Diversity of metabolic syndrome criteria in association with cardiovascular diseases – a family medicine-based investigation

Background

Metabolic syndrome (MetS) is defined as a cluster of risk factors [1,2] that identifies persons with increased risk of cardiovascular disease (CVD). The prediction of CVD onset does not have to be better than the Framingham score and Systematic Coronary Risk Evaluation (SCORE), based on the main factors of cardiovascular risk (age, sex, systolic blood pressure, smoking, total, and HDL and LDL cholesterol) [3,4]. The epidemiological proportions of MetS prevalence support its importance in the past 5 decades in countries where the population was found to have increased food consumption and insufficient physical activity [5]. Recent studies indicate that MetS is inferior in establishing rules for the prediction of either type 2 diabetes mellitus (DM2) or coronary heart disease (CHD) [6]. MetS, as a predictor of CVD, has also been studied due to the existence of different definitions [7]. The most frequently mentioned definitions are the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults, Adult Treatment Panel III, the new World Health Organization (WHO) definition, the modified International Diabetes Federation (IDF) definition [8–10], and that of the American College of Endocrinology (AACE) [11]. The WHO and the European Group for the Study of Insulin Resistance Guidelines (EGIR) [12] have been primarily proposed for the needs of research, and the NCEP and IDF definitions were designed for clinical use. This indicates that the role of MetS as a CVD predictor is still uncertain and insufficiently researched. Differences in the definition of MetS create confusion in the timely detection of persons with increased cardiovascular risk in general practitioners’ practices.

Although there are studies about MetS in the Republic of Croatia [13–19], there have been few population studies that analyzed the predictive relevance of the association of each definition with CVD been sparse. We aimed to determine if, defined on the basis of the WHO, NCEP, or IDF criteria, MetS was associated with an increased risk of developing CVD in the population under a family physicians’ care, living in various life conditions (region, rural vs. urban), and depending on age and sex.

Material and Methods

STUDY DESIGN:

This study was conducted within the randomized clinical research of the Cardiovascular Risk and Intervention Study in Croatia-family medicine (CRISIC-fm) in the Republic of Croatia and was registered as a clinical trial (International Standard Randomized Controlled Trial Number Register – ISRCTN31857696). It was a two-phase study that ran from May through July 2008.

PARTICIPANTS:

The study included 3245 participants of both sexes, aged ≥40. Exclusion criteria were the inability to communicate due to conditions such as dysphasia, aphasia, serious dementia or psychiatric decompensation, and an expected survival of less than 6 months.

SAMPLING:

The sample was two-stage, disproportionate, and mixed-sex. The first phase was to establish a quadruple stratified representative sample of family medicine physicians according to regions (coastal and continental), population size (up to 3999 inhabitants; 4000 to 9999; 10 000 to 29 999; 30 000 to 89 999; and 90 000 and over), rural area (<4000 inhabitants), urban area (>4000 inhabitants), and the number of the insured individuals contracted between family medicine practitioners and the Croatian Health Insurance (HZZO) in 2007 (up to 1399; 1400 to 1799; and ≥1800).

For each initially contacted physician, a reserve sample of 4 more GPs was made, according to the 4-fold stratum. If a GP declined to participate, the nearest GP from a reserve sample was invited. All GPs were verbally informed in detail about the study and then signed a consent form to participate in the research. The sample size needed to reach 95% confidence interval and the desired power of statistical tests. Of the 82 GPs invited to participate in this study, 64 of them accepted (78%), of which 5 declined participation at first follow-up, so the total number of GPs in the final sample was 59.

In the second stage, each GP chose a systematic, disproportionate sample of the first 55 patients who visited the practice for any reason from the day the study began, and who met the inclusion criteria and confirmed their consent by signing a written informed consent. All the participating GPs included the same number of patients (N=55), regardless of the total number of insured persons they have contracted with CIHI, and the total number of patients from the target population they examined. This was corrected by post-hoc weight factors prior to statistical analysis (Figure 1).

