8.77 (3.27) | <0.001 |
| SAT thickness (cm) | 2.33 ± 0.82 | 2.41 ± 0.74 | 2.16 ± 0.95 | 0.086 |
| VAT/SAT ratio (thickness) | 2.75 (2.75) | 2.40 (1.78) | 3.90 (3.59) | <0.001 |
| Cardiovascular parameters | ||||
| Total cholesterol (mg/dL) | 209.85 ± 33.05 | 213.89 ± 34.93 | 202.21 ± 27.99 | 0.056 |
| HDL‐c (mg/dL) | 55.00 (20.75) | 60.00 (19.75) | 45.50 (12.50) | <0.001 |
| LDL‐c (mg/dL) | 131.79 ± 29.43 | 133.33 ± 30.81 | 128.90 ± 26.81 | 0.450 |
| Triglycerides (mg/dL) | 101.50 (53.75) | 98.50 (42.25) | 113 (91.75) | 0.006 |
| Glucose (mg/dL) | 92.90 ± 11.21 | 90.55 ± 10.92 | 97.36 ± 10.48 | 0.002 |
| SBP (mmHg) | 123.87 ± 14.54 | 120.30 ± 14.60 | 130.64 ± 11.92 | <0.001 |
| DBP (mmHg) | 83.69 ± 9.87 | 83.41 ± 8.72 | 84.23 ± 11.86 | 0.677 |
| HR (mmHg) | 68 (13) | 69 (11.75) | 67 (18.50) | 0.540 |
Note: Significant results are expressed in bold. Values are expressed as mean ± standard deviation (SD) for variables with normal distribution or median and interquartile range (IQR) for variables with non‐normal distribution.
Abbreviations: BMI, body mass index; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; US, ultrasound; HDL‐c, high‐density lipoprotein cholesterol; LDL‐c, low‐density lipoprotein cholesterol; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate.
Sex differences by t test or Mann–Whitney U test, depending on the nature of the variable. p‐value <0.05 is statistically significant.
TABLE 2.
Frequencies of metabolic conditions of the participants.
| Total (n = 113) | Women (n = 74) | Men (n = 39) | p‐value a | |
|---|---|---|---|---|
| Smoking habit | 0.008 | |||
| Non‐smokers, % (n) | 82.3 (93) | 89.2 (66) | 69.2 (27) | |
| Smokers, % (n) | 17.7 (20) | 10.8 (8) | 30.8 (12) | |
| BMI | 0.165 | |||
| Underweight, % (n) | 0 (0) | 0 (0) | 0 (0) | |
| Healthy weight, % (n) | 1.8 (2) | 1.4 (1) | 2.6 (1) | |
| Overweight, % (n) | 47.8 (54) | 55.4 (41) | 33.3 (13) | |
| Obese, % (n) | 50.4 (57) | 43.2 (32) | 64.1 (25) | |
| Class 1, % (n) | 35.4 (40) | 29.7 (22) | 46.2 (18) | |
| Class 2, % (n) | 15 (17) | 13.5 (10) | 17.9 (7) | |
| Class 3, % (n) | 0 (0) | 0 (0) | 0 (0) | |
| Metabolic syndrome | <0.001 | |||
| No, % (n) | 65.5 (74) | 77 (57) | 43.6 (17) | |
| Yes, % (n) | 34.5 (39) | 23 (17) | 56.4 (22) | |
| Prediabetes | <0.001 | |||
| No, % (n) | 76.1 (86) | 86.5 (64) | 56.4 (22) | |
| Yes, % (n) | 23.9 (27) | 13.5 (10) | 43.6 (17) | |
| High blood pressure | 0.017 | |||
| No, % (n) | 51.3 (58) | 59.5 (44) | 35.9 (14) | |
| Yes, % (n) | 48.7 (55) | 40.5 (30) | 64.1 (25) | |
| Hypertriglyceridemia | 0.007 | |||
| No, % (n) | 80.5 (91) | 87.8 (65) | 66.7 (26) | |
| Yes, % (n) | 19.5 (22) | 12.2 (9) | 33.3 (13) | |
| Low HDL‐cholesterol | 0.210 | |||
| No, % (n) | 77.9 (88) | 74.3 (55) | 84.6 (33) | |
| Yes, % (n) | 22.1 (25) | 25.7 (19) | 15.4 (6) | |
Note: Significant results are expressed in bold.
Abbreviations: BMI, body mass index; HDL, high‐density lipoprotein; WHR, waist‐to‐hip ratio.
Sex differences by chi‐square test.
Regarding high blood pressure, 48.7% (n = 55/113) of the participants presented high blood pressure [40.5% (n = 30/74) in women and 64.1% (n = 25/39) in men], although only 18 participants were prescribed antihypertensive medication.
Regarding glucose, 23.9% (n = 27/113) of the participants presented prediabetes [13.5% (n = 10/74) in women and 43.6% (n = 17/39) in men]. Impaired fasting glucose (100–125 mg/dL) is classified as prediabetes according to the latest definition from the American Diabetes Association. 3
Additionally, according to the 2009 Harmonized Criteria, 25 56.4% of men presented metabolic syndrome, and 23% of women (p < 0.001).
3.2. Comparison of US and MRI measurements: Correlations and Bland–Altman analysis
Spearman correlation showed a very high correlation between WC measured by the researcher and MRI (r = 0.85, p < 0.001), as well as between VAT thickness (cm) (r = 0.82, p < 0.001) and SAT thickness (cm) (r = 0.81, p < 0.001) measured by US and MRI. A high correlation was also observed between VAT area (cm2) measured by MRI and VAT thickness (cm) measured by US (r = 0.71, p < 0.001). Additionally, SAT area (cm2) measured by MRI showed a moderate correlation with SAT thickness (cm) measured by US (r = 0.60, p < 0.001).
