Age-dependent increase in blood pressure in two different Native American communities in Brazil

Original article 1753 Age-dependent increase in blood pressure in two different Native American communities in Brazil Diana Meyerfreunda, Christine P. Gonçalvesa, Roberto S. Cunhaa, Alexandre C. Pereirab, José E. Kriegerb and José G. Milla Objective Cardiovascular risk factors were surveyed in two Indian populations (Guarani, n U 60; Tupinikin, n U 496) and in a non-Indian group (n U 114) living in the same reserve in southeast Brazilian coast. The relationship between an agedependent blood pressure (BP) increase with salt consumption was also investigated. in Guarani (7.5 W 1.4) than in Tupinikins (8.8 W 2.2) and non-Indians (8.4 W 2.0). Multiple regression analysis showed that age and waist-to-hip ratio (WHR) were independent predictors of SBP and DBP (r2 U 0.44) in Tupinikins, whereas the WHR was the unique independent predictor of BP variability in Guaranis (r2 U 0.22). Methods Overnight (12 h) urine was collected to evaluate Na excretion. Fasting glucose and lipids, anthropometry, BP, ECG and carotid-femoral pulse wave velocity (PWV) were measured in a clinic visit. Participation (318 men/352 women, age 20–94 years; mean U 37.6 W 14.9 years) comprised 80% of the eligible population. Conclusion Lower BP levels in Guaranis cannot be explained by low salt intake observed in other primitive populations. J Hypertens 27:1753–1760 Q 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins. Results The prevalence of hypertension, diabetes and high cholesterol was similar in Tupinikins and in non-Indians and higher than in Guaranis. The prevalence of smoking and obesity was higher in the latter group. Hypertension and diabetes were detected in only one individual of the Guarani group. Mean BP adjusted to age and BMI was significantly lower (P < 0.01) in Guaranis (82.8 W 1.6 mmHg) than in Tupinikins (92.3 W 0.5 mmHg) and non-Indians (91.6 W 1.1 mmHg). Urinary Na excretion (mEq/12h), however, was similar in the three groups (Guarani U 94 W 40; Tupinikin U 105 W 56; nonIndian U 109 W 55; P > 0.05). PWV (m/s) was lower (P < 0.01) Introduction Essential hypertension in humans is a complex and multifactorial disease influenced by a great number of genetic and environmental factors [1]. Observational and experimental studies have shown that high dietary sodium intake, obesity and excessive stress are conditions associated with blood pressure increase and hypertension development, mainly in individuals with genetic predisposition to develop this disease [2–7]. Thus, studies in animal models, human twins and families suggest that inherited markers contribute to more than 50% of the blood pressure variability in human populations [1,8]. Amerindian populations demonstrated low blood pressure levels when living in their traditional lifestyle and the occurrence of hypertension and other cardiovascular diseases was sporadic [9,10]. Yanomami Indians, for example, when investigated in the Intersalt Study two decades ago, had an average arterial pressure of only 96/61 mmHg. Hypertension and obesity cases were not found in that study. More importantly, it was also 0263-6352 ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins Journal of Hypertension 2009, 27:1753–1760 Keywords: arterial hypertension, blood pressure, ethnic groups, Indians Abbreviations: BP, blood Pressure; PWV, pulse wave velocity; WHR, waistto-hip ratio a Department of Physiological Sciences, Federal University of Espı́rito Santo, Vitória, Espı́rito Santo and bHeart Institute (InCor), University of São Paulo Medical School, São Paulo, Brazil Correspondence to José G. Mill, MD, PhD, Departamento de Ciências Fisiológicas, Centro de Ciências da Saúde, Universidade Federal do Espı́rito Santo, Av. Marechal Campos 1468, Vitória, Espı́rito Santo 29042-755, Brazil Tel: +55 27 3335 7335; fax: +55 27 3335 7330; e-mail: jgmill@npd.ufes.br Received 12 December 2008 Revised 18 April 2009 Accepted 7 May 2009 observed that SBP did not increase with age, a finding attributed to the very low urinary sodium excretion (<1 mmol Na/24 h) because salt was not used in food preparation process [3,10]. Similar results were found in other Amerindians groups living in the Brazilian Amazonia when an investigation was performed in individuals still living in their primitive life conditions [11,12]. Subsequent studies have shown that the age-dependent blood pressure increase in these primitive populations is timely related to the emergence of other chronic diseases, including dyslipidemias, type 2 diabetes and atherosclerosis [9,13–15]. Despite the good documentation of the epidemiological transition in some native communities, mainly in the North American Indians, the relative participation of genetic and environmental factors to the uprising of chronic degenerative diseases in primitive communities is not completely understood. Presumably, adoption of acculturated eating habits, such as the intake of foods rich in sodium and carbohydrates with elevated glicemic index, may trigger genetic markers that facilitate the expression of specific phenotypes favoring development DOI:10.1097/HJH.0b013e32832e0b2b Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 1754 Journal of Hypertension 