MEASUREMENTS AND DEFINITION OF METABOLIC SYNDROME:

A standardized, validated CRISIC-fm questionnaire with 140 questions, designed for the study, was administered. Participants’ height and weight were measured twice (standardized, identical measuring scales) as well as their waist circumference (WC) and hip circumference (with plasticized inelastic tape measure), and their waist-hip ratio (WHR) was then calculated. Mean arterial pressure (mercury sphygmomanometer) and pulse frequency were assessed. A blood sample for the analysis of total cholesterol concentration, HDL and LDL cholesterol, triglycerides, fasting glucose (FG), and uric acid [20] was taken from each participant. The modified WHO, IDF, and NCEP definitions (Table 1) were used for the MetS analysis. Coronary disease was defined by previous myocardial infarction, angina pectoris, and/or revascularisation of coronary arteries, and cerebrovascular disease was defined by previous cerebral insult and/or transitory ischaemic attack. Peripheral artery disease (PAD) was defined by anamnestic data of intermittent claudication (fatigue, cramping and pain during walking) and <0.8 of ankle brachial index (ABI). Overweight was defined as BMI ≥25, and obesity was defined as BMI ≥30 kg/m2[21].

Bias: We did not examine microalbuminuria, which is part of the WHO definition, due to its unavailability at primary healthcare level.

STATISTICAL ANALYSIS:

Descriptive statistics procedures were used to describe basic sample characteristics: continental/urban, continental/rural, coastal/urban, and coastal/rural, with differences examined using the χ² test. An independent t-test was used to analyze the differences in quantitative values of participants’ characteristics between regions and population sizes. We used by binary logistic regression to analyze the relationship between MetS and CVD, as well the association between MetS according to NCEP and socio-behavioral settings. All values were interpreted at the significance level P<0.05. Statistical analysis was performed using SPSS for Windows (11.5, SPSS Inc., Chicago, IL, 2002).

Results

Fifty-nine family medicine doctors joined the study (response rate, 71%) with 2467 participants (38.1% men and 61.9% women) (response rate, 78%). In Figure 2 prevalence of MetS, according to the NCEP, IDF definitions was 38.7% (15.9% men and 22.7% women), 45.9% (18.6% men and 27.3% women); in the continental region 28.4%, 33.9%; in the coastal region 10.2%, 10.9%; in urban residents 26.2%, 31.5%; and in rural residents 12.4%, 14.4%. According to the NCEP criteria, there were significantly more obese people in rural areas (P=0.002) with elevated LDL cholesterol (P<0.001) and systolic pressure (P=0.029) compared to people in urban areas. Participants were older in the urban areas (P=0.022) of the continental region, and they were older (P=0.048) and had a higher waist circumference (P=0.035) in rural residents of the coastal region. Persons with MetS engaged in moderate physical activity more in continental rural than in urban s. The prevalence of prehypertension was significantly higher in urban residents of both coastal and continental regions. MetS was most prevalent in persons aged 50 to 59, except in coastal urban residents, in whom MetS was prevalent in older age (Tables 2 and 3). Association between MetS according to NCEP and socio-behavioural settings was significant (χ2=103.4; df=8; P<0.001) and correctly classified with 60% of subjects. Older age, male sex, and residence in the continental area were positively associated with MetS diagnosis according to NCEP, and current smoking and Mediterranean diet adherence were protective factors (Table 4). The association between MetS and cardiovascular disease was determined according to all 3 definitions, but the NCEP definition was the most sensitive (Table 5). Persons with MetS were, according to NCEP, at a 2.48 times higher risk of coronary disease. The association between MetS and cerebrovascular disease, according to the modified WHO definition, has not been supported by evidence, and according to NCEP and IDF definitions, it was equally significant. The association between MetS and peripheral vascular disease was strongest when expressed according to the modified WHO definition, a weaker association according to the NCEP definition, but the association was negligible according to the IDF definition. Age and sex also show a statistically significant correlation with CVD (Table 6).

Discussion

LIMITATIONS:

There are 2 main limitations to this study. The sample’s subjects were patients registered by GPs, which does not entirely correspond to the general population sample. In addition, we did not examine microalbuminuria, which is part of the WHO definition, due to its unavailability at the primary health care level in Croatia.

Conclusions

Although all 3 definitions of MetS were associated with a higher risk for CHD, association was the greatest and most consistent when using the NCEP definition. According to that definition, the prevalence of MetS is higher in the continental region (possibly due to different diet type) and in urban residents (probably due to less physical activity and more sedentary lifestyle), in older people, and in males.