Prevalencia del síndrome metabólico, índice de insulino resistencia y niveles de leptina y hormona tiroidea en la población general de Mérida (Venezuela).

Prevalence of the metabolic syndrome, insulin resistance index, leptin and thyroid hormone levels in the general population
of Mérida (Venezuela). 

Euderruh Uzcátegui¹, Lenin Valery², Lilia Uzcátegui³, Roald Gómez Pérez³, David Marquina⁴ and Trino Baptista⁴. 

¹Department of Psychiatry, Los Andes University Medical School. Mérida, Venezuela.

²Department of Accounting and Finances, Economic and Social School,
Los Andes University. Mérida, Venezuela.

³Department of Endocrinology, Los Andes University Medical School. Mérida, Venezuela.

⁴Department of Physiology, Los Andes University Medical School. Mérida, Venezuela. 

Resumen. El síndrome metabólico (SM) es un factor de riesgo significativo en la ocurrencia de eventos cardiovasculares. Sin embargo, existe poca información sobre su frecuencia en Venezuela. En la presente investigación transversal, cuantificamos la prevalencia del SM es una muestra probabilística de 274 sujetos con edad mayor o igual a 18 años provenientes del municipio Libertador de la cuidad de Mérida, Venezuela. Como resultados secundarios, se cuantificaron los niveles de leptina, hormonas tiroideas (T₄ libre y TSH), y el índice de resistencia a la insulina (HOMA2-RI). La frecuencia de SM (porcentaje ± intervalo de confianza del 95%) de acuerdo a varios criterios diagnósticos fue la siguiente: Panel para el Colesterol (original): 27,4% (22,1-32,7); Panel para el Colesterol (modificado): 31,8% (26,3-37,3); Federación Internacional de Diabetes: 40,9% (35,1-46,7); Asociación Latino-Americana de Diabetes: 27% (21,7- 32,3) y criterios Venezolanos: 31,8% (26,3-37,3). Al utilizar la mayoría de los criterios, el SM fue más frecuente en hombres que en mujeres. La prevalencia estimada de diabetes mellitus tipo 2 fue 2,9% tanto mediante el auto-reporte como mediante la detección de glicemia en ayunas > 126 mg/dL. Se detectaron niveles anormales (por encima del percentil 95) del índice HOMA2-IR, T₄ libre y TSH en el 4,5%, 4,4% y 5,1% de la muestra respectivamente. Niveles de T₄ y TSH por debajo del percentil 5 se detectaron en el 4,4% y 4,7% de sujetos respectivamente. Los valores del HOMA2-RI y hormonas tiroideas también se presentan con el fin de permitir comparaciones con futuros estudios en poblaciones clínicas específicas. Mientras se realizan estudios adicionales en Venezuela sobre las diversas definiciones del SM, recomendamos analizar y publicar los resultados científicos locales con todos los criterios disponibles con el fin de permitir la comparación con los estudios que ya están disponibles. 

Palabras clave:  epidemiología, panel para el colesterol, Federación Internacional de Diabetes, Asociación Latino-Americana de Diabetes, criterios venezolanos. 

Recibido: 27-07-2014 Aceptado: 30-01-2015 

INTRODUCTION 

The metabolic syndrome (MetSyn) refers to the association of five factors (abdominal obesity, hypertension and abnormal glucose, triglyceride and high density lipoprotein cholesterol [HDL-c] levels) associated with increased risk for the development of type 2 diabetes (Type 2 DM), heart disease and stroke (1). The MetSyn is a significant predictive factor in relation to these diseases, even though the contribution of each of its constituting variables to elevated health risks is controversial (2, 3). 

Some studies have evaluated the prevalence of MetSyn in the Venezuelan general population (GP) and in specific groups (4-12). In one early study conducted in Zulia state, Flórez et al. (4), reported an age-adjusted prevalence of MetSyn of 31.2% among 3.108 subjects aged 20 or over. In this study as in most of those conducted in Venezuela, the authors used the National Cholesterol Education Program (NCEP)/ Adult Treatment Panel III (ATP III) (11) criteria to diagnose the MetSyn. These criteria have been criticized for using cut-off points too high in waist circumference (WC) and fasting glucose. In particular, the WC limits appeared inadequate for Hispanic populations. Therefore, new criteria for Latin-America, including Venezuela that lowered the cut-off points for WC and glucose levels have been formulated (13-17). 