2009, Vol 27 No 9 of hypertension, atherosclerosis and diabetes [2,9,10,15]. Therefore, a better understanding of the epidemiological transition in primitive populations may give additional information about the biological and sociocultural factors influencing the emergence of these chronic diseases in developing societies. Two native Indian communities of different ethnic origin live in the same indigenous reserve in the Southeast Brazilian Coast, at Espı́rito Santo State. The larger group is composed of Tupinikin descendants. This group is well integrated to the usual lifestyle of small Brazilian countryside communities. The men generally work as small farmers or employees carrying out manual functions (mainly in timber industry), whereas women are mostly dedicated to domestic activities. Part of the people living in the Tupinikin reserve is not ethnically Indian, but married to Tupinikins or to their descendants. The other group is formed by Guarani Indians, with a different ethnic origin. Despite living in the same land reserve, this group has resisted incorporating some of the western civilization habits. They maintain their original language, live in poor houses and are mostly dedicated to handcrafting and fishing. Primitive food intake habits, however, were lost and most of food consumed every day is obtained by public and private donations. Therefore, their dietary habits are fully westernized with a high prevalence of obesity, a finding previously observed in some North American Indian communities [9]. Traditional rules followed by the Guarani community do not facilitate interethnic marriage, so that individuals living in the Guarani settlement may be considered mostly unmixed. Therefore, the aim of the present study was to determine the prevalence and severity of the traditional cardiovascular risk factors in these two Indian communities. As salt intake seems to exert a key role for the emergence of high blood pressure in primitive communities [3,10], our purpose was also to investigate the association between blood pressure levels versus urinary sodium excretion and obesity in these two communities. Participants and methods Study population A community-based and cross-sectional study of risk factors for cardiovascular diseases was carried out in two Indian groups (Guarani and Tupinikin) living at the Aracruz Indian Reserve, Espı́rito Santo State, in the southeast Brazilian coast. All individuals (n ¼ 834; 86 Guarani and 748 Tupinikin) aged 20 years or more were eligible for the study. During small meetings in each of the five small settlements, the eligible individuals were invited to participate in the study. Data were collected from February 2003 to April 2004, and 670 (80.3% of the eligible population) attended the local health unit to perform clinical and laboratorial examinations necessary to identify cardiovascular risk factors. The study was approved by the local (Centro de Ciências da Saúde da Ufes, Resolution 3123/02) and national (CONEP Register Number 4599) ethic committees on the human research. Clinical and laboratory data Participants were enrolled to attend the health unit by local health attendants. Participants were informed that they should submit urine and blood for clinical and biochemical analysis necessary to detect presence of cardiovascular diseases (such as hypertension) and diabetes. Participants attended the clinic in the morning (07.00 h) to undergo venous blood collection after a 10– 12 h fasting period, anthropometric evaluation [height, weight and waist-to-hip ratio (WHR)], blood pressure measurement, standard 12-derivations ECG and measurement of pulse wave velocity (PWV) and also to answer a questionnaire related to personal data and lifestyle habits (tobacco and alcohol consumption, medicines under use, etc.). Urine was collected from participants during a 12-h period (from 19.00 h to 07.00 h) during the night before the clinic visit. Blood samples were used to determine fasting glucose, total cholesterol, the lipoproteins fractions low-density lipoprotein cholesterol (LDL-c), high-density lipoprotein cholesterol (HDLc), very low-density lipoprotein cholesterol (VLDL-c), triglycerides, creatinine and uric acid. The 12-h urine collection was supervised by local health assistants (Indian origin and living in the same community). After blood collection, participants had a small breakfast and the other examinations were initiated 30 min later. Na and K concentrations were measured in the 12-h urine by flame photometry. All measurements were performed in the same laboratory and the same commercial kits were used in all investigations. Estimation of daily salt consumption (as NaCl) was calculated according to a previously tested protocol, where the 12-h urinary sodium excretion (night) represented 45% of 24-h total sodium excretion [4]. The clinical examination consisted of a face-to-face interview to determine ethnic classification, smoking habits, use of medicines and food intake habits. Inclusion of individuals in one of three groups (Tupinikin, Guarani or non-Indian) was determined by only one investigator by observing phenotype characteristics and self-reported information about ascendants. In the Guarani settlement, all individuals were considered to be native to this ethnic group. In the Tupinikin community, however, miscegenation to other ethnic groups (mostly African descents) was easily determined by typical phenotypes and participant information. Only individuals with clear indigenous phenotypes (characteristic skin color, straight hair) and having self-reported Tupinikin ascendants were classified as ‘Tupinikin’ (n ¼ 496), whereas 114 individuals (18.7%) living in the Tupinikin settlement were classified as ‘non-Indians’. Most of them showed intermediate phenotypes of Brazilian natives and/or African or European ascendants. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Blood pressure in Brazilian Indian communities Meyerfreund et al. 1755 Blood pressure was measured three times in the left arm with an automatic and validated device (Omron 705-CP; Omron Corp., Kyoto, Japan) after a rest period of 5– 10 min in the sitting position and the mean of the two last recordings was considered as the clinic blood pressure. Participants with SBP 140 mmHg or more and/or diastolic blood pressure 90 mmHg or more, or using antihypertensive drugs, were considered hypertensive. Stages of hypertension were defined by the JNC-VI criteria [16]. Diabetes mellitus was diagnosed by the presence of fasting glucose higher than 125 mg/dl or the use of antidiabetic drugs. Obesity was defined as BMI  30 kg/m2 or more and being overweight as a BMI between 25 and 29.9 kg/m2. The presence of left ventricular hypertrophy (LVH) was determined in the conventional ECG recording by using either the Sokolow–Lyon–Rappaport index [S(V1 or V2) þ R(V5 or v6) 3.5 mV] [17] as the RomhiltEstes score [18]. PWV was measured with an automatic device (Complior, Colson, France) by simultaneous recording of the carotid and femoral pulse waves, according to a technique previously validated, and PWV was used as an indirect index of aortic rigidity [19,20]. Data analysis Differences in the prevalence of nominal variables in the ethnic groups were tested by the x2 test. Differences between means of continuous variables were tested by the Student’s t-test or by a one-way analysis of variance (ANOVA – followed by the post-hoc Tukey test) when appropriate. Linear correlations between blood pressure and measured parameters were obtained by the Pearson’s product moment analysis in the ethnic groups. The influence of the different variables on blood pressure was assessed using a forward stepwise multiple regression analysis. Student’s one-sided t-test was used to compare the slope of regression lines for SBP and age between racial-ethnic groups. Subgroup case–control analysis was also used to verify whether blood pressure differences Table 1 between Guarani and Tupinikin groups were due to different number and age composition of each group. Statistical significance was set at P less than 0.05. Statistical analysis was conducted in the SPSS software (version 11.5; SPSS Inc., Chicago, Illinois, USA). Results A total of 670 individuals (60 out of 86 living in Guarani and 610 out of 748 living in Tupinikin settlements) agreed to participate in the study. The participation rate was similar in both sexes (78.9% and 80.9% of eligible men and women, respectively). The Guarani group was slightly younger than the Tupinikin and the non native group. The education level was generally low, with most of the individuals with less than 4 years of schooling. Education level tended to be smaller in the Guarani group (Table 1). The prevalence of cardiovascular risk factors is shown in Table 1. There were evident differences in relation to the two native groups and nonsignificant differences between the Tupinikin and the non-Indian groups. Smoking was much more frequent in Guaranis (60.4% of regular smokers) than in the other two groups (around 25% of regular smokers). Conversely, hypertension and diabetes was almost absent in the Guarani community where nobody was using antihypertensive and/or antidiabetic drugs. Only one individual (man, 43 years) showed slightly elevated SBP (142 mmHg  86 mmHg) and another (woman, 42 years) had elevated fasting glucose (140 mg/dl). A positive familiar history of hypertension (awareness of at least one parent and/or one brother or sister with high blood pressure or using antihypertensive drug) was significantly lower (P < 0.05) in the Guarani group (6.6%) compared with the Tupinikin (55%) and nonnative (66%) groups. Hypertension was detected in 124 individuals and 80% of them were aware of being hypertensive. Forty-six percent of hypertensive Demographic and clinical characteristics of groups n Male/female (%) Age (years, mean  SD) Age range (years) Education (%) Illiterate 4 years 4–11 years 12 years Smoking (%) Obesity (%) Diabetes (%) Hypertension (%) Cholesterol 200 mg/dl (%) HDL-c <40 mg/dl (%) Triglycerides >150 mg/dl (%) Left ventricular hypertrophy (%) Tupinikin Guarani Nonnatives All 496 49.6/50.4 37.5  15.4 20–94 60 40.0/60.0 36.3  12.4M,þ 20–81 114 42.2/57.8 39.0  13.1 20–91 670 47.6/52.4 37.5  14.9 20–94 12.1 54.8 31.6 1.4 21.4 14.9 4.2 20.8 40.7 34.7 21.6 2.9 28.3 63.3 8.3 0.0 60.4M,þ 22.8M,þ 1.5 1,5M,þ 44.6 60.1M,þ 27.7 0.0 12.3 55.3 29.8 2.6 26.3 11.4 6.1 17.5 41.3 36.8 16.8 6.1 13.6 55.5 29.4 1.5 25.6 14.5 4.3 25.8 41.2 37.2 21.3 3.2 HDL-c, high-density