Arterial stiffness in black adults from Angola and Brazil

Abstract

Ethnicity is an important determinant of blood pressure levels, being black individuals affected more than any other ethnic group. Arterial stiffening, an independent risk factor for hypertension, is also influenced by ethnicity. However, whether black individuals from different continents would have different patterns of arterial stiffening is still unknown. Thus, the authors aimed to compare pulse wave velocity (PWV) in black subjects living in Angola and Brazil. A total of 677 black individuals from two independent cross‐sectional studies conducted in Brazil and Angola were included in this analysis. Carotid‐to‐femoral PWV was measured following the same protocols for both studies, as well as clinical and anthropometric variables. Adjusted PWV was higher in Brazilian blacks than in Angolans, regardless of sex (men from Brazil: 10.7 ± 1.8 vs men from Angola: 9.9 ± 1.8 m/s, P < .001; women from Brazil: 10.3 ± 1.5 vs women from Angola: 9.2 ± 1.3 m/s, P < .001). Although the cf‐PWV was higher in Brazilian blacks, the age‐related increase in cf‐PWV was higher in Angolan men compared to Brazilians, but not in women. SBP showed the strongest association with cf‐PWV, regardless of sex and country. However, age was associated with cf‐PWV in all groups, except in Brazilian men. Our results clearly show a difference in PWV between two black populations, and highlight for sex differences in the hemodynamic parameters that might affect blood pressure levels in these populations.

1. INTRODUCTION

In the arteries of young and healthy subjects, the pulse wave is conducted slowly along the arterial wall. The reflected component of this wave is coincident with the diastolic period of the cardiac cycle thus contributing to keeping the diastolic pressure at higher levels and improving subendocardial blood flow. When arterial compliance is reduced, however, several hemodynamic changes can occur. The pulse wave is conducted faster in stiffer arteries, the reflecting wave occurs earlier thus amplifying the systolic blood pressure (SBP). ¹ The latter effect is associated with elevated cardiac afterload leading to ventricular hypertrophy and microcirculation damage, particularly in kidneys and brain, thus increasing the likelihood of future negative cardiovascular outcomes. ¹ , ² Methods that seek to identify early changes and arterial compliance may contribute to predict the risk of future cardiovascular events. The measurement of the aortic stiffness by the carotid‐femoral pulse wave velocity (PWV) represents the gold standard method to indirectly assess the stiffness of large arteries (mainly aorta) being a strong predictor of future cardiovascular events, the incidence of arterial hypertension and damage to target organs. ³

Several factors are associated with an increase of the arterial stiffness, being aging and SBP the strongest ones. Other factors include diabetes, uric acid, obesity, smoking habit, sedentary lifestyle, albuminuria, increased salt intake, and race. ⁴ It is known that race influence blood pressure (BP) levels and several studies have shown that black individuals have higher BP levels than any other race/ethnic groups. ⁵ , ⁶ Some studies have suggested that black individuals show a steeper increase of PWV associated with the aging process. ⁷ , ⁸ A study developed with 1427 individuals in Brazil, including different ethnicities, showed that PWV and BP were higher in black adults than in other groups (whites and brown), even after adjustment for age and BP. ⁵ There are few studies, however, comparing these parameters in black populations living in Africa and outside Africa. Therefore, our purpose in this study was to compare hemodynamic parameters in black subjects living in Angola and Brazil.

2. METHODS

2.1. Study characteristics

The data included in this study were obtained in two independent cross‐sectional studies conducted in the urban population of Vitoria/Brazil and with civil servants of a public University of Luanda/Angola. In both studies, data acquisition followed a standardized protocol ⁹ and participants agreed to participate by signing written “Informed Consent” terms, which were approved by the Research Ethics Committees in the two institutions where data were collected.

Data from Brazilian participants were collected in a population‐based, cross‐sectional study in the urban population (25‐64 years) of Vitória, Brazil, following the WHO MONICA Project guidelines. ⁹ The study design, sampling, and data collection were published elsewhere. ¹⁰ , ¹¹ In summary, a group of 1662 individuals was selected based on a multistage probability sampling and invited to the University Hospital for clinical and laboratory investigation. For the present study, only black participants (N = 133) were included in the analyses.

