Abstract
Objective
This study aimed to assess the relationship between abdominal fat distribution (AFD) and urinary albumin/creatinine ratio (UACR) in Chinese adults.
Methods
823 patients with type 2 diabetes mellitus(T2DM) were selected. Abdominal visceral fat area (VFA) and subcutaneous fat area (SFA) were measured using bioelectrical impedance analysis (BIA). Patients were divided into four groups: low VFA /low SFA, low VFA /high SFA, high VFA /low SFA, and high VFA /high SFA based on the median values (low: < median, high: ≥median).
Results
In the multifactor analysis after adjusting for relevant factors, VFA, waist-to-height ratio (WHtR), waist-hip ratio (WHR), and waist circumference (WC) showed significant positive correlations with UACR, while SFA and BMI did not. AFD combinations were independent predictors of UACR. The risk of UACR30-300 mg/g was highest in the high VFA/low SFA group (OR = 3.12), and for UACR > 300 mg/g, it was highest in the high VFA/high SFA group (OR = 24.69). The areas under the receiver operating characteristic (ROC) curvefor VFA prediction of UACR ≥ 30 mg/g was 0.69, significantly greater than that for WHtR, WHR, and WC. Optimal cut-off values were 98.8 cm² for VFA. When analyzed by gender, the optimal cut-off values for VFA were 98.8 cm² for males and 102.3 cm² for females.
Conclusion
Central obesity indicators (VFA, WHtR, WHR, WC) were associated with UACR. VFA was the strongest predictor for UACR ≥ 30 mg/g.
Supplementary Information
The online version contains supplementary material available at 10.1186/s13098-025-01814-5.
Keywords: Type 2 diabetes mellitus, Abdominal fat distribution, Visceral fat area, Subcutaneous fat area, Urinary albumin/Creatinine ratio
Introduction
The global prevalence of T2DM is increasing every year, accompanied by an increase in diabetes-related complications [1]. Diabetic kidney disease (DKD) is the most common microvascular complication of diabetes. It is usually diagnosed clinically based on an increase in the random urinary albumin/creatinine ratio (UACR) of ≥ 30 mg/g, which can be retested once in 3–6 months and still reach the critical value, while excluding other influencing factors [2]. Increased UACR is significantly associated with all-cause mortality [3, 4]. Therefore, early identification of risk factors for UACR and active management is essential.
Among the many risk factors, obesity is a significant health problem in many countries due to its high prevalence and related metabolic disorders [5, 6]. The severity of obesity is usually defined clinically by body mass index (BMI). However, BMI only represents overall obesity and has certain limitations. Waist Circumference (WC), waist-hip ratio (WHR), and waist-to-height ratio (WHtR) are better predictors of metabolic disease than BMI as indicators for assessing central obesity, and they are simple to measure. The study found that normal-weight obese people (normal BMI but high WHR) had a much higher risk of developing chronic kidney disease than healthy overweight people (low WHR but high BMI) [7]. Nevertheless, these indicators do not differentiate between abdominal visceral fat and subcutaneous fat. Thus, the abdominal visceral fat area (VFA) and subcutaneous fat area (SFA) are more critical in distinguishing abdominal fat distribution (AFD), as AFD is a stronger predictor of metabolic disease and all-cause mortality risk than general obesity [8, 9]. Studies have found that increased VFA is significantly associated with insulin resistance and the risk of metabolic complications of obesity, while the risk of SFA is lower [10]. Some studies have even pointed to the “obesity paradox” that increased SFA may have a protective effect on DKD, metabolic syndrome, and cardiovascular disease [11–13]. Most studies have shown a significant positive association between VFA and UACR, but the relationship between SFA and UACR is still debatable. Therefore, this study aimed to investigate the relationship between AFD and UACR, predict the optimal cut-off for the development of UACR ≥ 30 mg/g for anthropometric indicators, and provide evidence for the clinical prevention and delay of UACR.
Materials and methods
Study population
A cross-sectional analysis was conducted involving 823 patients with type 2 diabetes mellitus (T2DM) who were hospitalized in the endocrinology department of Xi’an Ninth Hospital from June 2018 to August 2021. Inclusion criteria: [1] Patients must meet the 1999 World Health Organization diagnostic criteria for type 2 diabetes mellitus (T2DM); [2] Age between 20 and 75 years. Exclusion criteria: [1] Patients with type 1 diabetes, ketoacidosis, and hyperosmolar coma; [2] Patients with infectious diseases, immune disease-related nephropathy, malignant tumors, and severe liver and kidney dysfunction; [3] Individuals who have recently taken nephrotoxic drugs or other medications known to affect UACR; [4] History of glomerular and adrenal related diseases; [5] Individuals with acute or chronic urinary tract infections; [6] Mentally impaired, pregnant or lactating women.
All subjects provided written informed consent and were approved by the Ethics Committee of Xi’an Ninth Hospital (IRB number: 202275).
