Introduction
The World Health Organization (WHO) defines cardiovascular diseases (CVD) as a group of disorders of the heart and blood vessels. It is considered the leading cause of death worldwide. More than three quarters of CVD deaths occur in low- and middle-income countries and it is expected that by 2030 they will cause around 23.6 million deaths1. Most CVD can be prevented using strategies that cover the entire population and impact behavioral risk factors, such as tobacco use, unhealthy diets that induce obesity, physical inactivity, or harmful alcohol consumption. There are several scales to determine the cardiovascular risk of each patient, most of them requiring data that involve some cost and waiting time such as lipid profile and certain inflammatory markers. This means that, in patients from low-income countries, there is greater difficulty in estimating this risk and the interventions are performed late. In addition, access to health services is more difficult and scarcer, which means that, in many cases, this evaluation is not even carried out2. That is why, in these cases, it is pertinent to use anthropometric measures as a cost-effective alternative to predict cardiovascular risk in all population groups regardless of their socioeconomic status. Other advantages of these techniques are their easy application and the fact that the measurement protocols are part of the training of health personnel in countries such as Colombia. The objective of the present study, then, is to show the capacity of some anthropometric measures for the prediction of cardiovascular risk in an adult population of a Colombian State University.
Materials and methods
A prospective study was conducted with a sample of employees, both genders, from the University of Caldas in Manizales (Caldas, Colombia). The study consisted of two phases, the first in 2014 and the next in 2018. Ninety-eight people over 30 years of age were included who agreed to participate in this study and formed the informed consent form. They did not have a previous diagnosis of the pathologies under evaluation such as coronary heart disease, stroke, peripheral vascular disease, high blood pressure (HBP ―Hypertension―), type 2 diabetes mellitus (DM), heart failure, angina pectoris, or aneurysm.
The study variables were gender, age, height, weight, percentage body fat (%BF), body mass index (BMI), blood pressure (BP), percentage of hand grip strength (%HGS) of the dominant side, midline level waist circumference (MLW), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), fruit and vegetable portions (FVP) ingested per day, and cigarette smoking were recorded.
In addition, the calculation of the risk of suffering a cardiovascular episode in the next 10 years, fatal or not, was made taking into account the WHO/ISH Region B risk prediction table (AMR B) for the contexts, in which it is not possible to measure blood cholesterol in order to compare it with our prediction based on anthropometric variables. This table includes: sex, age, systolic BP, smoking, and presence or absence of DM (the WHO, 2008).
Four years later (2018) of the first evaluation, people were contacted again by telephone and were asked if they had had new diagnoses of DM or CVD during this period (2014-2018).
The weight was measured with an ICOB® digital scale model PP20.00. For the stature or size, a SECA® Heigthronic TM digital stadiometer was used. The measurements were made according to the protocol of Lohman et al.3. The BMI was determined with the weight ratio in kilograms (kg), over the height in meters squared (m²) and the WHO classification was used4.
A stethoscope and a LORD® brand tensiometer were used for the measurement of BP under the recommended semiological standards. Subsequently, individuals were categorized according to the classification proposed by JOINT VII5.
The %HGS was evaluated three times with a Baseline® digital dynamometer on the dominant side; each time, with a duration of 5 seconds (sec) with 30 sec rest intervals between them. The highest measurement of the three was compared with reference data according to Mathioweitz, 19856. A value ≥ 85% is classified as normal grip strength; in the same way, a lower result was considered as altered.
With the patient standing and an inelastic Lord® tape measure, the midpoint between the costal ridge and the iliac crest was obtained and the abdominal circumference was measured at this level. Values ≥ 90 cm in men and ≥ 80 cm in women were considered altered7. For the calculation of the WHtR, the value at the level of the largest hip circumference was recorded. The reference values used as altered were: ≥ 0.9 in men and ≥ 0.8 in women8. An altered WHtR value was considered as ≥ 0.5 for both genders9.
