Visceral adipose index as a predictor of new-onset diabetes and cardiovascular events in a geriatric community

doi:10.21203/rs.3.rs-1642405/v1

Download PDF

Research Article

Visceral adipose index as a predictor of new-onset diabetes and cardiovascular events in a geriatric community

https://doi.org/10.21203/rs.3.rs-1642405/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Abdominal obesity presents a higher risk of developing diabetes and future cardiovascular events. The Visceral Adiposity Index (VAI) can be an indicator of adipose distribution and function. This study aimed to evaluate VAI as a predictor of new-onset T2DM and cardiovascular events in the geriatric population.

Method

Community-dwelling elderly adults ages 60–80 received regular health examinations in a privately owned chain of clinics in Taiwan between 1999 and 2008. VAI was calculated with anthropometric measures and biochemical investigations. Longitudinal analyses were performed to examine whether VAI could predict T2DM and cardiovascular events. The relationships between VAI and predicted T2DM and cardiovascular events among elderly adults were analyzed in sexes separately in quartile-based analyses.

Result

The study population consisted of 39,590 geriatric patients recruited and the mean age was 65.38 ± 5.09 years. Participants with T2DM had significantly higher VAI (P-values < 0.05). Different subgroups based on sex and VAI quartiles were divided. The median length of follow-up was 8.5 years. The highest VAI level was associated with a significantly higher risk of T2DM compared with the lowest VAI in males (multivariate-adjusted HR 1.77; 95% CI 1.25–2.5; P = 0.001) and females (multivariate-adjusted HR 1.56; 95% CI 1.12–2.15; P = 0.008). The highest VAI quartile was associated with a significantly higher risk of cardiovascular events only in the unadjusted models in males (unadjusted HR 1.65; 95% CI 1.3–2.09; P < 0.001) and females (unadjusted HR 1.69; 95% CI 1.3–2.09, P < 0.001). No association was noted between the VAI quartile and the risk of cardiovascular events after multivariate adjustments.

Conclusion

We evaluate the association between VAI and new-onset T2DM and cardiovascular events in the Taiwanese geriatric population. Older adults with higher VAI had a higher incidence of new-onset T2DM after adjusting for covariates. Higher VAI was also associated with a higher risk of cardiovascular events; however, there was no significant difference between the sexes after adjusting for covariates. The risk of VAT accumulation increases with age and affects health. Treatments aimed at depleting VAT should be considered to reduce the risk of diseases.

Trial registration

Rretrospectively registered

visceral adipose index

diabetes mellitus

elderly

cardiovascular events

Obesity is a global problem that results in several diseases, such as metabolic dysfunction, type 2 diabetes mellitus (T2DM), and cardiovascular disease (CVD). Body fat distribution is often more emphasized than general or visceral adiposity. Visceral adiposity is significantly related to decreased insulin sensitivity and includes long-term effects on cardiac diseases.[1] Accumulation of visceral adipose tissue (VAT) not only increases the risk of metabolic syndromes[2] but also causes cardiometabolic abnormalities.[3] Evaluation of VAT is usually accomplished using computed tomography (CT) or magnetic resonance imaging (MRI). Although accurate, these imaging techniques are technically challenging, time consuming, and expensive.

Amato et al.[4] developed the visceral adiposity index (VAI), a novel sex-specific indicator derived from anthropometric measures (body mass index [BMI], waist circumference [WC]) and biochemical investigations (serum levels of high-density lipoprotein cholesterol [HDL-C], and triglycerides [TG]). Amato found that VAI was negatively associated with insulin sensitivity and strongly associated with cardiometabolic risk in an Italian population (average age 43.46 ± 14.30 years).

Ageing impairs β-cell function and insulin secretion, which leads to insulin resistance and contributes to the development of T2DM. Geriatric populations with new onset diabetes have a higher risk of diabetic complications as well as cardiovascular and renal morbidities and mortalities.[5, 6] Prevention of T2DM and treatment of hyperglycemia has numerous benefits for the overall health of older adults. Therefore, early detection of T2DM using a non-invasive and inexpensive method could help improve the overall geriatric health.

Previous studies have investigated the association between VAI and T2DM. VAI was found to correlate with future incidence of T2DM and was a risk factor for T2DM in Asian adults.[7] Despite a large body of literature on the association between VAI and T2DM in adults, few studies have focused on the geriatric population. Therefore, we conducted an observational study using cross-sectional and longitudinal analysis of eligible geriatric participants from community health screening centers in Taiwan between 1999 and 2008. We aimed to evaluate the utility of VAI as a predictor of new onset T2DM and cardiovascular events in the geriatric population.

Study approval was obtained from the Ethical Committee of MJ Health Screening Centers and the Ethical Committee of the Cardinal Tien Hospital. Each participant provided written informed consent. The described methods were conducted in accordance with the guidelines of the Declaration of Helsinki.

