Association of metabolic syndrome with carotid thickening and plaque in the general population: A meta‐analysis

Abstract

We performed a meta‐analysis of population studies reporting data on carotid intima‐media thickness and plaque in patients with and without metabolic syndrome (MetS) to provide a new piece of information on the relationship of MetS with both phenotypes of vascular damage. The Ovid MEDLINE, PubMed, and Cochrane CENTRAL databases were searched without time restriction up to December 31, 2016. Overall, 19 696 patients (22.2% with MetS) were included in eight studies. Common carotid intima‐media thickness was greater in patients with MetS compared with those without it (788 ± 47 μm vs 727 ± 44 μm), with a standard means difference of 0.28 ± 0.06 (P = .00003). Increased intima‐media thickness in patients with MetS was paralleled by a higher prevalence of plaques. The present meta‐analysis shows that MetS is associated with both ultrasonographic phenotypes of carotid damage. This finding is consistent with the view of MetS as a cluster of hemodynamic and nonhemodynamic factors promoting vascular hypertrophy and plaque.

1. INTRODUCTION

In the past 2 decades, carotid ultrasonography has been increasingly used for intima‐media thickness (IMT) evaluation and plaque identification. Both imaging alterations have progressively gained the role of surrogate biomarkers of subclinical atherosclerosis.1, 2

Cross‐sectional studies performed in a variety of clinical settings and ethnic groups have consistently documented an association between carotid atherosclerosis and traditional risk factors, as well as markers of subclinical organ damage and cardiovascular events.3, 4, 5, 6, 7 Major factors involved in the multifactorial pathogenesis of carotid atherosclerosis include hypertension, dyslipidemia, diabetes mellitus, obesity, and smoking. The metabolic syndrome (MetS) is a clinical trait characterized by the clustering of multiple risk factors such as abdominal obesity, high normal or elevated blood pressure (BP), hyperglycemia, elevated triglycerides, and reduced high‐density cholesterol levels. MetS has been shown to be associated with an adverse cardiovascular prognosis.8, 9

Increased susceptibility to atherosclerosis in multiple districts including carotid, coronary, cerebral, and peripheral vessels has been hypothesized to explain the increased cardiovascular risk related to MetS.10

The synergic impact of increased mechanical stress, such as circumferential wall stress and flow‐mediated shear stress caused by high BP with metabolic‐driven alterations of vessel properties may result in changes in carotid artery structure and function such as wall thickening, plaque formation, and arterial stiffening.10 Whether arterial wall thickening and plaque constitute a common trait of atherosclerosis, the first being the earliest manifestation and the second a more advanced expression of the process, is still undefined.11 Recent evidence suggests that these alterations represent distinct vascular phenotypes with different pathogenic mechanisms and prognostic significance. According to this hypothesis, carotid wall thickening represents an adaptive response to increased shear stress related to aging and hypertension rather than an atherosclerotic phenotype. Findings provided by recent studies support the view that carotid plaque is a stronger predictor of cardiovascular risk compared with IMT.12

Since the mid‐90s, the impact of metabolic risk factors on the extracranial carotid tree has been studied in hundreds of ultrasonographic studies analyzing the relationship of MetS with IMT or plaque.

Despite the abundance of data on this issue, findings are not univocal and the association of MetS with both markers of vascular damage remains undefined.3, 13 Because of the unfavorable impact of MetS components on arterial wall, we hypothesized that both markers of vascular damage are more pronounced in patients with MetS than in their counterparts. Thus, we performed a meta‐analysis of population‐based studies reporting data on carotid IMT and plaque in patients with and without MetS to provide a new piece of information on the relationship between MetS and these different phenotypes of subclinical vascular damage in members of the general population.

2. METHODS

2.1. Search strategy and study selection

The present study was performed according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses) guidelines.14 Medical literature was reviewed to identify all articles investigating the association of MetS with subclinical carotid damage as assessed by ultrasonographic assessment of IMT and plaques. The Ovid MEDLINE, PubMed, and Cochrane CENTRAL databases were searched for English‐language articles without time restriction up to December 31, 2016, through focused, highly sensitive search strategies.

Studies were identified by crossing the following search terms: “metabolic syndrome,” “general population,” “carotid intima‐media thickness,” “carotid atherosclerosis,” “carotid damage,” “carotid plaque,” and “ultrasonography.”