The %BF was estimated with a digital Skyndex System® adipometer from the sum of the triceps, biceps, subscapular, and suprailiac subcutaneous fat folds using the Durnin and Womersley formula10. The reference value considered normal in women was < 30% and < 20% in men.
Fruit and vegetable consumption was recorded as the number of servings of fruit and vegetables ingested per day (FVP). Adequate consumption was considered when it was ≥ 5 servings of 1 or both types of food11.
Characteristics of the subjects included were described as average and standard deviation. A logistic regression with progressive stepwise method was performed using the SPPS version 24 program, which licensed to the University of Caldas. The explanatory variables were gender, age, tobacco consumption, %HGS, BMI, WHR, WHtR, %BF, FVP, systolic BP (SBP), and diastolic BP (DBP). This analysis was univariate. Response variables included the appearance of coronary heart disease, stroke, peripheral vascular disease, HBP, DM, or heart failure after the first evaluation in 2014. Composite unbalance was defined as the probability of having CVD or DM if an alteration was found in the anthropometric variables. Odds ratio and 95% CI were calculated to estimate the probability of suffering DM and CVD associated to the evaluated variables. Significance was stated as p < 0.05.
Results
In 2018, when patients were contacted a second time, 15 people (three men and 12 women) were excluded, because their telephone numbers or location had changed and contact with them was not possible. Finally, 83 people were included in the study, of the 98 people evaluated in 2014.
Table 1 describes the characteristics of the subjects included for the final analysis of 2018. Table 2 shows the results of the values of %HGS, WHR, MLW, %BF, FVP, WHtR, and BP. Only 7.2% of the subjects evaluated reported smoking.
Sex | n | Indicator | Age (years) | Height (cm) | Weigth (Kg) | BMI (kg/m2) | %BF | MLW (Cm) | WHtR | WHR (Cm) | %HGS | SBP (mmhg) | DBP (mmhg) | FVP (portion/day) |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Men | 16 | Average | 50.6 | 166.9 | 72.3 | 26 | 29.5 | 90 | 0.5 | 0.9 | 76.8 | 114.25 | 76 | 2.3 |
Standard deviation | 7.5 | 5.6 | 9.9 | 3.2 | 4.9 | 8.6 | 0.1 | 0.1 | 16.2 | 12.88 | 11.1 | 1.6 | ||
Women | 67 | Average | 43.2 | 155.7 | 62.1 | 25.6 | 37.4 | 78.9 | 0.5 | 0.8 | 89.8 | 112.25 | 71.03 | 2.19 |
Standard deviation | 9.8 | 5.2 | 10.7 | 4.4 | 4.4 | 10.1 | 0.1 | 0.1 | 14.7 | 13.74 | 8.78 | 2 | ||
Total | 83 | Average | 44.6 | 157.85 | 64.04 | 25.67 | 35.91 | 81.7 | 0.5 | 0.8 | 87.3 | 112.6 | 72 | 2.2 |
Standard deviation | 9.82 | 6.87 | 11.29 | 4.18 | 5.46 | 10.75 | 0.06 | 0.1 | 15.77 | 13.53 | 9.4 | 1.9 |
BMI: Body mass index, MLW: midline level waist circumference. WHtR: waist-to-height ratio. WHR: waist-to-hip ratio, %HGS: percentage of handgrip strength, PAS: Systolic blood pressure, PAD: diastolic blood pressure, and PFV: fruit portions and vegetables.