In the present study, we evaluated the prevalence of MetSyn and its constituting variables according to five diagnostic criteria (13-17) in a probabilistic sample of adults from the GP in Mérida, a university town up in the Venezuelan Andes. As secondary outcomes, in all the subjects we quantified the homeostatic model for the insulin resistance index (HOMA2-IR) (18, 19), and the serum levels of tetraiodothyroxine (T₄), thyroid-stimulating hormone (TSH). We also assessed the serum leptin levels in a probabilistic subsample of the interviewed subjects. This secondary data aim to register normative data by including abnormal values located out of the 5-10^th and 90-95^th percentiles. 

SUBJECTS AND METHODS 

We used a multistage design to obtain a probabilistic sample out of a total of 204.879 subjects aged ³18 years according to the 2001 Venezuelan National Census living at the Libertador municipality in Mérida city. The primary selection units were randomly selected segments of the Libertador borough, where individual houses were randomly selected as well. 

Sample size was calculated according to a simple random model with a two-tailed 95% confidence interval, a sampling error of 5% and a 24% of estimated prevalence. A minimum sample size of 280 subjects was estimated and invited to enter the study. 

For the assessment of leptin levels, we selected every fifth consecutive subject from a randomly organized list. 

The basic sampling units were households that were randomly chosen in each selected district segment. In each basic unit, one subject out of all the adults that were present at the time of the visit was randomly selected. The inclusion criteria were: (1) the subjects had to be 18 or more years old and permanent residents in the area; and (2) they must have signed an informed consent for voluntary participation. 

The selected subjects were assessed in a single session in fasting conditions where the following evaluations were conducted: a) anthropometric variables were measured in light clothes. Abdominal circumference was measured in an equidistant point to the lower ribcage border and the anterior-superior iliac spine, at the end of a normal expiration; b) blood pressure was calculated as the average of two measures in standing, sitting and horizontal positions in a calm environment; c) a cubital venous blood sample was taken and immediately processed for posterior serum analysis in a single assay, and d) a detailed interview for collecting personal, demographic and medical data. 

Analysis 

Glucose and lipids levels were measured with an enzymatic method from Human (Berlin, Germany). Insulin and leptin levels were assessed in duplicate using enzyme-linked immunosorbent assay (DRG, Germany). Thyroid hormones were quantified in by chemiluminiscence (Siemens, Germany). All the analyses were conducted in a single assay, with intra essay variations £10%. 

The body mass index (BMI) was calculated as follows: weight (kg)/height (m²). Obesity was defined as a BMI ³30 kg/m². The insulin resistance index (HOMA2-IR) was calculated with an online computer program (19). 

The MetSyn was diagnosed according to the following criteria (Table I): 1) National Cholesterol Education Panel (NCEP, original and modified) (13, 15); 2) International Diabetes Federation (IDF, 14); 3) Latin-American Diabetes Association (ALAD, 16), and; 4) harmonized criteria of the Venezuelan Society of Endocrinology and Metabolism (17). 

The frequency of the MetSyn and its constituting variables was expressed as percentage and 95% confidence interval. Crude and age-adjusted frequencies according to the most recent Venezuelan census and based on the procedure used by Florez et al. (4), are provided. The influence of the demographic features on the prevalence of the syndrome was investigated by binary logistic regression analysis. Age and race were analyzed with the two-tailed t test for unrelated samples and chi-squared test, respectively. Bivariate correlation was conducted with the Pearson “r” coefficient analysis. Results were considered significant when p £0.05. 

RESULTS 

The study was conducted between January and March of 2006 in 274 subjects (63.9% females), representing 99.3% of the calculated sample size (n = 280). Four individuals did not attend the examination session, and two were not included in the final analysis due to incomplete anthropometric data. The present sample had a 10.5 % excess of female representation (the local census reported 53.4% of women). This was due to a higher proportion of women present at home when the survey was conducted. 

Age, race, personal history of hypertension, dyslipidemia and sedentary life were similar in females and males, but alcohol abuse and tobacco consumption were significantly more frequent in males (Table II). 

Metabolic syndrome 

The age-adjusted and unadjusted frequency of the MetSyn diverged according to the diagnostic criteria, ranging from 40.9% to 27% in the unadjusted figures, as follows (Table III): IDF > NECP (modified) > Venezuelan criteria > NECP (original) > ALAD. Glucose levels, WC and systolic pressure were the MetSyn constituting variables showing the highest variability, whereas triglyceride, HDL-c and blood pressure levels showed small or no variability (Table III). 

