Valmore Bermúdez1*, Joselyn Rojas1, Juan Salazar1, Robys Gonzalez1, Maria Sofía Martinez1, Carmen Chavez1, Marcos Palacio2,3, Edward Rojas1, Mayela Cabrera1 and Jose Lopez Miranda4
1Endocrine and Metabolic Diseases Research Center, School of Medicine, The University of Zulia, Maracaibo, Zulia - Venezuela
2Fellow of the Advanced Endocrinology Master Program. Alcalá de Henares University. Madrid, Spain
3Medicine Department, IVSS "Dr. Pedro García Clara" Hospital. Ciudad Ojeda, Zulia - Venezuela
4Lipid and Atherosclerosis Unit, Department of Medicine. IMIBIC/Reina Sofia University Hospital/University of Cordoba, and CIBER Fisiopatología Obesidad y Nutrición (CIBEROBN), Instituto de Salud Carlos III, Spain
Received: 17 September, 2014; Accepted: 22 October, 2014; Published: 24 October, 2014
Valmore J. Bermúdez, MD, MSc, MPH, PhD, The University of Zulia, Endocrine and Metabolic Diseases Research Center, 20th Avenue, Maracaibo 4004, Venezuela, Tel/Fax: 58-261-7597279; Email:
Bermudez V, Rojas J, Salazar J, Gonzalez R, Martinez MS, et al. (2014) Coronary Risk Estimation according to the Framingham-Wilson Score and Impaired Fasting Glucose in Adult Subjects from Maracaibo city, Venezuela. Glob J Obes Diabetes Metab Syndr 1(1): 020-029. DOI: 10.17352/2455-8583.000004
© 2014 Bermúdez V, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Impaired fasting glucose; Coronary risk; Framingham-Wilson equation; Prediabetes; Type 2 Diabetes Mellitus
Introduction: Type 2 Diabetes Mellitus (T2DM) is an independent risk factor for coronary artery disease. Nevertheless, the coronary risk rendered by pre-diabetes states such as Impaired Fasting Glucose (IFG) has not been thoroughly explored. The purpose of this study was to evaluate the influence of glycemic status on coronary risk estimated by the recalibrated Framingham-Wilson equation in our population.
Materials and Methods: A total 1,378 of subjects were selected from the Maracaibo City Metabolic Syndrome Prevalence Study. A complete medical history was taken, next to laboratory workup and anthropometric measurement. All subjects were classified according to glycemic status in normal fasting glucose (NG), impaired fasting glucose (IFG), and T2DM. Coronary risk estimation was calculated using the Framingham-Wilson equation recalibrated for our population.
Results: An increase in coronary risk was observed in IFG (p=3.78x10-6) and DM (p=4.34x10-13) when compared to NG; pattern also observed within genders: men (p=1.57x10-4) and women (p=2.37x10-4). Coronary risk also increased according to age in all categories of glycemic status. Within the IFG group, occupational status (p=0.004), smoking (p=2.29x10-5) and alcohol consumption (p=0.013) were associated with higher coronary risk. Finally, subjects with IFG exhibited greater probability for allocation in high coronary risk categories [OR: 1.46 (1.06 - 2.14); p=0.05].
Conclusions: Coronary risk scores from the recalibrated Framingham-Wilson equation were significantly higher in subjects with IFG. In our population, this increase may be influenced by sociodemographic and psychobiological traits, such as marital and occupational status, smoking and alcohol consumption.
Type 2 Diabetes Mellitus (T2DM) is one of the most prevalent diseases worldwide, with over 382 million people affected by the year 2013 , and a solemn prediction that by 2050, 1 out of every 5 adults will be diagnosed as diabetic . In the United States, there are currently 25.8 million people with diabetes, 11,3% being over 20 years of age and 26.9% over the age of 65 . There is also an increase in the annual cost of this disease, which has been calculated to be around 174 million dollars in the United States by 2013 . The burden of this disease has also been observed in Venezuela, with 1.2 million patients by 2013, representing a prevalence of 6,61% .
The situation regarding prediabetes is a growing alarm because by 2010 a total of 79 million people in the United States were diagnosed , and it is estimated that by 2030, 472 million people worldwide will be diagnosed . In our country, 1.5 million people had prediabetes by 2013, representing 8.25% of the general population . The prediabetic state was first recognized in 1997 , encompassing two types of patients: those with impaired fasting glucose (IFG) and those with impaired glucose tolerance (IGT). Currently, this is diagnosed by either glucose measurement during fasting or after an oral glucose tolerance test (75 g) or the glycated hemoglobin percentage (HbA1c) determination . It has been previously established that a prediabetic state precedes the development of full-blown glucose intolerance or diabetes mellitus , therefore prediabetes is a high-risk state , characterized by b-cell dysfunction and insulin resistance, even before hyperglycemia levels reach diabetic levels [8,9]. The risk of developing diabetes is 4-6% when compared with normoglycemic subjects , and year by year this risk increases by a factor of 10 .