Regarding the VAT/SAT ratio (thickness), Spearman correlation showed a very high correlation between VAT/SAT ratio (thickness) measured by US and MRI (r = 0.84, p < 0.001).
The Bland–Altman analysis revealed a nonsignificant mean difference between VAT thickness (cm) measured by US and MRI (mean difference: −0.23 ± 1.42, p = 0.100). However, a significant mean difference was observed between US and MRI measurements of SAT thickness (cm) (p < 0.001) and VAT/SAT ratio (thickness) (p = 0.007). Also, a significant mean difference was observed between WC measured by the researchers and by MRI (p = 0.025).
When segregated by sex, a nonsignificant mean difference in VAT thickness (cm) measured by US and MRI was presented for both men (mean difference: −0.08 ± 1.50, p = 0.742) and women (mean − 0.30 ± 1.38, p = 0.070) (Figure 1).
FIGURE 1.
Bland–Altman analysis of VAT for (A) all population, (B) women, and (C) men. In all three plots, the Y‐axis represents the difference between VAT from MRI and US. The X axis represents the mean VAT from the two methods. VAT, visceral adipose tissue; MRI, magnetic resonance imaging; US, ultrasound.
Additionally, a nonsignificant mean difference in VAT/SAT ratio (thickness) measured by US and MRI (mean difference: −0.29 ± 1.88, p = 0.202) was observed, as well as a nonsignificant mean difference in WC measured by the researchers and MRI (mean difference: 0.70 ± 3.37, p = 0.077) in women (Figure S1, Supporting Information).
3.3. ROC analysis and cut‐off points to confirm metabolic syndrome diagnosis via abdominal fat distribution measured by US and MRI
Prior to ROC analysis, correlation and linear regression analysis were performed (Tables S1–S3). After linear regression, statistically significant associations were observed between glucose and VAT thickness measured by US (b = 0.257, p = 0.013), VAT thickness measured by MRI (b = 0.333, p = 0.002), and VAT area measured by MRI (b = 0.323, p = 0.009).
Regarding VAT thickness measured by US, ROC analysis revealed poor diagnostic accuracy for metabolic syndrome in women (AUC = 0.69, p < 0.001) and fair diagnostic accuracy in the total population (AUC = 0.70, p < 0.001). Similarly, VAT thickness measured by MRI showed poor diagnostic accuracy for metabolic syndrome in women (AUC = 0.66, p = 0.038) and fair diagnostic accuracy in the total population (AUC = 0.70, p < 0.001). VAT area measured by MRI demonstrated similar results, poor accuracy in women (AUC = 0.68, p = 0.020) and fair accuracy in the total population (AUC = 0.71, p < 0.001).
According to Youden's index, optimal cut‐off points for confirming metabolic syndrome in women were 4.83 cm for VAT thickness measured by US, 6.78 cm for VAT thickness measured by MRI, and 117.90 cm2 for VAT area measured by MRI (Table 3 and Figure 2). No statistically significant diagnostic accuracy by ROC analysis was found in men (p > 0.05) (Table 3 and Figure 2).
TABLE 3.
ROC analysis for the diagnostic accuracy of metabolic syndrome and prediabetes by abdominal obesity employing VAT measured by US and MRI.
| AUC | 95% CI | p‐value | Sensitivity | Specificity | Youden's index | Cut‐off | |
|---|---|---|---|---|---|---|---|
| Metabolic syndrome diagnosis | |||||||
| VAT thickness (cm) by US | |||||||
| All | 0.70 | 0.60–0.79 | <0.001 | 0.816 | 0.465 | 0.351 | 5.77 |
| Women | 0.69 | 0.56–0.81 | <0.001 | 1.00 | 0.589 | 0.411 | 4.83 |
| Men | 0.58 | 0.39–0.77 | 0.430 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (thickness) by US | |||||||
| All | 0.68 | 0.58–0.78 | 0.001 | 0.658 | 0.324 | 0.334 | 2.90 |
| Women | 0.69 | 0.56–0.83 | 0.004 | 0.813 | 0.411 | 0.402 | 2.52 |
| Men | 0.52 | 0.33–0.71 | 0.853 | ‐ | ‐ | ‐ | ‐ |
| VAT thickness (cm) by MRI | |||||||
| All | 0.70 | 0.60–0.80 | <0.001 | 0.667 | 0.297 | 0.369 | 6.56 |
| Women | 0.66 | 0.51–0.81 | 0.038 | 0.471 | 0.158 | 0.313 | 6.78 |
| Men | 0.54 | 0.36–0.73 | 0.654 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (thickness) by MRI | |||||||
| All | 0.64 | 0.54–0.75 | 0.009 | 0.410 | 0.151 | 0.260 | 3.92 |
| Women | 0.64 | 0.48–0.80 | 0.082 | ‐ | ‐ | ‐ | ‐ |
| Men | 0.50 | 0.31–0.69 | 1.000 | ‐ | ‐ | ‐ | ‐ |
| VAT area (cm2) by MRI | |||||||
| All | 0.71 | 0.61–0.81 | <0.001 | 0.538 | 0.189 | 0.349 | 214.37 |
| Women | 0.68 | 0.53–0.83 | 0.020 | 0.882 | 0.579 | 0.303 | 117.90 |
