The effect of exercise training on blood pressure in African and Asian populations: A systematic review and meta-analysis of randomized controlled trials

Abstract

Background

Exercise is key in the primary prevention and management of hypertension. Yet, current exercise recommendations are predominantly based on meta-analyses involving populations of European descent. Since blood pressure (BP) responses to pharmaceutical interventions are known to differ among ethnic groups, we aimed to investigate the BP responses to exercise training in non-European descendants.

Purpose

The aim of this study was to systematically summarize the available literature on the efficacy of exercise on BP in healthy adults (age ≥18 years) of African or Asian origin.

Methods

We searched the MEDLINE database for randomized controlled trials that evaluated the effect of exercise training on BP in healthy African and Asian adults with optimal BP, elevated BP or hypertension and published in a peer-reviewed journal up to May 2019. Random effect models were fitted to estimate the effect sizes.

Results

We identified 22 trials involving individuals of Asian origin (n = 931; mean age: 44 years; 41% male) and four trials involving individuals of African origin (n = 510; mean age: 56.7 years; 80% male). Aerobic exercise training significantly (p < 0.001) reduced systolic and diastolic BP in each ethnic group. Resistance training did not affect the BP of Asian participants with optimal BP. The effect of resistance training in Asians with elevated BP or hypertension and Africans could not be determined due to lack of data. Sub-analyses suggested somewhat larger reductions in systolic BP following aerobic training in hypertensive Africans compared with hypertensive Asians.

Conclusions

We found favorable effects of aerobic exercise training on BP in the African and the Asian populations. However, the overall low number of studies and especially the lack of data on resistance training and combined training in African and Asian populations warrant more research to improve the quality of evidence.

Introduction

Worldwide, hypertension is responsible for almost 19% of all deaths¹ and remains the leading risk factor for cardiovascular disease and all-cause mortality.² Over the last four decades the number of individuals with raised blood pressure (BP) almost doubled.³ This increase was largest in low- and middle-income countries in South Asia and Sub-Saharan Africa and stemmed largely from growth and ageing of the population.⁴ In order to address the burden of cardiovascular disease associated with elevated BP, governments have been urged to invest in population-based strategies and promote individual lifestyle management.⁵ In this context, present national and international guidelines recommend regular physical activity and exercise as part of the first-line therapy of primary and secondary prevention of hypertension.⁶ Bountiful evidence supports participation in at least 30 min of moderate-intensity aerobic exercise (walking, jogging, cycling or swimming) on 5–7 days per week for all pre- and hypertensive patients.^6,7 Performance of resistance exercises on 2−3 days per week has also been recommended as an adjunct therapy to lower BP.^6,7 These exercise guidelines, which are uniform across all different ethnic groups, are based mainly on meta-analytic research demonstrating clinically significant reductions in systolic and diastolic BP following aerobic exercise^8,9 (–4.7 mmHg/−3.2 mmHg) or dynamic resistance training⁷ (−2.8 mmHg/−2.3 mmHg). Unfortunately, the included studies in these meta-analyses involve predominantly populations of European descent. Yet, prevalence rates of hypertension are known to differ among ethnic groups. For instance, in the United States, at any decade of life, African descendants have a higher prevalence rate of hypertension than Hispanic Americans, Whites and Native Americans.¹⁰ Similar observations were found in Suriname¹¹ and the Netherlands.¹² Moreover, the efficacy of pharmacological antihypertensive treatments is heterogeneous among ethnic groups, resulting in distinct pharmacological management in African populations.⁶ Namely, the blood pressure lowering efficacy of renin–angiotensin system inhibitors is attenuated in African descendants due to greater likelihood of having a low renin profile.⁶ Earlier, we showed that one of the potential BP lowering mechanisms of exercise is a reduction in plasma renin activity (i.e. –19.8% (95% confidence interval (CI), –35.0 to –4.7); p < 0.05).¹³ Hence, it could be hypothesized that the effect of exercise on BP is different in non-European descendants.

