Metabolic syndrome components determine the presence of subclinical atherosclerosis in obese and overweight

Abstract

Metabolic syndrome (MS) is frequently associated with an increased body mass index (BMI), and related to an adverse cardiovascular prognosis. The purpose of this study is to evaluate the prevalence and association between MS, obesity and subclinical atherosclerosis (SA).

This cross-sectional study included healthy adults, allocated to normal weight (NW) when BMI <25 kg/m², overweight (OW) BMI ≥25 and <30 kg/m², or obese (OB) BMI ≥30 kg/m² groups. Presence of MS was defined according to National Cholesterol Education Program (NCEP) criteria. SA was evidenced with vascular ultrasound. Association between SA, obesity and MS, was evaluated by logistic regression models.

There were 3,716 patients studied (female 66.7%, mean age 47 ± 17.5 years). According to BMI, NW represented 28.2%, OW 39.4% and OB 32.4%. MS showed a strong correspondence with BMI (NW 4.9%, OW 21.4%, OB 49.7%; p<0.001). SA was more prevalent in each group when MS was present: NW (25.4% vs. 45.1%, p<0.005), OW (43.2% vs. 58.9%, p<0.0001) and OB (44.2% vs. 57.8%, p<0.0001). Logistic-regression models showed an independent association of SA with MS criteria (arterial hypertension p<0.001; high-density lipoprotein [HDL] p<0.05; and triglycerides p<0.005) adjusted by gender, age and BMI.

In conclusion, overweight and obesity are frequent and strongly linked with MS and SA. Prevalence of SA is high, and is independently associated with MS components. However, BMI could not retain statistical significance in the multivariate analyses.

Introduction

Overweight and obesity are a global pandemic.¹ These evermore frequent conditions are associated with serious chronic diseases (e.g. diabetes mellitus, hyperlipidaemia, cancer), and incremented risk for cardiovascular events.^2,3 Metabolic syndrome (MS), a constellation of anthropometric and metabolic anomalies, is frequently associated with an increased body mass index (BMI) and related to an adverse cardiovascular prognosis.⁴ Despite the well-established association between BMI and cardiovascular prognosis, the concept of healthy obesity has emerged in more recent years, resembling a phenotype of obesity without metabolic disturbances.⁵ Whether this condition is associated with better cardiovascular prognosis is under debate. To date, few studies have addressed the association of MS, body weight and subclinical atherosclerosis (SA). This relationship holds major implications in cardiovascular risk prediction, since SA is associated with incident cardiovascular events.⁶ Hence, the aim of the present study is to evaluate the importance of MS components across BMI categories, and on the prevalence of SA as a manifestation of arterial vascular disease.

Materials and method

This study was of cross-sectional design and conducted at the Hospital Universitario René G Favaloro in Buenos Aires, Argentina. Participants were recruited from a cardiovascular health prevention programme if they were above 18 years, with no history of cardiovascular disease (acute coronary syndrome, transient ischaemic attack [TIA] or stroke, any arterial revascularisation). Those with untreated hypothyroidism, history of acute or chronic hepatic or kidney disease, were excluded. Medical records were used as the source for general data collection. This study was approved by the institutional ethics committee (Comité de Bioética del Hospital Universitario Fundación Favaloro).

Subjects were allocated according to BMI to normal weight (NW) when BMI <25 kg/m², overweight (OW) BMI ≥25 and <30 kg/m², or obese (OB) BMI ≥30 kg/m² groups.

MS was diagnosed according to the NCEP ATP III (National Cholesterol Education Program Adult Treatment Panel III) criteria.⁷ SA was diagnosed with vascular ultrasound using an Affinity 50 ultrasound system (Philips HealthCare, USA) with a 9–12 MHz vascular probe. Plaques were defined as a focal protrusion into the arterial lumen of >50% of the surrounding intima-media thickness or a thickness >1.5 mm measured between the media-adventitia and intimalumen interfaces.⁸

Quantitative data are presented as mean ± standard deviation [SD] and categorical variables with number and percentage. For comparison between groups of quantitative variables, one-way ANOVA, with post-hoc Tukey test was performed. Categorical variables were analysed with Chi-square and exact Fisher test. The association between SA, BMI and MS components, was evaluated by multi-nomial logistic-regression models.