Study design and participants

We obtained the participants’ data from MJ Health Screening Centers, a privately owned chain of clinics that provided regular health examinations in Taiwan between 1999 and 2008. Data were collected from the participants anonymously and provided for research purposes only.

There were 40,398 eligible participants of 60–80 years of age for this study. First, the participants with missing data were excluded (n = 808). Second, we explored the association between VAI and T2DM using cross-sectional analysis. The first stage of this study was cross-sectional. For observations of the relationships between VAI and T2DM, we included 39,590 participants of 60–80 years of age and categorized them into three groups: healthy, pre-T2DM, and T2DM groups. Next, we excluded the participants with a history of T2DM who underwent only one health examination. Longitudinal analyses were performed to examine whether VAI could predict T2DM and cardiovascular events in sexes separately at the final stage using quartile-based analyses.

The second stage of this study was longitudinal. Overall, 12,875 participants without T2DM were followed up annually for 2–10 years (median 8.5 years). We grouped the participants according to quartiles of VAI in both the sexes. The incidences of subsequent T2DM and cardiovascular events were calculated for the two groups. T2DM was defined as a change in the fasting plasma glucose level to ≥ 126 mg/dL or the introduction of antidiabetic medications over the subsequent years. Cardiovascular events were defined as the development of non-fatal cardiovascular events, including cerebral and cardiac events, based on the history during a subsequent health check-up. Cerebral events included stroke and transient ischemic attacks. Cardiac events included angina (stable and unstable) and myocardial infarction after balloon or stent placement.

General data and anthropometric measurements

The senior nursing staff used a questionnaire to record the drinking habits and medical history of each participant. Complete physical examinations were then performed.

Anthropometric measurements, including WC, weight, height, systolic blood pressure, and diastolic blood pressure were measured described previously.[8] After 10 h of fasting, blood samples were drawn from the antecubital vein for biochemical analysis. Plasma was separated from the blood within an hour and stored at − 30°C until fasting plasma glucose (FPG) and lipid profile could be analyzed. FPG was detected using the glucose oxidase method (YSI 203 glucose analyzer, Scientific Division, Yellow Springs Instruments, Yellow Springs, OH, USA). Total cholesterol, TG, and low-density lipoprotein cholesterol levels were estimated with the enzymatic colorimetric method using Roche Cobas C501 Chemistry Analyzer (Diamond Diagnostics, USA). Serum levels of HDL-C were determined using enzymatic colorimetric assay following dextran sulphate precipitation. Participants with three or more of the following abnormalities were diagnosed with metabolic syndrome according to the revised definition of the National Cholesterol Education Program Adult Treatment Panel III: (1) high fasting glucose level (≥ 100 mg/dL) or use of antidiabetic medication(s); (2) low HDL-C level (< 40 mg/dL in men or < 50 mg/dL in women) or use of specific pharmacological intervention(s); (3) high TG level (≥ 150 mg/dL) or use of lipid-lowering medication(s); (4) central obesity with WC ≥ 90 cm in men or ≥ 80 cm in women; and (5) blood pressure ≥ 130/85 mmHg or use of antihypertensive medication(s).[9]

Definition of T2DM and hypertension

T2DM was defined as FPG ≥ 126 mg/dL, history of T2DM or use of antidiabetic agents. Hypertension was defined as blood pressure > 140/90 mmHg or the use of antihypertensive agents.

Calculation of VAI

The formula to calculate VAI requires four parameters: WC, BMI, and serum TG and HDL-C levels. The following equations are used to calculate VAI in men and women:

Men: VAI\(=\left[\frac{\text{W}\text{C}}{39.68+(1.88\times \text{B}\text{M}\text{I})}\right]\times \left[\frac{\text{T}\text{G}}{1.03}\right]\times \left[\frac{1.31}{\text{H}\text{D}\text{L}}\right]\)

Women: VAI\(=\left[\frac{\text{W}\text{C}}{36.58+(1.89\times \text{B}\text{M}\text{I})}\right]\times \left[\frac{\text{T}\text{G}}{0.81}\right]\times \left[\frac{1.52}{\text{H}\text{D}\text{L}}\right]\)

Statistical analysis

Anthropometric and biochemical data are expressed as mean ± standard deviation. The Pearson chi-square test was used to analyze categorical data, and the analysis of variance (ANOVA) test was used to examine the differences between groups in terms of the demographic and laboratory data. A two-sided P-value ≤ 0.05 was considered as the threshold for statistical significance.

Odds ratios (ORs) calculated using logistic regression were used to examine the association between VAI and T2DM in the cross-sectional analysis. From a clinical standpoint, demographic factors can influence the results; therefore, covariate adjustments were used. Three multiple logistic regression models were constructed: one was an unadjusted model and the other two were adjusted models. Model 1 was unadjusted. Model 2 included Model 1 and adjustments for age, sex, blood pressure, hypertension, and alcohol consumption. Model 3 included Model 2 and serum levels of fasting glucose, total cholesterol, alanine aminotransferase (ALT), uric acid, creatinine, and C-reactive protein (CRP).