However, to the best of our knowledge, BP responses to lifestyle interventions, and in particular exercise therapy, among non-European descendants have been scarcely investigated. Therefore, we performed a systematic review with meta-analysis to 1) evaluate the effectiveness of exercise interventions to reduce BP in healthy adults of African or Asian origin and 2) compare the efficacy of exercise interventions between both ethnic groups.

Methods

We followed the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines to conduct this systematic review and meta-analysis¹⁴ and registered the protocol a priori with PROSPERO, the International Prospective Register of Systematic reviews (trial registration number CRD42018102077).

Search strategy

A database of RCTs on the effect of exercise training on BP was initiated in 1985 and updated in 1994, 1999, 2003 and 2012.^8,13,15–17 From this database, we retrieved trials investigating the effect of exercise on BP in African descendants or individuals from African-origin (henceforward called Africans) or Asian descendants or from Asian-origin (henceforward called Asians). We then performed a new search in the MEDLINE database, covering the period from March 2012 up to May 2019. The full search strategy for this database is provided in Supplementary Material file 1 online. Additionally, we manually searched the bibliographies of recent meta-analyses on the topic for other potentially eligible papers. We did not impose any language restriction.

Study identification and selection

After removal of duplicates, two independent researchers (MB and SMB) screened the titles and abstracts of all retrieved articles according to predefined inclusion criteria. A study was considered eligible for inclusion if it was 1) a RCT of exercise training for a minimum of four weeks; 2) performed in healthy adults (age ≥18 years) of African or Asian origin; 3) with optimal BP (systolic BP (SBP): <120 mmHg and diastolic BP (DBP): <80 mmHg), elevated BP (SBP: 120–129 mmHg or DBP: <80 mmHg) or hypertension (stage 1, SBP: 130–139 mmHg or DBP: 80–89 mmHg; stage 1, SBP: ≥140 mmHg or DBP: ≥90 mmHg) using the 2017 American College of Cardiology/American Heart Association classification;⁶ without cardiovascular, respiratory or metabolic disease; 4) reporting mean and standard deviation (or standard error) of BP at baseline and follow-up for the exercise and control group or the difference in means between the two groups and standard deviation of the difference (or standard error); and 5) published in a peer-reviewed journal. Following title and abstract screening, the same two researchers reviewed full text articles to determine study eligibility. Disagreements were resolved by discussion with a third reviewer (VAC). If data were not available, one reviewer (MB) contacted the corresponding author by email to obtain missing outcome data.

Data extraction

A standardized extraction sheet was used by both reviewers (MB and SMB) to extract the following data: publication year, country of origin, participant characteristics (gender, age, ethnicity, weight, height, body mass index, BP category), details on the intervention (exercise type, exercise frequency, intensity, duration, session duration, total duration of the intervention, dropouts) and details on the primary outcomes (BP) and secondary outcomes (i.e. BP lowering mechanisms). Presence of any disagreement was resolved through discussion with a third reviewer (VAC).

Assessment of risk of bias

The possibility of publication bias was assessed by plotting net changes in BP against sample size for each study group and interpreting p-values using Egger's regression. The quality of each study that met the inclusion criteria was evaluated using the Tool for the assEssment of Study qualiTy and reporting in EXercise (TESTEX).¹⁸

Statistical analysis

Descriptive data are reported as the mean and standard deviation (SD) or median and range. Meta-analyses were conducted using Review Manager, version 5.3 statistical software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark, 2014) for outcome parameters reported in at least three trials. The effect sizes were calculated by subtracting the pre-exercise values from the post-exercise values for both the intervention (Δ1) and control groups (Δ2). The net treatment effect was then obtained by subtracting Δ1 from Δ2. Inverse-variance weighting was applied to obtain weighted mean effect sizes using random-effects models to incorporate heterogeneity between studies. The following a priori determined subgroup analyses were conducted according to the following covariance: ethnicity (African versus Asian participants), BP category (optimal BP, elevated BP and hypertension), type of exercise training (aerobic exercise training (AET), resistance training, combined training). Tests for subgroup differences were conducted with statistical significance test at p < 0.05 using Chi². A 5% level of significance and 95% CIs was used for all outcomes. Heterogeneity was analyzed using Cochran's Q-test. Additionally, the degree of consistency between the trials in the meta-analysis was assessed using the I² test.