References from relevant studies were screened for supplementary articles. We included any observational (either cross‐sectional or longitudinal), general population‐based study comparing the extent of subclinical atherosclerosis, as assessed by echo‐measured carotid IMT and plaque detection, in patients with MetS compared with their non‐MetS counterparts, without restriction concerning MetS definition and IMT measurement or definition of plaque. Studies were excluded if: (1) they did not specifically deal with the adult general population (ie, MetS in specific settings such as diabetes mellitus, chronic renal disease, cardiac disease, or pediatric patients); (2) subclinical atherosclerosis was not assessed by both carotid IMT and plaque; and (3) actual numbers on IMT and plaque prevalence according to presence or absence of MetS were not provided (ie, data reported according to the components of MetS or narrative description of changes in IMT over time). Titles and abstracts were screened independently by two authors (CS and EG) who discarded all studies not relevant for the topic. Case reports, reviews, editorials, and letters were excluded from qualitative analyses but screened for potential additional references. Two authors (CS and EG) independently assessed the retrieved abstracts and the full text of these studies to determine eligibility according to the inclusion criteria. A third reviewer (CC) solved disagreements on study judgments. Data extraction was performed by one reviewer (CS) and independently verified by another reviewer (EG).

Only updated or the largest reports were considered when multiple publications by the same research group were found to avoid double counting patients.

The flow diagram of the selection process is shown in Figure 1. A total of 3051 potentially relevant references were initially retrieved. By screening titles and abstracts, a total of 2926 citations were excluded because of search overlap, wrong population or topic, and review article design.

Figure 1.

Schematic flow chart for the selection of studies. IMT indicates intima‐media thickness.

Among the 125 studies selected for full text examination, 117 studies were excluded because data on IMT (n = 47) and plaque (n = 13) were not available, no quantitative comparisons with individuals without MetS were performed (n = 8), the wrong population was studied (n = 44), or because they were editorials/reviews (n = 5). A total of eight full studies (19 696 participants) were therefore reviewed in detail and included in the analysis.15, 16, 17, 18, 19, 20, 21, 22

2.2. Statistical analysis

The primary aim of the meta‐analysis was to compare subclinical alterations in carotid structure, expressed as a continuous variable (ie, common carotid IMT) or categorical variable (ie, prevalence of carotid plaque) as assessed by ultrasonography in patients with MetS compared with their counterparts without MetS. To this purpose, a pooled analysis of common carotid IMT was performed using fixed or random effects meta‐analysis by Comprehensive Meta‐Analysis Version 2 (Biostat). To calculate the average prevalence of carotid plaque in the pooled population, we considered the occurrence of plaque as an event rate. Standard means difference with 95% confidence interval (CIs) were calculated to evaluate the statistical difference of the above‐mentioned continuous variables between patients with and without Mets. The limit of statistical significance was set at P < .05. Demographic and clinical data provided by selected studies are expressed as absolute numbers, percentage, mean ± standard deviation, or mean ± standard error. Heterogeneity was estimated using I ², Q, and Tau² values; random effect models were applied when the heterogeneity across studies was high (I ² > 75). Publication bias was assessed using the funnel plot method according to the trim and fill test. Observed and adjusted values, their lower and upper limits, and the classic fail‐safe N (ie, the number of studies that would be required to nullify the effect) were calculated.

3. RESULTS

Table 1 shows the main characteristics of the analyzed studies, including sample size, mean age, sex distribution, prevalence rates of MetS, mean common carotid IMT, and prevalence rates of carotid plaque. Three of eight studies provided data according to sex‐based analysis.18, 19, 22

Table 1.

Summary of studies reporting data on carotid IMT and plaque in patients with MetS compared with their non‐MetS counterparts