In the longitudinal analysis, we grouped the participants into quartiles of VAI for both sexes. Survival analysis was performed to examine the association between VAI and subsequent T2DM and cardiovascular events according to the sex. Kaplan–Meier survival curves were plotted to investigate the relationship between VAI quartiles in participants and subsequent DM and cardiovascular events. Associations between VAI quartiles and endpoints were evaluated using multivariable Cox proportional hazard models. The P-values for the trend tests were determined by treating the quartiles of VAI as a continuous variable (1–4) to observe the associations between an increase in the number of quartiles and future T2DM and cardiovascular events. The statistical analyses were performed using SPSS (version 18.0 for Windows; IBM, Armonk, NY, USA). Two-sided P-values < 0.05 were considered statistically significant.

Characteristics of the study population in the cross-sectional analysis

The study population consisted of 39,590 geriatric patients who completed with available anthropometric measurements and laboratory data. Table 1 summarizes the demographic information of the eligible participants categorized into groups. Overall, 18,828 (47.6%) participants were male, and the mean age was 65.38 ± 5.09 years. Participants with T2DM had significantly higher VAI, BMI, and serum levels of glucose, total cholesterol, TG, and creatinine and significantly lower HDL (All P-values < 0.05).

Table 1

**Characteristics of all study participants.**
Characteristic	Healthy group n = 19166	Prediabetes group n = 14715	Diabetes group n = 5709	Total N = 39590	p-value
Continuous variables ^a.
Age, years	65.14(4.97)	65.35(5.14)	66.22(5.3)	65.38(5.09)	< 0.001
Visceral Adipose Index	2.49(1.38)	2.82(1.49)	3.32(1.58)	2.73(1.48)	< 0.001
Body mass index (kg/m²)	23.6(3.19)	24.48(3.26)	25.05(3.36)	24.14(3.29)	< 0.001
Waist circumference (cm)	79.88(9.04)	82.65(9.16)	85.31(9.31)	81.69(9.33)	< 0.001
SBP(mmHg)	131.71(20.34)	136.23(20.25)	139.01(20.49)	134.44(20.52)	< 0.001
DBP(mmHg)	75.55(11.57)	77.69(11.66)	77.15(11.8)	76.57(11.69)	< 0.001
Fasting glucose, mg/dL	93.4(5.03)	108.23(6.12)	162.53(55.69)	108.88(31.69)	< 0.001
Total cholesterol, mg/dL	206.32(37.4)	209.8(37.2)	206.3(40.54)	207.61(37.84)	< 0.001
Triglyceride, mg/dL	120.07(62.49)	133.44(68.9)	158.76(81.57)	130.62(69.2)	< 0.001
High density lipoprotein, mg/dL	55.96 (15.81)	53.53(14.68)	49.47(13.69)	54.12(15.26)	< 0.001
Low density lipoprotein, mg/dL	126.38(33.32)	129.64(32.9)	125.16(35.52)	127.41(33.54)	< 0.001
ALT, IU	25.63(24.39)	28.04(25.41)	32.42(28.5)	27.51(25.5)	< 0.001
Uric acid, mg/dL	6.03(1.55)	6.29(1.58)	6.07(1.65)	6.12(1.58)	< 0.001
Creatinine, mg/dL	0.99(0.39)	1.01(0.35)	1.06(0.54)	1.01(0.4)	< 0.001
Categorical variables^b
Male	8639(45.1)	7372(50.1)	2817(49.3)	18828(47.6)	0.713
MetS	3304(17.2)	7399(50.3)	3696(64.7)	14399(36.4)	< 0.001
Never drinking	2098(10.9)	1625(11)	619(10.8)	4342(11)	0.911
Hypertension	2857(14.9)	3237(22)	2238(39.2)	8332(21)	< 0.001
^aValues in the continuous variables were expressed as mean (standard deviation) unless otherwise specified.
^bValues in the categorical variables were expressed as number (percent).
SBP, systemic blood pressure; DBP, diastolic blood pressure; ALT, alanine transaminase; MetS, Metabolic syndrome

[Table 1 near here]

Association between VAI and T2DM in the cross-sectional analysis

To examine the association between VAI and T2DM, logistic regression analysis was performed (Table 2). In the total population, VAI was significantly associated with the presence of T2DM (OR 1.345; 95% confidence interval [CI] 1.32–1.37). After adjusting for other covariates, the association persisted.

Table 2

**Multivariate logistic regression model of diabetes mellitus and VAI levels with the other factors.**
Models	Odds Ratio (95% CI)	p-value
Model 1	1.345 (1.32–1.37)	< 0.001
Model 2	1.355 (1.33–1.38)	< 0.001
Model 3	1.145 (1.1–1.18)	< 0.001

* Adjusted covariates: Model 1 = unadjusted; Model 2 = Model 1 + age, gender, blood pressure, hypertension, alcohol drinking; Model 3 = Model 2 + fasting glucose, total cholesterol, ALT, Uric acid and creatinine; ALT: alanine aminotransferase

Characteristics of the study population in the longitudinal analysis

We divided the participants into different subgroups based on sex and VAI quartiles (Table 3). Participants with higher VAI had higher BMI and higher serum levels of glucose, total cholesterol, TG, creatinine, uric acid, and CRP, and lower serum level of HDL.

