Cardiovascular reactivity and oxidative stress in young and older adults: the African-PREDICT and SABPA studies

Abstract

Background: Oxidative stress and increased cardiovascular reactivity are associated with endothelial dysfunction and cardiovascular disease development. These factors along with early vascular compromise are more pronounced in black populations. We aimed to compare cardiovascular reactivity and investigate associations thereof with oxidative stress in two bi-ethnic cohorts (younger: 25.0 ± 3.19yrs; older: 44.7 ± 9.61yrs).

Methods: Cardiovascular reactivity using the color-word conflict test was measured with the Finometer device. Oxidative stress markers included superoxide dismutase (SOD), γ-glutamyl transferase (γ-GT) and reactive oxygen species (ROS).

Results: Black groups displayed greater cardiovascular responses to stress than white groups. In younger white participants, diastolic blood pressure (DBP) (β = 0.31; p = 0.001) and mean arterial blood pressure (MAP) (β = 0.28; p = 0.002) associated with ROS. In older black participants, DBP (β = 0.23; p = 0.009), MAP (β = 0.18; p = 0.033), stroke volume (β = −0.20; p = 0.023) and arterial compliance (β = −0.25; p = 0.005) associated with γ-GT. In older white participants, systolic blood pressure (β = −0.20; p = 0.006) and MAP (β = −0.19; p = 0.009) associated with SOD.

Conclusions: In the older black group, cardiovascular reactivity associated with markers of glutathione metabolism, suggesting a possible compensatory up-regulation thereof in order to correct their heightened responses to stress. Independent of age, findings in the white groups support a regulatory role of ROS to maintain vascular tone during stress.

Trial registration: ClinicalTrials.gov identifier: NCT03292094.

Keywords:

• Acute stress
• color-word conflict test
• age
• reactive oxygen species
• glutathione metabolism

Background

Cardiovascular disease is a growing concern in the developing world [1,2], with South Africa displaying an alarming increase in the incidence of hypertension, especially amongst urban black populations [3–5]. One of the key role players in the development of hypertension is oxidative stress, which occurs due to either an increased production of reactive oxygen species (ROS), or a decrease in ROS scavenging by the antioxidant system [6]. Physiologically, ROS play an important role in vascular biology and cell signaling as it is involved in control of vascular tone and endothelial function, cell growth, differentiation and apoptosis [6–9]. However, oxidative stress aggravates endothelial dysfunction, vascular remodeling and inflammation which may result in elevated blood pressure, peripheral resistance, arterial stiffness and atherosclerosis [6,7,9], all of which worsen with aging [10,11]. Acute stress has also been associated with both endothelial dysfunction as well as oxidative stress [12,13], and a heightened cardiovascular response to stress is a risk factor for the development of cardiovascular disease, including hypertension [14–16]. Upon acute stress, black populations have shown increased cardiovascular reactivity when compared to white populations [17–19]. Black South Africans are also more prone to early vascular changes [20], and previous results linked increased blood pressure to oxidative stress [21,22]. However, it has not been previously investigated whether oxidative stress plays a role in a heightened cardiovascular response in black and white South Africans, and whether this link is age dependent. Therefore we aimed to compare oxidative stress and cardiovascular reactivity profiles between two age-stratified South African cohorts, and to determine whether relationships exist between cardiovascular reactivity and oxidative stress markers.

Methods

Study populations

This study is embedded in the baseline phases of the African Prospective study on the Early Detection and Identification of Cardiovascular disease and Hypertension (African-PREDICT) study (mean age: 25.0 years) and the Sympathetic activity and Ambulatory Blood Pressure in Africans (SABPA) study (mean age: 44.7 years).