First author and reference	Sex or men, %	Sample size, No.	Age, y	MetS, %	IMT MetS, μm	IMT, no MetS, μm	Plaque, MetS, %	Plaque, no MetS, %
Ahluwalia15	53	896	35–65	15	640 ± 110	580 ± 90	25.4	11.2
Kovaite16	46	186	46–65	33	900 ± 390	700 ± 570	49.2	23.2
Paras17	48	796	47 ± 9	17	680 ± 481	650 ± 915	55.0	50.0
Escobedo18	Men	5291	25–64	20	695 ± 116	645 ± 166	9.5	6.4
	Women	6211	25–64	22	682 ± 113	631 ± 177	11.2	6.8
Lee19	Men	634	66 ± 8	15	760 ± 143	745 ± 141	38.1	37.8
	Women	1096	63 ± 8	31	733 ± 139	685 ± 126	20.6	14.5
Della‐Morte20	39	1133	65 ± 9	49	920 ± 90	920 ± 90	74.4	72.2
Antonini‐Canterin21	62	479	59 ± 12	17	880 ± 160	810 ± 190	59.0	37.0
Herder22	Men	1442	56 ± 9	19	870 ± 170	830 ± 170	49.3	42.1
	Women	1532	57 ± 10	18	850 ± 160	770 ± 150	48.4	31.0

Abbreviations: IMT, intima‐media thickness; MetS, metabolic syndrome.

Data are shown as absolute numbers, percentages, or mean ± standard deviation.

Overall, 19 696 patients (22.2% with MetS) of both sexes were included in eight studies (total sample size range, 186–11 502) performed in three geographical areas (Europe, 4; United States, 3; Asia, 1). Of note, 13 431 patients (68.1%) were examined in the United States, 4535 (23.1%) in Europe, and 1730 in Asia (8.8%).

The mean age ranged from 47 ± 9 to 66 ± 8 years (data provided by five of eight studies),17, 19 and a total of 9047 participants (45.9%) were men.

Patients included in selected studies were apparently healthy, with no history of heart or vascular diseases. Prevalence of MetS among studies varied from 15%15 to 49%.20

Ultrasonographic assessment of carotid IMT and plaque as markers of subclinical atherosclerosis associated with MetS was the primary aim of all studies.

3.1. Definition of MetS and assessment of carotid atherosclerosis by ultrasonography

Participants were classified as having MetS as defined by the National Cholesterol Education Program Adult Treatment Panel III in all but one study.16 Two or more alternative definitions of MetS based on International Diabetes Federation World Health Organization recommendations were used in two studies.15, 17

In all of the studies, IMT was measured at the common carotid artery level and calculated on a two‐dimensional longitudinal section as the distance between the leading edge of the lumen‐intima echo and the leading edge of the media‐adventitia echo in a plaque‐free site at 10 to 20 mm proximal to the carotid bulb by using high‐resolution ultrasound systems (Table 2).

Table 2.

Ultrasound methodology in studies reporting data on carotid IMT and plaque in patients with MetS compared with their non‐MetS counterparts

First author and reference	Device	Probe, MHz	Site of measurement	Distance from bifurcation	Manual/computerized
Ahluwalia15	NA	NA	Far wall	NA	Manual
Kovaite16	General Electric LOGIQ 700 system (Milwaukee, WI)	12	Near and far wall	10 mm	Manual
Paras17	NA	10	Far wall	Within 20 mm	Manual
Escobedo18	NA	7.5	Far wall	NA	Computerized
Lee19	Medison SONOACE 9900 (Seoul, Korea)	7.5	Far wall	10 mm	Manual
Della‐Morte20	General Electric LOGIQ 700 system (Milwaukee, WI)	9–13	Near and far wall	10–20 mm	Computerized
Antonini‐Canterin21	NA	7.5–10	Near and far wall	10–15 mm	Manual
Herder22	Siemens Acuson Xp10 128 (Parkway, PA)	7.5	Near and far wall	10 mm	Computerized

Abbreviations: IMT, intima‐media thickness; MetS, metabolic syndrome; NA, not available.

IMT was measured exclusively in the far wall of both common carotid arteries in four of eight studies.15, 17, 18, 19 In the remaining studies, IMT was defined as the average IMT value of the near and far wall of both common carotid arteries. Full or semiautomated electrocardiographically triggered measurements of IMT were performed in three studies.

All studies investigated the presence of plaque in the right and left common carotid artery and internal carotid arteries (including carotid bulbs). Seven of eight studies provided details on carotid plaque definition. In two studies, plaque was defined as a focal lesion thicker than 1.2 mm20 or 1.3 mm.16 In the remaining studies, focal lesions were defined as any focal protrusion into the lumen of the vessel at least 50% to 100% greater than the surrounding wall thickness.

3.2. Carotid IMT and plaque in patients with and without MetS

A total of 4372 patients who met the diagnostic criteria for MetS were compared with patients without MetS (n = 15 324).

