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.