Sex | N | Indicator | %BF | WHtR | WHR | MLW | %HGS | FVP | BP |
---|---|---|---|---|---|---|---|---|---|
Men | 16 | Normal | 37.5% (6) | 6.25% (1) | 56.25% (9) | 56.25% (9) | 6.25% (1) | 6.25% (1) | 68.75% (11) |
Pathological | 62.5% (10) | 93.75% (15) | 43.75% (7) | 43.75% (7) | 93.75% (15) | 93.75% (15) | 31.25% (5) | ||
Women | 67 | Normal | 65.67% (44) | 11.94% (8) | 79.1% (53) | 64.18% (43) | 20.9% (14) | 17.9% (12) | 76.12% (51) |
Pathological | 34.33% (23) | 88.06% (59) | 20.9% (14) | 35.82% (24) | 79.1% (53) | 82.1% (55) | 23.88% (16) | ||
Total | 83 | Normal | 60.24% (50) | 10.84% (9) | 74.7% (62) | 62.65% (52) | 18.07% (15) | 15.66% (13) | 74.7% (62) |
Pathological | 39.76% (33) | 89.16% (74) | 25.3% (21) | 37.35% (31) | 81.93% (68) | 84.34% (70) | 25.3% (21) |
%HGS: Percentage of handgrip strength, WHtR: waist-to-height ratio, WHR: waist-to-hip ratio, MLW: midline level waist circumference, %BF: percentage body fat, FVP: number of fruit and vegetable portions, and BP: blood pressure.
According to the WHO/ISH AMR B risk prediction table WHO 200812, the probability of suffering a cardiovascular episode, fatal or not, in a period of 10 years, was less than 10% for all evaluated, indicating a very low cardiovascular risk.
Despite this, the information obtained concluded that 17 individuals had been diagnosed with HBP, four people had been diagnosed with DM, one other person had coronary infarction, and another individual had peripheral arterial disease from the first evaluation.
After the statistical analysis, it was established the MLW as a predictor of DM, this measurement had a statistical significance of 0.010 and an odds ratio of 1.214 (95% CI) indicating that the probability of suffering diabetes increased by 12.14% for each unit that increased the measure of the hip with respect to an individual with a lower value.
With regard to the HBP response variable, statistical significance of 0.028 was found for the %BF as a predictor of this pathology. The odds ratio in this case was 1.137 (95% CI). This result indicated a probability of suffering from HBP of 11.37% for each unit of percentage of fat above the value established as normal.
In this study, the other variables showed no relationship with the onset of DM or HBP. Nor was a relationship found between these anthropometric variables and any of the other CVD taken into account in this study.
Discussion
Our results suggest that there is an association between the variable MLW and DM, as well as the %BF and HBP. The other variables were not related to the pathologies that were taken into account in the present analysis.
That is, in less than 10 years after the first evaluation, 23 subjects (27.7%) had already presented a CVD or had been diagnosed with DM. This leads us to suggest that in a 10-year follow-up, this percentage could greatly exceed this figure.
%BF and BP
The findings of our study are related to others previously published, as is the case of the relationship between the %BF and BP. In a cross-sectional study, conducted by Thang et al.,13 50.436 individuals between men and women were analyzed. The population was divided into five groups according to their BP status, as follows: normotensive individuals, with undetected HBP, with controlled HBP, with uncontrolled HBP and those with untreated HBP. Subsequently, each group was compared with %BF. The authors concluded that BP was inversely associated with muscle mass, as well as positively with fat mass13.
Trinidades et al.14 evaluated 408 individuals between the ages of 20 and 59 years old in Brazil and determined that the prevalence of HBP was 23.03%, being higher among women. Likewise, they found a significant association between HBP and smoking, obesity, DM, and dyslipidemia. About 40.38% of hypertensive individuals 50-59 years old had five or more cardiovascular risk factors. Another study, conducted by Alvarez Cortez et al., showed that those who have HBP were at risk of suffering an acute myocardial infarction approximately 3.8 times more compared to those who did not suffer from this condition. Thus, if HBP was controlled, 73.5% of cardiovascular accidents in the studied population would be avoided15.
MLW and DM
A study conducted in Bogotá (Colombia) by Buendía et al.16 evaluated 2200 patients attending endocrinology consultation and found, using a multivariate analysis, that male sex and increased MLW were independent risk factors related to diabetes. In this study, the same reference values were considered as in our analysis (90 cm for men and 80 cm for women)16.
A study, including 982 patients, determined that there was a positive correlation between MLW and blood glucose, uric acid and insulinemia values. MLW in both genders was the variable with the highest predictive power of hyperglycemia, with waist cutoff points of 86.75 cm in men and 80.5 cm in women. Above this value, the risk of type 2 diabetes increased 8.7 times17.
