Hierarchical Analysis of Hypertension with the Polymorphic Variant of the ACE Gene and Other Risk Factors in the Elderly

Hypertension is one of the main clinical conditions harming the health of elderly individuals, and in hypertensive elderly individuals, the cardiovascular risk is higher than that in normotensive ones of similar age.3 Genetic and environmental factors are associated with the chronic maintenance of elevated blood pressure (BP) levels, and this correlation in the hypertensive population is 30% and 70%, respectively.4 Hypertension may be considered a polygenic clinical condition in which changes in multiple genes interact with each other and with various other risk factors.5 The angiotensin-converting enzyme (ACE), a key component of the renin-angiotensin system 52 International Journal of Cardiovascular Sciences. 2017;30(1):52-60

(RAS), converts the decapeptide angiotensin I into the octapeptide angiotensin II, the most powerful circulating vasoconstrictor in the human body. This way, ACE plays a fundamental role in BP control. 6 Intraindividual serological levels of ACE remain stable, but its interindividual variation is high and is attributed to polymorphic ACE gene variants (DD > DI > II). 7 It should be pointed out the lack of recent studies evaluating the association of these polymorphic variants with hypertension exclusively in elderly individuals. 8 In addition to genetic factors, other factors have been reported as being a risk for hypertension: age above 60 years, gender (postmenopausal women), ethnicity (non-white skin color), diet, sedentary lifestyle, obesity, alcoholism, and smoking, among others. 9 The interaction of these various factors over time contributes to the development and worsening of hypertension, and the greater the number of risk factors to which individuals are exposed, the higher their chance of becoming hypertensive. 10 Therefore, the objective of this study was to assess the prevalence of hypertension among elderly individuals, and its possible correlation with the polymorphic variant I/D of the ACE gene and with other associated risk factors.

Study design
This was an epidemiological, cross-sectional, and population study including elderly individuals aged 60 years or more, resident and domiciled in the municipality of Ibiaí, north of Minas Gerais (MG), Brazil.

Procedure and instrument for data collection
The data collection was carried out between 2011 and 2012. The elderly individuals were evaluated by health professionals who were trained and calibrated (kappa = 0.63) to the demographic, socioeconomic, b e h a v i o r a l , a n t h r o p o m e t r i c , a n d l a b o r a t o r y characteristics, and genotypic variant of the polymorphic ACE. According to the 2010 census, a total of 505 elderly individuals aged 60 or more years were registered in the Primary Care Information System. Of the 479 names obtained from the Family Health Strategy (FHS) list, 449 (93.74%) were found. The registration also included 27 elderly people whose names were not listed, but who were nonetheless also added, yielding a total of 476 elderly individuals. Of these, 387 (84.31%) answered the questions and/or were assessed in regards to the main variable of interest in this study (hypertension).
The measurement of BP levels was conducted by the auscultatory technique with a calibrated sphygmomanometer. The BP measurements were performed with the individual seated and after a 5-minute rest. The individuals were questioned about being with an empty bladder, having avoided physical activity, food, smoking, ingestion of alcoholic beverages or coffee (at least 30 minutes before the measurement) and if they had taken drugs that could interfere with the mechanisms of BP regulation. 11 A total of three BP measurements were obtained; the first was discarded and the mean of the second and third measurements was considered. The patients were instructed about the procedure beforehand, with an explanation of its steps and objectives. The diagnosis of hypertension was established in the presence of a mean diastolic BP ≥ 90 mmHg and/ or mean systolic BP ≥ 140 mmHg and/or use of systemic antihypertensive medication. 11 To achieve the objective of the study, the variables were grouped into individual, socioeconomic, behavioral, metabolic, and health-related, as described below: Individual: age (60 to 69 years and ≥ 70 years), sex (male and female), self-declared skin color (White/ Yellow and Brown/Black), I/D polymorphic variant (rs 4646994) of the ACE gene (II, DI, and DD).
Socioeconomic: marital status (with a partner [married and stable union] and without a partner [single, widowed/divorced]), education (≥ 4 years and 0 to 3 years), household crowding (less than one person/room and more than one person/room), material goods (at least one material goods [house and/or automotive] or lack of material goods), work status (yes [employed, self-employed, informal work, retired/employed] and no [retired/without work, does not work/never worked, and unemployed]), and income (assessed by the monthly individual income in Brazilian real).
Behavioral disorders: dietary factors (consumption of fruits and vegetables [more than 3 servings/ day and less than 3 servings/day] and salt intake [one portion per day and more than one portion per day]), 12 smoking (non-smokers and smokers), alcoholism (never consumed versus consumption of alcohol), physical activity (active [moderately active, active, and very active] and not active [inactive and little active]). Metabolic and health-related: optimal total cholesterol (< 200 mg/dL), optimal high-density lipoprotein (HDL; women ≥ 50 mg/dL and men ≥ 40 mg/dL), optimal low-density lipoprotein (LDL; < 130 mg/dL), optimal triglycerides (< 150 mg/dL), and optimal waistto-hip ratio (WHR; women < 0.85 and men < 0.95). We evaluated the glomerular filtration rate (GFR) using the Cockcroft-Gault formula: [GFR (mL/min) = (140 -age) x weight (kg)/72 x serum creatinine (multiply by 0.85 for female individuals)], in which values < 60 mL/min/1.73 m² are considered abnormal and ≥ 60 mL/min/1.73 m² are considered normal. 13 The anthropometric variables weight and height were used to calculate the body mass index (BMI), with results < 25 kg/m 2 considered as optimal.