TABLE II ]]> UNWEIGHTED SOCIODEMOGRAPHIC CHARACTERISTICS 

The percentage of subjects with abnormal values of the MetSyn constituting variables who were aware or under specific pharmacological treatment was: fasting glucose (0/24 = 0.0%); HDL-c (41/231 = 17.7%); triglycerides (21/105 = 20%; hypertension: 41/87 = 47.21%). In the whole sample, 2.9% reported being treated for type 2 diabetes mellitus or were found with fasting glucose levels above 126 mg/dL. This may be cautiously considered as an estimation of the prevalence of that disease. 

Influence of demographic variables 

The MetSyn frequency was higher in males than in females, but the figures did not reach statistical significance after adjusting for the other co-variables: IDF (45.5 % vs. 38.3%), NECP modified (42.4% vs. 26.9%), Venezuelan criteria (40.4% vs. 25.1%), NECP original (29.3% vs. 26.3%), ALAD (34.3% vs. 22.9%) (Table IV). The frequency increased with age and BMI, and such a difference was statistically significant with almost all the diagnostic criteria (Table IV). For example, with the Venezuelan criteria, the age-related figures were: 18-29 yrs. (21.2%), 30-39 yrs. (15.8%), 40-49 yrs. (34.6%), 50-59 yrs. (47.3%), 60-69 yrs. (59.1%), >70 yrs. (61.5%). As regards BMI, the MetSyn frequencies were: <20 (0.0%), 20-24.9 (6.6%), 25-29.9 (36.5%), 30-34.9 (63.5%), ³35 (85.7%). As for race, no significant differences were observed with any criteria (Table IV). 

No subject had either a previous diagnosis of, or was under treatment for a thyroid disease. Abnormally low T₄ levels (below the 10^th or 5^th percentiles) were detected in 24 (8.8%) and 12 (4.4%) subjects of the whole sample respectively. Abnormally high T₄ levels (above the 90^th or 95^th percentiles) were detected in 26 (9.5%) and 12 (4.4%) subjects respectively. Abnormally low TSH levels (below the 10^th or 5^th percentiles) were detected in 24 (8.8%) and 13 (4.7%) subjects of the whole sample respectively. Abnormally high TSH levels (above the 90^th or 95^th percentiles) were detected in 27 (9.9%) and 14 (5.1%) subjects respectively. 

When combining both data sets analyzed with the 5-95^th percentile cut-off limits, hyperthyroidism was detected in approximately 4.4-4.7% of the population, whereas hypothyroidism was observed in approximately 4.4-5.1% of the sample. Seven out of fourteen subjects with high TSH levels (50% of the abnormal TSH group and 2.5% of the whole sample) had normal T4 levels. Therefore, this value should be considered with caution as an estimation of subclinical hypothyroidism. 

The frequency of MetSyn was similar in subjects with or without abnormal TSH levels: 25.9% vs. 27.1 %, c² (1) = 0.000, p = 1.0. 

Correlation analysis 

Several positive and negative correlations were observed among the anthropometric and biochemical variables (Table VII). This information is presented for comparative analysis in future studies. 

TABLE V  LEPTIN SERUM LEVELS (pg/mL) 

DISCUSSION 

This study described the frequency of MetSyn and related variables according to several diagnostic criteria in Merida city. On the one hand, it will allow comparisons with studies conducted in other countries and in Venezuelan cities with different ethnic and nutritional features. On the other hand, the results presented in this GP sample will be compared with those obtained in a set of studies conducted in specific clinical populations in this city, particularly in drug-treated subjects with psychiatric and/or neurological disorders. 

The study has the limitation of a higher proportion of females than that reported in the local census (63.9% vs. 53.4%). Besides, the blood pressure quantification included standing and lying down measures in addition to the sitting position, the formers not being included in previous studies. And lastly, the sample was not adjusted by race and socioeconomic status. These features hinder the strict comparison of this data with previous studies. 

The MetSyn is a pathological condition affecting almost one third of the general population in Latin America (11). By including anthropometric, blood pressure, lipid and carbohydrate variables in its definition and given its predictive value for cardiovascular and metabolic disorders, the MetSyn is the focus of numerous preventive and therapeutic strategies. 