Others variables and CVD/DM
In the present study, 30.76% of the individuals evaluated had a %HGS altered. Despite this, no relationship was found with the subsequent onset of CVD. However, a systematic review highlights that the measurement of %HGS has prognostic value for cardiovascular disease, as well as being an important indicator of fragility and vulnerability. The authors concluded that this measurement could be a simple and useful tool to identify individuals at risk of CVD that requires closer clinical attention18.
The Prospective Urban Rural Epidemiology study, conducted in 2015, Corsi et al.,19 confirmed the prognostic importance of the %HGS in adult populations of all social strata in 17 countries. It was found that low grip strength was associated with a high lethality rate for a variety of incident diseases, including CVD. This result suggests that %HGS could be an important marker of the ability to resist or recover from a disease. These data support the measurement of muscle strength as a key indicator of frailty20.
The %BF is not only associated with an increased risk of HBP as evidenced in a study that included 182 subjects with a mean age of 38 years. The authors demonstrated that excess BF increased the individual risk of CVD and metabolic syndrome in the population of subjects with no history of CVD. They also demonstrated that the %BF was significantly related to biochemical variables such as serum lipids, fasting blood glucose, and BP21.
We found that 84.34% of those evaluated do not consume the recommended daily servings of fruits and vegetables. It has been described that diets rich in fruit and vegetables are associated with a reduced risk of chronic diseases. In our country, healthy eating guidelines include the recommendation to consume at least 5 VFP daily, although the exact amount of recommended portions and descriptions of portion sizes vary by country22. A meta-analysis suggests an association between increased consumption of FVP intake and reduced risk of CVD and stroke23. Another study showed that the intake of more than 8 FVP for 8 weeks decreased SBP by 2.8 mm Hg and DPB in 1.1 mm Hg24.
An investigation conducted by Valenzuela et al.24 included 998 people between 22 and 28 years old. In this study, the MLW and the WHtR index were directly associated with both systolic and diastolic BP. The WHtR and the WHR have been classified as better predictors of cardiovascular risk than the BMI; however, in our study, no significant relationship was found between them and the subsequent presentation of HBP25.
The study by Muñoz-Muñoz et al.26 supports that this relationship is that of who determined that high WHtR was strongly associated with increased BP, excess weight, increased MLW, fasting hyperglycemia, hypertriglyceridemia, and LDL-cholesterol > 2.6 mmol. Likewise, elevated WHtR was associated with the presence of metabolic syndrome26.
The study by Luengo-Pérez et al.27 showed a much more important relationship. The research showed that in men up to 50 years, the WHR remains the one that shows the highest correlation with cardiovascular risk, followed by WHtR. However, for men 50 years of age and older, none of the indices showed a significant correlation. In the case of women under 50, the WHtR and BMI showed a greater correlation with cardiovascular risk. For women 50 years of age and older, only WHR presented a significant correlation27.
The association between obesity and HBP is a perfectly established fact, although not all obese people develop HBP. However, obesity has been indicated as a contributing factor to the increase in BP, being more frequent among overweight people than among those with normal weight. This relationship is evidenced in the study by Fernandez Díaz et al.28 who demonstrated a predominance of obesity in hypertensive patients who also had high values in the %BP. This relationship was not observed in our study28.
Despite the current evidence, based on other research, some of the variables evaluated were not related to the probability of developing CVD. We believe that this could be related to the small sample that we used for the present analysis and for the moderate period of time that the study lasted.
Conclusion
High MLW values may indicate an increased risk of developing DM. Furthermore, high values in the %BF would be associated with a higher risk of HBP, 4 years after an initial evaluation.
The evaluated population is at risk of developing CVD, although the WHO/ISH AMR, risk prediction tables have classified it as a very low risk.
The MLW and %BF measurements are economical, fast, and safe to assess cardiovascular risk in individuals and should be used in routine medical consultations. We recommend its application, especially in remote regions of large urban centers, where access to technology and diagnostic techniques are more difficult.