| Men | 0.51 | 0.32–0.70 | 0.934 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (area) by MRI | |||||||
| All | 0.66 | 0.56–0.77 | 0.003 | 0.641 | 0.311 | 0.330 | 0.54 |
| Women | 0.62 | 0.47–0.78 | 0.117 | ‐ | ‐ | ‐ | ‐ |
| Men | 0.51 | 0.33–0.70 | 0.887 | ‐ | ‐ | ‐ | ‐ |
| Prediabetes diagnosis | |||||||
| VAT thickness (cm) by US | |||||||
| All | 0.65 | 0.54–0.77 | 0.010 | 0.630 | 0.317 | 0.313 | 7.35 |
| Women | 0.71 | 0.52–0.90 | 0.029 | 0.800 | 0.371 | 0.429 | 5.87 |
| Men | 0.43 | 0.24–0.61 | 0.435 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (thickness) by US | |||||||
| All | 0.65 | 0.53–0.77 | 0.014 | 0.667 | 0.366 | 0.301 | 2.90 |
| Women | 0.72 | 0.52–0.91 | 0.030 | 0.700 | 0.258 | 0.442 | 3.04 |
| Men | 0.42 | 0.23–0.61 | 0.411 | ‐ | ‐ | ‐ | ‐ |
| VAT thickness (cm) by MRI | |||||||
| All | 0.70 | 0.59–0.80 | <0.001 | 0.815 | 0.442 | 0.373 | 6.11 |
| Women | 0.77 | 0.64–0.90 | <0.001 | 0.800 | 0.297 | 0.503 | 6.11 |
| Men | 0.44 | 0.25–0.63 | 0.519 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (thickness) by MRI | |||||||
| All | 0.66 | 0.55–0.77 | 0.006 | 0.889 | 0.612 | 0.277 | 1.58 |
| Women | 0.69 | 0.53–0.86 | 0.021 | 0.700 | 0.302 | 0.398 | 2.27 |
| Men | 0.48 | 0.30–0.67 | 0.844 | ‐ | ‐ | ‐ | ‐ |
| VAT area (cm2) by MRI | |||||||
| All | 0.72 | 0.62–0.82 | <0.001 | 0.889 | 0.523 | 0.366 | 150.84 |
| Women | 0.71 | 0.54–0.88 | 0.014 | 0.700 | 0.344 | 0.356 | 161.93 |
| Men | 0.52 | 0.33–0.71 | 0.846 | ‐ | ‐ | ‐ | ‐ |
| VAT/SAT ratio (area) by MRI | |||||||
| All | 0.69 | 0.58–0.80 | 0.001 | 0.667 | 0.314 | 0.353 | 0.58 |
| Women | 0.61 | 0.44–0.78 | 0.195 | ‐ | ‐ | ‐ | ‐ |
| Men | 0.51 | 0.33–0.69 | 0.909 | ‐ | ‐ | ‐ | ‐ |
Note: Significant results are expressed in bold. ROC analysis is significant when AUC is >0.5 and the lower 95% CI value is >0.5. A p‐value <0.05 is statistically significant.
Abbreviations: AUC, area under the curve; CI, confidence interval; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; US, ultrasound.
FIGURE 2.
ROC analysis for the accuracy diagnosis of metabolic syndrome and prediabetes of VAT by US and MRI. US, ultrasound; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; AUC, area under the curve; CI, confidence interval.
Regarding the VAT/SAT ratio, ROC analysis revealed poor diagnostic accuracy for metabolic syndrome when using VAT/SAT ratio (thickness) measured by US in the total population (AUC = 0.68, p = 0.001) and in women (AUC = 0.69, p = 0.004). Similarly, VAT/SAT ratio (thickness) measured by MRI also demonstrated poor diagnostic accuracy in the total population (AUC = 0.64, p = 0.009). Also, VAT/SAT ratio (area) measured by MRI showed poor diagnostic accuracy in the total population (AUC = 0.66, p = 0.003).
According to Youden's index, optimal cut‐off points for confirming metabolic syndrome for all populations were 2.90 for VAT/SAT thickness ratio measured by US, 3.92 for VAT/SAT thickness ratio measured by MRI, and 0.54 for VAT/SAT area ratio measured by MRI. In women, the proposed cut‐off point for VAT/SAT thickness ratio measured by US was 2.52 (Table 3 and Figure S2). No statistically significant diagnostic performance by ROC analysis was observed in men for any VAT/SAT ratios, including thickness measured by US or MRI and area ratio measured by MRI (p > 0.05) (Table 3 and Figure S2).
3.4. ROC analysis and cut‐off points to confirm prediabetes diagnosis via abdominal fat distribution measured by US and MRI
Regarding VAT thickness measured by US, the ROC analysis demonstrated poor diagnostic accuracy for prediabetes in the total population (AUC = 0.65, p = 0.010) but fair diagnostic accuracy in women (AUC = 0.71, p = 0.029).
Similarly, VAT thickness measured by MRI revealed fair diagnostic accuracy for prediabetes in both the total population (AUC = 0.70, p < 0.001) and women (AUC = 0.77, p < 0.001). VAT area measured by MRI yielded comparable results, with AUC = 0.72 (p < 0.001) for the total population and AUC = 0.71 (p = 0.014) for women.
According to Youden's index, the optimal cut‐off points for these parameters to confirm prediabetes in women were 5.87 cm for VAT thickness measured by US, 6.11 cm for VAT thickness measured by MRI, and 161.93 cm2 for VAT area measured by MRI (Table 3 and Figure 2). No statistically significant diagnostic accuracy by ROC analysis was observed for men in any of the measurements (p > 0.05) (Table 3 and Figure 2).