Diabetes is a known important risk factor for cardiovascular disease , especially when considering that there is evidence of myocardial damage in recently diagnosed diabetic patients  and in those subjects with non-diabetic hyperglycemia , both related to microvascular dysfunction  and accelerated calcification of arterial vessels . In a clinical study conducted by Selvin et al. , it was revealed that prediabetes and diabetes were associated with subclinical myocardial damage, with cumulative probabilities of elevated troponin T of 6.4% for prediabetes and 10.8% for diabetes patients. Even though prediabetes offers some degree of cardiovascular risk, the two clinical states (IFG and IGT) show different patterns on influence, being linear for IGT and “J-shaped” for IFG , both associated with higher cardiovascular associated morbidity and mortality [18,19].
Consequently, proper assessment of cardiovascular risk is a priority considering the elevated cardiovascular risk observed during the progression of hyperglycemic states towards diabetes. Several cardiovascular scores have been applied to determine cardiovascular risk in subjects with prediabetes and diabetes [20,21], and amongst them is the Framingham-Wilson Score [20,22]. Although this equation has been criticized due to its lack of applicability in populations such as Hispanics [23,24], we have recalibrated this formula to enhance its detection power in order to apply it in our locality [25,26]. In this context, estimation of coronary risk in patients with prediabetes or diabetes should be a basic tool in preventive epidemiology, especially in a population such as ours with high prevalence of obesity  and hyperlipoproteinemia (a) , and physical inactivity .
Therefore, the purpose of this study was to assess the epidemiologic behavior of coronary risk (using the Framingham-Wilson Equation) regarding classification of glycemic status, and the association between glycemic disturbances and risk stratification categories.
Materials and Methods
The Maracaibo City Metabolic Syndrome Prevalence Study (MMSPS) was a cross-sectional study, with a randomized, multi-staged sampling methodology which has been published previously . MMSPS enrolled 2,230 individuals, and for this branch of the study, the following exclusion criteria were applied: those under 30 or older than 74 years of age, and personal history of acute coronary syndrome. The final working sample was 1,378 subjects. After written consent was obtained, each individual was subjected to a complete physical examination and anamnesis, where information concerning ethnic group, occupational status, and educational status was obtained, while socioeconomic status was assessed through the Méndez-Castellano-modified Graffar´s scale .
Framingham-wilson equation calibration for maracaibo city
To calibrate the Framingham-Wilson equations , coefficients from the original formula were employed along with the major coronary event rate (lethal and non-lethal myocardial infarction and angor pectoris) obtained from the 2008 Zulia State Vital Statistics Yearbook . From this data, risk was estimated with the following equation:
where S0(t) is the rate of individuals without coronary events; β iXi is the coefficient that results from the multiplication of each risk factor value by its β constant obtained from the Cox regression model from the Framingham study ; and β iXm is the coefficient rendered by the multiplication of the mean of each risk factor by its β constant, also obtained from the original Framingham study . Afterwards, we calculate S0(t) in two steps using the following formulas [23,24,33,34]:
Step A: H0(t)/FramAll determination
where H0(t)/FramAll represents the rate of all types of coronary events - including silent Myocardial Infarction (MI) and angor pectoris, H0(t)/FramMajor is the rate of lethal and non-lethal symptomatic MI, and t is the time of follow-up, which in our case is 10 years. The coefficients 1.41 for men and 1.91 for women are risk variables obtained in the original Framingham study, and these proportions were assumed to be similar to our population. In this regard, based on local information, H0(t)/FramMajor can be substituted in the equation and H0(t)/FramAll calculated for each gender by resolving the equation as follows:
Obtaining the final results:
Step B: S0(t) determination
At this point, the rate of subjects without coronary events could be calculated through a subtraction, using H0(t)/FramAll values:
Once S0(t) determination was finalized, it was included in the original equation. Final results were classified in 4 categories: Low Risk (<5%), Moderate Risk (5-9.9%), High Risk (10-19.9%) and Very High Risk (≥20%) (32).
Blood pressure was determined using the auscultatory method, with calibrated mercury sphygmomanometers, selecting Korotkoff`s phase I and V as systolic and diastolic blood pressures respectively. Blood pressure was taken with the subject sitting down, feet resting on the ground after a resting period of 15 minutes. The pressure was taken 3 times, with 15 minutes in between each take on two different days. Blood pressure was classified using the Seventh Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7) .