Results

A total of 3,716 patients were included (table 1). According to BMI, NW represented 28.2%, OW 39.4% and OB 32.4%. MS was present in 25.9% (n=964), with a strong correspondence with BMI (NW 4.9%, OW 21.4%, OB 49.7%; p<0.001). The OW and OB groups showed higher prevalence of men and older age. The proportion of patients with dyslipidaemia, arterial hypertension (HT) and diabetes was higher in OW and OB groups. Triglycerides (TG) were increased in OW and OB patients with a lower high-density lipoprotein (HDL)-cholesterol in these two groups compared with NW. In addition, a higher use of statins was observed in OW and OB groups.

Table 1. Participants’ general characteristics according to body mass index (BMI) group.

	NW group (n=1,047)	OW group (n=1,464)	OB group (n=1,205)	p value for trend
Female gender, n (%)	698 (66.7)	548 (37.4)a	429 (35.6)a	<0.0001
Mean age ± SD, years	47 ± 17.5	55.6 ± 14.6a	57.2 ± 13.5a,b	<0.001
Mean weight ± SD, kg	61.4 ± 8.9	77.4 ± 9.8a	96.1 ± 15.1a,b	<0.001
Mean BMI ± SD, kg/m2	22.5 ± 2.0	27.4 ± 1.4a	34.3 ± 4.1a,b	<0.001
Mean waist circumference ± SD, cm	79.8 ± 8.6	93.3 ± 8.5a	108.7 ± 11.9a,b	<0.001
Mean systolic BP ± SD, mmHg	113.3 ± 13.5	120.9 ± 14.0a	126.7 ± 14.8a,b	<0.001
Mean diastolic BP ± SD, mmHg	72.1 ± 10.0	76.6 ± 9.0	80.1 ± 8.8	<0.001
Hypertension, n (%)	156 (14.9)	430 (29.4)a	591 (49)a,b	<0.0001
Type 1 diabetes, n (%)	2 (0.2)	6 (0.4)	3 (0.2)	ns
Type 2 diabetes, n (%)	21 (2)	109 (7.4)a	174 (14.4)a,b	<0.0001
Current smoking, n (%)	200 (19.1)	258 (17.6)	198 (16.4)	ns
Former smoking, n (%)	162 (15.5)	317 (21.6)a	315 (26.1)a,b	<0.001
Family history of CVD, n (%)	90 (8.6)	118 (8.1)	80 (6.6)	ns
Dyslipidaemia, n (%)	310 (29.6)	665 (45.4)a	621 (52.6)a,b	<0.0001
Mean glycaemia ± SD, mg/dL	93.7 ± 17.0	99.0 ± 22.1a	105.6 ± 33.7a,b	<0.001
Mean HbA1c ± SD, mg/dL	4.3 ± 2.2	5.0 ± 1.8a	5.2 ± 2.0a,b	<0.001
Mean HDL-cholesterol ± SD, mg/dL	61.5 ± 15.4	53.7 ± 13.6a	51.3 ± 13.1a,b	<0.001
Mean total cholesterol ± SD, mg/dL	197.8 ± 39.9	199.4 ± 39.6a	200.1 ± 41.2a,b	ns
Mean triglycerides ± SD, mg/dL	99.0 ± 54.5	129.5 ± 82.4a	148.2 ± 98.0a,b	<0.001
Mean LDL-cholesterol ± SD, mg/dL	116.5 ± 34.5	119.7 ± 35.2a	119.2 ± 39.9a	ns
Treatment with statins, n (%)	106 (10.1)	308 (21)	263 (21.8)	<0.001

BMI categories are divided into NW (normal weight group) when <25 kg/m², OW (overweight group) with BMI ≥25 and <30 kg/m², and OB (obese group) for BMI >30 kg/m².

^a p<0.001 vs. NW; ^b p<0.05 vs. OW.

Key: BMI = body mass index; BP = blood pressure; CVD = cardiovascular disease; HbA_1c = glycated haemoglobin; HDL = high-density lipoprotein; LDL = low-density lipoprotein; ns = not significant; SD = standard deviation

Prevalence of SA was also associated with BMI (OB 50.9%, OW 46.6%, NW 26.4%; p<0.001). MS was associated with a higher SA prevalence in the NW group (without MS 25.4% vs. 45.1% with MS; p<0.005), and was also increased in OW (without MS 43.2% vs. 58.9% with MS; p<0.0001) and OB (without MS 44.2% vs. 57.8% with MS; p<0.0001) groups.