Concerning VAT/SAT ratio (thickness) measured by US, the ROC analysis revealed poor diagnostic accuracy for prediabetes in the total population (AUC = 0.65, p = 0.014) and fair diagnostic accuracy in women (AUC = 0.72, p = 0.030). Similarly, VAT/SAT ratio (thickness) measured by MRI showed poor diagnostic accuracy in both the total population (AUC = 0.66, p = 0.006) and women (AUC = 0.69, p = 0.021). VAT/SAT ratio (area) measured by MRI also showed poor diagnostic accuracy in the total population (AUC = 0.69, p = 0.001).
According to Youden's index, optimal cut‐off points for confirming prediabetes for all population were 2.90 for VAT/SAT ratio (thickness) measured by US, 2.58 for VAT/SAT ratio (thickness) measured by MRI, and 0.58 for VAT/SAT ratio (area) measured by MRI. For women, the proposed cut‐off point was 3.04 for VAT/SAT ratio (thickness) measured by US and 2.27 for VAT/SAT ratio (thickness) measured by MRI (Table 3 and Figure S3). No statistically significant diagnostic accuracy by ROC analysis was found for men in any of the VAT/SAT ratio measurements (p > 0.05) (Table 3 and Figure S3).
4. DISCUSSION
The present cross‐sectional study adds new evidence supporting the use of US as a non‐invasive method to assess abdominal fat distribution. The present results demonstrate a strong correlation between US and MRI measurements, particularly for VAT thickness. Furthermore, US‐measured VAT thickness and VAT/SAT ratio showed fair diagnostic accuracy for identifying not only metabolic syndrome but also prediabetes, especially in women.
According to Youden's Index, the optimal cut‐off points for confirming metabolic syndrome in women by US are 4.83 cm for VAT thickness and 2.52 for VAT/SAT ratio (thickness). Additionally, the optimal cut‐off points to confirm prediabetes in women by the US are 5.87 cm for VAT thickness and 3.04 for VAT/SAT ratio (thickness). These proposed cut‐off points of VAT thickness and VAT/SAT ratio (thickness) measured by US may be useful for initial metabolic conditions screening in clinical practice.
Our results support abdominal US as a valid method to determine VAT for all populations and by sex, although significant differences between US and MRI measurements of SAT and VAT/SAT ratio were found. These discrepancies were consistent with earlier studies, which also found significant mean differences for SAT and VAT/SAT ratio. 28 , 29
Since women generally have more SAT and less VAT than men, this may partly explain discrepancies between US and MRI. 30 Furthermore, menstrual cycle–related weight changes in premenopausal women were not considered, which could be a limitation of this study. 31 Banack et al. reported that in 10 184 post‐menopausal women from the Women's Health Initiative (WHI), those with more than 10 years post‐menopause exhibited higher levels of VAT and lower levels of SAT, resulting in an elevated VAT/SAT ratio (area) compared to those within 10 years of menopause. 32 Fat distribution was measured by dual‐energy x‐ray absorptiometry (DXA) scans in the 1990s, and Hologic APEX software was used to re‐analyse archived DXA scans to obtain measures of abdominal adipose tissue. 32 Their findings indicate that alterations in VAT and SAT can occur without changes in conventional anthropometric measures, such as BMI or WC, among older women.
Mean researcher‐measured WC was 100.6 ± 8.4 cm for women and 111.1 ± 7.7 cm for men. Considering sex differences in fat distribution, the present results suggest that women with WC values similar to men may have higher VAT levels and, consequently, a greater risk for metabolic syndrome. In the Framingham Heart Study, VAT and VAT/SAT ratio were more strongly associated with CMD risk and cardiovascular events in women than BMI and WC, whereas in men, higher BMI and WC were similarly associated with increased CMD risk as VAT. 30
Our ROC analysis showed that VAT thickness measured by US or MRI served as a fair predictor of metabolic syndrome diagnosis. Similarly, the VAT area measured by MRI also presented a fair diagnostic accuracy for metabolic syndrome diagnosis.
Regarding VAT/SAT ratio, the ROC analysis showed a poor diagnostic accuracy for VAT/SAT ratio (thickness) measured by US and MRI, and VAT/SAT ratio (area) measured by MRI. When segregated by sex, only the VAT/SAT ratio (thickness) measured by the US showed significant diagnostic accuracy for metabolic syndrome in women.
Previous studies have focused on the association of VAT and VAT/SAT ratio and CMD risk. 5 , 30
In our study, the VAT/SAT ratio (thickness) measured by US emerged as a key indicator of fat distribution and CMD risk. A higher VAT/SAT ratio suggests a greater accumulation of VAT relative to SAT, indicating a tendency to store fat preferentially in the visceral compartment which is associated with increased CMD risk. Conversely, a lower VAT/SAT ratio reflects a reduction in VAT, an increase in SAT, or a combination of both. 32
Regarding the generalisability of our findings, it is important to acknowledge that this study was conducted in a single‐centre Spanish population from the Camp de Tarragona region, which may influence the observed VAT/SAT ratio distribution due to specific ethnic, lifestyle, and dietary characteristics. The participants were predominantly of Caucasian European descent, a group that generally exhibits lower VAT/SAT ratios compared to Asian populations but similar or slightly higher ratios than other Western cohorts. 32 , 33 Additionally, the population of Mediterranean regions typically shows distinctive lifestyle and dietary patterns, such as moderate physical activity and high adherence to the Mediterranean diet, which have been associated with more favourable fat distribution profiles in previous studies. 34 , 35 Although our study was descriptive and not designed to assess these influences directly, such regional and cultural factors could partly explain the VAT/SAT ratio values observed in our Spanish population and should be considered when comparing our findings with populations of different ethnic or lifestyle backgrounds.