Results

Results of search

A PRISMA flowchart of our literature search and selection is presented in Figure 1. In the original database, we found 16 trials investigating the effect of exercise on BP among non-European descendants. Our new search identified 877 additional potentially relevant publications from which only five trials could be included. Five trials were identified through other sources. Finally, we included 26 RCTs involving 32 study groups published between 1989 and 2018.

Study design and participant characteristics

Table 1 gives a detailed summary of the study characteristics of the 26 included trials.^19–44 Twenty-two trials^23-44 evaluated the effect of exercise training on BP in Asians (n = 931, mean age: 44 (18–78) years; 41% male), whereas only four trials^19-22 were performed in Africans (n = 510, mean age: 56.7 (50–58) years; 80% male). The trials involving Asians were conducted in Japan (17 trials; n = 717), Korea (three trials; n = 84), Taiwan (one trial; n = 102) and India (one trial; n = 18) whereas the trials involving Africans were performed in Nigeria (two trials; n = 305) and in the USA (two trials; n = 205). Based on the average baseline BP reported in the original papers, 14 trials included a population with hypertension (n = 4 African; n = 10 Asian); three trials a population with an elevated BP (n = 3 Asian) and nine trials a population with an optimal BP (n = 9 Asian). The use of antihypertensive drug treatments was only reported in three trials, all including a hypertensive African population.

Outcome assessment and intervention characteristics

As shown in Table 1, office BP was measured in the supine (n = 9 trials) or sitting position (n = 10 trials); seven trials did not report the measurement position. Home BP measurements were performed in only one trial. For trials that reported on the type of instrument (n = 19), 13 used a (semi) automated BP measuring device, five used a standard sphygmomanometer and one used a plethysmographic device to record the SBP and DBP.

Eighteen study groups delivered AET (n = 13 Asian; n = 5 African), eight study groups applied resistance training (n = 8 Asian) and six study groups used combined training (n = 5 Asian). Among the AET study groups, 16 involved continuous training and two involved interval training (n = 150). The intervention duration ranged from eight to 26 weeks of training (median 12 weeks) with the frequency of exercise ranging between two and seven days per week (median 3 days/week). The intensity of AET was expressed as a percentage of the maximum heart rate (between 55% and 80%), as a percentage of heart rate reserve (between 40 and 80%), as a percentage of the maximal oxygen consumption (between 40% and 75%) or training at the anaerobic threshold. resistance training was performed at intensities between 50% and 100% of one repetition maximum with 8–20 repetitions per exercise. Additional details about the exercise interventions can be found in Table 1. The majority of trials advised their control participants to continue their normal daily lifestyle.

Publication bias and study quality

The funnel plots show that the net changes in SBP and DBP were symmetrical around the mean effect size lines in the Asian and the African populations (Supplementary file 2, Figures S1–S6). The quality assessment results are provided in Supplementary file 4 with an overall median score of 8 (range 6–12) for quality.

Intervention effect on blood pressure

Figure 2 shows the forest plots for systolic and diastolic BP following AET interventions in Asian and African participants. A significantly larger reduction in BP was observed in individuals assigned to an AET intervention compared with a control group (–8.69 (–12.07, −5.31) mmHg, p < 0.0001/–4.47 (–7.18, –2.48) mmHg, p < 0.0001) in the total group of non-Europeans, and in the subgroups of Asians (–7.23 (–10.59, –3.86) mmHg, p < 0.0001/–4.66 (–7.08; –2.23); p < 0.001) and Africans (–12.45 (–17.96, –6.94) mmHg, p < 0.0001/−4.12 (–7.18; –1.05) mmHg, p < 0.01). Furthermore, in the Asian subpopulation, BP reductions following AET were largest in the participants with elevated BP and hypertension compared with participants with optimal BP (Supplementary file 3 – Figure S7). Yet, this difference did not reach statistical significance (p = 0.11 for SBP; p = 0.33 for DBP).