The age-adjusted frequency of abnormal values of the following variables of the MetSyn was higher in the Zulia state (4) than in Mérida: WC (42.9% vs. 27.5%), glycaemia (10.9% vs. 4.9%) and blood pressure (38.1% vs. 26.6%), whereas triglycerides (32.3% vs. 36.4%) and HDL-c levels (65.3% vs. 84.2%) were higher in our sample. 

Like Flórez et al. (4), we found that the MetSyn was more frequent in males, but sex differences did not reach statistical significance after covariate analysis. In fact, the difference reached its lowest value in the analysis using the original NECP criteria (29.3% vs. 26.3%, see Results section). 

Flórez et al., (4) indeed reported that the frequencies of abnormal WC and HDL-C levels were significantly higher in women than in men, whereas an opposite pattern was reported for glucose, triglyceride and blood pressure values. In our analysis with the original NCEP criteria and age and BMI as covariates in the logistic regression, the following figures were observed (females vs. males): triglycerides (38.9% vs. 41.1%, p = 0.9), HDL-c (86.3% vs. 81.8%, p = 0.3), blood pressure (26.9% vs. 41.4%, p = 0.06), WC (35.4% vs. 21.2%, p = 0.000) and glycaemia (2.3% vs.12.1%, p = 0.005). 

The study of Becerra et al., (9) reported a higher frequency of MetSyn than we did (38% and 43% according to the ATP-III and IDF criteria respectively) and a slightly higher frequency in females than in males is worth discussing, because it was conducted in the same state of this investigation. However, that study was carried out in a rural area, of a high altitude (more than 2.000 meters of difference between both villages) and with a higher proportion of ethnically mixed subjects (100% vs. 5.5% in our study). It is thus possible that these differences in methods explain the studies’ discrepancies. 

While it is not the objective of this study, it is worthwhile comparing the Venezuelan and the Latin-American criteria to enrich the discussion about which criteria should be adhered to in future comparative studies. The cut-off points for WC, glucose and HDL-c levels are similar in both lists, whereas the systolic pressure has a higher cut-off point (³135 vs. ³130 mm/Hg), and the triglyceride definition includes 150 mg/dL in the Venezuelan criteria. An important difference between the two criteria is that the Latin-American criteria require abdominal obesity to be mandatory in order to diagnose MetSyn. Since the difference in the triglyceride criteria is negligible, and since the diastolic pressure criteria, in contrast to the systolic ones, are similar, it appears that it is the abdominal obesity requirement that better discriminates between both sets of criteria. Therefore, the Venezuelan criteria consider abdominal obesity as having the same diagnostic weight as the lipid level, carbohydrate and blood pressure abnormalities, whereas the ALAD criteria propose a more prominent role for the WC status. 

To clarify which set of criteria predicts the highest health risk is a major scientific and clinical issue. A robust finding is that the risk of type 2 diabetes and CVD significantly increases as more components of the MetSyn definitions are included, for example from 3 to 4 or more (21). 

According to most MetSyn definitions, the relative risk for incident diabetes is higher than for cardiovascular diseases (CVD) (22) Impaired fasting glucose (IFG) is the strongest predictor for type 2 diabetes development; the other criteria components are similar or better than IFG to predict CVD (23). 

An index of central obesity (WC/ height) was tested in China as a substitute for the single WC assessment according to the NECP and IDF criteria, but its risk predictive value for CVD was not better than that of the standard WC criteria (24). Besides, empirically-derived MetSyn criteria were not better predictors of CVD than did the already existing definitions (25). These last authors concluded that “the existing metabolic syndrome definitions will be sufficient for use in clinical practice and research studies should the MetSyn debate conclude in a recommendation for its continued use” (25). 

Several significant correlations were observed in the bivariate analysis of the MetSyn constituting variables. This reinforces the postulated physiological association among them and indirectly supports the conceptual validity of the syndrome. 

The thyroid hormone and the HOMA-IR data presented here are expected to be used for future comparisons in ongoing studies. By combining the frequencies of abnormal TSH and T4 levels, we report an estimated prevalence of hypothyroidism of 4.4-4.7% (2.5% of the subclinical type) and hyperthyroidism of 4.4-5.1%. These values must be cautiously interpreted because they were based only on laboratory data and no clinical examination was conducted. Unfortunately, there are no other published data about thyroid dysfunction in the Venezuelan adult GP that would allow a comparison with the results reported here. The third National Health and Nutrition Examination Survey in the United States reported hypothyroidism in 4.6% of the sample (0.3% clinical and 4.3% subclinical) and hyperthyroidism in 1.3% (0.5% clinical and 0.7% subclinical) (26). We expect to compare our data with future studies conducted in Venezuela. 