Data collection and measurements
General patient clinical information was collected by trained and certified medical professionals [1]. General information: gender, age, duration of diabetes, family history of diabetes, history of smoking (current/ever smoking), and history of hypertension (current/ever systolic blood pressure ≥ 140 mmHg and (or) diastolic blood pressure ≥ 90 mmHg) [2]. Anthropometric indicators: Measurements were performed according to World Health Organization recommended methods [14]. Patients should be fasted for at least 8 h, after urination and defecation, barefoot, and in thin clothes. Height Measurement: Patients maintained an upright posture with their heels, sacrum, and both scapulae against the height meter; Weight Measurement: Patients stood naturally on the scale; Waist Circumference (WC): the patient stands with both upper limbs naturally lowered, the feet are shoulder-width apart, and an inelastic soft ruler was placed flat against the skin at the level of 1/2 of the line between the lowest rib margin and the iliac crest; Hip circumference (HC): circumference of the most protruding hip backward. All result readings were accurate to 0.1 cm or 0.1 kg. BMI was calculated as Weight (kg)/Height squared (m²), WHR as WC (cm)/HC (cm), and WHtR as WC (cm)/Height (cm) [3]. Laboratory measurement indicators: the patient should have blood collected from a vein in the morning after fasting for at least 8 h. Glycated hemoglobin A1c (HbA1c) measurement using a glycated hemoglobin meter (Norway, Nyco Card ReaderI); Detection of total cholesterol (TC), triglycerides (TG), low-density lipoprotein-cholesterol (LDL-c), high-density lipoprotein-cholesterol (HDL-c), uric acid (UA), blood urea nitrogen (BUN) by fully automatic biochemical instrument (IDEXX, Catalyst One). Patients fasted for at least 8 h and morning urine specimens were obtained. Urine albumin and urinary creatinine concentrations were measured using a protein scattering turbidimetry and colorimetric method, respectively (Hitachi Automatic Analyzer 7600). The UACR was calculated as Urinary Albumin / Creatinine. UACR < 30 mg/g is considered normal albuminuria, microalbuminuria ranges from 30 to 300 mg/g, and macroalbuminuria > 300 mg/g [4]. Determination of VFA and SFA by BIA: The VFA and SFA of patients at umbilical levels were measured by BIA according to the DUALSCAN HDS-2000 (Omron Healthcare Co. Ltd, Kyoto, Japan) instrument manual. Compared with Computed Tomography(CT), BIA has the advantages of no radiation, simplicity, and low cost [15, 16]. The working principle of BIA is to use the differences in the resistance of various tissues and organs of the human body, by applying a low-frequency current or voltage to the body, and then measuring the changes in current and voltage, so as to calculate the resistance of the human body, due to the differences in the resistance of fat and muscle (fat with little moisture content, electrolyte conductivity is poor, resistance is high; muscle with high moisture content, electrolyte conductivity is good, resistance is low).These differences in electrical conductivity are used to determine body composition and estimate the area of abdominal fat.
Grouping
Based on the values of VFA and SFA measured by BIA, median of VFA and SFA were calculated, respectively. Participants were then divided into four groups (low VFA/low SFA, low VFA/high SFA, high VFA/low SFA, high VFA/high SFA group) according to the relationship with the median (low: < median, high: ≥median).
Statistical analysis
Statistical analysis was performed using the SPSS statistical software (version 27, IBM Corporation, Armonk, NY, USA). Quantitative data obeying normal distribution were expressed as mean ± standard deviation, and ANOVA was used for the comparisons between groups. Quantitative data obeying skewed were expressed as median (first quartile, third quartile), Kruskal-Wallis H test was performed to compare between-group differences. Qualitative data were expressed as percentages and analyzed using the chi-square test. Single-factor linear regression and ridge regression were used to analyze the relationship between each index and UACR. Multivariable logistic regression models were used to assess the predictors of AFD for UACR30-300 mg/g and UACR > 300 mg/g. The capacity of anthropometric indices to predict UACR ≥ 30 mg/g was evaluated by plotting the receiver operating characteristic (ROC) curve and calculating the area under the curve (AUC). The optimal anthropometric cut-offs were calculated using the Youden Index. A two-sided test with P < 0.05 indicates a significant difference.
Results
Baseline characteristics of the study population
Table 1 shows the baseline characteristics based on the four groups. There were differences in the duration of diabetes, HbA1c, UA, TG, TC, HDL-c, LDL-c, UACR, gender, history of smoking and hypertension, BMI, WC, WHR, WHtR, VFA, and SFA between the four groups. However, there were no significant differences in age, BUN, or family history of diabetes.
Table 1.