In five studies reporting unified data of both sexes, mean carotid IMT ranged from 640 ± 110 μm15 to 920 ± 90 μm20 in MetS and from 580 ± 90 μm15 to 920 ± 90 μm20 in patients without Mets. In the remaining three studies reporting data on carotid IMT by sex, IMT tended to be higher in men than in women regardless of MetS status.

Mean common carotid IMT was 788 ± 47 μm in all patients with MetS examined in eight studies and 727 ± 44 μm in their non‐MetS counterparts.

Figure 2 reports the results of the meta‐analysis from all selected studies: standard means difference of common carotid IMT was positive in favor of patients with MetS (0.28 ± 0.06; CI, 0.16–0.40 [P = .00003]).

Figure 2.

Forest plot for unadjusted standard means difference of common carotid intima‐media thickness (IMT) in patients with and without metabolic syndrome (MS) (P < .0001). Data from eight studies. CI indicates confidence interval. Random model, I ² = 0.86, Q = 51.7, Tau²=0.021.

Overall, 1228 of 4372 patients with MetS (28.1%) and 2804 of 15 324 patients without MetS (18.3%) were found to have at least a discrete carotid plaque.

As reported in Figure 3, all but one study²⁸ revealed higher prevalence rates of carotid plaque in participants with MetS compared with their non‐MetS counterparts (odds ratio, 1.61; CI, 1.29–2.01 [P < .0001]).

Figure 3.

Odd ratio and confidence interval (CI) for the prevalence of carotid plaque in 4372 patients with metabolic syndrome (MS) vs 15 324 patients without MS (data from eight studies). Random model, I ² = 0.79, Q = 32.9, Tau² = 0.072.

A funnel plot excluded the presence of relevant publication bias of studies comparing carotid IMT (Figure S1) and plaque prevalence (Figure S2) in patients with and without MetS. Adjustment for publication bias did not abolish the difference in carotid IMT and plaque prevalence between the groups.

4. DISCUSSION

The present meta‐analysis performed on eight studies published during the past decade provides comprehensive and updated information on the association between MetS and two well‐established markers of subclinical carotid damage such as arterial wall thickening and plaque, as assessed by high‐resolution ultrasonography in a large multiethnic pooled sample (>19 500 participants) of patients in the general population from three continental areas.

The main findings of our investigation were as follows: (1) common carotid IMT was higher in participants with MetS than in those without it; (2) increased IMT in patients with MetS was paralleled by an increased prevalence of carotid plaques; and (3) differences in carotid IMT and plaque between patients with MetS and controls were unaffected by publication bias or single study effect. Many aspects of our data and related issues deserve to be further addressed.

First, the pooled prevalence of MetS in our analysis approximates prevalence estimates obtained in population‐based surveys in the United States (23%–27%). In contrast, lower prevalence rates have been reported in South European population samples. For, instance, MetS prevalence determined in 2013 patients from a Northern Italian population aged 25 to 74 years in the PAMELA (Pressioni Arteriose Monitorate E Loro Associazioni) study, an epidemiologic trial designed to determine normal values and prognostic significance of ambulatory and home BP in the population, was approximately 16%.23 In that survey, office BP elevation was found to be the most frequent (95.4%) and blood glucose abnormality the least frequent (31.5%) component. Prevalence of the syndrome was slightly greater in men than in women (17.6% vs 14.8%). In line with the PAMELA study, two of the three reports included in the present meta‐analysis and providing data by sex, documented a higher frequency of MetS in women.18, 19

Second, our meta‐analysis supports the view that patients with MetS have a more pronounced burden of subclinical arterial damage of established prognostic value, as reflected by a thicker common carotid IMT and higher prevalence of focal lesions. Of note, in all but one study included in the meta‐analysis, the difference in IMT and plaque between participants with and without MetS was corrected for major confounding factors such as age, sex, total cholesterol, physical activity, smoking status, and alcohol intake. In the only study in which data adjustment for potential confounders was not explicitly mentioned in the methods, average age and sex distribution were superimposable as in the remaining studies.21 Therefore, the quality of the statistical methodology of selected studies supports the conclusion that association between MetS and vascular damage is a consistent unfavorable biological phenomenon. Furthermore, no evidence of publication bias or of a single study effect significantly affecting the main results was present. In particular, average common carotid IMT of the pooled population was 61 μm higher (+9%) in patients with MetS than in their non‐MetS counterparts. Moreover, the risk of having a carotid plaque was approximately 61% higher in participants with MetS. Thus, both phenotypes of vascular damage appear to be more pronounced in the presence of MetS.