Data from the Angolan participants were obtained in a cross‐sectional study of a non‐probabilistic sample of civil servants of Agostinho Neto University (UAN) in Luanda, the Angolan capital. The UAN is a public institution of higher education with 1458 workers, (50.3% of which were nonteaching staff). The sampling methods and the characteristics of the sample can be found in detail in previous publications. ¹² , ¹³

2.2. Racial classification

In the Brazilian sample, racial stratification was based on phenotypic parameters (skin color, hair texture, and nose and chin characteristics) and in the information provided by participants regarding their ancestors. Based on this information, a trained researcher classified the individual as white, black, brown, indigenous, or Asiatic. In the Angolan sample, all participants were classified as blacks.

2.3. Clinical investigation

Systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) were measured in the left arm after a resting period of 5‐10 minutes in the sitting position. Three consecutive readings with a minimum interval of 5 minutes between measurements were taken from each subject. The levels of SBP and DBP were calculated as the arithmetic mean of the last two measurements.

Biochemical blood parameters were obtained in participants instructed to fast for at least 10 hours before the examination. The blood samples were drawn after venipuncture in the upper limb, performed by a previously trained laboratory technician. Biochemical analyses were done using commercially available kits. The LDL‐c fraction was calculated indirectly by the Friedewald equation for triglycerides <400 mg/dL. ¹⁴ The fasting glucose level was measured from blood collected in a tube containing fluoride as an anticoagulant, and diabetes was defined as having a fasting blood glucose ≥126 mg/dL.

Anthropometric parameters were collected by trained technicians. Bodyweight was obtained on a calibrated scale with an accuracy of 0.1 kg. Height was measured on a wall stadiometer with an accuracy of 0.5 cm. The body mass index (BMI) was calculated as the ratio of body weight (kg) to the squared height (m²). Waist circumference (WC) was measured at the midpoint between the last costal arch and the iliac crest, considering the maximum point of normal expiration, with the individual in the standing position. The hip circumference (HC) was measured with an accuracy of 0.1 cm around the thighs, at the height of the greater trochanter, with the individual standing.

2.4. Pulse wave velocity

The carotid‐to‐femoral pulse wave velocity (cf‐PWV) was obtained by simultaneous recording of the arterial pulse at the right common carotid artery and the right femoral artery. ¹⁰ Recordings were obtained after 10‐minute rest in the supine position according to procedure guidelines. Distance between the two sites of pulse recording was measured with an inelastic tape. The PWV was automatically calculated as the distance travelled by the pulse wave between the two points (in meters) divided by the time interval (in seconds) between the two pulse waves. ⁶ The same equipment (Complior) was used in the two studied groups.

2.5. Statistical analysis

Data are presented as mean ± standard deviation (SD) for continuous variables and as frequency and percentages for categorical variables. Differences in clinical and anthropometric characteristics of the two samples were determined by a two‐tailed Student t test for independent samples.

Pearson correlation was used to fit univariate associations, and the slope of the linear regression was compared with the Student t test. Analysis of covariance (ANCOVA) was used to compare PWV among groups after adjustment for selected clinical and anthropometric covariables. We used multiple linear regressions to create prediction models for PWV based on selected clinical and anthropometric characteristics. All analyzes were performed with the Statistical Package for the Social Sciences—SPSS, version 22.0 (IBM Corp.). The significance level was set at P < .05.

3. RESULTS

This study comprises 677 black participants, being 133 from Brazil (48.9% women) and 544 from Angola (49.8% women). The clinical and anthropometric characteristics of the two samples are presented in Table 1 stratified by sex. Age means were similar, regardless of sex (men from Brazil: 42.7 ± 10.6 vs men from Angola: 45.1 ± 11.1 years, P = .107; women from Brazil: 42.6 ± 10.7 vs women from Angola: 43.9 ± 10.1 years; P = .341). When hemodynamic parameters were evaluated, we observed that SBP was similar in the two groups, without differences between men and women. However, DBP was higher in both men and women from Brazil compared to those from Angola. The cf‐PWV was higher in Brazilian blacks than in Angolans, regardless of sex (men from Brazil: 10.39 ± 2.15 vs men from Angola: 10.00 ± 2.17 m/s, P < .026; women from Brazil: 10.41 ± 2.47 vs women from Angola: 9.20 ± 1.82 m/s, P < .001).

TABLE 1.