After an initial screening phase (evaluating blood pressure, cholesterol, glucose levels, anthropometry, HIV status and a general health questionnaire) apparently healthy normotensive or pre-hypertensive (systolic blood pressure (SBP) <140 and diastolic blood pressure (DBP) <90mmHg) black and white men and women between 20 and 30 years of age were invited to participate in the African-PREDICT study [23]. Participants who are not permanent residents of Potchefstroom or surrounding areas (or plan regular trips to the area) and who are unable to read or understand English were excluded from the study. Participants with elevated glucose levels >5.6 mmol/L (or glycated haemoglobin ≥6.5%), who were HIV infected and those with ear temperature >37.5 °C were also excluded. Further exclusion criteria included previous diagnoses of (or took medication for) type 1 or 2 diabetes mellitus, liver disease, cancer, tuberculosis, renal or cardiovascular diseases, who recently had surgery or trauma and who was pregnant or breastfeeding. This study included the first 387 participants with complete antioxidant enzyme activity data divided into black (N = 191) and white groups (N = 196).

Detail on the SABPA study was published elsewhere [24]. In summary, participants were between 20 and 65 years of age, and participants who were pregnant, lactating, exhibited ear temperature >37 °C, used alpha and beta blockers, confirmed psychotropic substance abuse and those who donated blood or were vaccinated 3 months prior to their participation were excluded. We included 409 participants of the SABPA study divided into black (N = 200) and white (N = 209) groups.

Both studies complied with the Declaration of Helsinki regarding investigations among human participants, and were further approved by the Health Research Ethics Committee of the North-West University. The African-Predict study was registered as a clinical trial as required (ClinicalTrials.gov Identifier: NCT03292094). All participants were informed of each study in their home language, including stressor test protocols, and written informed consent was obtained before commencement.

Questionnaires

Participants completed a general health questionnaire to obtain demographic data and information on alcohol usage.

Anthropometric and physical activity measurements

All anthropometric measurements were performed using standardized methods [25]. These measurements included body height measured with a stadiometer (SECA 213 SECA, Hamburg, Germany and Invicta, IP 1465, Invicta, London, UK), body weight measured with an electronic scale (SECA 813, SECA, Hamburg, Germany and Precision Health Scale, A & D Company, Tokyo, Japan) and waist circumference measured with a metal tape (Lufkin Steel Anthropometric Tape, W606PM, Lufkin, Apex, MD, USA and Holtain unstretchable flexible 7mm wide metal tape, Crosswell, Wales). Body mass index (BMI) was calculated as kg/m² and body surface area was calculated as m² according to the Mosteller formula [26]. Total energy expenditure (TEE) was monitored over 7 days in the African-PREDICT study (ActiHeart physical activity monitor, CamNtech Ltd., England, UK) and over 24 hours in the SABPA study (Actical® activity monitor, Mini Mitter Co., Inc., Bend, OR; Montreal, Quebec, Canada).

Cardiovascular reactivity

Cardiovascular measurements were made non-invasively using the validated Finometer device (FMS, Finapres Medical Systems, Amsterdam, Netherlands). The Finometer device provides a beat-to-beat arterial blood pressure waveform after adjusting for hydrostatic height from the heart while computing additional cardiac parameters by way of the Modelflow method [27]. The finger cuff of the Finometer was fitted on the left middle-finger, and after a 10 minute resting period, a 5 minute continuous measurement of resting variables was recorded.

During the recording, after 2 minutes, a return-to-flow systolic calibration was performed to provide an individual subject-level adjustment of the finger arterial pressure with the brachial artery pressure, which allows for the highest precision in measurements. Resting measurements included SBP, DBP, mean arterial pressure (MAP), stroke volume (SV), total peripheral resistance (TPR) and Windkessel arterial compliance (Cwk). Thereafter the color-word conflict test was applied for 1 minute during which the cardiovascular variables were measured. The mean of the last minute of the resting values, and the last 20 seconds during exposure were used for the calculations. Cardiovascular reactivity was calculated as the percentage change from resting values [28].

Biochemical measurements

Fasting blood samples were obtained from each participant by a registered nurse from the antebrachial vein. Serum and plasma samples were prepared using standardized procedures, aliquoted into cryovials and stored in bio-freezers at −80 °C until analysis to maintain efficient long-term stability of antioxidant enzymes [29]. Antioxidant enzyme activities measured included glutathione reductase (GR), glutathione peroxidase (GPx), superoxide dismutase (SOD) and γ-glutamyl transferase (γ-GT). Additional oxidative stress markers included total glutathione (tGSH) and reactive oxygen species (ROS) (measured as serum peroxides and reported in units, where 1 mg H₂O₂/L equates to 1 unit)[30]. In addition, fluoride plasma glucose and serum total cholesterol, triglycerides, cotinine, interleukin-6 (IL-6) and cortisol levels were determined. All apparatus used are summarized in Table S1.