Table 3

**Characteristics of all study participants without diabetes.**
	Quartiles of visceral adipose index (Total N = 12875)
	Male(N = 6075)					Female(N = 6800)
	Q1	Q2	Q3	Q4		Q1		Q2	Q3	Q4
Characteristic	(< 2.8) n = 1499	( 2.8 to < 3.26) n = 1531	(3.26 to < 3.87) n = 1526	(≥3.87) n = 1519	p-value	(< 0.97) n = 1699		( 0.97 to < 1.5) n = 1698	(1.5 to < 2.39) n = 1703	(≥2.39) n = 1700	p-value
Continuous variables ^a.
Age, years	65.46 (5.58)	65.26 (5.45)	65.28 (5.5)	64.9 (5.29)	0.037	63.59 (4.62)		63.87 (4.5)	64.35 (4.8)	64.98 (5.17)	< 0.001
Body mass index (kg/m²)	21.97 (2.68)	23.32 (2.71)	24.18 (2.78)	24.95 (2.69)	< 0.001	22.65 (3.11)		23.64 (3.06)	24.45 (3.17)	25.14 (3.21)	< 0.001
Waist circumference (cm)	78.48 (7.75)	83.16 (7.79)	85.89 (7.7)	88.22 (7.72)	< 0.001	73.31 (7.38)		76.44 (7.34)	78.97 (7.88)	81.29 (8.15)	< 0.001
SBP(mmHg)	126.49 (19.28)	128.95 (18.77)	129.96 (19.16)	132.4 (18.07)	< 0.001	128.09 (19.18)		131.33 (19.71)	132.51 (19.4)	135.57 (19.73)	< 0.001
DBP(mmHg)	74.09 (10.79)	76.02 (11.2)	76.59 (11.35)	78.32 (11.35)	< 0.001	72.27 (10.91)		73.96 (10.85)	74.84 (10.88)	76.2 (11.24)	< 0.001
Fasting glucose, mg/dL	99.11 (12.77)	101.53 (16.68)	103.03 (21.72)	105.81 (24.2)	< 0.001	97.69 (12.16)		98.82 (12.34)	101.17 (16.26)	105.87 (26.01)	< 0.001
Triglyceride, mg/dL	65.82 (15.21)	92.93 (17.99)	125.17 (23.19)	213.04 (71.37)	< 0.001	67.77 (15.67)		97.36 (18.73)	129.49 (25.4)	211.68 (66.6)	< 0.001
Total cholesterol, mg/dL	196.1 (31.96)	197.53 (34.02)	202.99 (36.43)	207.47 (37.56)	< 0.001	210.92 (34.09)		212.36 (36.98)	215.14 (37.78)	218.01 (37.88)	< 0.001
High density lipoprotein, mg/dL	63.98 (12.57)	51.52 (9.68)	45.44 (8.81)	39.28 (8.81)	< 0.001	74.67 (14.28)		61.9 (10.69)	53.67 (9.15)	45.15 (8.57)	< 0.001
Low density lipoprotein, mg/dL	118.97 (28.86)	127.44 (30.28)	132.52 (32.68)	125.82 (33.82)	< 0.001	122.7 (30.7)		131 (33.39)	135.58 (34.01)	130.58 (34.32)	< 0.001
ALT, IU	23.79 (16.9)	26.45 (20.7)	27.72 (19.64)	30.96 (24.64)	< 0.001	22.69 (20.94)		24.88 (25.78)	25.27 (25.74)	26.56 (29.1)	< 0.001
Uric acid, mg/dL	6.27 (1.39)	6.47 (1.4)	6.75 (1.47)	7.08 (1.55)	< 0.001	5.1 (1.2)		5.38 (1.29)	5.7 (1.38)	6.11 (1.49)	< 0.001
Creatinine, mg/dL	1.08 (0.18)	1.12 (0.34)	1.13 (0.2)	1.17 (0.34)	< 0.001	0.83 (0.39)		0.83 (0.14)	0.85 (0.17)	0.88 (0.39)	< 0.001
CRP, mg/dL	0.25 (0.58)	0.28 (1)	0.3 (0.67)	0.34 (1.1)	0.022	0.25 (0.61)		0.25 (0.56)	0.3 (0.68)	0.34 (0.58)	< 0.001
Categorical variables^b
MetS	26 (1.7)	151 (9.9)	348 (22.8)	1028 (67.7)	< 0.001	89 (5.3)	214 (12.6)		558 (32.8)	1317 (77.5)	< 0.001
Never drinking	120(8)	131(8.6)	146(9.6)	140(9.2)	0.437	196(11.5)	199(11.7)		192(11.4)	224 (13.2)	0.316
Hypertension	388(25.9)	472(30.8)	503(33)	597(39.3)	< 0.001	484(28.5)	588(34.6)		637(37.4)	785(46.2)	< 0.001
New onset DM	62(4.1)	103(6.7)	131(8.6)	187(12.3)	< 0.001	72(4.2)	117(6.9)		138(8.1)	216(12.7)	< 0.001
CVD events	110(7.7)	136(9.3)	149(10.1)	178(12.3)	< 0.001	122(7.6)	153(9.5)		190(11.8)	207(13.2)	< 0.001
^a Values in the continuous variables were expressed as mean (standard deviation) unless otherwise specified.
^b Values in the categorical variables were expressed as number (percent).
SBP, systemic blood pressure; DBP, diastolic blood pressure; ALT, alanine transaminase; CRP, C-reactive protein; MetS, Metabolic syndrome; New onset DM, New onset Diabetes mellitus; CVD events, cardiovascular events.