ACE polymorphism evaluation -polymerase chain reaction
In order to investigate the genotypic frequencies of the polymorphic variant I/D (rs 4646994) of the ACE gene (II, DI, and DD), saliva samples were obtained. Gene samples were obtained with a swab from buccal mucosa smears, which was stored at a temperature of -20 ºC in test tubes with Krebs solution. The samples were isolated with silica particles, which absorb the DNA. Subsequently, the DNA was washed to remove impurities and suspended with TE buffer. The genomic DNA was amplified with the polymerase chain reaction (PCR) technique.

Statistical analysis
All data were tabulated and analyzed using the program Statistical Package for the Social Sciences for Windows, version 20.0. The variables investigated were described as distributions of frequencies. In the univariate analysis, we estimated the crude odds ratios values with 95% confidence intervals. The variables with descriptive level p ≤ 0.20 were selected for the multivariate analysis.
In the multivariate analysis, we adopted the model of hierarchical logistic regression, and the order of entry of the blocks was determined from a theoretical model and involved known factors associated with hypertension. 11 For the hierarchical analysis in this study, we followed the diagram shown in Figure 1, which guided the order of entry of the variables blocks in the model. The block "individual factors" was the first to be included in the model and remained as a factor of adjustment for the other variables. We then included "socioeconomic factors", and only those presenting a descriptive level p < 0.05 remained in the model after adjustment for the variables in the "individual factors" block. After that, we included the "behavioral habits" block and maintained in the model only those variables with descriptive level p < 0.05 after adjustment for "individual factors" and "socioeconomic factors". For inclusion of the remaining blocks, we adopted the same procedure of the previous blocks.

Ethical aspects
The project was conducted in accordance with the Resolution 196/96 of the National Health Council of the Ministry of Health and presented a favorable opinion of the Research Ethics Committee at the Universidade Estadual de Montes Claros (CEP/Unimontes) with the unified opinion number 2903/11. The elderly subjects signed a free and informed consent form agreeing to participate in the study.

Discussion
The hierarchical approach is an alternative applicable Our findings corroborate those of another Brazilian epidemiological study that showed a prevalence of hypertension of approximately 65% in the elderly, with a higher prevalence among women older than 75 years, potentially reaching 80%. 15 In the present research, this clinical condition was similar to that observed in other population studies. 9,16 The greatest chance of presenting hypertension was observed in women (p = 0.009). The literature highlights a greater prevalence of hypertension among elder women when compared with men at the same age. 9 A previous study had already demonstrated this greater probability of association with hypertension. 17 Up until menopause, women are hemodynamically younger than men at the same age and, thus, less vulnerable to BP increases. However, after menopause and due to weight gain and hormonal changes, women become more vulnerable to increased BP. 18 Despite the direct and linear relationship of BP with increased age, this association did not remain, possibly due to the fact that the age range in this population was more homogeneous. The literature highlights a greater chance of hypertension in Blacks. 19 In the present study, the skin color was not associated with hypertension, which was comparable with another study that included this variable in its analyses. 17 This shows that the extension of the impact of the Brazilian miscegenation on hypertension is still unclear.
A recent study on the genomic ancestry in our country observed that the Brazilian genome is much more uniform than otherwise expected with its large racial mixture over the past two centuries. 20    After control for socioeconomic factors, the variable gender lost the strength of its association (p = 0.067), while marital status (p = 0.032) was shown to be statistically significant. There are few studies specifically correlating marital status with hypertension. 21,22 In the present study, elderly individuals without a partner had a greater chance of hypertension and one of the possible explanations may be the fact that the elderly living alone has a greater probability of having emotional disturbances, influencing the BP increase. 23 However, it is worth noting that in another study, the greatest risk of hypertension was observed among married individuals or among those living with a partner. 17 The WHR presented a positive correlation (p = 0.050) with hypertension. There are few studies specifically investigating the relationship between different fat indicators (body/corporal and/or abdominal) with hypertension. 24,25 Excess weight among older individuals is still a reality, especially among women. 26 Munaretti et al. 24 published the first populationbased and domici liar study verifying an association between hypertension and anthropometric indicators of fat (corporal and abdominal) in elderly Brazilians. In that study, fat anthropometric indicators (corporal and abdominal) were associated with hypertension, corroborating the results of other studies conducted with individuals from different po pulations and age groups, which found that excess fat, regardless of the anthropometric indicator used, is one of the main risk factors for hypertension. [27][28][29] Although no consensus exists on an adequate criteria and values to define obesity among the elderly, BMI is the indicator most often used in epi demiological studies at this age range. 28 However, some authors have suggested that BMI alone is unable to identify the association between body fat and hypertension. 30 Similarly, in a study conducted with 9,936 men and 12,154 women aged 45 to 79 years, the indicators WHR and BMI showed an association with hypertension. 28 Such fact can probably be explained by the physiological changes that occur in obese individuals, such as the activation of the sympathetic nervous system, the SRA, endothelial dysfunction, and insulin resistance (increasing tubular sodium reabsorption). 28 It should be emphasized that the predominance of obesity tends to be higher in lower socioeconomic classes, as was the case in the population investigated. 30 The fact that the other study variables showed no association strength with hypertension may be due to the sample size and the possibilities of selection biases and/or information involving confounding factors and occasional findings from the analysis, which constitute limitations of observational studies in general.

Conclusion
Although the polymorphic ACE variant did not influence the prevalence of hypertension in elderly individuals, this condition was associated with risk factors such as female gender and marital status (absence of a partner), and modifiable risk factors such as excessive salt consumption and alterations in WHR and body weight. The study proposes the adoption of a healthy lifestyle to prevent the disease, since modifiable factors correlated significantly with hypertension.
In addition, studies addressing this issue in population groups at greatest risk are fundamental in improving the knowledge of genetic polymorphisms in the etiology of the disease.