Statistical analyses

Statistica version 13.2 (Dell, TX, USA) was used to perform the statistical analyses of this study. Normal distribution of data was assessed and the central tendency and spread for normally distributed variables were expressed as arithmetic mean and standard deviation. Skewed variables were logarithmically transformed (triglycerides, IL-6, total cholesterol, SOD and γ-GT in the African-PREDICT study and glucose, triglycerides, IL-6, GR, SOD and γ-GT in the SABPA study) and expressed as the geometric mean and the 5th and 95th percentile intervals.

Interactions of race and sex were tested on the relationships between main cardiovascular reactivity variables and oxidative stress markers using multiple regression analyses. Continuous variables were compared using independent T-tests, while proportions were compared using Chi-square tests. Single regression analyses were performed to evaluate unadjusted associations between cardiovascular variables and oxidative stress related variables. Thereafter, partial correlations were performed to evaluate the associations between variables while adjusting for age, sex and BMI. Multiple regression analyses were performed to evaluate the independent associations between cardiovascular reactivity variables and oxidative stress markers. Covariates entered into the models included the relative oxidative stress marker, age, sex, BMI, TEE, self-reported alcohol use, cotinine, glucose, IL-6, triglycerides and cortisol. Sensitivity analyses were performed by repeating the multiple regression analyses after including statin usage and anti-hypertension medication usage into the models.

In post hoc analyses the achieved power (1−β) was determined for multiple regression models with 11 covariates, with a group size of N = 191 for black (1−β = 0.97) and N = 196 for white African-PREDICT (1−β = 0.97), and N = 200 for black (1−β = 0.97) and N = 209 for white SABPA (1−β = 0.98) (G*power v3.1.9.2) [31].

Results

Characteristics of the study population

No interactions of sex on the relationships between cardiovascular reactivity and oxidative stress markers were found. Meanwhile, interactions of race were found on the association of Cwk with SOD (β = 0.285; p = 0.026) in the African-PREDICT study, and on the associations of SBP with SOD (β = −0.477; p = 0.018), MAP with SOD (β = −0.578; p = 0.004) and TPR with tGSH (β = 0.635; p = 0.015) in the SABPA study. Stratification was done accordingly to compare black and white groups.

Black participants of the African-PREDICT study were younger (24.4 ± 3.39; p < 0.001) with a lower BMI (24.5 ± 5.41; p = 0.013) than their white counterparts, while in the SABPA study the black participants displayed a higher BMI (30.1 ± 7.00; p < 0.001) than the white group (Table 1).

Table 1. Characteristics of black and white groups.