Patient baseline characteristics
| Parameters | All (n = 823) | Low VFA/Low SFA (n = 303) | Low VFA/High SFA (n = 108) | High VFA/Low SFA (n = 108) | High VFA/High SFA (n = 304) | P |
|---|---|---|---|---|---|---|
| Age (years) | 59(52, 64) | 61(52, 65) | 60(53, 63) | 58.5(52, 63) | 58(51, 64) | 0.054 |
| Duration of diabetes (year) | 6(2, 12) | 7(2, 13) | 6(1, 13) | 7.5(3, 13) | 5(1.25, 10) | 0.033* |
| HbA1c (%) | 7.8(6.8, 9.4) | 7.6(6.6, 9.3) | 7.6(6.53, 9.18) | 8.2(6.93, 9.9) | 8(7, 9.5) | 0.028* |
| UA (umol/L) | 318(273, 377) | 298(257, 356) | 324(265.75, 375.5) | 312.5(280.25, 371.25) | 340(279, 409.5) a | 0.000* |
| BUN (mmol/L) | 4.85(4.05, 5.64) | 4.89(4, 5.57) | 4.79(3.96, 5.69) | 4.75(4.07, 5.57) | 4.88(4.11, 5.76) | 0.672 |
| TG (mmol/L) | 1.54(1.07, 2.29) | 1.24(0.88, 1.72) c | 1.43(1.05, 1.87) | 1.595(1.14, 2.3) a | 1.88(1.35, 2.98) a b | 0.000* |
| TC (mmol/L) | 4.59(3.89, 5.35) | 4.41(3.7, 5.08) b | 4.83(4.06, 5.6) a | 4.62(3.99, 5.43) | 4.69(4.01, 5.41) a | 0.001* |
| HDL-c (mmol/L) | 1.09(0.95, 1.27) | 1.16(0.99, 1.37) c* | 1.16(1, 1.38) c* | 1.07(0.91, 1.23) a* b* | 1.04(0.89, 1.78) a b | 0.000* |
| LDL-c (mmol/L) | 2.67 ± 0.86 | 2.53 ± 0.83 b | 2.87 ± 0.91 a | 2.71 ± 0.92 | 2.73 ± 0.81 a* | 0.000* |
| Male (%) | 567(69) | 202(67)b c | 42(39)a c | 88(81)a b | 235(77)a b | 0.000* |
| History of smoking (%) | 376(47) | 113(37) c | 33(31) c | 60(56) a b | 170(56) a b | 0.000* |
| History of hypertension (%) | 381(46) | 105(35) c | 43(40) | 56(52) a | 177(58) a b | 0.000* |
| Family history of diabetes (%) | 244(30) | 86(28) | 29(27) | 29(27) | 100(33) | 0.462 |
| BMI (kg/m2) | 25.39(23.7, 27.4) | 23.4(22.1, 24.5) b c | 25.45(24.4, 26.9) a | 25(24, 26.39) a | 27.95(26.41, 29.88) a b c | 0.000* |
| WC (cm) | 92.59 ± 9.31 | 85.82 ± 6.65 b c | 91.12 ± 7.37 a | 91.5 ± 4.77 a | 100.25 ± 7.61 a b c | 0.000* |
| WHR | 0.93(0.89, 0.97) | 0.9(0.87, 0.93) c | 0.92(0.87, 0.95) c | 0.94(0.91, 0.96) a b | 0.96(0.93, 0.99) a b c | 0.000* |
| WHtR | 0.55(0.52, 0.59) | 0.52(0.49, 0.54) b c | 0.55(0.53, 0.58) a | 0.55(0.53, 0.58) a | 0.59(0.57, 0.62) a b c | 0.000* |
| VFA (cm2) | 105.9(82.6,128.4) | 79.4(67.8, 91.9) b* c | 93.95(78.3, 101.95) a* c | 116.5(112, 125.23) a b | 136.85(119.73, 153) a b c | 0.000* |
| SFA (cm2) | 184.6(151, 225.4) | 145(122.9, 164) b c | 207(194.55, 229.93) a c | 164.55(149.5, 176.53) a b | 233.85(209.23, 268.63) a b* c | 0.000* |
| UACR (mg/g) | 15.8 (7.3, 41) | 10.4 (5.2, 23.1) c | 12 (5, 26.5) c | 28 (15.5, 73.2) a b | 21.7 (9.7, 64.6) a b | 0.000* |
Data are presented as mean ± SD, medians (interquartile ranges), or n (%).
*: Statistically significant differences. a P < 0.01 a* P < 0.05: vs. Low VFA/Low SFA group; b P < 0.01 b* P < 0.05: vs. Low VFA/High SFA group; c P < 0.01 c*P < 0.05: vs. High VFA/Low SFA group.
HbA1c, Glycated hemoglobin A1c; UA, uric acid; BUN, blood urea nitrogen; TG, triglycerides; TC, total cholesterol; LDL-c, low-density lipoprotein-cholesterol; HDL-c, high-density lipoprotein-cholesterol; BMI, body mass index; WC, waist circumference; WHR, waist-hip ratio; WHtR, waist-to-height ratio; VFA, visceral fat area; SFA, subcutaneous fat area; UACR, urinary albumin/creatinine ratio.