An electrical bioelectric scale was used to obtain weight (Tanita, TBF-310 GS Body Composition Analyzer, Tokyo - Japan). Height was measured using a calibrated metric measurement tape, with the subject standing up barefoot. Body Mass Index (BMI) was calculated using the formula (weight/height2) expressed as kg/m2. Subjects were classified as normal weight, overweight, and obese according to the classification proposed by the WHO .
Diabetes and prediabetes: T2DM was diagnosed using one of the following criteria: 1) previous diagnosis of T2DM; 2) those without such previous diagnosis, but who over the course of this study were found to display fasting glucose levels ≥126 mg/dL in 2 different measurements . Prediabetes was determined evaluating fasting glucose levels, considering IFG when it resulted between ≥100 and <126 mg/dL. Individuals with <100 mg/dL glucose levels were considered normoglycemic (NG). Known T2DM patients were considered to have satisfactory metabolic control when fasting glucose <130 mg/dL.
Smokers were classified as follows : a) Non-Smoker, any individual which has never tried a cigarette or has smoker less than 100 cigarettes in his life; b) Current Smoker, any individual which has smoked over 100 cigarettes in his life, or that refers to current habit, or that has quit smoking less than a year previous to this anamnesis; and c) Past Smoker, any individual that has consumed over 100 cigarettes in his life but has quit the habit over a year previous to this questioning.
During anamnesis, individuals were asked how many drinks were consumed monthly. Those who consumed at least 1 drink per month were considered as Drinkers, and those who denied such habit were catalogued as Non-Drinkers . Daily alcohol intake was calculated using the following formula .
where 0.8 is a constant representing ethanol density in drinks. Based on this quantification estimation, Drinkers were defined as subjects who consumed ≥1 gram of alcohol daily [38,39].
Serum glucose levels were determined following 8-12 hours of fasting through the glucose oxidase technique using automated equipment (Human Gesellschoft Biochemica and Diagnostica MBH, Magdeburg, Germany). Likewise, fasting serum insulin concentration was determined with a commercial kit based on ultrasensitive ELISA methodology (DRG internacional. Inc. USA. New Jersey); insulin detection limit is <1 µU/mL. Lipoprotein(a) [Lp(a)] was measured using the latex turbidimetric method (Human Gesellschaftfür Biochemica and Diagnostica, Germany), in which the presence of Lp(a) in the sample causes agglutination of latex particles coated with antibodies against Lp(a); the cut-off value for the consideration as elevated Lp(a) levels was ≥30 mg/dL .
Qualitative variables were expressed as absolute and relative frequencies, assessed through the χ2 test and the Z test for Proportions. Distribution of continuous variables was evaluated by the Geary’s test. Results were expressed as medians (p25-p75), assessing differences through Mann-Whitney’s U Test or One-Way ANOVA with Bonferroni correction as required. Additionally, an ordinal logistic regression model was constructed, wherein the dependent variable included coronary risk categories for our population (low, moderate, high and very high), while the independent variables selected were: Gender, age groups, ethnic groups, occupational status, socioeconomic status, marital status, alcohol consumption (drinker and non-drinker), BMI, presence of insulin resistance (HOMA2-IR≥2), elevated Lp[a] (Lp[a] ≥30mg/dL) and fasting glycemic status. Regression coefficients (β) were calculated with their corresponding confidence intervals (CI95%), along with Odds Ratios (eβ) and their CI95%; as well as goodness of fit parameters and parallel line testing. The database analysis was done using the Statistical Package for the Social Sciences (SPSS) v. 20 for Windows (IBM Inc. Chicago, IL). Results were considered statistically significant when p<0,05.
General characteristics of the population selected for coronary risk estimation
The overall sample was 2,230 individuals, but after applying the exclusion criteria a total of 1,378 subjects were used in this study, where 55.9% (n=770) were females and 44.1% (n=608) were males, with an arithmetic mean for age of 46.9±10.9 years. Sociodemographic and psychobiological characteristics according to glycemic status are presented in table 1.
Prevalence of type 2 diabetes mellitus and impaired fasting glucose
The overall (n=2,230) prevalence of DM2 in this investigation was 8.4% (n=187), of which 5.8% (n=130) had previous diagnoses of DM2 and 2.6% (n=57) were new cases (Figure 1A). A total of 435 (19.5%) individuals had IFG. However, in the subsample selected for coronary risk estimation (n=1,378), a greater prevalence of DM2 was observed (11.4%; n=157), as well as a larger proportion of subjects with IFG (23.5%; n=324); Figure 1B.