lipoprotein cholesterol; n, number of individuals. continuous variables and x2 for categorical variables). M P < 0.05, Guarani vs. Tupinikin. þ P < 0.05, Guarani vs. nonnative (ANOVA and Tukey test for Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 1756 Journal of Hypertension Table 2 2009, Vol 27 No 9 Anthrophometric and biochemical variables recorded in blood and urine according to groups Weight (kg) Height (cm) BMI (kg/m2) Waist-to hip-ratio Glucose (mg/dl) Cholesterol (mg/dl) LDL-cholesterol (mg/dl) HDL-cholesterol (mg/dl) Triglycerides (mg/dl) Uric acid (mg/dl) Creatinine (mg/dl) Urine12 h volume (ml) Urinary Na12h (mEq) Urinary K12 h (mEq) Salt consumption (g/day) Tupinikin Guarani Nonnative All 65.9  11.9 160  17 25.4  4.2 0.89  0.07 92  16 166  43 95  37 50  15 108  80 4.4  1.6 0.86  0.31 720  340 105  56 17.6  8.2 13.7  7.2 63.1  14.1 152  22M,þ 26.3  5.0 0.87  0.06 79  12M,þ 147  43M,þ 83  35,þ 41  10M,þ 119  79M,þ 4.0  1.0 0.82  0.25 620  360 90  40 13.4  7.9M,þ 11.7  5.3 64.2  13.6 161  19 24.9  5.0 0.86  0.07# 91  21 180  51 112  46# 50  10 91  52 4.2  1.2 0.83  0.22 770  390 109  55 17.1  8.0 14.2  7.2 65.4  12.3 159  17 25.4  4.4 0.88  0.07 91  17 167  45 97  39 49  14 106  75 4.3  1.5 0.84  0.31 730  350 105  54 17.2  8.7 13.7  7.1 Data are means  SD. HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol. nonnative. þ P < 0.05 Guarani vs. nonnative (ANOVA and Tukey test). individuals (n ¼ 57) were under medication and only 28% (n ¼ 16) showed normal blood pressure levels. According to JNC-VI classification [16], 69% showed hypertension phase 1, 19% phase 2 and 12% phase 3. As expected from blood pressure data, the presence of left ventricular hypertrophy according to the Romhilte–Estes index was not found in the Guaranis investigated in this study and it was similar in Tupinikins and nonnatives (x2 ¼ 3.2; P > 0.05). Left ventricular hypertrophy was much more frequent if the Sokolow–Lyon index was used instead. In this case, the prevalence of this condition was higher (P < 0.05) in Tupinikins (31.6%) and nonnatives (27.2%) than in the Guarani group (13%). The prevalence of low HDL-c was more frequent in Guaranis than in the other two groups. Overall prevalence of diabetes was 4.3%. Obesity (BMI 30 kg/m2) was found in 14.3% of the individuals, with higher prevalence (P < 0.05) in the Guarani group. Anthropometric and biochemical parameters are shown in Table 2. Height was significantly lower in Guarani. This group also showed smaller mean values of glucose, total cholesterol, LDL-c and HDL-c and higher values of triglycerides compared with the other two groups. The 12-h urine volume, creatinine clearance and Na excretion were similar in the three groups, despite a tendency to lower values observed in the Guarani group. Potassium excretion, however, was significantly lower in this group. Salt consumption calculated from the 12-h Na excretion was high in the three studied groups and no statistical difference in relation to ethnicity was found. Hemodynamic parameters are shown in Table 3. Heart rate was similar in the three groups, whereas SBP, DBP, mean and pulse pressure were significantly lower (P < 0.01) in the Guarani compared with the Tupinikin and nonnative groups. This difference remains with a high degree of statistical significance (P < 0.01) after adjustment of the mean blood pressure for age and BMI as covariates (Guarani ¼ 83  9 mmHg; Tupinikin ¼ 92  11; M P < 0.05 Tupinikin vs. Guarani. # P < 0.05 Tupinikin vs. nonnatives ¼ 92  13 mmHg). SBP and DBP also remain significantly lower in the Guarani group after adjustment for these two covariates. PWV, a variable related to aortic stiffness, was also significantly lower in the Guarani group, even after adjusting for age and SBP. Univariate analysis correlating SBP and DBP with age, anthropometric and biochemical variables in the two native groups is shown in Table 4. The strongest correlation coefficients in the Guarani group were found in relation to anthropometric variables, either for SBP or DBP. In the Tupinikins, however, the strongest predictor of the SBP was age. The regression analysis for SBP and DBP by age in the three groups is shown in Fig. 1. SBP was significantly related to age in Tupinikins (r ¼ 0.598; P < 0.01), in nonnatives (r ¼ 0.454; P < 0.05) and in Guaranis (r ¼ 0.272; P < 0.05). The rate of SBP increase (beta coefficient) in Tupinikins (7.9  0.5 mmHg/decade) and nonnatives (6.6  1.2 mmHg/decade) was higher Table 3 Hemodynamic values according to ethnic groups Tupinikin SBP (mmHg) Mean  SD 123  20 Range 89–202 DBP (mmHg) Mean  SD 77  11 Range 53–119 Mean BP (mmHg) Mean  SD 92  13 Range 67–142 Pulse pressure (mmHg) Mean  SD 46  14 Range 22–115 Heart rate (beats/min) Mean  SD 69  11 Range 40–135 PWV (m/s) Mean  SD 8.8  2.2 Range 4.8–18.2 Guarani Nonnative All 108  14MM 83–140 120  21 87–191 121  20 83–202 70  10MM 48–88 78  13 58–127 76  9 48–127 81  10MM 61–105 92  15 70–147 91  13 61–147 39  8MM 24–59 42  12 20–80 45  13 20–115 70  9 55–98 71  11 48–113 69  11 40–135 7.5  1.4MM 5.3–12.0 8.4  2.0 5.1–13.2 8.6  2.2 4.8–18.2 BP, blood pressure; PWV, pulse wave velocity. MM P < 0.01, Guarani vs. Tupinikin and nonnative groups (ANOVA and Tukey post-hoc test). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Blood pressure in Brazilian Indian communities Meyerfreund et al. 1757 Pearson’s correlation coefficient (r) between anthropometric and biochemical variables and SBP and DBP in Tupinikin and Guarani groups Table 4 Tupinikin Independent variables Age (years) Weight (kg) Body mass index (kg/m2) Waist circumference (cm) Waist-to-hip ratio Glucose (mg/dl) Cholesterol (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Urinary Na excretion12h (mEq) Guarani SBP DBP SBP DBP 0.598MM 0.087 0.209MM 0.348MM 0.537MM 0.305MM 0.262MM 0.004 0.241MM 0.048 0.368MM 0.249MM 0.312MM 0.395MM 0.460MM 0.258MM 0.287MM 0.057 0.278MM 0.100M 0.314M 0.340M 0.386MM 0.372MM 0.447MM 0.055 0.083 0.015 0.016 0.035 0.122 0.483MM 0.514MM 0.485MM 0.473MM 0.169 0.262 0.038 0.231 0.192 HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol. M P < 0.05. MM P < 0.01. Fig. 1 (P < 0.01) than in Guaranis (3.1  1.5 mmHg/decade). A significant (P < 0.01) age-dependent increase of DBP was observed in the Tupinikins and nonnatives (2.5  0.3 and 4.0  0.7 mmHg/decade, respectively). This parameter, however, was not different from zero in Guaranis (1.2  1.1 mmHg/decade; P ¼ 0.28). A forward multiple regression analysis was tested to determine main predictors of blood pressure variability in the two Indian groups. The model included all variables with significant correlation depicted in Table 4. In Tupinikins, age and WHR remained in the model, and these two variables explained 44% of total SBP variability (SBP ¼ 36.6 þ 0.70  age þ 68.6  WHR; r2 ¼ 0.44; P < 0.01). In the Guarani group, only WHR remained in the model, explaining 22% of total SBP variability (SBP ¼ 22.1 þ 95.9  WHR; r2 ¼ 0.22; P < 0.01). The DBP was dependent on WHR, age and body weight in Tupinikins (r2 ¼ 0.28; P < 0.01), whereas only WHR remained in the model in the Guarani group (r2 ¼ 0.22; P < 0.01). The previous analysis was performed including all individuals of our data bank, and some differences found between Guaranis and Tupinikins might be due to differences in sample size or age distribution. Therefore, all data were re-evaluated after sorting 50 Guaranis and 50 Tupinikins paired by sex and age. Mean values of SBP and DBP remained similar to those shown in Table 3 (Guarani: 107  13/69  9 mmHg vs. Tupinikin: 119  18/75  8 mmHg). Twelve-hour urinary sodium excretion was also similar in subgroups (Guarani: 91  43 mEq and Tupinikin: 98  57 mEq; P ¼ 0.52), as observed considering all components of the two groups. Considering only the data obtained in individuals older than 45 years (eight Guaranis and 101 Tupinikins), a significant difference (P < 0.05) in blood pressure between groups is still observed. Thus, SBP and DBP were smaller in this Guarani subgroup (120/74 mmHg) compared with the Tupinikin group (143/83 mmHg). In the latter group, 58% were hypertensive, whereas all Guaranis had normal blood pressure values. Despite these clear differences in blood pressure values, urinary sodium excretion (mEq/12 h) was similar (Guarani: 88  39 mEq vs. Tupinikin: 99  56 mEq; P ¼ 0.68). Discussion Relationship between SBP and DBP and age in Tupinikins, Guaranis and non-Indians. Age-related increase in SBP was steeper in Tupinikins (SBP ¼ 93.3 þ 0.79  age; r ¼ 0.598) and non-Indians (SBP ¼ 94 þ 0.66  age; r ¼ 0.455) than in Guaranis (SBP ¼ 97 þ 0.31  age; r ¼ 0.272). Significant age-related increase in DBP was observed only in Tupinikins (DBP ¼ 67 þ 0.25  age; r ¼ 0.37) and in non-Indians (DBP ¼ 62 þ 0.41  age; r ¼ 0.41). The main finding of this study was the different behavior of blood pressure as a function of age in the Guarani group compared with the other two groups (Tupinikin and nonnatives). The food intake habits of these communities were not investigated in detail in the present study. However, according to the data obtained from the urine analysis, the food intake habits seem to be similar, mainly in relation to sodium and potassium intake. In many aspects, the Tupinikin and non-Indians showed similar characteristics, despite the fact that the latter is composed Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 1758 Journal of Hypertension 2009, Vol 27 No 9 of an inhomogeneous group of individuals living together with Tupinikins. An age-dependent increase in blood pressure was observed in almost all populations investigated until now, except for some isolated and primitive groups. The age-dependent blood pressure increase has been ascribed to diet habits and body weight gain, as studies in isolated populations with very low salt intake in different parts of the world have shown no increase of blood pressure with age [5,10,12,15,21–24]. These primitive populations are usually characterized by a Naþ-poor and Kþ-rich diet, because it is highly dependent on vegetable (roots) and fruits. Other investigators have argued that the lack of an age-increase in blood pressure in these isolated populations could be due to the high prevalence of malnutrition and/or chronic diseases [25]. None of these situations was found in the Guarani group investigated in this study. Both Indian populations of our study have characteristics of the so-called occidental diet including high sodium content and adequate calories. The levels of 12-h urinary sodium excretion were similar