[Table 3 near here]

Association between VAI and the incidence of DM and CVD

The median length of follow-up was 8.5 years. During this period, 1,026 patients (483 males and 543 females) developed T2DM, and 1,245 patients developed CVD (573 males and 672 females). The hazard ratio (HR) for developing T2DM and CVD per quartile of VAI, as indicated by the unadjusted and multivariate-adjusted Cox proportional regression analyses, are summarized in Table 4. The highest VAI level was associated with a significantly higher risk of T2DM (multivariate-adjusted HR 1.77; 95% CI 1.25–2.5; P = 0.001) compared with the lowest VAI in males. Similar results were noted in females (multivariate-adjusted HR 1.56; 95% CI 1.12–2.15; P = 0.008). The highest VAI quartile was associated with a significantly higher risk of cardiovascular events only in the unadjusted models in males (unadjusted HR 1.65; 95% CI 1.3–2.09; P < 0.001) and females (unadjusted HR 1.69; 95% CI 1.3–2.09, P < 0.001). No association was noted between VAI quartile and the risk of cardiovascular events after multivariate adjustments.

Table 4

**Multivariate Cox Regression model using new onset diabetes events and cardiovascular events.**
Male
New events	Model 1				Model 2			Model 3
	VAI quartiles	Hazard Ratio (95% CI)	p-value		VAI quartiles	Hazard Ratio (95% CI)	p-value	VAI quartiles	Hazard Ratio (95% CI)	p-value
New onset diabetes	Q2 v.s. Q1	1.68 (1.22–2.31)	0.002		Q2 v.s. Q1	1.61 (1.16–2.22)	0.004	Q2 v.s. Q1	1.53 (1.1–2.11)	0.01
	Q3 v.s. Q1	2.11 (1.56–2.87)	< 0.001		Q3 v.s. Q1	1.77 (1.29–2.43)	0.003	Q3 v.s. Q1	1.7 (1.24–2.34)	0.001
	Q4 v.s. Q1	3.14 (2.35–4.2)	< 0.001		Q4 v.s. Q1	1.85 (1.32–2.61)	0.002	Q4 v.s. Q1	1.77 (1.25–2.5)	0.001
	p for trend	< 0.001			p for trend	0.001		p for trend	0.002
Cardiovascular events	Q2 v.s. Q1	1.25 (0.97–1.6)	0.085		Q2 v.s. Q1	1.18 (0.92–1.52)	0.201	Q2 v.s. Q1	1.15 (0.89–1.49)	0.266
	Q3 v.s. Q1	1.28 (1-1.64)	0.049		Q3 v.s. Q1	1.13 (0.88–1.46)	0.326	Q3 v.s. Q1	1.09 (0.85–1.42)	0.47
	Q4 v.s. Q1	1.65 (1.3–2.09)	< 0.001		Q4 v.s. Q1	1.27 (0.95–1.7)	0.101	Q4 v.s. Q1	1.19 (0.89–1.6)	0.227
	p for trend	< 0.001			p for trend	0.148		p for trend	0.309
Female
New events	VAI quartiles	Hazard Ratio (95% CI)		p-value	VAI quartiles	Hazard Ratio (95% CI)	p-value	VAI quartiles	Hazard Ratio (95% CI)	p-value
New onset diabetes	Q2 v.s. Q1	1.52 (1.13–2.04)		0.006	Q2 v.s. Q1	1.38 (1.02–1.86)	0.035	Q2 v.s. Q1	1.35 (0.99–1.82)	0.051
	Q3 v.s. Q1	1.89 (1.42–2.53)		< 0.001	Q3 v.s. Q1	1.44 (1.06–1.94)	0.018	Q3 v.s. Q1	1.39 (1.02–1.88)	0.036
	Q4 v.s. Q1	2.99 (2.28–3.91)		< 0.001	Q4 v.s. Q1	1.63 (1.18–2.24)	0.003	Q4 v.s. Q1	1.56 (1.12–2.15)	0.008
	p for trend	< 0.001			p for trend	0.005		p for trend	0.014
Cardiovascular events	Q2 v.s. Q1	1.18 (0.93–1.5)		0.17	Q2 v.s. Q1	1.09 (0.86–1.39)	0.465	Q2 v.s. Q1	1.09 (0.86–1.38)	0.489
	Q3 v.s. Q1	1.5 (1.2–1.89)		< 0.001	Q3 v.s. Q1	1.31 (1.03–1.66)	0.026	Q3 v.s. Q1	1.29 (1.02–1.64)	0.037
	Q4 v.s. Q1	1.69 (1.35–2.12)		< 0.001	Q4 v.s. Q1	1.24 (0.94–1.61)	0.125	Q4 v.s. Q1	1.2 (0.91–1.58)	0.184
	p for trend	< 0.001			p for trend	0.061		p for trend	0.098
*Adjusted covariates: Model 1 = unadjusted; Model 2 = Model 1 + age, MetS blood pressure, hypertension, alcohol drinking; Model 3 = Model 2 + fasting glucose, total cholesterol, ALT, Uric acid and creatinine, CRP; MetS: metabolic syndrome; ALT: alanine aminotransferase; CRP, C-reactive protein