A sub-analysis involving only AET interventions performed in hypertensive study groups showed non-significant larger reductions in SBP (p = 0.14) among Africans versus Asians. The results of AET on DBP (p = 0.57) were not different for hypertensive Asian versus hypertensive Africans (Supplementary file 3, Figure S8).

Since resistance and combined training interventions were not applied in the African population, the analyses of their effectiveness could be performed only for the Asian population. No BP lowering effect of resistance training could be demonstrated in six trials (nine study groups) of Asians with optimal BP (+0.4 (−2.47, 3.27)/–1.56 (–4.44, 1.32) mmHg, I² = 0%), whereas combined training significantly lowered SBP (–5.17 (–9.55, –0.79) mmHg, I² = 56%) and DBP (−5.27 (–7.37; –3.16) mmHg, I² = 0%) in two trials (Supplementary file 3, Figure S9). Only one trial with 39 participants evaluated the effect of combined training in Asians with elevated BP. The results were not significantly different between exercise and no intervention (Supplementary file 3 – Figure S10). Three trials (n = 162 participants) reported BP following combined training in hypertensive Asians and found a significant improvement in SBP (–9.36– (–19.53; 0.80) mmHg; I² = 83%) and DBP (–5.93 (–8.71; –3.15) mmHg; I² = 30%) (Supplementary file 3 – Figure S11). No trial has investigated the effect of resistance training in Asians with elevated BP or Asians with hypertension. An overview of all effect sizes is presented as a summary in Figure 3.

Discussion

This is the first review to systematically summarize the benefits of exercise in non-European descendants, including Asians and African descendants. Our findings show favorable effects of aerobic exercise training on BP in both ethnic groups. Limited data suggest a tendency towards larger reductions in SBP following aerobic exercise training in i) hypertensive patients of African origin compared with hypertensive patients of Asian origin and ii) in Asians with elevated BP or hypertension versus Asians with optimal BP. Further, we found no effect of resistance training on BP in Asians with optimal BP. Yet, the overall paucity of studies investigating the effects of different exercise modalities on BP in non-European descendants was striking and calls for further research in the field of exercise therapy for BP treatment in non-European descendants.

Aerobic exercise training lowered BP by –7.2/−4.7 mmHg in the Asian population. These finding corroborate the results of a recent meta-analysis by Igarashi et al. 2018,⁴⁵ who pooled data from studies performed by East Asians or in East Asia and documenting the potential of exercise as a viable BP lowering intervention (–4.7/–3.2 mmHg). The BP lowering effect of aerobic exercise in Asians seems a little larger compared with the results of the 2013 systematic review by Cornelissen and Smart 2013,⁸ who found that aerobic exercise training significantly lowered BP (–3.5/−2.5 mmHg) in a mixed population. Yet, in line with their observations, BP reduction in the Asian participants was also more pronounced in individuals with elevated BP or hypertension compared with individuals with optimal BP.

These results are also in line with Gorostegi-Anduaga et al. 2018,⁴⁶ who found a 16-week lifestyle intervention (aerobic exercise + diet) to significantly reduce pharmaceutical therapy in European descendants with hypertension stage 1 or 2. Future research is needed to confirm this in other ethnic groups.

Whereas Cornelissen and Smart 2013⁸ found similar reductions in BP following aerobic exercise training or dynamic resistance training in normotensive individuals and in individuals of predominantly European descent, we found no change in BP in response to dynamic resistance training in Asians with optimal BP. Regrettably, whether resistance training is able to reduce BP in Asian participants with elevated BP or hypertension could not be investigated given the lack of available data.