to that found in Brazilian urban populations [4]. Other alimentary habits were not systematically investigated. However, the main meals usually have a high content of cereals (rice, beans and manioc powder) and saturated fat. Fish consumption was less common even in the Guarani group. Therefore, the low blood pressure observed in the Guarani seems not to depend on the low sodium diet as observed in other primitive Brazilian populations, such as the Yanomami, Amondava and Suruı́ Indians living in the Amazonian region [11–13]. Malnutrition, characterized by low body weight, also cannot contribute to low blood pressure levels as was found in other Brazilian indigenous populations [15] because the prevalence of being overweight and obesity found in the Guarani group was similar to that observed in other Brazilian urban populations [4]. By contrast, a high prevalence of being overweight and obesity was found in Guaranis, and the multivariate analysis suggests that this finding accounts for most of the blood pressure variability observed in this Indian group. Interestingly, Cardoso [26] also studied some cardiovascular risk factors in a Guaranı́-Mbyá settlement in Angra dos Reis, Rio de Janeiro State. Blood pressure was measured in 145 individuals (15 years, age mean 36.2  20.8 years) and the mean values of SBP and DBP (109.3  13.6 and 69.7  9.8 mmHg, respectively) were similar to the ones found in our study. The prevalence of hypertension was only 4.1%, a value lower than those found in most of the studies conducted in North American Indian populations [27]. Our study, however, was the first to show that the low pressure levels seem not to depend on low-salt diet as previously stated in others primitive societies. The high prevalence of obesity in Guaranis agrees with the sedentary lifestyle of this group because the main commercial activity is still handcrafting, although farming was the main occupational activity declared either by Tupinikins and by the nonnatives integrated to the Tupinikin community. The seemingly lower energy cost of work activities, therefore, in Guaranis may explain the higher BMI and obesity prevalence found in this group. According to other studies, this fact would facilitate an age-dependent blood pressure increase, particularly related to DBP [28]. Our data agree with this view because, in the multivariate regression analysis in the Guarani group, the WHR, an indirect index of central obesity, was the only independent predictor of blood pressure variability. Our findings support the view that differences in blood pressure patterns found in Guaranis compared with Tupinikins and non-Indians are not due to different nutritional habits. Studies have shown an association between alcohol ingestion and blood pressure [29]. We did not perform a systematic investigation of alcohol consumption in our study. Poor language understanding (mainly in Guaranis) makes it difficult to obtain reliable data on this subject. However, alcohol use and abuse was roughly investigated during the medical interview, and our general view is that alcohol consumption is high (mainly in men) and similar in all three studied groups. Therefore, it is unlikely that alcohol consumption may be responsible for the differences in blood pressure levels between groups. Therefore, our data suggest that differences in the genetic structure modulating the age-increase in blood pressure (such as salt sensitivity) or some gene– environment interaction not explored in the study may account for the some differences found between the two native groups. The finding that a familiar history of hypertension was more frequent in Tupinikins reinforces this view. It is interesting to note that, despite the high prevalence of being overweight/obesity in the Guarani group, mean values of blood glucose were also lower in these individuals than in Tupinikins and nonnatives. This finding reinforces the view that deregulation of carbohydrate metabolism and blood pressure increase may depend on common pathophysiological mechanisms. It is well described that in North American Indians, obesity is strongly associated with type 2 diabetes and lifestyle modifications, including nutritional transition [30,31]. Lifestyle modifications also may increase stress levels, thus contributing to age-dependent blood pressure increase, a fact observed in some primitive and isolated populations after emigration to urban areas [32,33]. However, several studies have suggested that models used to explain the increase in hypertension and diabetes in North American Indians may not be reproduced in South American Indians [30]. Thus, the Aymara and Mapuche populations living in Chile also showed a high degree of obesity and a very low prevalence of diabetes and hypertension [34], as well as low resting insulin levels [35]. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Blood pressure in Brazilian Indian communities Meyerfreund et al. 1759 Some degree of dissociation between obesity, hypertension and diabetes was also observed in the Guarani group investigated in our study, as well as in another study in a Guarani-Mbyá group living in the Brazilian southeast cost [26]. In overall, our data do not explain the origin of the evident differences in relation to the blood pressure pattern in these two Indian communities living in the same native reserve and having similar habits related to food consumption. However, they suggest that the genetic background of different communities may help to understand how lifestyle habits may contribute to the surge of complex diseases, such as hypertension and diabetes. However, we cannot discard cultural influences accounting for some of differences in physiological parameters, such as blood pressure, between Guaranis and Tupinikins. Chronic stress may influence blood pressure levels and may contribute to change in values of this variable along time [32,33]. However, psychosocial characteristics were not determined in the individuals included in this study. PWV was significantly lower in Guarani compared with the Tupinikin group, suggesting the existence of less stiff arteries in the former group [19]. Age contributes to increase the collagen-to-elastin ratio in large arteries, thus increasing vascular stiffness, SBP and PWV. As PWV in the Guarani group remained significantly lower that in Tupinikins even after adjustment for age and SBP, we may speculate that differences in arterial rigidity expressed by the lower PWV values in Guaranis may contribute to the lower blood pressure levels observed in this group. Genetic determinants may contribute to phenotypes differences observed in different populations. Thus, a polymorphism of the angiotensin II type 1 receptor is an independent determinant of aortic stiffness in hypertensive patients [36]. Therefore, genetic markers may contribute to the occurrence of reduced artery stiffness in this group. However, additional studies in other Guarani communities, including the investigation of candidate genes affecting blood pressure, may give new insights into this area. Previous studies have suggested that the age-dependent increase in blood pressure in primitive communities was closely related to the use of salt in food preparation [3]. Several studies in primitive Amazonian tribes as well as in other primitive communities were concordant with this view. The present study, however, points in a different direction. The Guarani community investigated in this study has adopted several occidental habits, including those related to salt use and the age-dependent increase in blood pressure is significantly attenuated compared with Tupinikins, so that hypertension was almost absent in the former group, even considering the individuals with more than 45 years, when high blood pressure is highly prevalent. Therefore, our study suggests that salt addition to food preparations seems not to be the only factor explaining age-dependent SBP increase in primitive populations after changing their primitive habits. Social stress seems to also contribute to this phenomenon. However, additional studies are necessary to quantify the relative contributions of the different factors contributing to the progressive increase of blood pressure with age, a factor predisposing to the worldwide development of hypertension. Study limitations A major limitation in our study is the different number of individuals in the two Indian communities. However, the subgroup analysis showed similar results to those obtained when entire groups were considered. In addition, we did not evaluate psychosocial characteristics that may influence time-dependent changes in physiological variables such as blood pressure. The Tupinikin community is more integrated to the surrounding communities than Guaranis, and such isolation of Guaranis may reduce exposure to some stressors such as daily labor and the necessity to obtain money to survive. Apart from genetic factors, we cannot exclude contributions of such variables to explain blood pressure differences between groups. Acknowledgements We express our acknowledgments to Funasa/ES and Funai for continuous support during data collection. We also acknowledge financial support from CNPq (304248/2005-3) and Fapesp. References 1 2 3 4 5 6 7 8 9 10 Marteu JB, Zaiou M, Siest G, Visvikis-Siest S. Genetic determinants of blood pressure regulation. J Hypertens 2005; 23:2127–2143. Kaufmann EE, Owaaje SA, James SA, Rotini CN, Cooper RS. Determinants of hypertension in West Africa: contribution of anthropometric and dietary factors to urban–rural and socioeconomic gradients. Am J Epidemiol 1992; 143:1203–1218. INTERSALT (International Study of Electrolyte Excretion and Blood Pressure), 1988 Intersalt: An International Study of Electrolyte Excretion and Blood Pressure. Results for 24 h urinary sodium and potassium excretion. Br Med J 1988; 297:319–328. Molina MCB, Cunha RS, Herkenhoff LF, Mill JG. Hypertension and salt intake in an urban population. Rev Saude Publica 2003; 37:743–750. Tesfaye F, Nawi NG, Van Minh H, Byass P, Berhane Y, Bonita R, et al. Association between body mass index and blood pressure across three populations in Africa and Asia. J Hum Hypertens 2007; 21:28–37. Schneider GM, Jacobs DW, Gevirtz RN, O’Connor DT. Cardiovascular haemodynamic response to repeated mental stress in normotensive subjects at genetic risk of hypertension: evidence of enhanced reactivity, blunted adaptation, and delayed recovery. J Hum Hypertens 2003; 17:829–840. Fuchs FD, Wannmacher CM, Wannmacher