Cuatrecasas et al. compared VAT measured by US and CT; however, VAT was divided into preperitoneal and omental fat. 5 Moreover, a proposed cut‐off point of omental fat thickness by US (54 mm for men and 37 mm for women) was proposed as a predictive marker of metabolic syndrome. 5 However, in our study, VAT thickness was measured as the distance from the posterior face of the linea alba to the anterior wall of the aorta, including preperitoneal and omental fat in the same measurement. Taking into account the difference in measuring VAT between the present study and theirs, our proposed cut‐off point of 48.3 mm for women would be new data to the proposed cut‐off by Cuatrecasas et al.
For metabolic syndrome diagnosis, following the 2009 Harmonized Criteria, 25 23% (n = 17/74) of women had metabolic syndrome. Applying Youden's cut‐offs increased detection to 59.2% by VAT thickness (US), 45.1% by VAT/SAT ratio, 54.1% by WC, and 43.2% by WHR (Table S4). This shows VAT thickness (US) notably improves diagnostic yield, identifying over twice as many cases.
Regarding prediabetes, initially 23.9% (n = 27/113) of all population presented prediabetes, however, applying proposed Youden's cut‐off points, 39.4% by VAT thickness (US), and 44% by VAT/SAT ratio (thickness) measured by US presented prediabetes. When segregated by sex, in women, initially 13.5% (n = 10/74) presented prediabetes, and applying Youden's cut‐off points, 38% by VAT thickness (US) and 45.1% by VAT/SAT ratio (thickness) measured by US (n = 32/74) presented prediabetes (Table S4). This represents a nearly twofold increase in detection, highlighting the potential of US‐measurements to enhance early identification of individuals at risk of metabolic alterations. These results align with findings conducted by Philipsen et al., who reported an association between VAT via US and indices of glucose metabolism, such as fasting plasma glucose and HbA1C. 36 Recent evidence supports the utility of US VAT measurements as early predictors of impaired glucose metabolism. 36 Particularly, reduction of perirenal fat has been shown to correlate with improvements in glucose metabolism. 16 , 37 Also, our study provides new evidence supporting abdominal US as a non‐invasive technique for assessing abdominal fat distribution. Given its widespread use in clinical practice, US offers a practical tool for evaluating metabolic syndrome‐related risk factors. 18 Incorporating US into traditional WC and WHR assessments may enable earlier detection of prediabetes and metabolic syndrome by providing real‐time evaluation of VAT, unlike MRI, which requires post‐acquisition image analysis. 18 Routine use of US could lower healthcare costs, improving patient outcomes and reducing CVD risk and mortality. 38 , 39
A limitation of our study is that all participants presented abdominal obesity, which may have influenced the relationship between waist circumference and metabolic markers. Consequently, the proposed ultrasound cut‐off points are applicable only to populations with abdominal obesity and cannot be extrapolated to individuals with normal waist circumference. Future studies including adults with and without abdominal obesity are needed to validate and refine these thresholds. Furthermore, the use of a single anatomical site for WC measurement represents a limitation, as recent evidence from Kim et al. demonstrated that the specific measurement location can alter the absolute WC values and consequently affect the classification of abdominal obesity. 40 Nevertheless, measuring WC 2 cm above the umbilicus, as performed in our study, is a widely accepted and standardised method in adult populations and has been shown to provide valid estimates of central adiposity and associations with cardiometabolic risk factors. 23 Although multi‐site or alternative measurements may yield slightly different absolute values, the overall associations between WC and metabolic outcomes are expected to remain consistent. 40
Another limitation is that all US measurements were carried out by a single operator. To minimize variability, the operator acquired specific training, following standardised protocols for the US‐measurements. Although it reduces variability, it does not address inter‐observer reproducibility, limiting its generalizability. Despite the fact that abdominal US is a potentially cost‐effective imaging technique, its benefits rely on proper operator training and standardization of reliability of measurement protocols and cut‐off points.
Furthermore, SAT and VAT were evaluated as heterogeneous compartments, instead of evaluating each fat layer and its metabolic implications. Although our study assessed VAT as the combined thickness of preperitoneal, omental, and mesenteric fat, US allows differentiation of distinct visceral fat layers, including perirenal fat, each of which may have unique physiological roles and implications for cardiometabolic risk. Future studies should consider evaluating the individual VAT layers separately and incorporate perirenal fat, which could provide more precise insights into the relationship between VAT distribution and CVD risk.
Nevertheless, our proposed cut‐off points for total VAT, including preperitoneal and omental fat, and the VAT/SAT ratio are promising. However, these thresholds should be considered exploratory and require confirmation in future studies before implementation in guidelines.
Lastly, our sample size was modest, 113 participants with a predominant female population (74/113, 65.5%) from Spain. This limited overall statistical power and severely limits sex‐stratified analyses, particularly for men, where associations between ultrasound‐derived VAT measures and metabolic outcomes were non‐significant. These findings likely reflect both the smaller male sample and potential sex‐specific differences in fat distribution and metabolic risk. Therefore, our results should be interpreted with caution regarding their applicability to men and to adults with abdominal obesity from other countries. Moreover, considering the mean age of women (53 ± 11.6 years), information on hormonal changes in women, such as peri‐menopause and post‐menopause, would be relevant to fully interpret the clinical significance of the observed US‐measured VAT and VAT/SAT ratio values. The absence of these data represents a limitation, as hormonal variations may partly influence fat distribution in women. Future studies, including larger and more balanced samples, are warranted to confirm these findings and to determine whether sex‐specific thresholds should be established.
Nonetheless, our findings support and provide new evidence to the results already presented by Cuatrecasas et al. 5 and Philipsen et al., 36 suggesting that US could be a practical tool for early detection of metabolic conditions, simplifying risk assessment and reducing healthcare costs. To the best of our knowledge, this is the first study to suggest cut‐off points specifically for VAT/SAT ratio (thickness) measured by US in adults with abdominal obesity, highlighting the potential of this measure as a non‐invasive tool for early risk stratification.