[Table 4 near here]

In this study, we investigated a Taiwanese geriatric population to evaluate the association between VAI and new onset T2DM and cardiovascular events. We found that older adults with higher VAI had a higher incidence of new onset T2DM.

Adipose tissue plays an important role in longevity.[10] Obesity, one of the most important global health care problems, is associated with many pathologies, such as metabolic dysfunction, insulin resistance, T2DM, and CVD.[11] It is also associated with high mortality,[12] and strategies directed at the management of adipose tissue may affect the life span.[10]

Increased risk of metabolic disease is closely linked to fat distribution, and different fat compartments may cause different metabolic risks.[13] Fat distribution varies with many factors, including ageing.[14] Circulating levels of fatty acids increase when subcutaneous fat cannot store lipids, which results in ectopic deposition and toxicity to most cells. They result in aged pre-adipocytes and potentially exacerbate fat dysfunction and contribute to metabolic dysfunction. WC alone cannot help in distinguishing between subcutaneous and VAT.[15] Although both subcutaneous abdominal adipose tissue and VAT are associated with metabolic risk factors,[3] VAT is more prevalent in obesity and T2DM because of chronic, low-grade inflammation.[16]

Epidemiologic studies have revealed that VAT is a common but preventable risk factor for metabolic disease. Huffman et al. reported that the mean and maximum lifespans in rats improved following surgical removal of VAT. This was the first evidence to suggest that VAT depletion may be a strong factor underlying improved life span.[14] In contrast, Klein et al.[17] surgically removed large volumes of subcutaneous abdominal fat in 15 participants and found insufficient improvements in their metabolic parameters. Therefore, subcutaneous fat may not be a key component of obesity-related metabolic disorders in humans.

The mechanism linking VAT and metabolic syndrome involves VAT’s anatomical location. VAT leads to a “portal” effect of glycerol release and greater blood levels of free fatty acids.[18] Adipose tissue is an active endocrine organ and secretes adipokines that can interfere with insulin action and promote inflammation.[19] VAT also drives cardiac aging through modulation of fibroblast senescence via osteopontin.[20] Cells of the immune system play varying roles in adipose tissue infiltrates. For example, macrophages were recognized in obesity-induced inflammation of the adipose tissue.[21] Ageing adipose tissue undergoes significant changes in the cellular composition, abundance, distribution, and endocrine signaling. Therefore, ageing plays a central role in metabolic dysfunction, insulin resistance, impaired regenerative capacity, and inflammation.[22] Accumulation of VAT is a hallmark of ageing in humans.[23] Ageing is associated with fat tissue redistribution from subcutaneous to visceral depots, which may cause not only insulin resistance but also metabolic complications as well as functional decline and mortality in older adults.

Amato et al.[24] demonstrated that VAI was closely correlated with body visceral adiposity estimated using MRI. They found that VAI was independently associated with cardiovascular events, which was not observed with BMI, WC, and other classical cardiovascular risk factors. VAI indirectly reflects other non-classical risk factors, such as elevated lipolysis, altered production of adipocytokines, and plasma free fatty acids. Therefore, VAI could be a valuable index of both fat distribution and function.[24]

Other studies have investigated the relationship between VAI and several health-related outcomes. A cross-sectional epidemiological survey in China revealed that higher VAI was positively associated with hypertension in both sexes.[25] Another cross-sectional analysis of 2,754 Chinese community-dwelling people, aged 20–50 years, revealed that higher VAI was positively associated with pre-T2DM and T2DM.[26] Bozorgmanesh et al. followed up 5,964 non-diabetic adult participants in Iran for a median of 6 years and found that VAI could be a prognostic tool for incident diabetic events.[27] Nusrianto et al. reviewed seven studies to determine whether VAI could be used as a predictor of T2DM in Asian populations.[7] The review included six studies from China and one from Iran. Four studies were prospective cohort studies and three were cross-sectional studies. They found that VAI was a predictor of T2DM in Asian populations with better prediction values compared with that in Caucasian populations.[7] The largest cross-sectional study included 7,639 participants.[28] The longest study lasted 15 years but only included 687 participants aged 35–65 years.[29] Therefore, most studies have included participants aged 18–65 years.