Clinical and anthropometric characteristics of black adults from Angola and Brazil

	Men			Women
	Brazil (n = 68)	Angola (n = 273)	P value	Brazil (n = 65)	Angola (n = 271)	P value
Age (y)	42.7 ± 10.6	45.1 ± 11.1	.107	42.6 ± 10.7	43.9 ± 10.1	.341
Weight (kg)	74.7 ± 13.3	68.0 ± 14.9	.001	67.2 ± 15.3	69.2 ± 15.7	.356
Height (cm)	169.1 ± 6.0	167.4 ± 7.1	.066	155.4 ± 6.3	159.6 ± 6.6	<.001
BMI (kg/m2)	26.0 ± 4.09	24.1 ± 4.3	.001	27.8 ± 5.9	27.1 ± 5.8	.422
WC (cm)	86.6 ± 10.2	80.1 ± 12.9	<.001	87.2 ± 13.6	83.9 ± 13.5	.068
WHR	0.90 ± 0.06	0.87 ± 0.08	.022	0.87 ± 0.08	0.84 ± 0.10	.025
Uric acid (mg/dL)	5.19 ± 1.31	6.12 ± 1.72	<.001	4.48 ± 1.42	4.78 ± 1.39	.120
Glucose (mg/dL)	103.8 ± 31.2	94.9 ± 19.9	.004	106.7 ± 46.2	93.2 ± 21.8	<.001
Cholesterol (mg/dL)	200.3 ± 39.4	189.5 ± 41.3	.052	212.2 ± 42.7	193.2 ± 36.4	<.001
LDL‐c (mg/dL)	131.9 ± 36.9	125.0 ± 41.7	.216	145.5 ± 37.2	125.8 ± 38.6	<.001
HDL‐c (mg/dL)	44.9 ± 13.3	44.1 ± 10.3	.613	47.4 ± 11.1	47.6 ± 11.2	.905
LDL/HDL (mg/dL)	3.20 ± 1.29	3.06 ± 1.43	.450	3.24 ± 1.28	2.86 ± 1.31	.040
TG (mg/dL)	124.5 ± 91.7	101.8 ± 41.6	.002	114.7 ± 91.9	98.7 ± 38.3	.022
SBP (mmHg)	134.3 ± 19.2	136.7 ± 22.7	.420	134.0 ± 24.3	133.3 ± 26.6	.845
DBP (mmHg)	90.0 ± 13.5	82.9 ± 14.2	<.001	90.2 ± 16.9	82.7 ± 13.8	<.001
MBP (mmHg)	104.8 ± 14.8	100.7 ± 16.4	.062	104.8 ± 18.8	99.4 ± 17.4	.026
HR (bpm)	63.5 ± 7.5	67.0 ± 10.1	.230	70.3 ± 13.1	69.4 ± 9.9	.688
cf‐PWV (m/s)	10.4 ± 2.2	10.0 ± 2.2	.026	10.4 ± 2.5	9.2 ± 1.8	.001

Abbreviations: BMI, body mass index; cf‐PWV, carotid‐to‐femoral pulse wave velocity; DBP, diastolic blood pressure; HDL‐c, high‐density lipoprotein; HR, heart rate; LDL‐c, low‐density lipoprotein; MBP, mean blood pressure; SBP, systolic blood pressure; TG, triglycerides; WC, waist circumference; WHR, waist‐to‐hip ratio.

To evaluate the changes in SBP and cf‐PWV according to age in men and women from Brazil and Angola, we performed a linear regression analysis and the slopes of each regression line were compared (Figure 1). Changes in SBP with aging were steeper in Angolans that in Brazilians, regardless of sex (Figure 1A‐C). Although the cf‐PWV is higher in Brazilian blacks, the age‐related increase in cf‐PWV was higher in Angolan men compared to Brazilians men (Figure 1D,F). This difference was not found in women (Figure 1, panels E and F).

FIGURE 1.

Linear regression for the association between SBP (A and B) or cf‐PWV (D and E) with age in men and women from Angola and Brazil. The linear slope was used to compare the magnitude of each association (C and F). *Represents P < .05

We next controlled the cf‐PWV values for confounders, such as age, SBP, WC, glucose, and uric acid. As observed for crude values (Table 1), adjusted cf‐PWV values were kept significantly higher for Brazilians than for Angolans, regardless of sex (Figure 2A).

FIGURE 2.