From a clinical perspective, these thresholds may help clinicians to identify individuals at higher metabolic risk, even in primary care or nutrition settings, using an accessible, low‐cost, and radiation‐free technique. Incorporating these measures into routine assessment could facilitate early preventive interventions in patients at risk of prediabetes or metabolic syndrome, particularly in women. Nevertheless, further research is needed to confirm these results and to establish sex‐specific cut‐off points for VAT thickness and VAT/SAT ratio (thickness) measured by US to be implemented in guidelines and clinical settings.
5. CONCLUSION
In conclusion, the present study provides new evidence supporting the use of abdominal US for evaluating abdominal fat distribution, particularly VAT thickness and VAT/SAT ratio (thickness).
VAT thickness and VAT/SAT ratio (thickness) measured by US demonstrated a significant positive association with metabolic conditions, particularly metabolic syndrome and prediabetes, especially in women.
These findings highlight the potential of abdominal US as a non‐invasive and accessible technique for the early detection of individuals at risk of metabolic conditions. While the proposed US thresholds are promising, they should be considered exploratory and interpreted within the context of adults with abdominal obesity; whereas validation in independent cohorts, including individuals with normal waist circumference and larger male samples, is needed to provide more information to be translated to clinical implementation. Overall, the applicability of US to characterise obesity in routine clinical settings should be considered, as it can serve as a valuable tool to complement and potentially standardise risk assessment, ultimately supporting more timely and personalised preventive strategies.
AUTHOR CONTRIBUTIONS
Conceptualization: C.J.‐t.H., E.L., and A.P. Methodology: C.J.‐t.H., E.L., and A.P. Validation: E.L., A.P., and R.S. Formal analysis: C.J.‐t.H. Investigation: C.J.‐t.H., M.B.‐M., and J.Q. Resources: E.L., R.S., and A.P. Data curation: C.J.‐t.H., A.P., E.L., and R.S. Writing – original draft preparation: C.J.‐t.H., E.L., A.P., and R.S. Writing – review and editing: R.M.V., M.B.‐M., and J.Q. Visualization: C.J.‐t.H., E.L., A.P., R.S., R.M.V., M.B.‐M., and J.Q. Supervision: E.L., R.S., and A.P. All authors have read and agreed to the published version of this manuscript.
FUNDING INFORMATION
This research has been funded by Biopolis S.L. and Chic‐kles Gum S.L.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
Supporting information
Figure S1. Bland–Altman analysis of (a) VAT/SAT ratio (thickness) and (b) WC for women. The Y‐axis represents de difference between WC from MRI and US. In figure (a) the X axis represents the mean VAT/SAT ratio (thickness) from the two methods. In figure (b) the X axis represents the mean WC from the two methods. VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; WC, waist circumference; MRI, magnetic resonance imaging; US, ultrasound.
Figure S2. ROC analysis for the accuracy diagnosis of metabolic syndrome of VAT/SAT ratio by US and MRI. US, ultrasound; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; AUC, area under the curve, CI, confidence interval.
Figure S3. ROC analysis for the accuracy diagnosis of prediabetes of VAT/SAT ratio by US and MRI. US, ultrasound; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; AUC, area under the curve, CI, confidence interval.
Table S1. Spearman correlation coefficients between results from the US and MRI techniques.
Table S2. Spearman correlation coefficients between adipose tissue, anthropometric and cardiovascular parameters.
Table S3. Linear regression between adipose tissue and anthropometric and cardiovascular parameters.
Table S4. Frequencies of metabolic syndrome and prediabetes according to Youden's cut‐off points.
ACKNOWLEDGEMENTS
The authors would like to thank the Centre MQ Reus and the Centre ALOMAR for their support. The NFOC‐Salut group is a consolidated research group of Generalitat de Catalunya, Spain (2021 SGR 00817).
Jiménez‐ten Hoevel C, Besora‐Moreno M, Queral J, et al. Ultrasound and MRI abdominal fat distribution and its associations with metabolic conditions in adults with abdominal obesity. Diabetes Obes Metab. 2026;28(3):2061‐2074. doi: 10.1111/dom.70390
Contributor Information
Elisabet Llauradó, Email: elisabet.llaurado@urv.cat.
Rosa Solà, Email: rosa.sola@urv.cat.
DATA AVAILABILITY STATEMENT
The data presented in this study are available upon request from the corresponding author due to privacy reasons.