Finally, 2.9% of the subjects either reported type 2 diabetes mellitus or were found to have glucose levels similar or above 126 mg/dL. Hoffman et al., (27) reported newly diagnosed type 2 diabetes in 6.5% of Hispanic subjects but only 46% of these subjects had fasting glucose levels ³126 mg/dL. Hence, the values reported here may underestimate the actual prevalence of that disease. 

This descriptive study corroborates that significant differences exist in the frequency of MetSyn and its constituting variables according to the diagnostic criteria selected. The Latin American Diabetes Association criteria give a prominent status to abdominal obesity, whereas the Harmonized Venezuelan criteria attribute the same weight to the five MetSyn variables. This may account for the differences observed between both criteria schedules. This is why the authors recommend analyzing the MetSyn frequency data with both (or more) diagnostic criteria in the research about the construct validity of the MetSyn that is currently being conducted in Venezuela. 

ACKNOWLEDGMENTS 

Part of this study was the thesis of Dr. E. Uzcátegui to apply for the Doctorate in Medical Sciences at Zulia University, Maracaibo, Venezuela, and was funded by the Scientific, Humanistic, Artistic and Technological Council of Los Andes University, Mérida (Venezuela), Grant M-838-05- 07-C, and by the National Council for Science and Technology (Caracas, Venezuela), Grant 2005-000-384. We are grateful to Françoise Salager-Meyer for her editorial assistance. 

REFERENCES 

1.  Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes 1988; 37:1595-1607.         [ Links ] 

2.  Schmidt C, Bergström GML. The metabolic syndrome predicts cardiovascular events: results of a 13-year follow-up in initially healthy 58-year-old men. Metabol Synd Relat Dis 2012, 10(6): 394-399.         [ Links ] 

3.  Dusseault-Belanger F, Cohen AA, Hivert MF, Courteau J, Vanasse A. Validating metabolic syndrome through principal component analysis in a medically diverse, realistic cohort. Metab Syndr Relat Disord 2013; 11(1): 21-28.         [ Links ] 

4.  Florez H, Silva E, Fernandez V, Ryder E, Sulbaran T, Campos G, Calmon G, Clavel E, Castillo-Florez S, Goldberg R. Prevalence and risk factors associated with the metabolic syndrome and dyslipidemia in White, Black, Amerindian and Mixed Hispanics in Zulia State, Venezuela. Diabetes Res Clin Pract 2005; 69(1):63-77.         [ Links ] 

5.  Ryder E, Silva E, Sulbaran T, Fernandez V, Campos G, Calmon G, Clavell E, Raleigh X, Florez H. Black Hispanics have a worse cardiovascular risk profile than mixed Hispanics in Venezuela. Invest Clin 2007; 48: 45-55.         [ Links ] 

6.  Sirit Y, Acero C, Bellorin M, Portillo R. Metabolic syndrome and other factors cardiovascular risk in workers of a plant of vinyl polychloride. Rev Salud Pública (Bogotá) 2008; 10: 239-249.         [ Links ] 

7.  Ruiz-Fernandez N, Espinoza M, Barrios E, Reigosa A. Cardiometabolic factors in a comunity located at Valencia city, Venezuela. Rev Salud Pública (Bogotá) 2009; 11:383-394.         [ Links ] 

8.  Escobedo J, Schargrodsky H, Champagne B, Silva H, Boissonnet CP, Vinueza R, Torres M, Hernandez R, Wilson E. Prevalence of the metabolic syndrome in Latin America and its association with sub-clinical carotid atherosclerosis: the CARMELA cross sectional study. Cardiovasc Diabetol. 2009; 26; 8:52.         [ Links ] 

9.  Becerra A, Arata- Bellabarba G, Velázquez-Maldonado E. Prevalencia de síndrome metabólico en la población urbana de Mucuchíes, Mérida. Rev Venez Endrocrinol Metab 2009; 7 (3): 16-22.         [ Links ] 