Panel A showed cf‐PWV values for black men and women from Angola and Brazil, adjusted by age, SBP, BMI, glucose, and uric acid. Panel B and C are adjusted by SBP, BMI, glucose, and uric acid. *Represents P < .05

To verify the sex‐specific influence of age in the slope of cf‐PWV (Figure 1D,E), we stratified our sample in two different age intervals (<45 or ≥45 years) for both men and women from Angola and Brazil. As observed in Figure 2B, adjusted cf‐PWV was higher in Brazilian men than in Angolans below 45 years, but not for those over 45 years. For women, however, adjusted cf‐PWV was higher in Brazilians than in Angolans, regardless of the age interval (Figure 2C).

Multiple linear regression was used to evaluate the main determinants of cf‐PWV in men and women from Brazil and Angola. In men, SBP showed the strongest association with cf‐PWV, regardless of country. However, age was associated with cf‐PWV only in Angolans, while BMI demonstrated a significant and independent association with cf‐PWV only in Brazilians (Table 2). In women, age and SBP showed the strongest association in Angolans and Brazilians. Moreover, fasting glucose was significantly associated with PWV in Brazilian women but not in Angolan women (Table 2).

TABLE 2.

Multiple linear regression for the association of some classical risk factors with cf‐PWV in men and women from Angola and Brazil

	Men						Women
	Brazil (r 2 = .33)			Angola (r 2 = .37)			Brazil (r 2 = .43)			Angola (r 2 = .43)
	B ± SE	β	P	B ± SE	β	P	B ± SE	β	P	B ± SE	β	P
Intercept	2.95 ± 2.04	‐	.153	2.17 ± 0.91	‐	.018	0.44 ± 1.63	‐	.790	2.36 ± 0.70	‐	.001
Age	0.03 ± 0.02	0.13	.229	0.07 ± 0.01	0.32	<.001	0.06 ± 0.03	0.28	.014	0.05 ± 0.01	0.30	<.001
SBP	0.07 ± 0.01	0.61	<.001	0.04 ± 0.01	0.42	<.001	0.05 ± 0.01	0.45	<.001	0.03 ± 0.004	0.42	<.001
Uric acid	0.09 ± 0.17	0.05	.603	0.02 ± 0.07	0.01	.796	0.20 ± 0.18	0.12	.272	0.03 ± 0.07	0.02	.669
Glucose	0.01 ± 0.01	0.10	.374	−0.003 ± 0.005	−0.03	.606	0.01 ± 0.01	0.25	.012	−0.002 ± 0.005	−0.02	.671
BMI	−0.16 ± 0.06	−0.30	.010	−0.02 ± 0.03	−0.03	.590	−0.04 ± 0.05	−0.11	.307	0.03 ± 0.01	0.08	.091

4. DISCUSSION

Racial differences in arterial stiffness have been frequently reported in the literature. ⁵ , ⁸ , ¹⁵ Epidemiological and clinical studies conducted mostly outside Africa have reported greater arterial stiffness in black individuals compared to white individuals and that difference remained significant even after controlling for classic cardiovascular risk factors. ⁸ , ¹⁶ However, despite interracial differences being constantly reported in the last decade, the comparison of the arterial stiffness in black Africans and the black population outside Africa was not previously reported. Such comparison is important because, beyond racial characteristics, arterial stiffness is also influenced by different phenotypes and lifestyles, such as BP, salt intake, and others.

A study by Salciccioli et al ¹⁷ compared the arterial stiffness of Afro‐American and Afro‐Caribbean individuals, all living in the United States, and they did not find a difference in this parameter between these two black populations. In our study, we evaluated two black populations, from Brazil and Angola, with different cultural and sociodemographic profiles. Our results showed that black individuals from Brazil show higher PWV compared with their counterparts from Angola, regardless of sex. The discordant results between our data and Salciccioli's ¹⁷ might be attributed to different methods of assessing arterial stiffness, as they measured the radial‐to‐carotid PWV whereas in our study we measured de cf‐PWV, which depends on the aortic stiffness. However, differences also may depend on the participant's characteristics, considering that in the Salciccioli study all participants lived in the same country where lifestyle and exposure to environmental factors were more uniform than that occurring in Brazil and Angola.