REFERENCES
- 1. Obesity and Overweight. Accessed February 1, 2024. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight
- 2. Fahed G, Aoun L, Zerdan MB, et al. Metabolic syndrome: updates on pathophysiology and management in 2021. Int J Mol Sci. 2022;23(2):786. doi: 10.3390/IJMS23020786 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. ElSayed NA, McCoy RG, Aleppo G, et al. Diagnosis and classification of diabetes: standards of care in diabetes‐2025. Diabetes Care. 2025;48(Suppl 1):S27‐S49. doi: 10.2337/DC25-S002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Swainson MG, Batterham AM, Tsakirides C, Rutherford ZH, Hind K. Prediction of whole‐body fat percentage and visceral adipose tissue mass from five anthropometric variables. PLoS One. 2017;12(5):e0177175. doi: 10.1371/JOURNAL.PONE.0177175 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Cuatrecasas G, de Cabo F, Coves MJ, et al. Ultrasound measures of abdominal fat layers correlate with metabolic syndrome features in patients with obesity. Obes Sci Pract. 2020;6(6):660‐667. doi: 10.1002/OSP4.453 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Piché ME, Tchernof A, Després JP. Obesity phenotypes, diabetes, and cardiovascular diseases. Circ Res. 2020;126(11):1477‐1500. doi: 10.1161/CIRCRESAHA.120.316101 [DOI] [PubMed] [Google Scholar]
- 7. Khawaja T, Nied M, Wilgor A, Neeland IJ. Impact of visceral and hepatic fat on cardiometabolic health. Curr Cardiol Rep. 2024;26(11):1297‐1307. doi: 10.1007/S11886-024-02127-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lee MJ, Wu Y, Fried SK. Adipose tissue heterogeneity: implication of depot differences in adipose tissue for obesity complications. Mol Aspects Med. 2012;34(1):1‐11. doi: 10.1016/J.MAM.2012.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Abraham TM, Pedley A, Massaro JM, Hoffmann U, Fox CS. Association between visceral and subcutaneous adipose depots and incident cardiovascular disease risk factors. Circulation. 2015;132(17):1639‐1647. doi: 10.1161/CIRCULATIONAHA.114.015000 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Tchernof A, Després JP. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93(1):359‐404. doi: 10.1152/PHYSREV.00033.2011 [DOI] [PubMed] [Google Scholar]
- 11. Sezer S, Demirci S. Preperitoneal fat thickness as a potential noninvasive marker for metabolic‐associated liver disease: a retrospective study. Medicine. 2025;104(30):e43584. doi: 10.1097/MD.0000000000043584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Manzotti C, Colombo F, Zurleni T, Danelli P, Maconi G. Prognostic role of intestinal ultrasound in Crohn's disease. World J Gastroenterol. 2023;29(23):3595‐3605. doi: 10.3748/WJG.V29.I23.3595 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Hua Y, Li MH, Lou YX, et al. Perirenal fat as a potential marker and therapeutic target for metabolic syndrome: insights from a multicenter randomized controlled trial. Front Endocrinol (Lausanne). 2025;16:1557701. doi: 10.3389/FENDO.2025.1557701/FULL [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Brand T, Van Den Munckhof ICL, Van Der Graaf M, et al. Superficial vs deep subcutaneous adipose tissue: sex‐specific associations with hepatic steatosis and metabolic traits. J Clin Endocrinol Metab. 2021;106(10):E3881‐E3889. doi: 10.1210/CLINEM/DGAB426 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. van den Munckhof ICL, Bahrar H, Schraa K, et al. Sex‐specific association of visceral and subcutaneous adipose tissue volumes with systemic inflammation and innate immune cells in people living with obesity. Int J Obes (Lond). 2024;48(4):523‐532. doi: 10.1038/S41366-023-01444-9 [DOI] [PubMed] [Google Scholar]
- 16. Cuatrecasas G, De Cabo F, Coves MJ, et al. Dapagliflozin added to metformin reduces perirenal fat layer in type 2 diabetic patients with obesity. Sci Rep. 2024;14(1):10832. doi: 10.1038/S41598-024-61590-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Emamat H, Jamshidi A, Farhadi A, Ghalandari H, Ghasemi M, Tangestani H. The association between the visceral to subcutaneous abdominal fat ratio and the risk of cardiovascular diseases: a systematic review. BMC Public Health. 2024;24(1):1827. doi: 10.1186/S12889-024-19358-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zieff G, Cornwall J, Blue MN, Smith‐Ryan AE, Stoner L. Ultrasound‐based measurement of central adiposity: key considerations and guidelines. Obes Rev. 2024;25(5):e13716. doi: 10.1111/obr.13716 [DOI] [PubMed] [Google Scholar]
- 19. Marín Baselga R, Teigell‐Muñoz FJ, Porcel JM, Ramos Lázaro J, García Rubio S. Ultrasound for body composition assessment: a narrative review. Intern Emerg Med. 2024;20:23‐34. doi: 10.1007/S11739-024-03756-8 [DOI] [PubMed] [Google Scholar]
- 20. Hoffmann J, Thiele J, Kwast S, et al. A new approach to quantify visceral fat via bioelectrical impedance analysis and ultrasound compared to MRI. Int J Obes (Lond). 2024;48(2):209‐217. doi: 10.1038/S41366-023-01400-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Stoner L, Chinn V, Cornwall J, et al. Reliability tests and guidelines for B‐mode ultrasound assessment of central adiposity. Eur J Clin Invest. 2015;45(11):1200‐1208. doi: 10.1111/ECI.12540 [DOI] [PubMed] [Google Scholar]
- 22. Schaudinn A, Linder N, Garnov N, et al. Predictive accuracy of single‐ and multi‐slice MRI for the estimation of total visceral adipose tissue in overweight to severely obese patients. NMR Biomed. 2015;28(5):583–590. doi:10.1002/nbm.3286 [DOI] [PubMed] [Google Scholar]
- 23. Seyedhoseinpour A, Barzin M, Mahdavi M, et al. BMI category‐specific waist circumference thresholds based on cardiovascular disease outcomes and all‐cause mortality: Tehran lipid and glucose study (TLGS). BMC Public Health. 2023;23(1):1297. doi: 10.1186/S12889-023-16190-W [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Krist AH, Davidson KW, Mangione CM, et al. Screening for hypertension in adults: US preventive services task force reaffirmation recommendation statement. JAMA. 2021;325(16):1650‐1656. doi: 10.1001/JAMA.2021.4987 [DOI] [PubMed] [Google Scholar]