In our study, we included 12,875 patients aged 60–80 years over a 10-year period. Notably, this is the first study to demonstrate the relationship between VAI and new onset T2DM and cardiovascular events in geriatric patients. After adjusting for covariates, a higher VAI was associated with a higher risk of new onset T2DM in both males and females with up to 77% higher risk in males and 56% in females. Higher VAI was also associated with a higher risk of cardiovascular events; however, there was no significant difference between the sexes after adjusting for covariates. Only few participants developed CVD in this study.

There are several plausible explanations for this positive association between VAI and new onset T2DM in older adults. Fundamental ageing mechanisms occur in adipose tissue, including progenitor cell dysfunction and cellular senescence. A previous review[10] reported that ageing and obesity were related to elevated systemic inflammatory cytokines. Ageing and obesity are associated with adipose tissue dysfunction and reduced generation of anti-inflammatory factors such as adiponectin and interleukin-10. Recent evidence has demonstrated that ageing and obesity could aggravate metabolic dysfunction and that ageing is related to fat tissue redistribution and insulin resistance. Therapeutic interventions targeted at both obesity and ageing should be considered in the future. The risk of VAT accumulation increases with age and affects the health. Treatments aimed at depleting VAT should be considered to reduce the risk of diseases[14].

There were several limitations in this study. First, self-reported medical conditions may be affected by recall bias or misclassification or have led to over-reporting. Second, data regarding potential confounding factors, such as adiponectin and interleukin-10, were not available. Third, this was a mixed design study and may not demonstrate the cause-and-effect relationship. Finally, despite adjustments for several confounding factors, other unmeasured confounders cannot be ruled out.

This study demonstrated that in a representative geriatric population, a significant association was noted between VAI and new onset T2DM. Easy and non-invasive assessments of VAI in community settings and hospitals suggest that it could be used as a clinical prognostic indicator of new onset T2DM. Additional studies are warranted to elucidate their underlying mechanism further. We demonstrated that ageing should not be neglected in the study of VAI and diabetes because older patients are predominantly in need of such care or therapies.[10] Therefore, early and accurate identification of visceral obesity in geriatric patients is urgently needed to prevent dysglycaemia.

VAI

Visceral Adiposity Index

T2DM

type 2 diabetes mellitus

CVD

Cardiovascular disease

VAT

Visceral adipose tissue

Computed tomography

MRI

Magnetic resonance imaging

BMI

Body mass index

Waist circumference

HDL-C

High-density lipoprotein cholesterol

Triglycerides

ALT

Alanine aminotransferase

Confidence interval

CRP

C-reactive protein

Hazard ratio

Declarations

Not Applicable.

Ethics approval and consent to participate

Consent for publication

Not applicable

Availability of data and materials

The datasets and/or analyzed during the current study available from the corresponding author on reasonable request due to privacy concerns. ([email protected]).

Competing interests

The authors declare that they have no competing interests.

Funding

This study did not receive funding from any organization or institution.

Authors’ contributions

PD designed the study. FYL and YLC carried out the study. TWK and FYC analyzed the data. FYL, YLC and LYH interpreted the findings and handled the writing.

Acknowledgements

We thank the physicians who participated and everyone who invited others to participate.