Linear and ridge regression analysis of UACR
Univariate linear regression showed that both VFA (β-coefficients = 0.20, P = 0.000) and SFA (β-coefficients = 0.11, P = 0.002) were significantly positively associated with UACR (Table 2). In contrast, the correlation between VFA and UACR was stronger. Furthermore, gender, duration of diabetes, history of smoking, history of hypertension, BMI, WC, WHR, WHtR, HbA1c, UA, BUN, TG, TC, and LDL-c were significantly positively associated with UACR. HDL-c was negatively associated with UACR. However, age and family history of diabetes were insignificant with UACR.
Table 2.
Linear and ridge regression analysis of UACR
| Parameters | Linear regression analysis | Ridge regression analysis | ||
|---|---|---|---|---|
| β-coefficients | P | β-coefficients | P | |
| Male | 0.10 | 0.005* | 0.01 | 0.308 |
| Age(years) | 0.00 | 0.906 | 0.00 | 0.589 |
| BMI (kg/m2) | 0.16 | 0.000* | 0.02 | 0.097 |
| WC (cm) | 0.19 | 0.000* | 0.02 | 0.035* |
| WHR | 0.20 | 0.000* | 0.05 | 0.001* |
| WHtR | 0.18 | 0.000* | 0.03 | 0.018* |
| VFA (cm2) | 0.20 | 0.000* | 0.04 | 0.001* |
| SFA (cm2) | 0.11 | 0.002* | -0.01 | 0.611 |
| HbA1c (%) | 0.08 | 0.025* | 0.03 | 0.099 |
| UA (umol/L) | 0.09 | 0.011* | 0.01 | 0.592 |
| BUN (mmol/L) | 0.16 | 0.000* | 0.07 | 0.000* |
| TG (mmol/L) | 0.17 | 0.000* | 0.06 | 0.000 |
| TC (mmol/L) | 0.09 | 0.013* | 0.03 | 0.039* |
| HDL-c(mmol/L) | -0.12 | 0.001* | -0.03 | 0.058 |
| LDL-c (mmol/L) | 0.08 | 0.018* | 0.04 | 0.015* |
| Duration of diabetes (year) | 0.11 | 0.001* | 0.07 | 0.000* |
| History of smoking | 0.10 | 0.005* | 0.02 | 0.315 |
| History of hypertension | 0.14 | 0.000* | 0.05 | 0.003* |
| Family history of diabetes | 0.04 | 0.215 | ||
| History of taking glucose-lowering drugs that affect UACR | 0.00 | 0.981 | ||
*: Statistically significant differences
HbA1c, Glycated hemoglobin A1c; UA, uric acid; BUN, blood urea nitrogen; TG, triglycerides; TC, total cholesterol; LDL-c, low-density lipoprotein-cholesterol; HDL-c, high-density lipoprotein-cholesterol; BMI, body mass index; WC, waist circumference; WHR, waist-hip ratio; WHtR, waist-to-height ratio; VFA, visceral fat area; SFA, subcutaneous fat area; UACR, urinary albumin/creatinine ratio
The variance inflation factor (VIF) for multiple linear regression revealed the presence of multicollinearity among BMI, WC, and WHR variables. Therefore, ridge regression was used to solve the covariance problem. It was concluded that VFA, WHtR, WHR, WC, BUN, TG, TC, LDL, duration of diabetes, and history of hypertension were still significantly and positively correlated with UACR. However, gender, BMI, SFA, history of smoking, HbA1c, UA, HDL, and UACR are no longer relevant.
Association between AFD combinations and UACR
We analyzed the predictive value of AFD combinations for UACR30-300 mg/g (Table 3 Model 1). The multivariate analysis adjusted for gender, age, duration of diabetes, BUN, TG, TC, LDL-c, and history of hypertension. AFD combinations were standalone predictors of UACR30-300 mg/g. In particular, patients in the high VFA/low SFA group had the highest risk of UACR30-300 mg/g (OR 3.12, 95% CI: 1.90–5.14, P < 0.01), followed by the high VFA/high SFA group (OR 2.19, 95% CI: 1.46–3.30, P < 0.01), using the low VFA/low SFA group as the reference category.
Table 3.