- 25. Alberti KGMM, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640‐1645. doi: 10.1161/CIRCULATIONAHA.109.192644 [DOI] [PubMed] [Google Scholar]
- 26. Harris M, Taylor G. Medical statistics made easy. Ulster Med J. 2015;84(1):58. [Google Scholar]
- 27. Nahm FS. Receiver operating characteristic curve: overview and practical use for clinicians. Korean J Anesthesiol. 2022;75(1):25‐36. doi: 10.4097/KJA.21209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. De Lucia Rolfe E, Sleigh A, Finucane FM, et al. Ultrasound measurements of visceral and subcutaneous abdominal thickness to predict abdominal adiposity among older men and women. Obesity. 2010;18(3):625‐631. doi: 10.1038/OBY.2009.309 [DOI] [PubMed] [Google Scholar]
- 29. Stolk RP, Wink O, Zelissen PMJ, Meijer R, Van Gils APG, Grobbee DE. Validity and reproducibility of ultrasonography for the measurement of intra‐abdominal adipose tissue. Int J Obes Relat Metab Disord. 2001;25(9):1346‐1351. doi: 10.1038/SJ.IJO.0801734 [DOI] [PubMed] [Google Scholar]
- 30. Kammerlander AA, Lyass A, Mahoney TF, et al. Sex differences in the associations of visceral adipose tissue and cardiometabolic and cardiovascular disease risk: the Framingham Heart Study. J Am Heart Assoc. 2021;10(11):e019968. doi: 10.1161/JAHA.120.019968 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kanellakis S, Skoufas E, Simitsopoulou E, et al. Changes in body weight and body composition during the menstrual cycle. Am J Hum Biol. 2023;35(11):e23951. doi: 10.1002/AJHB.23951 [DOI] [PubMed] [Google Scholar]
- 32. Banack HR, Bea JW, Chen Z, et al. Longitudinal patterns of abdominal visceral and subcutaneous adipose tissue, total body composition, and anthropometric measures in postmenopausal women: results from the Women's Health Initiative. Int J Obes (Lond). 2023;47(4):288‐296. doi: 10.1038/S41366-023-01266-9 [DOI] [PubMed] [Google Scholar]
- 33. Eastwood SV, Tillin T, Wright A, et al. Estimation of CT‐derived abdominal visceral and subcutaneous adipose tissue depots from anthropometry in Europeans, South Asians and African Caribbeans. PLoS One. 2013;8(9):e75085. doi: 10.1371/JOURNAL.PONE.0075085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Alvarez‐Alvarez I, de Rojas JP, Fernandez‐Montero A, et al. Strong inverse associations of Mediterranean diet, physical activity and their combination with cardiovascular disease: the Seguimiento Universidad de Navarra (SUN) cohort. Eur J Prev Cardiol. 2018;25(11):1186‐1197. doi: 10.1177/2047487318783263 [DOI] [PubMed] [Google Scholar]
- 35. Papadaki A, Nolen‐Doerr E, Mantzoros CS. The effect of the Mediterranean diet on metabolic health: a systematic review and meta‐analysis of controlled trials in adults. Nutrients. 2020;12(11):1‐21. doi: 10.3390/NU12113342 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Philipsen A, Jørgensen ME, Vistisen D, et al. Associations between ultrasound measures of abdominal fat distribution and indices of glucose metabolism in a population at high risk of type 2 diabetes: the ADDITION‐PRO study. PLoS One. 2015;10(4):e0123062. doi: 10.1371/JOURNAL.PONE.0123062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Cuatrecasas G, Calbo M, Rossell O, et al. Effect of liraglutide in different abdominal fat layers measured by ultrasound: the importance of perirenal fat reduction. Obes Facts. 2024;17(4):347‐354. doi: 10.1159/000538996 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Chong KS, Chang YH, Yang CT, Chou CK, Ou H, Kuo S. Longitudinal economic burden of incident complications among metabolic syndrome populations. Cardiovasc Diabetol. 2024;23(1):246. doi: 10.1186/S12933-024-02335-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Li S, Wen CP, Tu H, et al. Metabolic syndrome including both elevated blood pressure and elevated fasting plasma glucose is associated with higher mortality risk: a prospective study. Diabetol Metab Syndr. 2025;17(1):72. doi: 10.1186/S13098-025-01628-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Kim Y, Kuk JL, Park KH, Lee SJ. Influence of waist circumference measurement site on the cardiometabolic risk factors in Korean adults. Obes Res Clin Pract. 2025;19(1):28‐33. doi: 10.1016/j.orcp.2025.01.007 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Bland–Altman analysis of (a) VAT/SAT ratio (thickness) and (b) WC for women. The Y‐axis represents de difference between WC from MRI and US. In figure (a) the X axis represents the mean VAT/SAT ratio (thickness) from the two methods. In figure (b) the X axis represents the mean WC from the two methods. VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; WC, waist circumference; MRI, magnetic resonance imaging; US, ultrasound.
Figure S2. ROC analysis for the accuracy diagnosis of metabolic syndrome of VAT/SAT ratio by US and MRI. US, ultrasound; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; AUC, area under the curve, CI, confidence interval.
Figure S3. ROC analysis for the accuracy diagnosis of prediabetes of VAT/SAT ratio by US and MRI. US, ultrasound; MRI, magnetic resonance imaging; VAT, visceral adipose tissue; AUC, area under the curve, CI, confidence interval.
Table S1. Spearman correlation coefficients between results from the US and MRI techniques.
Table S2. Spearman correlation coefficients between adipose tissue, anthropometric and cardiovascular parameters.
Table S3. Linear regression between adipose tissue and anthropometric and cardiovascular parameters.
Table S4. Frequencies of metabolic syndrome and prediabetes according to Youden's cut‐off points.
Data Availability Statement
The data presented in this study are available upon request from the corresponding author due to privacy reasons.