L, Guimarães FS, Rosito GA, Gastaldo G, et al. Effect of sodium intake on blood pressure, serum levels, and renal excretion of sodium and potassium in normotensives with and without familial predisposition to hypertension. Braz J Med Biol Res 1987; 20:25–34. Hong Y, de Faire U, Heller DA, McCleam GE, Pedersen N. Genetic and environmental influences on blood pressure in elderly twins. Hypertension 1994; 24:663–670. Mendlein JM, Freedman DS, Peter DG, Allen B, Percy CA, Ballew C, et al. Risk-factors for coronary heart disease among Navajo Indians: findings from the Navajo Health and Nutrition Survey. J Nutr 1997; 127:2099S– 2105S. Mancilha-Carvalho JJ, Baruzzi RG, Howard PF, Poulter N, Alpers MP, Franco LJ, et al. Blood pressure in four remote populations in the INTERSALT Study. Hypertension 1989; 14:2388–2446. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 1760 Journal of Hypertension 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 2009, Vol 27 No 9 Fleming-Moran M, Santos RV, Coimbra GA Jr. Blood pressure levels of the Suruı́ and Zoró Indians of the Brazilian Amazon: group and sex specific effects resulting from body composition, health status and age. Hum Biol 1991; 63:835–861. Pavan L, Casiglia E, Braga LMC, Winnid M, Perato M, Pauleto P, et al. Effects of a traditional lifestyle on Cardiovascular risk profile: The Amondava populations of the Brazilian Amazon. Comparisons with matched Africans, Italian and Polish populations. J Hypertens 1999; 17:749–756. Mancilha-Carvalho JJ, Crews DE. Lipid profiles of Yanomame Indians of Brazil. Prev Med 1990; 19:66–75. Bloch KV, Coutinho ESF, Lobo MEC, Oliveira JE, Milech A. Blood pressure, capillary glucose, and anthropometric measurements in a Yanomámi population. Cad Saude Publica 1993; 9:428–438. Santos RV, Coimbra CEA Jr. Socioeconomic transition and physical growth of Tupı́-Mondê Amerindian children of the Aripuanã Park, Brazilian Amazon. Hum Biol 1991; 63:795–819. JNC-VI. The Sixth Report of the Joint National Committee on Prevention, Detection, and Treatment of high blood pressure. Arch Int Med 1997; 157:2413–2446. Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J 1949; 37:161–186. Romhilt DW, Estes EH Jr. A point score system for the ECG diagnosis of left ventricular hypertrophy. Am Heart J 1968; 75:752–758. Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B, Brisac AM, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement. Hypertension 1995; 26:485–490. Ferreira AVL, Viana MC, Mill JG, Asmar RG, Cunha RS. Racial differences in aortic stiffness in normotensive and hypertensive adults. J Hypertens 1999; 17:631–637. Mancilha-Carvalho JJ, Souza-e-Silva NA. The Yanomami Indians in the INTERSALT Study. Arq Bras Cardiol 2003; 80:295–300. Priox AM, Grimley-Evans J, Harvey HPB, Davidson F, Lyndsey M. Sodium intake and blood pressure in two Polynesian populations. N Engl J Med 1968; 279:515–520. Oliver WJ, Cohen EL, Neel JV. Blood pressure, sodium intake and sodiumrelated hormones in the Yanomami Indians, a ‘no salt’ culture. J Hum Hypertens 1975; 52:146–151. Truswell AS, Kennely BM, Hansen JDL, Lee RB. Blood pressure of Kung bushman in northern Botswana. Am Heart J 1972; 84:5–12. Maddocks I, Vine AP. The influence of chronic infection on blood pressure in New Guinea males. Lancet 1966; 2:262–264. Cardoso AM, Mattos IE, Koifman RJ. Prevalence of risk factors for cardiovascular diseases in the Guarani-Mbyá population of the State of Rio de Janeiro. Cad Saude Publ 2001; 17:345–354. Howard BV, Lee ET, Yeh JL, Go O, Fabsitz RR, Devereux RB, Welty TK. Hypertension in adult American Indians. The Strong Heart Study. Hypertension 1996; 28:256–264. Kannel WB. Fifty years of Framingham Study contributions to understanding hypertension. J Hum Hypertens 2000; 14:795–819. Miller PM, Anton RF, Egan BM, Basile J, Nguyen SA. Excessive alcohol consumption and hypertension: clinical implications of current research. J Clin Hypertens (Greenwich) 2005; 7:346–351. Howard BV, Lee ET, Cowan LD, Devereux RB, Galloway JM, Go OT, et al. Rising tide of cardiovascular disease in American Indians. The Strong Heart Study. Circulation 1999; 99:2389–2395. Percy C, Freedman DS, Gilbert TJ, White L, Ballew C, Mokdad A. Prevalence of hypertension among Navajo Indians: findings from the Navajo Health and Nutrition Survey. J Nutr 1997; 127:2114S–2119S. Steptoe A. Psychosocial factors in the development of hypertension. Ann Med 2000; 32:371–375. Hollenberg NK, Martinez G, McCullough M, Meinking T, Passan D, Preston M, et al. Aging, acculturation, salt intake, and hypertension in the Kuna of Panama. Hypertension 1997; 29:171–176. Pãrez F, Carrasco E, Santos JL, Calvilla M, Albala C. Prevalence of obesity, hypertension and dyslipidemia in rural aboriginal groups in Chile. Rev Med Chil 1999; 127:1169–1175. Santos JL, Pãrez-Bravo F, Carrasco E, Calvilla M, Albala C. Low prevalence of type 2 diabetes despite a high average body mass index in the Aymara natives from Chile. Nutrition 2001; 17:305–309. Benetos A, Topouchian J, Ricard S, Gautier S, Bonnardeaux A, Asmar R, et al. Influence of angiotensin II type 1 receptor polymorphism on aortic stiffness in never treated hyperertensive patients. Circulation 1996; 94:698–703. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.