Association between AFD combinations and UACR30-300 mg/g (Model 1) and uacr > 300 mg/g (Model 2)
| Parameters | Model 1 | Model 2 | ||
|---|---|---|---|---|
| OR (95%CI) | P | OR (95%CI) | P | |
| Low VFA/Low SFA | 1 [reference] | 1 [reference] | ||
| Low VFA/High SFA | 1.10(0.62, 1.97) | 0.741 | 6.49(0.53, 79.21) | 0.143 |
| High VFA/Low SFA | 3.12(1.90, 5.14) | 0.000* | 12.15(1.30, 113.87) | 0.029* |
| High VFA/High SFA | 2.19(1.46, 3.30) | 0.000* | 24.69(2.99, 204.30) | 0.003* |
*: Statistically significant differences. Model 1 and Model 2 were adjusted for gender, age, duration of diabetes, BUN, TG, TC, LDL-c, and history of hypertension, the predictive value of combinations of AFD for UACR 30–300 mg/g and UACR ≥ 300 mg/g were analyzed separately
OR, odds ratios; CI, confidence intervals
VFA, visceral fat area; SFA, subcutaneous fat area
Further analyses the predictive value of AFD combinations for UACR > 300 mg/g (Table 3 Model 2). AFD combinations were also independent predictors of UACR > 300 mg/g. Differently, the risk of UACR > 300 mg/g was higher in the high VFA/high SFA group (OR 24.69, 95% CI: 2.99–204.30, P < 0.01) than in the high VFA/low SFA group (OR 12.15, 95% CI: 1.30-113.87, P < 0.05), using the low VFA/low SFA group as the reference category. Compared to the low VFA/low SFA group, the low VFA/high SFA group predicted an increased risk of UACR 30–300 mg/g (OR 1.10, 95% CI: 0.62–1.97, P = 0.741) and UACR > 300 mg/g (OR 6.49, 95% CI: 0.53–79.21, P = 0.143), although there were no statistical differences. To further investigate the impact of some key factors (including duration of diabetes and whether taking glucose-lowering drugs affecting UACR) on the association between VFA and UACR, we conducted stratified analysis. The results were summarized in Supplementary Table S1 and S2. The effect direction of VFA on UACR is consistent across different diabetes duration groups; however, statistical significance was observed only in the group with a longer disease duration (≥ 6years). On the other hand, the direction of the association between VFA and UACR was consistent across groups of drugs taking. However, statistical significance was observed only in patients who were not taking such medications. These results may be attributed to differences in statistical power after stratification, or to the potential effects of these key factors. The current sample size of the present study is insufficient to draw a definitive conclusion.
ROC curves and optimal cut-offs for anthropometric indicators to predict uacr ≥ 30 mg/g
The ability to predict UACR ≥ 30 mg/g was assessed using ROC curve analysis of anthropometric indicators (Fig. 1). Overall, the AUC rankings were as follows: VFA: 0.69 (95% CI: 0.66–0.73), WHR: 0.64 (95% CI: 0.60–0.68), WC: 0.64 (95% CI: 0.60–0.68), and WHtR: 0.63(95% CI: 0.59–0.67). In males, the AUC rankings were: VFA: 0.66 (95% CI: 0.61–0.70), WHtR: 0.62 (95% CI: 0.57–0.67), WHR: 0.62 (95% CI: 0.57–0.66), and WC: 0.61 (95% CI: 0.56–0.66). In Females, the AUC rankings were: VFA: 0.76 (95% CI: 0.69–0.82), WHtR: 0.67 (95% CI: 0.60–0.75), WHR: 0.67 (95% CI: 0.60–0.74), and WC: 0.66 (95% CI: 0.58–0.73). The AUC of VFA was highest in both males and females compared to other indicators, suggesting that VFA is a superior predictor of UACR ≥ 30 mg/g compared to WHR, WHtR, and WC. Overall, the optimal cut-off for VFA, WHtR, WHR, and WC were 98.8cm2, 0.59, 0.95, and 92.2 cm. By gender, the optimal cut-off for VFA, WHtR, WHR, and WC were 98.8cm2, 0.56, 0.95, and 92.2 cm in males, and 102.3cm2, 0.52, 0.89, and 82.5 cm in females (Table 4).
Fig. 1.
ROC curves and AUC of VFA, WHR, WHtR, WC to predict UACR ≥ 30 mg/g. (a) Overall (b) Male (c) Female (d) AUC of VFA, WHR, WHtR, WC
Table 4.