	African-PREDICT study			SABPA study
	Black	White	p values	Black	White	p values
N	191	196		200	209
Sex, females (%)	56.0	60.7	0.349	50.5	48.3	0.660
Age (years)	24.4 ± 3.39	25.5 ± 2.87	<0.001	44.4 ± 8.11	45.0 ± 10.9	0.495
Anthropometric measurements
Body mass index (kg/m^2)	24.5 ± 5.41	25.9 ± 5.82	0.013	30.1 ± 7.00	27.6 ± 5.94	<0.001
Body surface area (m^2)	1.72 ± 0.18	1.91 ± 0.27	<0.001	1.92 ± 0.23	2.00 ± 0.28	0.001
Waist circumference (cm)	77.2 ± 10.9	82.3 ± 15.4	<0.001	93.6 ± 15.5	93.0 ± 16.1	0.707
Biochemical analyses
Glucose (mmol/l)	3.87 ± 0.84	4.71 ± 0.85	<0.001	5.41 (4.04; 10.4)	5.62 (4.70; 6.90)	0.055
Total cholesterol (mmol/l)	3.78 (2.70; 5.48)	4.66 (3.30; 6.43)	<0.001	4.60 ± 1.19	5.54 ± 1.28	<0.001
Triglycerides (mmol/l)	0.77 (0.41; 1.56)	0.98 (0.45; 2.18)	<0.001	1.16 (0.49; 3.66)	1.01 (0.44; 2.79)	0.024
Interleuken-6 (pg/ml)	0.97 (0.40; 3.29)	0.74 (0.27; 2.55)	<0.001	1.13 (0.34; 3.07)	0.92 (0.30; 3.04)	0.002
Cortisol (nmol/l)	422 ± 186	489 ± 277	0.009	355 ± 152	384 ± 160	0.063
Oxidative stress markers
Reactive oxygen species (Units)¹	197 ± 70.9	180 ± 89.3	0.033	192 ± 63.0	158 ± 58.2	<0.001
Total glutathione (µM)	1231 ± 295	931 ± 248	<0.001	895 ± 189	820 ± 173	<0.001
Glutathione peroxidase (nmol/min/ml)	18.5 ± 1.76	19.9 ± 1.34	<0.001	33.3 ± 14.0	36.2 ± 7.97	0.010
Glutathione reductase (nmol/min/ml)	52.9 ± 17.1	54.8 ± 16.1	0.263	7.06 (2.55; 16.8)	2.54 (0.25; 7.64)	<0.001
Superoxide dismutase (U/ml)	2.33 (1.27; 3.71)	2.19 (1.38; 3.83)	0.114	4.25 (0.83; 22.9)	4.13 (1.65; 7.92)	0.723
γ-glutamyl transferase (U/l)	25.4 (10.6; 82.8)	17.5 (7.40; 52.7)	<0.001	47.4 (20.1; 184)	19.3 (7.00; 76.0)	<0.001
Lifestyle
Cotinine (ng/ml)	64.5 ± 115	33.7 ± 86.7	0.003	27.3 ± 60.9	22.7 ± 77.5	0.514
Total energy expenditure (kcal/day)	2183 ± 364	2383 ± 504	<0.001	2685 ± 796	3113 ± 1597	0.001
Self-reported alcohol use (% yes)	64.9	65.8	0.845	26.0	49.0	<0.001
Statin use (% yes)	0	0	–	1.00	4.31	0.039
Anti-hypertension medication use (% yes)	0	0	–	34.5	12.9	<0.001

Data expressed as arithmetic mean ± standard deviation or geometric mean with 5^th and 95^th percentile boundaries, % or n.

Bold text indicate statistical significance.

¹Reactive oxygen species measured as serum peroxides where 1 unit = 1.0 mg/L H₂O₂.

Oxidative stress markers including ROS, tGSH, GPx and γ-GT revealed similar profiles in both studies with black participants having higher ROS (African-PREDICT: 197 ± 70.9; p = 0.033; SABPA: 192 ± 63.0; p < 0.001), tGSH (African-PREDICT: 1231 ± 295; p < 0.001; SABPA: 895 ± 189; p < 0.001) and γ-GT (African-PREDICT: 25.4 (10.6; 82.8); p < 0.001; SABPA: 47.4 (20.1; 184; p < 0.001) with lower GPx activity (African-PREDICT: 18.5 ± 1.76; p < 0.010; SABPA: 33.3 ± 14.0; p = 0.010) than their white counterparts. Although no differences were found in the African-PREDICT study, GR activity was higher in the black participants of the SABPA study (7.06 (2.55; 16.8); p < 0.001).

Cardiovascular reactivity

In the younger cohort, black participants displayed higher DBP, MAP and TPR but lower SV and Cwk at rest and after acute stress when compared to the younger white participants (Figure 1). Similarly, black participants of the older cohort also displayed higher DBP and MAP with lower Cwk at rest and after acute stress than the white participants. Additionally, this group also displayed higher SBP at rest and after acute stress along with a higher TPR than their white counterparts after acute stress only.

Figure 1. Rest and change in cardiovascular variables after the color-word conflict test showing significant differences between black and white groups of the African-PREDICT and SABPA studies.