Association between SA, obesity and MS, was evaluated by three logistic-regression models with SA as an independent variable (table 2). In model 1, adjusted by gender and age, BMI (p<0.05), as a continuous variable, was independently associated with SA. Addition of MS as a dichotomic variable in model 2, reveals MS is also associated with SA (p<0.001), when adjusted by age, gender and BMI. In this second model, BMI loses statistical significance. Model 3 was further adjusted for age, gender and BMI (kg/m²), triglycerides (mg/ dL), HDL (mg/dL), HT (yes or no), waist circumference (cm), and glycaemia (mg/ dL). An independent association of SA with MS criteria was found: HT (p<0.001), HDL (p<0.05) and TG (p<0.005).

Table 2. Multi-variate analysis of association between BMI, metabolic syndrome (MS) and subclinical atherosclerosis.

Variables in equation	B	Standard error	Significance	Exp(B)	95%CI for Exp(B)
Model 1
Age (years)	0.117	0.004	<0.001	1.124	1.115 to 1.134
Gender (m or f)	0.730	0.087	<0.001	2.075	1.751 to 2.459
BMI (kg/m2)	0.16	0.008	0.044	1.017	1 to 1.033
Model 2
Age (years)	0.117	0.004	<0.001	1.124	1.114 to 1.133
Gender (m or f)	0.735	0.087	<0.001	2.086	1.759 to 2.473
BMI (kg/m2)	0.003	0.009	ns	1.003	0.986 to 1.021
MS (yes or no)	0.36	0.1	<0.001	1.433	1.178 to 1.744
Model 3
Age (years)	0.113	0.004	<0.001	1.12	1.11 to 1.13
Gender (m or f)	0.558	0.1	<0.001	1.747	1.436 to 2.126
BMI (kg/m2)	–0.014	0.015	ns	0.986	0.958 to 1.015
HT (yes or no)	0.494	0.091	<0.001	1.638	1.371 to 1.957
Glycaemia (mg/dL)	–0.002	0.002	ns	0.998	0.995 to 1.001
HDL (mg/dL)	–0.007	0.003	0.049	0.993	0.987 to 1
Triglycerides (mg/dL)	0.002	0.001	0.004	1.002	1 to 1.003
WC (cm)	0.006	0.006	ns	1.006	0.995 to 1.017

Key: BMI = body mass index; CI = confidence interval; Exp(B) = odds ratio; f = female; HDL = high-density lipoprotein; HT = hypertension; m = male; MS = metabolic syndrome; ns = not significant; WC = waist circumference

Discussion

Overweight and obesity were frequent in our study, reaching almost 72% of otherwise healthy primary prevention individuals. As expected, the increase in BMI was accompanied by a higher prevalence of cardiovascular risk factors, as well as MS diagnosis, which is concordant with other studies.⁹ Overall, there was a 40% prevalence of SA in our population, which reproduces the results of other cohorts.¹⁰ In our study, OB and OW individuals were more prone than their NW counterparts to present SA. However, including MS in the second logistic regression model caused BMI to lose its statistical significance. In addition, in model 3, HT, HDL, and TG were found to have an independent association to SA. The loss of independent prediction of BMI over SA is not new in literature. Prediction models of cardiovascular risk often attenuate significant associations between BMI and incidental events when adjusted by other risk factors as covariates.¹¹ This interaction between BMI, SA and MS, could partly be attributed to ‘metabolically healthy obesity’. Without a homogeneous definition, this concept developed from the observation that some individuals with obesity do not exhibit metabolic disorders and have a lower incidence of cardiovascular events, reflecting that obesity itself would show less influence on vascular damage in the absence of metabolic factors.^12,13

To conclude, we propose that the association of weight disorders and SA could mainly be attributed to the effect of MS components. Future research, focused on freedom from clinical events, will elucidate the influence of these findings on prognosis.

Key messages

• This study was a cross-sectional design, which evaluated 3,716 healthy patients to evaluate the role of metabolic syndrome (MS), overweight (OW) and obesity (OB) in the presence of subclinical atherosclerosis (SA)

• MS showed a strong correspondence with body mass index (BMI) (normal weight 4.9%, OW 21.4%, OB 49.7%; p<0.001)

• Logistic-regression models showed an independent association of SA with MS criteria (arterial hypertension p<0.001, HDL p<0.05, and triglycerides p<0.005) adjusted by gender, age and BMI

• BMI could not retain statistical significance in the multi-variate analyses

Study approval

This study was approved by the institutional ethics committee (Comité de Bioética del Hospital Universitario Fundación Favaloro).

Funding Statement

Funding None.