Although plaque formation is a dynamic process involving a complex cascade of metabolic/inflammatory events, and carotid thickening represents a more adaptive hypertrophy of the media layer rather than an atherosclerotic lesion, our findings are in line with the concept that both phenotypes coexist as a result of common pathophysiological mechanisms. For instance, the investigators of the Northern Manhattan Study, a population‐based observational survey including 1788 participants, have shown that each 10‐μm increase in baseline carotid IMT was associated with 1.72‐fold increased odds of plaque presence, after adjusting for demographics and vascular risk factors.12 Our meta‐analysis suggests that MetS simultaneously promotes vascular hypertrophy and carotid plaques. Unfortunately, our results do not identify which of the two phenotypes is preferably associated with MetS.

Third, to quantify the burden of MetS on subclinical target organ damage at the community level, we excluded from our analyses heterogeneous groups characterized by different degrees of vascular injury, such as cohorts of patients with hypertension, diabetes mellitus, and coronary and chronic kidney disease, and analyzed only studies conducted in members of the general population across a wide range of age. This selective approach, in our opinion, constitutes a strength of the present analysis, as the bias related to different cardiovascular risk profiles among the clinical settings was minimized.

Fourth, our results indicate a large variability in carotid plaque prevalence across studies (approximately from 10% to 70%). This heterogeneity is likely related to diagnostic criteria and to demographic/clinical characteristics of the samples, such as age, sex, body size, ethnicity, and prevalence of the MetS. As for the ultrasonographic method, carotid plaque was defined according to as many as six different criteria; this means that only two studies included in this meta‐analysis applied the same criterion. This crucial aspect partly explains the marked difference among studies in carotid atherosclerosis prevalence both in patients with MetS and in controls.

A comprehensive discussion addressing the pathophysiology of carotid damage in the MetS was beyond the aim of the present review. It can be assumed, however, that the synergic impact of increased mechanical stress, such as circumferential wall stress and flow‐mediated shear stress caused by high BP with metabolic‐driven alterations of vessel properties (ie, impaired nitric oxide release from endothelium, lipid accumulation, and glycation of matrix proteins) may result in substantial changes in carotid artery structure and function such as wall thickening, plaque formation, and arterial stiffening.

5. STUDY LIMITATIONS AND STRENGTHS

The limitations and strengths of our meta‐analysis merit recognition. All data included in the present analysis were cross‐sectional and a causal relationship between MetS and carotid damage could not be definitively established. A further limitation of cross‐sectional studies that evaluate the impact of MetS on target organ damage is the scarce evidence supporting the persistence of the syndrome over time.24 Ultrasonographic methods varied at each study site in terms of equipment, type (manual or computerized), site of IMT, and plaque measurement, as well as definition of plaque. Definition of MetS based on the presence of at least three components identifies patients with different clusters. Scuteri and colleagues25 have shown that not all clusters of MetS components are equally associated with subclinical carotid damage. Furthermore, the number of MetS components has been reported to predict subclinical atherosclerosis more reliably than the binary presence/absence of MetS.26

The strengths are represented by the large number of patients included in our analysis and homogeneity of the clinic setting (ie, general population) and definition of MetS. Finally, parameters of carotid subclinical damage were adjusted for major confounders in seven of the eight studies including approximately 98% of the pooled population.

6. CONCLUSIONS

The present meta‐analysis shows that MetS is associated with both ultrasonographic phenotypes of subclinical carotid damage, ie, medial wall thickening and plaques. This finding is consistent with the view that hemodynamic and nonhemodynamic factors clustering in the MetS may promote vascular hypertrophy and focal lesions. Therefore, to prevent carotid atherosclerosis, a biomarker of increased cardiovascular risk, prevention, and reversal of the MetS should be planned at the community level.

DISCLOSURE

The authors report no conflicts of interest.

Supporting information

 

 

Cuspidi C, Sala C, Tadic M, Gherbesi E, Grassi G, Mancia G. Association of metabolic syndrome with carotid thickening and plaque in the general population: A meta‐analysis. J Clin Hypertens, 2018;20:4–10. 10.1111/jch.13138