In the younger black participants, a larger decrease in SV (p < 0.001) and smaller decrease in TPR (<0.001) were noted in comparison to the younger white participants (Figure 2). In the older black group, cardiovascular reactivity after applying the stressor revealed greater increases in DBP (p = 0.002), and MAP (p = 0.031) than their white counterparts. When comparing change in TPR the older black group experienced an increase in TRP whereas the white group experienced a decrease in TPR, and these responses differed significantly between the two groups (p = 0.014). The older black group also showed a significantly larger decrease in Cwk (p = 0.001) and SV (p < 0.001) when compared to their white counterparts.

Figure 2. Cardiovascular reactivity following the color-word conflict test showing significant differences between black and white groups of the African-PREDICT and SABPA studies. % Δ, Percentage change.

Regression analyses

In single and partial regression analyses (Table S2), we found an association of change in TPR with SOD in the young black group, however, this lost significance after full adjustments were made (R² = −0.004; β = −0.12; p = 0.145). In the younger white group, changes in DBP (R² = 0.05; β = 0.31; p = 0.001) and MAP (R² = 0.10; β = 0.28; p = 0.002) associated positively with ROS, while change in Cwk associated positively with SOD activity (R² = 0.09; β = 0.18; p = 0.023) (Table 2).

Table 2. Summary of multiple regression analyses of cardiovascular reactivity with markers of oxidative stress in participants of the African-PREDICT and SABPA studies.

	African-PREDICT study
	Black (n = 191)			White (n = 196)
	Adjusted R^2	β (95% CI)	p	Adjusted R^2	β (95% CI)	p
Diastolic blood pressure (% Δ)
Reactive oxygen species	−0.02	0.05 (−0.16; 0.25)	0.643	0.05	0.31 (0.13; 0.49)	0.001
Mean arterial pressure (% Δ)
Reactive oxygen species	−0.004	0.01 (−0.19; 0.21)	0.899	0.10	0.28 (0.10; 0.45)	0.002
Windkessel arterial compliance (% Δ)
Superoxide dismutase	0.01	−0.02 (−0.18; 0.15)	0.860	0.09	0.18 (0.03; 0.32)	0.023
	SABPA study
	Black (n = 200)			White (n = 209)
	Adjusted R^2	β (95% CI)	p	Adjusted R^2	β (95% CI)	p
Systolic blood pressure (% Δ)
Superoxide dismutase	0.06	0.03 (−0.12; 0.17)	0.708	0.04	−0.20 (−0.33; −0.06)	0.006
Diastolic blood pressure (% Δ)
γ-glutamyl transferase	0.03	0.23 (0.06; 0.40)	0.009	−0.04	0.08 (−0.10; 0.26)	0.361
Mean arterial pressure (% Δ)
Superoxide dismutase	0.05	0.10 (−0.04; 0.25)	0.178	0.01	−0.19 (−0.33; −0.05)	0.009
γ-glutamyl transferase	0.06	0.18 (0.02; 0.35)	0.033	−0.02	0.11 (−0.07; 0.29)	0.223
Stroke volume (% Δ)
Glutathione peroxidase	0.06	0.16 (0.02; 0.31)	0.032	−0.02	0.07 (−0.08; 0.22)	0.376
Superoxide dismutase	0.04	−0.09 (−0.23; 0.06)	0.233	0.0004	−0.14 (−0.28; 0.00)	0.052
Total glutathione	0.05	0.12 (−0.03; 0.28)	0.116	0.02	0.20 (0.05; 0.34)	0.009
γ-glutamyl transferase	0.06	−0.20 (−0.36; −0.03)	0.023	−0.02	0.02 (−0.16; 0.20)	0.813
Total peripheral resistance (% Δ)
Total glutathione	0.04	−0.15 (−0.30; 0.01)	0.062	0.01	−0.01 (−0.15; 0.14)	0.920
Windkessel arterial compliance (% Δ)
γ-glutamyl transferase	0.07	−0.25 (−0.41; −0.08)	0.005	−0.02	−0.04 (−0.22; 0.14)	0.652

Covariates included in each model included the relative oxidative stress marker, age, sex, body mass index, total energy expenditure, self-reported alcohol use, cotinine, glucose, interleukin-6, triglycerides and cortisol.