Optimal cut-offs and AUC for anthropometric indicators (VFA, WHR, whtr, WC) to predict uacr ≥ 30 mg/g
| Parameters | Optimal cut-off | SE | AUC (95% CI) | Sensitivity | Specificity |
|---|---|---|---|---|---|
| VFA (cm2) | |||||
| Overall | 98.8 | 0.02 | 0.69(0.66–0.73) | 0.81 | 0.50 |
| Male | 98.8 | 0.02 | 0.66(0.61–0.70) | 0.83 | 0.42 |
| Female | 102.3 | 0.03 | 0.76(0.69–0.82) | 0.72 | 0.71 |
| WHtR | |||||
| Overall | 0.59 | 0.02 | 0.63(0.59–0.67) | 0.38 | 0.81 |
| Male | 0.56 | 0.03 | 0.62(0.57–0.67) | 0.60 | 0.58 |
| Female | 0.52 | 0.04 | 0.67(0.60–0.75) | 0.94 | 0.30 |
| WHR | |||||
| Overall | 0.95 | 0.02 | 0.64(0.60–0.68) | 0.55 | 0.66 |
| Male | 0.95 | 0.03 | 0.62(0.57–0.66) | 0.63 | 0.56 |
| Female | 0.89 | 0.04 | 0.67(0.60–0.74) | 0.75 | 0.50 |
| WC (cm) | |||||
| Overall | 92.2 | 0.02 | 0.64(0.60–0.68) | 0.64 | 0.70 |
| Male | 92.2 | 0.03 | 0.61(0.56–0.66) | 0.72 | 0.46 |
| Female | 82.5 | 0.04 | 0.66(0.58–0.73) | 0.92 | 0.33 |
AUC: area under the curve; CI: confidence intervals; SE: standard error
WC, waist circumference; WHR, waist-hip ratio; WHtR, waist-to-height ratio; VFA, visceral fat area
Discussion
DKD is the main contributor to end-stage renal disease (ESRD) and has a high mortality and disability rate, resulting in a significant economic burden for individuals and society [2]. Obesity is a risk known factor for UACR. Regarding the relationship between various anthropometric indicators and UACR, previous studies have found an association between BMI and proteinuria [17]. A study of the relationship between various obesity indicators and the prevalence of chronic kidney disease in 125 overweight T2DM patients found that WHtR was significantly and positively correlated with urinary albumin excretion, while the relationship between WC and urinary albumin excretion was insignificant [18]. The present study found that WC, WHR, and WHtR, which represent central obesity, were significantly and positively correlated with UACR, while BMI, which represents general obesity, was insignificant with UACR. This suggests that abdominal obesity is more likely to lead to proteinuria than generalized obesity in T2DM patients. Moreover, previous studies have demonstrated that Chinese are more likely to develop central obesity and have a higher metabolic risk than whites [19]. Therefore, Chinese patients with T2DM should be more concerned about abdominal fat.
The relationship between AFD and UACR has been controversial for years [10, 11]. In the present study, after correcting for relevant influences, we found that VFA was an independent risk factor for UACR, but SFA was not. These results are consistent with a previous Japanese study demonstrating that microalbuminuria was independently related to increased visceral fat but not to subcutaneous fat [20]. However, this Japanese study did not group abdominal fat and was limited to exploring the association with microalbuminuria. Furthermore, previous studies were usually based on VFA grouping only and did not consider SFA, treating low VFA/low SFA, low VFA/high SFA, high VFA/low SFA, and high VFA/high SFA as only two groups [21]. Therefore, the present study continued to use the median of VFA and SFA, and the subgroup revealed that the AFD combinations were independent risk factors for UACR. In a multivariate logistic regression analysis adjusting for relevant influences, we assessed the risk of UACR occurrence in four groups of patients. The results found that the risk of UACR30-300 mg/g was higher in the high VFA/low SFA group, while the risk of UACR > 300 mg/g was higher in the high VFA/high SFA group.
The mechanism of increased VFA leading to proteinuria is likely due to the large amounts of free fatty acids (FFA), cellular and inflammatory factors directly into the liver through the portal vein [22], which activate the renin-angiotensin system, exacerbating insulin resistance and promoting renal injury [23]. The potential harm of increased SFA on proteinuria may also be related to insulin resistance; however VFA is more closely associated with insulin resistance [24, 25].
A Croatian study on the application of central obesity indicators in predicting microvascular complications in obese T2DM patients found that WHtR was the strongest predictor of CKD, followed by WC, with WHR being the weakest predictor. However, the sample size of this study was small, 125 cases were limited to T2DM patients with BMI ≥ 35 kg/m2, and predictions were made only for the overall population, not by gender; thus, the conclusions may be limited [26]. Previous studies in China have also shown that the optimal cut-off for abdominal obesity indicators (WHtR, WHR, and WC) to predict microalbuminuria were 0.52, 0.88, and 89.6 cm in men, and 0.52, 0.84, and 84.5 cm in women [27]. Both of these studies mentioned central obesity indicators, but neither examined the predictive value of VFA for UACR. In the present study, in addition to central obesity indicators, VFA was included, and the prediction of VFA for UACR ≥ 30 mg/g was better than that of WHR, WHtR, and WC based on AUROC. The optimal cut-off for VFA, WHtR, WHR, and WC were 98.8cm2, 0.56, 0.95, and 92.2 cm in males, and 102.3cm2, 0.52, 0.89, and 82.5 cm in females. The optimal cut-off for predicting proteinuria or DKD vary slightly among studies regarding different obesity indicators, and the main reasons may be related to differences in gender and age. In addition, sample size may also affect the reliability of the results. Therefore, it is clinically recommended that patients above this threshold should control their diet and strengthen their physical activity, which is essential to prevent and delay the progression of UACR.