Among individuals of African origin, AET was the sole applied exercise intervention. Moreover, these five AET trials involved only hypertensive participants. The observed BP reduction (–12.5/–4.1 mmHg) confirms the findings of Whelton et al. 2002⁴⁷ and supports the application of AET to lower BP in Africans with hypertension. To our knowledge Whelton et al.⁴⁷ were the last to analyze the effectiveness of aerobic exercise interventions on BP in a subgroup of African descendants. Since their meta-analyses, published in 2002, only one additional trial could be identified. In comparison to the Asian and European descendants, their results indicated a significantly greater reduction in SBP in the African adults. Yet, given that all BP categories were pooled within each ethnic group and knowing that baseline BP impacts on magnitude of the BP response we compared in our meta-analysis hypertensive Africans with hypertensive Asians. Though non-significant, we confirmed a slightly greater effect on systolic BP in the hypertensive Africans compared with hypertensive Asians. Yet, our results still indicated a large difference among the baseline BP values, with the African-origin group having a baseline BP of 162.5 (±12.9)/96.4 (±5.7) whereas the mean baseline BP of the hypertensive Asian-origin group was only 141.2 (±10.4)/89.5 (±7.1). As such, it is likely that the more pronounced BP-lowering response in the Africans with hypertension could partly be attributed to the higher baseline BP. Moreover, 80% of the African-origin group were men in contrast to only 41% in the Asian-origin group. Earlier, it was shown that in a mixed population, male participants achieved greater than twice the reduction in SBP and DBP than female participants.⁸ Overall, with a mean age of 56.7 years, the African descendants were also older compared with the Asians (mean age 44 years). Yet, currently there is no evidence suggesting that the BP response to exercise therapy would be more pronounced with increasing age.⁸

A likely explanation for these differences between ethnic groups could be attributed to potential variation in physiological responses due to differences in genotype and/or phenotype.^48. However, given the scant data on the effect of exercise on known BP lowering mechanisms (i.e. renin–angiotensin–aldosterone system and/or autonomic function¹³) in these non-European descendants this could not be confirmed in this meta-analysis.

A number of limitations should be considered in our meta-analysis. Heterogeneity was high and could be explained by the presence of the different types of exercise interventions, ranging from resistance to aerobic exercises with different intensities, durations and frequencies. The effectiveness of resistance training and combined training could not be evaluated in the Africans in the absence of trials. Similarly, the lack of data for each of the BP categories restricted our comparison with the hypertensive population following aerobic training. Obesity, known to be highly associated with hypertension,⁴⁹ could not be taken into account because there were only scant data on weight and body mass index. The lack of data on the BP mechanisms made it impossible to draw any conclusions on potential differences in BP regulation that could explain the different BP response. Finally, due to lack of data on ethnicity, many studies had to be excluded and the lack of consensus in defining and classifying the racial/ethnic groups also resulted in difficulties when selecting studies. It is not always clear which national groups are included under terms such as ‘Asians’ and ‘Africans’ or subgroups such as ‘South Asians’ or ‘East Asians’. Especially, in studies performed in multi-racial and multi-ethnic countries such as the USA or Brazil, it was difficult to interpret the racial/ethnic classification. It is recommended to future researchers to clearly specify the ethnicity to prevent misinterpretation and misclassification of the groups.

Conclusion

We found favorable effects of aerobic exercise training on BP in the African and the Asian populations. However, the overall low number of studies and especially the lack of data on resistance training and combined training in non-European descendants warrant more research to improve the quality of evidence.

Author contribution

MB and SMB are equal first author. MB and SMB contributed to the conception of the work. MB and SMB performed searches, extracted data and conducted statistical analyses. VAC contributed to the conception of design of the work and interpretation of data. MB and SMB drafted the manuscript and contributed to the acquisition, analyses and interpretation. VAC, NC and JT critically revised the manuscript and contributed to interpretation. All authors gave final approval and agree to be accountable for all aspects of the work ensuring integrity and accuracy.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Author notes