Bold text indicate statistical significance.

In the older black group, single, partial (Table S3) and multiple regression analyses (Table 2) indicated an increase in DBP (R² = 0.03; β = 0.23; p = 0.009) and MAP (R² = 0.06; β = 0.18; p = 0.033) to be positively associated with γ-GT levels. Meanwhile, a decrease in SV (R² = 0.06; β = −0.20; p = 0.023) and Cwk (R² = 0.07; β = −0.25; p = 0.005) was negatively associated with γ-GT levels. In the same group, an increase in TPR was borderline negatively associated with tGSH (R² = 0.04; β = −0.15; p = 0.062). The borderline association of decreased SV with GPx in partial regression analyses (p = 0.054) became significant after multiple regression analyses (R² = 0.06; β = 0.16; p = 0.032) in the older black group. In the older white group, increased SBP (R² = 0.04; β = −0.20; p = 0.006), increased MAP (R² = 0.01; β = −0.19; p = 0.009) were inversely associated with SOD. In the same group a decrease in SV was positively associated with tGSH (R² = 0.02; β = 0.20; p = 0.009) and negatively associated with SOD (R² = 0.0004; β = −0.14; p = 0.052). All other associations lost significance after full adjustments.

Sensitivity analyses

The African-PREDICT study excluded participants using statins and anti-hypertension medication. After adding these variables as covariates into multiple regression analyses in the SABPA study, only the borderline negative association of TPR with tGSH in the black group became significant after including anti-hypertension medication into the model (R² = 0.05; β = −0.17; p = 0.039).

Multiple regression analyses of SV with oxidative stress markers in the older group were repeated after substituting stroke volume with stroke volume index (SVI) (calculated as stroke volume/body surface area) as the main dependent variable in the model. Similar results were obtained in sensitivity analyses showing a significant negative association of SVI with γ-GT (R² = 0.06; β = −0.20; p = 0.023) in black groups, and a positive association of SVI with tGSH (R² = 0.02; β = 0.20; p = 0.009) in white groups. The negative association of SVI with SOD (R² = 0.0004; β = −0.14; p = 0.052) was borderline significant in the older white group after full adjustments.

Discussion

We found age-related disparities in the link between cardiovascular reactivity and markers of oxidative stress in black populations. In the young black group TPR decreased less and SV decreased more, but we found no link between cardiovascular reactivity and markers of oxidative stress. In the older black group with a heightened cardiovascular response to stress, increased TPR and decreased SV and Cwk was associated with markers of glutathione metabolism. In contrast, our findings in the white groups indicated that cardiovascular reactivity was associated with ROS and important determinants of ROS availability, namely SOD and tGSH, independent of age.

Our findings in the older black group confirm results from a previous study by our research team, in which older black men displayed increased TPR reactivity to acute stress [18]. Increased TPR reactivity may enhance the afterload and subsequently decrease SV, which related to end-organ damage in this cohort [18]. Similar findings were noted in the younger black adults described in the present study, where TPR decreased less and SV decreased more when compared to the white adults during stress. This is in agreement with previous findings in another young black group (29.9 ± 2.4 years) where it was suggested that increased TPR may play a role in the pathogenesis of hypertension in blacks [19]. Despite evidence of an altered cardiovascular response in the young black group, we were unable to link this to any oxidative stress markers investigated in this study. This finding may suggest that factors other than ROS, such as enhanced arterial stiffness and premature vascular aging, may play a more prominent role in cardiovascular reactivity in young black South Africans [20,32].