The present study has several advantages. Firstly, we included indicators related to obesity (VFA, SFA, WC, WHR, and WHtR), explored their correlation with UACR, and derived the optimal cut-off for predicting UACR ≥ 30 mg/g. Secondly, subgroup analysis yielded the AFD combinations as independent predictors of UACR. However, this study also has some limitations. First, this study used BIA to measure VFA and SFA. Although BIA is an alternative to CT for abdominal fat measurement [28], CT is still the gold standard. Although abdominal fat can be reliably measured using BIA and shows a significant correlation with measurements from CT, the absolute values and the differences may be underestimated. Future studies should consider using CT for abdominal fat assessment. Second, this study was a cross-sectional study with a population selected from the same hospital, the results may not be generally representative, and the causal relationship between AFD and UACR cannot be established, which needs to be further confirmed by a multicenter, prospective study.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
We thank all researchers and study participants for their contributions.
Abbreviations
- BIA
Bioelectrical impedance analysis
- BMI
Body mass index
- BUN
Blood urea nitrogen
- CKD
Chronic kidney disease
- CT
Computed Tomography
- DKD
Diabetic kidney disease
- eGFR
estimated glomerular filtration rate
- ESRD
end-stage renal disease
- HbA1c
Glycated hemoglobin A1c
- HC
Hip circumference
- HDL-c
High-density lipoprotein
- LDL-c
Low density lipoprotein
- ROC
receiver operating characteristic
- SFA
Subcutaneous fat area
- T1DM
Type 1 diabetes mellitus
- T2DM
Type 2 diabetes mellitus
- TC
Total cholesterol
- TG
Triglycerides
- UA
Uric acid
- UACR
Urinary albumin/creatinine ratio
- VFA
Visceral fat area
- WC
Waist circumference
- WHR
Waist-hip ratio
- WHtR
Waist-to-height ratio
Author contributions
YN, SW, and TZ: contributed to the study concept and design. TS and YL: participated in the literature search and data collection. TS, YN, and SW: drafted the manuscript. TS, YY and TZ: analysis and interpretation of the data. YN, YY, SW, TZ, YL, MZ, and JS: critically reviewed and edited the manuscript. All authors revised the paper and approved the final manuscript.
Funding
This work was supported by Xi’an Science and Technology Plan Project (2023JH-YXYB-0243).
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
The research adhered to the guidelines of the Helsinki Declaration and was approved by the Ethics Committee of Xi’an Ninth Hospital (IRB number: 202275). Each participant provides written informed consent.
Consent for publication
Not applicable.
Conflict of interest
All the authors declare that there is no conflict of interest.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work we used the Large Language Model (ChatGPT) in order to refine grammar and language. After using this tool, we reviewed and edited the content as needed and take full responsibility for the content of the publication.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Tianlu Shi and Ying Yang contributed equally to this work.
Contributor Information
Tianxiao Zhang, Email: joshuaz@mail.xjtu.edu.cn.
Yu Niu, Email: 15802954724@163.com.
References
- 1.Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the international diabetes federation diabetes atlas, 9(th) edition. Diabetes Res Clin Pract. 2019;157:107843. [DOI] [PubMed] [Google Scholar]
- 2.Nephrology EGoCSo. Chinese guidelines for diagnosis and treatment of diabetic kidney disease. Chin J Nephrol. 2021;37(3):255–304. [Google Scholar]
- 3.Hillege HL, Fidler V, Diercks GF, van Gilst WH, de Zeeuw D, van Veldhuisen DJ, et al. Urinary albumin excretion predicts cardiovascular and noncardiovascular mortality in general population. Circulation. 2002;106(14):1777–82. [DOI] [PubMed] [Google Scholar]
- 4.Greco M, Chiefari E, Mirabelli M, Salatino A, Tocci V, Cianfrone P et al. Plasma or Urine Neutrophil Gelatinase-Associated Lipocalin (NGAL): Which Is Better at Detecting Chronic Kidney Damage in Type 2 Diabetes? 2022;3(2):175– 86.