Lalia AZ, Dasari S, Johnson ML, Robinson MM, Konopka AR, Distelmaier K, et al. Predictors of whole-body insulin sensitivity across ages and adiposity in adult humans. J Clin Endocrinol Metab. 2016;101:626–34.
Goldani H, Adami FS, Antunes MT, Rosa LH, Fassina P, Quevedo Grave MT, Morelo Dal Bosco S. Applicatility of the visceral adiposity index (Vai) in the prediction of the components of the metabolic syndrome in elderly. Nutr Hosp. 2015;32:1609–15.
Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116:39–48.
Amato MC, Giordano C, Galia M, Criscimanna A, Vitabile S, Midiri M, et al. Visceral Adiposity Index: a reliable indicator of visceral fat function associated with cardiometabolic risk. Diabetes Care. 2010;33:920–2.
Twito O, Ahron E, Jaffe A, Afek S, Cohen E, Granek-Catarivas M, et al. New-onset diabetes in elderly subjects: association between HbA1c levels, mortality, and coronary revascularization. Diabetes Care. 2013;36:3425–9.
Bethel MA, Sloan FA, Belsky D, Feinglos MN. Longitudinal incidence and prevalence of adverse outcomes of diabetes mellitus in elderly patients. Arch Intern Med. 2007;167:921–7.
Nusrianto R, Tahapary DL, Soewondo P. Visceral adiposity index as a predictor for type 2 diabetes mellitus in Asian population: A systematic review. Diabetes Metab Syndr. 2019;13:1231–5.
Huang CL, Chang HW, Chang JB, Chen JH, Lin JD, Wu CZ, et al. Normal fasting plasma glucose predicts type 2 diabetes and cardiovascular disease in elderly population in Taiwan. Q J M. 2016;109:515–22.
Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112:2735–52.
Palmer AK, Kirkland JL. Aging and adipose tissue: potential interventions for diabetes and regenerative medicine. Exp Gerontol. 2016;86:97–105.
González N, Moreno-Villegas Z, González-Bris A, Egido J, Lorenzo Ó. Regulation of visceral and epicardial adipose tissue for preventing cardiovascular injuries associated to obesity and diabetes. Cardiovasc Diabetol. 2017;16:44.
Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005;293:1861–7.
Poirier P, Després JP. Waist circumference, visceral obesity, and cardiovascular risk. J Cardiopulm Rehabil. 2003;23:161–9.
Huffman DM, Barzilai N. Role of visceral adipose tissue in aging. Biochim Biophys Acta. 2009;1790:1117–23.
Pouliot MC, Després JP, Lemieux S, Moorjani S, Bouchard C, Tremblay A, et al. Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol. 1994;73:460–8.
Burhans MS, Hagman DK, Kuzma JN, Schmidt KA, Kratz M. Contribution of adipose tissue inflammation to the development of type 2 diabetes mellitus. Compr Physiol. 2018;9:1–58.
Klein S, Fontana L, Young VL, Coggan AR, Kilo C, Patterson BW, Mohammed BS. Absence of an effect of liposuction on insulin action and risk factors for coronary heart disease. N Engl J Med. 2004;350:2549–57.
Björntorp P. ‘Portal’ adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis. 1990;10:493–6.
Trayhurn P, Wood IS. Signalling role of adipose tissue: adipokines and inflammation in obesity. Biochem Soc Trans. 2005;33:1078–81.
Sawaki D, Czibik G, Pini M, Ternacle J, Suffee N, Mercedes R, et al. Visceral adipose tissue drives cardiac aging through modulation of fibroblast senescence by osteopontin production. Circulation. 2018;138:809–22.
Mathis D. Immunological goings-on in visceral adipose tissue. Cell Metab. 2013;17:851–9.
Tchkonia T, Morbeck DE, Von Zglinicki T, Van Deursen J, Lustgarten J, Scrable H, et al. Fat tissue, aging, and cellular senescence. Aging Cell. 2010;9:667–84.
Matsuzawa Y, Shimomura I, Nakamura T, Keno Y, Kotani K, Tokunaga K. Pathophysiology and pathogenesis of visceral fat obesity. Obes Res. 1995;3 Suppl 2:187S–94S.
Amato MC, Giordano C, Galia M, Criscimanna A, Vitabile S, Midiri M, et al. Visceral Adiposity Index: a reliable indicator of visceral fat function associated with cardiometabolic risk. Diabetes Care. 2010;33:920–2.
Ding Y, Gu D, Zhang Y, Han W, Liu H, Qu Q. Significantly increased visceral adiposity index in prehypertension. PLOS ONE. 2015;10:e0123414.
Liu PJ, Ma F, Lou HP, Chen Y. Visceral adiposity index is associated with pre-diabetes and type 2 diabetes mellitus in Chinese adults aged 20–50. Ann Nutr Metab. 2016;68:235–43.
Bozorgmanesh M, Hadaegh F, Azizi F. Predictive performance of the visceral adiposity index for a visceral adiposity-related risk: type 2 diabetes. Lipids Health Dis. 2011;10:88.
Du T, Sun X, Huo R, Yu X. Visceral adiposity index, hypertriglyceridemic waist and risk of diabetes: the China Health and Nutrition survey 2009. Int J Obes (Lond). 2014;38:840–7.
Wang Y, He S, He J, Wang S, Liu K, Chen X. Predictive value of visceral adiposity index for type 2 diabetes mellitus: A 15-year prospective cohort study. Herz. 2015;40 Suppl 3:277–81.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Visceral adipose index as a predictor of new-onset diabetes and cardiovascular events in a geriatric community

Status:

Version 1

Abstract

Background

Methods

Study design and participants

General data and anthropometric measurements

Definition of T2DM and hypertension

Calculation of VAI

Statistical analysis

Results

Characteristics of the study population in the cross-sectional analysis

[Table 1 near here]

Association between VAI and T2DM in the cross-sectional analysis

Characteristics of the study population in the longitudinal analysis

[Table 3 near here]

Association between VAI and the incidence of DM and CVD

[Table 4 near here]

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1