Article Text
PDF
Abstract
Objective We aimed to compare the efficacy of isometric exercise training (IET) versus high-intensity interval training (HIIT) in the management of resting blood pressure (BP).
Design Systematic review and meta-analysis.
Data sources PubMed (MEDLINE), the Cochrane library and SPORTDiscus were systematically searched.
Eligibility criteria Randomised controlled trials published between 1 January 2000 and 1 September 2020. Research trials reporting the effects of IET or HIIT on resting BP following a short-term intervention (2–12 weeks).
Results 38 studies were analysed (18 IET and 20 HIIT), including 1583 (672 IET and 911 HIIT) participants, of which 612 (268 IET and 344 HIIT) were controls.
IET produced significantly greater reductions in resting BP compared with HIIT with systolic, diastolic and mean BP effect sizes of 8.50 mm Hg vs 2.86 mm Hg (Q=17.10, p<0.001), 4.07 mm Hg vs 2.48 mm Hg (Q=4.71, p=0.03) and 6.46 mm Hg vs 3.15 mm Hg (Q=4.21, p=0.04) respectively. However, HIIT reduced resting heart rate significantly more than IET (3.17bpm vs 1.34bpm, Q=7.63, p=0.006).
Conclusion While both modes are efficacious, IET appears to be the superior mode of exercise in the management of resting BP. However, HIIT may achieve wider physiological benefits, with greater reductions in resting heart rate.
- exercise
Statistics from Altmetric.com
Introduction
As the leading modifiable risk factor for mortality, arterial hypertension remains a global problem with important implications for cardiovascular health.1 Although largely efficacious, antihypertensive medication has a number of substantial limitations including adverse side effects,2 considerable economic strain3 and poor adherence rates, typically reported at less than 50% 1 year following prescription.4 Additionally, current exercise guidance is unlikely to benefit long-term cardiovascular risk5 due to poor global adherence.6 Therefore, establishing novel lifestyle approaches to the management of hypertension is pivotal.
In recent years, isometric exercise training (IET) and high-intensity interval training (HIIT) have emerged as convenient and highly time-efficient modes of exercise, which can produce clinically significant reductions in resting clinic and ambulatory blood pressure (BP) following short-term interventions.7–10 This is crucial as behavioural psychology research has identified perceived lack of time and motivation as the most common barriers to the adoption of physical activity.11 Importantly, both IET and HIIT interventions have previously been shown to produce a magnitude of BP reduction similar to, or greater than the current recommended international physical activity guidelines.12 13 Thus, given the efficacy and potential for greater adoption and adherence, these novel exercise modes may have significant clinical utility in the control of arterial BP.
Despite this, no research to date has attempted to compare the antihypertensive efficacy of these two distinct exercise modes. Furthermore, previous investigations have only examined the efficacy of HIIT against moderate intensity continuous training or as a subgroup to separate primary analyses, resulting in the omission of a plethora of comprehensive trials and thus an insufficient understanding of the overall effects of HIIT on BP. Additionally, several new IET studies have been published since the latest meta-analysis,14 many of which use a lower-body methodology, establishing a need for a lower versus upper body IET analysis. Therefore, we aimed to perform an independent meta-analysis on randomised controlled trials investigating the effects of IET on BP, with subgroup analysis of the implemented IET method. Separately, we also aimed to perform an independent meta-analysis measuring the effects of HIIT on BP. Finally, following both independent analyses, we performed a meta-analytical comparison to measure the efficacy of IET versus HIIT in the management of BP.
Methods
Search strategy
This systematic review and meta-analysis was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. To identify potential studies, a systematic search was performed using PubMed (MEDLINE), the Cochrane library and SPORTDiscus for research trials reporting on the effects of IET or HIIT on BP. The search strategy was developed by a research librarian experienced in scientific database searching and included MeSH terms, key words and word variants. The IET key Boolean search terms included: “Isometric training” OR “Isometric Exercise” OR “Static contraction” OR “Exercise training” AND “Blood pressure” OR Hypertension. The HIIT key Boolean search terms included: “High intensity interval training” OR HIIT OR “anaerobic training” OR “Sprint training” OR “Exercise training” AND “Blood pressure” OR Hypertension. Journal articles written in English and published between the 1 January 2000 and 1 September 2020 were considered. Where possible, corresponding study authors were contacted to ascertain whether non-published data was available or in the preprint stage, and studies found through the systematic protocol were screened for exterior citations and their respective reference lists searched for eligible studies.
Study eligibility
Screening was carried out following the exclusion of duplicate articles by two authors (JE and ADC) who resolved any confliction via consensus. Studies were initially screened by title and then by abstract for relevance. Studies retained for the next step of evaluation were assessed by full text and considered if they reported pre–post intervention BP changes following either an IET intervention or a HIIT intervention with a control group and sufficient randomisation. Only studies of intervention groups performing IET individually or HIIT individually with no other concurrent exercise were included; however, research designs using moderate intensity continuous training as a control arm were included. HIIT was defined in accordance with the EXPERT tool,15 in which >89% heart rate maximum or >80% VO2 peak was considered a sufficient intensity. Randomised controlled trials, including cross-over design studies with an intervention duration of 2–12 weeks were eligible. Participants were required to be ≥18 years of age with no limitations on health or baseline BP status.