- 5.Collaborators GBDO, Afshin A, Forouzanfar MH, Reitsma MB, Sur P, Estep K, et al. Health effects of overweight and obesity in 195 countries over 25 years. N Engl J Med. 2017;377(1):13–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.de Vries AP, Ruggenenti P, Ruan XZ, Praga M, Cruzado JM, Bajema IM, et al. Fatty kidney: emerging role of ectopic lipid in obesity-related renal disease. Lancet Diabetes Endocrinol. 2014;2(5):417–26. [DOI] [PubMed] [Google Scholar]
- 7.Song Y, Hwang JA, Shin J, Cho E, Ahn SY, Ko GJ, et al. Waist-hip ratio measured by bioelectrical impedance analysis as a valuable predictor of chronic kidney disease development. BMC Nephrol. 2022;23(1):349. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hu J, Yang S, Zhang A, Yang P, Cao X, Li X, et al. Abdominal obesity is more closely associated with diabetic kidney disease than general obesity. Diabetes Care. 2016;39(10):e179–80. [DOI] [PubMed] [Google Scholar]
- 9.Lee SW, Son JY, Kim JM, Hwang SS, Han JS, Heo NJ. Body fat distribution is more predictive of all-cause mortality than overall adiposity. Diabetes Obes Metab. 2018;20(1):141–7. [DOI] [PubMed] [Google Scholar]
- 10.Lee MJ, Wu Y, Fried SK. Adipose tissue heterogeneity: implication of depot differences in adipose tissue for obesity complications. Mol Aspects Med. 2013;34(1):1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Foster MC, Hwang S-J, Massaro JM, Hoffmann U, DeBoer IH, Robins SJ, et al. Association of subcutaneous and visceral adiposity with albuminuria: the Framingham heart study. Obesity. 2011;19(6):1284–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Matsha TE, Ismail S, Speelman A, Hon GM, Davids S, Erasmus RT, et al. Visceral and subcutaneous adipose tissue association with metabolic syndrome and its components in a South African population. Clin Nutr ESPEN. 2019;32:76–81. [DOI] [PubMed] [Google Scholar]
- 13.Tanaka T, Kishi S, Ninomiya K, Tomii D, Koseki K, Sato Y, et al. Impact of abdominal fat distribution, visceral fat, and subcutaneous fat on coronary plaque scores assessed by 320-row computed tomography coronary angiography. Atherosclerosis. 2019;287:155–61. [DOI] [PubMed] [Google Scholar]
- 14.Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Medicine: J Br Diabet Association. 1998;15(7):539–53. [DOI] [PubMed] [Google Scholar]
- 15.Song P, Li X, Bu Y, Ding S, Zhai D, Wang E, et al. Temporal trends in normal weight central obesity and its associations with cardiometabolic risk among Chinese adults. Sci Rep. 2019;9(1):5411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jeon HH, Lee YK, Kim DH, Pak H, Shin SY, Seo JH. Risk for metabolic syndrome in the population with visceral fat area measured by bioelectrical impedance analysis. Korean J Intern Med. 2021;36(1):97–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Todd JN, Dahlström EH, Salem RM, Sandholm N, Forsblom C, McKnight AJ, et al. Genetic evidence for a causal role of obesity in diabetic kidney disease. Diabetes. 2015;64(12):4238–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Blaslov K, Bulum T, Duvnjak L. Waist-to-height ratio is independently associated with chronic kidney disease in overweight type 2 diabetic patients. Endocr Res. 2015;40(4):194–8. [DOI] [PubMed] [Google Scholar]
- 19.He W, Zhang S, Song A, Yang M, Jiao J, Allison DB, et al. Greater abdominal fat accumulation is associated with higher metabolic risk in Chinese than in white people: an ethnicity study. PLoS ONE. 2013;8(3):e58688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hanai K, Babazono T, Nyumura I, Toya K, Ohta M, Bouchi R, et al. Involvement of visceral fat in the pathogenesis of albuminuria in patients with type 2 diabetes with early stage of nephropathy. Clin Exp Nephrol. 2010;14(2):132–6. [DOI] [PubMed] [Google Scholar]
- 21.Hua Y, Zhang L. Application value of visceral fat area in screening risk factors of diabetic nephropathy. Chin J Prim Med Pharm. 2020;27(5):605–7. [Google Scholar]
- 22.Ibrahim MM. Subcutaneous and visceral adipose tissue: structural and functional differences. Obes Reviews: Official J Int Association Study Obes. 2010;11(1):11–8. [DOI] [PubMed] [Google Scholar]
- 23.Wang Y, Chen F, Wang J, Wang T, Zhang J, Han Q, et al. The relationship between increased ratio of Visceral-to-Subcutaneous fat area and renal outcome in Chinese adults with type 2 diabetes and diabetic kidney disease. Can J Diabetes. 2019;43(6):415–20. [DOI] [PubMed] [Google Scholar]
- 24.Xiong H, Zhang P, Li X, Zhang H, Yang L, Wu Y et al. Relationship between abdominal fat distribution and diabetic complications in patients with type 2 diabetes mellitus. Chin J Endocrinol Metabolism. 2020(4):299–303.
- 25.Preis SR, Massaro JM, Robins SJ, Hoffmann U, Vasan RS, Irlbeck T, et al. Abdominal subcutaneous and visceral adipose tissue and insulin resistance in the Framingham heart study. Obes (Silver Spring Md). 2010;18(11):2191–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bulum T, Blaslov K, Duvnjak L. The use of anthropometric measurements of obesity in prediction of microvascular complications in obese type 2 diabetic patients. Acta Clin Croatica. 2016;55(2):217–23. [DOI] [PubMed] [Google Scholar]
- 27.Xu M, Liu C, Rao X. Study on the correlation between anthropometric obesity index and microalbuminuria. Int J Urol Nephrol. 2017;37(2):217–20. [Google Scholar]
- 28.Gao B, Liu Y, Ding C, Liu S, Chen X, Bian X. Comparison of visceral fat area measured by CT and bioelectrical impedance analysis in Chinese patients with gastric cancer: a cross-sectional study. BMJ Open. 2020;10(7):e036335. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.