In the older black group, on the other hand, increased TPR and decreased SV and Cwk were associated with markers of glutathione metabolism. Glutathione, a powerful antioxidant, exists in either reduced (GSH) or oxidized (GSSG) form [33]. The synthesis of glutathione is dependent on the intracellular availability of its amino acid constituents. While γ-GT was originally considered a marker for alcohol abuse and liver dysfunction, the enzyme acts as an oxidative stress marker since it is involved in the breakdown of extracellular glutathione to increase the availability of amino acids for intracellular glutathione synthesis [34]. In this regard γ-GT has a dual function since it is also able to produce ROS directly [35–38]. The link between heightened cardiovascular reactivity and markers of glutathione metabolism may be explained by two possible mechanisms; firstly that acute stress may induce increased shear stress which may promote oxidative stress [12,39], and secondly that oxidative stress may leads to a deficiency in nitric oxide, thus further diminishing endothelial function [6,7,40]. During conditions of oxidative stress, glutathione may be consumed at a higher rate, leading to possible increased expression of γ-GT in an effort to increase intracellular glutathione levels [36]. The elevated γ-GT in this group may then also perhaps be explained as compensatory in an effort to stabilize the redox balance, which is further strengthened by the presence of higher tGSH in this group.

In contrast, the findings in the white cohorts indicated that cardiovascular reactivity was associated with ROS and important determinants of ROS availability, namely SOD and tGSH, independent of age. This was also indicated in an experimental study in which a role for ROS in the regulation of blood pressure during and after acute stress was confirmed. Here it was suggested that a lack of ROS is implicated in a greater MAP reactivity and a longer recovery period from acute stress [41]. This finding supports the important role of physiologically regulated ROS signaling in the maintenance of vascular tone during stress, while suggesting a possible role for an adequate antioxidant system, especially SOD activity, in the prevention of a heightened cardiovascular response to stress in white participants regardless of age.

This study has to be interpreted within the context of its strengths and limitations. Both our study populations included participants from the Potchefstroom area, and may not represent the South African population as a whole. This being a cross-sectional study, causality cannot be inferred, and although our results were consistent after multiple adjustments, we cannot exclude confounding of unknown factors. Measurements of ROS were made using an indirect biomarker and we cannot discern the sources of ROS generation in this study. Participants were not screened or excluded based on previous exposure to the color-word conflict test, which may be considered for future studies in this field. However, to the best of our knowledge, this is the first study investigating the associations of oxidative stress with cardiovascular reactivity in 2 age-stratified cohorts of black and white South Africans. Both studies were well executed under controlled conditions in a fully-equipped research facility. This study offers further insight into the role of redox signaling in cardiovascular responses to acute stress.

In conclusion, age-related differences were found in the link between cardiovascular reactivity and markers of oxidative stress in the black cohorts. Heightened cardiovascular reactivity during the color-word conflict test associated with markers of glutathione metabolism in older black participants, suggesting a possible compensatory up-regulation of this system in order to correct their unfavourable cardiovascular responses to stress. Findings in the white groups support a possible regulatory role of ROS in the maintenance of vascular tone during stress, independent of age.

Acknowledgements

The authors are grateful towards all individuals participating voluntarily in the study. The dedication of the support and research staff as well as students at the Hypertension Research and Training Clinic at the North-West University are also duly acknowledged.

Disclosure statement

The authors have no conflicts of interest to disclose. Any opinions expressed and conclusions arrived at, are those of the authors alone.

Additional information

Funding

The research funded in this manuscript is part of an ongoing research project financially supported by the South African Medical Research Council (SAMRC) with funds from National Treasury under its Economic Competitiveness and Support Package; the South African Research Chairs Initiative (SARChI) of the Department of Science and Technology and National Research Foundation (NRF) of South Africa (GUN 86895); the Strategic Health Innovation Partnerships (SHIP) Unit of the SAMRC with funds received from the South African National Department of Health, GlaxoSmithKline R&D, the UK Medical Research Council and with funds from the UK Government’s Newton Fund; as well as corporate social investment grants from Pfizer (South Africa), Boehringer-Ingelheim (South Africa), Novartis (South Africa), the Medi Clinic Hospital Group (South Africa) and in kind contributions of Roche Diagnostics (South Africa).

The SABPA study was supported by the North-West University, Potchefstroom; the South African Medical Research Council (SAMRC), the National Research Foundation (NRF); the NRF Thuthuka (80643); ROCHE Diagnostics, South Africa, the Department of Education, North-West Province and the Metabolic Syndrome Institute, France.

Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors, and therefore, the NRF does not accept any liability in this regard.