10.  Bermudez V, Marcano RP, Cano C, Arraiz N, Amell A, Cabrera M, Reyna N, Mengual E, Vega L, Finol F, Luti Y, Sanchez D, Sanchez W, Gonzalez J, Montes J, Cano J, Cano R, Velasco M, Miranda JL. The Maracaibo city metabolic syndrome prevalence study: design and scope. Am J Ther 2010; 17:288-294.         [ Links ] 

11.  Márquez-Sandoval F, Macedo-Ojeda G, Viramontes-Hörner D, Fernández J, Salas Salvadó J, Vizmanos B. The prevalence of metabolic syndrome in Latin América: A systematic review. Public Health Nutr 2011; 14:1702-1713.         [ Links ] 

12.  Bermudez V, Pacheco M, Rojas J, Cordova E, Velazquez R, Carrillo D, Parra MG, Toledo A, Anez R, Fonseca E, Marcano RP, Cano C, Miranda JL. Epidemiologic behavior of obesity in the Maracaibo City metabolic syndrome. PLoS One 2012; 7(4):e35392.         [ Links ] 

13.  Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of 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), J Am Med Assoc 2001; 285: 2486-2496.         [ Links ] 

14.  Alberti KG, Zimmet P, Shaw J. Metabolic Syndrome new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med 2006; 23 (5): 469-480.         [ Links ] 

15.  Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, Fruchart JC, James WP, Loria CM, Smith S. Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009; 120: 1640-1645.         [ Links ] 

16.  Guzman JR, Gonzalez-Chavez A, Aschner P, Bastarrachea R. Consenso Latinoamericano de la Asociacion Latinoamericana de Diabetes (ALAD). Epidemiologia, diagnóstico, control, prevención y tratamiento del síndrome metabólico en adultos. Consensos ALAD. 2010; Vol. XVIII (1): 25-44.         [ Links ] 

17.  Brajkovich IE, Caminos QR, Aure G, Nieto R. Síndrome Metabólico. Sociedad Venezolana de Endocrinología y Metabolismo. Publicación de GlaxoSmithKline Venezuela C.A. 2010. 15-18.         [ Links ] 

18.  Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998; 21(12): 2191-2192.         [ Links ] 

19.  University of Oxford. HOMA calculator. Downloaded on November 1th 2014. (http://www.dtu.ox.ac.uk/homacalculator/download.php).         [ Links ] 

20.  Marquina D, Peña R, Fernández E, Baptista T. Abnormal correlation between serum leptin levels and body mass index may predict metabolic dysfunction irrespective of the psychopharmacological treatment. Int Clin Psychopharmacol 2011; 26 (3): 169-172.         [ Links ] 

21.  Shin JA, Lee JH, Lim SY, Ha HS, Kwon HS, Park YM, Lee WC, Kang MI, Yim HW, Yoon KH, Son HY. Metabolic syndrome as a predictor of type 2 diabetes, and its clinical interpretations and usefulness. J Diab Invest 2013; 4 (4): 334-343.         [ Links ] 

22.  Ford ES, Li C, Sattar N. Metabolic syndrome and incident diabetes: current state of the evidence. Diabetes Care 2008; 31: 1898-1904.         [ Links ] 

23.  Wilson PW, D’Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and Type 2 diabetes mellitus. Circulation 2005; 112: 3066-3072;         [ Links ] 

24.  Luo W, Guo Z, Wu M, Hao C, Zhou Z, Yao X. Index of central obesity as a parameter to replace waist circumference for the definition of metabolic syndrome in predicting cardiovascular disease. J Cardiovasc Med (Hagerstown) 2014; 15 (10): 738-744.         [ Links ] 

25.  Wild RP, McGinn AP, Kim M, Muntner P, Wang D, Cohen HW, Ogorodnikova AD, Reynolds K, Fonseca V. Empirical derivation to improve the definition of the metabolic syndrome in the evaluation of cardiovascular disease risk. Diabetes Care 2011; 34:746-748.         [ Links ] 

26.  Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, Braverman LE. Serum TSH, T4, and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III. J Clin Endocrinol Metab 2002; 87(2):489-499.         [ Links ] 

27.  Hoffmann IS, Alfieri AB, Cubeddu LX. Prevalence of glucose abnormalities and diagnostic value of fasting glucose and fasting insulin in urban Hispanics. Am J Ther 2013; 20 (4): 369-375.         [ Links ]