Another study also conducted in Brazil ⁶ including only workers of the public sector showed that racial differences in arterial stiffness are mainly determined by differences in BP and age, being age the major determinant of arterial stiffness in large arteries in that sample. However, black individuals were more prevalent in the highest quartile of PWV. ⁶ Aging causes a set of changes in the arterial wall, with increased collagen production and reduced elastin deposition, with a consequent acceleration of arterial stiffness. ¹⁸ In stiffer arteries, pulse wave travels faster contributing to higher values of the SBP and pulse pressure. ¹⁸ Our data confirm the association between PWV and age, showing that PWV increases with age in both Brazilians and Angolans, regardless of sex. However, although black participants from Brazil have higher PWV, the increase in PWV with age was greater in Angolan than in Brazilian men. This fact was confirmed even when participants were stratified into two age categories (<45 or ≥45 years), and the data were adjusted for confounders.

Considering that the Brazilian black population is mainly derived from the Angolan population, we can suppose a high degree of common ancestry in the two populations included in our study. Therefore, differences seem to result from the environment existing in the two countries. ¹⁹ , ²⁰ Accordingly, a study showed that 85% of black individuals from Southeast Brazil (Sao Paulo city) had a similar genetic origin to those from sub‐Saharan African countries, while only 1.7% had genetic origin from native Indians of Brazil. ²¹ From this perspective, we can consider that differences in arterial stiffness observed in this study may be weakly explained by differences in their genetic background, but mainly by environmental, nutritional, and behavioral factors.

The SBP is a strong determinant of arterial stiffness in the aging process. In our study, SBP was similar between Brazilians and Angolans men and women. However, changes in SBP were more pronounced in Angolans than in Brazilians, regardless of sex. When we analyzed the increase in cf‐PWV with age controlling for SBP, men from Angola showed a stepper increase in cf‐PWV in those participants aged over 45 years. By adjusting the PWV values for age, SBP, BMI, glucose, and uric acid, these values remained higher for Brazilians compared to Angolans, regardless of sex. Thus, the difference in the rate of increase of cf‐PWV can be explained by other factors that might not be considered in this study, such as genetic determinants, socioeconomic factors, behavioral, and lifestyle patterns related to differences in the aging process.

Regarding women, we observed that PWV was higher in Brazilians than in Angolans at any of the age categories and the slope obtained from the linear regression was similar between Brazilian and Angolan women, unlike the results obtained in men. Although the influence of sex on cardiovascular risk profile is well described, ²² , ²³ the influence of sex on the arterial stiffening is still under debate, as the results of several studies are controversial. Some studies have reported a marked increase in age‐dependent arterial stiffening in women than in men, mainly due to menopause ²⁴ , ²⁵ while other studies have not found significant differences between men and women. ²⁶ Rossi et al ²⁷ published a review synthesizing the results from intersexual comparative studies on arterial structure and function at different stages of life and suggested that women had arteries more rigid than men. Also, the functional manifestation of this difference is attenuated by sex steroids over the years of women's reproductive life.

The present study should be seen in light of some limitations. Although the methods for data collection have been the same between these two studies, the use of two independent studies with different designs may limit data comparisons between these two samples. Contrary to the study conducted in Brazil, which selected a random sample of the adult population, in the Angolan study, participants were civil servants from a University. It may reflect in unbalanced socioeconomic, educational, and behavioral characteristics between these two samples. Although our multiple models were designed to control for possible confounders, we believe that such differences might influence the magnitude of certain associations. Besides, sample size differences may affect the power of the study for some analyzes. However, the comparison of two different black populations, from different continents, provides new insight into the study of arterial stiffness in different racial groups.

In summary, we evaluated two black populations from Brazil and Angola, with different cultural and sociodemographic profiles. We showed that black individuals from Brazil have higher PWV compared to those living in Angola. However, the increase in PWV with age is higher in Angolan men than in Brazilians, but not for women, highlighting for sex differences in the arterial stiffening in different black populations.

CONFLICT OF INTEREST

The authors have no conflict of interest to declare.

AUTHOR'S CONTRIBUTION

VPDC, CWOG, VBS, and AFPC participated in the data acquisition, analysis and interpretation of data, and drafting the manuscript article; ABTS, DPC, PM, DAZ, and ALSG participated to the conception and design, data acquisition, analysis and interpretation of data, and the final approval of the version to be published; RSC, JGM, and MPB participated in the conception and design, data acquisition, analysis and interpretation of data, drafting of the article and critical revision for important intellectual content, and the final approval of the version to be published.

Cruz VPD, Gonzaga CWO, da Silva VB, et al. Arterial stiffness in black adults from Angola and Brazil. J Clin Hypertens. 2020;22:1469–1475. 10.1111/jch.13962