Study quality
Study quality and risk of bias was evaluated using the TESTEX scale16 which is a 15 point (12 item) system designed for the specific application to exercise science research. Two reviewers (JE and ADC) independently scored all eligible articles. When disputes were detected in quality analyses, the reviewers met to discuss any conflicts and a third reviewer was consulted (JO’D).
Statistical analysis
Derivation of outcome statistics
Data were extracted from the studies independently by the first author. The key variables were four continuous measures of cardiovascular health before and after the study intervention; systolic BP (sBP), diastolic BP (dBP), mean BP (mBP) and heart rate with SDs. Where available we also extracted the exercise intervention duration, percentage of medicated participants, and percentage of hypertensive participants for use in the meta-regression analyses. If these data were not available, efforts were made to contact the authors of the paper to obtain them.
Pooling of results
Data were analysed with the meta17 packages in the statistical software R V.4.0.3 (R Core Team, 2020). Effect sizes were calculated based on the change score (mean difference, MD) between baseline and follow-up measures to demonstrate the clinical impact of each exercise type. Additionally, it has been suggested18 that change score provides a greater degree of accuracy when the baseline and follow-up scores are highly correlated and SDs are similar, as was the case in our dataset. The MDs with 95% CIs were computed individually for the four outcome measures. The effect sizes were weighted according to the intrastudy variability (calculated from reported means and SDs), with larger, more precise studies having a greater weighting on the overall effect size. Separate random-effects meta-analysis were run for each variable using CMA’s meta-regression function to establish if any moderator variables influenced the change in BP following IET or HIIT and explain any inter-study variance in outcomes. The moderators assessed independently were: hypertension diagnosis, medication status, TESTEX score, and intervention duration. This yielded the summary measures, significance levels (P), and the between-study heterogeneity (Cochran’s Q and I 2) and variability (τ2 ).
Exploration of publication bias
Publication bias was assessed via funnel plot inspection, with the expectation that studies would be evenly distributed between both sides of the average effect size, with high-precision studies close to the mean. The individual funnel plots are displayed in online supplemental figure S1. Quantitative assessments of Egger’s regression are reported for the four effect sizes alongside the funnel plots.
Supplemental material
PRISMA systematic review and meta-analysis flow chart. HIIT, high-intensity interval training; IET, isometric exercise training; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Exploration of heterogeneity and meta-regression analyses
Heterogeneity of the three effect sizes were assessed by Cochran’s Q and I2 values calculated in the random effects model. Statistical significance for Q was p<0.0119 and I2 of 25, 50% and 75% were interpreted as small, medium and large degrees of heterogeneity. In the event of significant levels of heterogeneity in any of the meta-analyses, exploratory meta-regression analyses were conducted to determine how participant and exercise characteristics accounted for heterogeneous effect sizes. The residual plots were inspected for suitability before reporting the R2 value of the fitted models. In the event of non-random residual plots, further variables were added to the model to capture possible significant interactions. All covariates were analysed separately and then together to assess the individual contribution of each variable. A method of moment’s estimator for the between-study covariance matrix in the random effect models meta-regression was used. A sub-group analysis was conducted based on the MD between study exercise-mode, and two groups were formed, which were IET and HIIT. On the IET data we performed additional subgroup analyses for the four outcome measures on upper versus lower body IET training.
Sensitivity analysis
We conducted a sensitivity analysis by removing outlying studies using the leave-one-out influence analysis method in the dmetar package.20 The study by May et al was found to contribute the most to the overall heterogeneity for all four outcome measures but did not influence the overall effect size to a large degree and was left in the final analysis.
Results
As presented in figure 1, our initial search identified 5220 manuscripts (2793 IET and 2427 HIIT). Following all exclusions, 38 studies were analysed (18 IET and 20 HIIT), including 1583 (672 IET and 911 HIIT) participants, of which 612 (268 IET and 344 HIIT) were controls. Study characteristics for the included IET and HIIT articles can be found in tables 1 and 2, respectively. Detailed TESTEX scoring for each category and the full reference list of all analysed studies are presented in online supplemental tables S1 and S2)
IET study characteristics
HIIT study characteristics
Isometric exercise training
IET significantly reduced sBP and dBP by a MD of 8.50 mm Hg (95% CI 6.49 to 10.52, p<0.001) and 4.07 mm Hg (95% CI 3.04 to 5.10, p<0.001) respectively, with no significant changes observed in the control group. mBP also significantly decreased following IET by 6.46 mm Hg (95% CI 4.18 to 8.74, p<0.001) with no significant change in the control group. There were significant differences between groups for all BP variables (all p<0.001). IET also had a significant effect on resting heart rate (RHR) by a MD of 1.34bpm (95% CI 0.24 to 2.44, p=0.0169) with no changes in the control group and a significant between-groups effect (p=0.006). The subgroup analyses for the effect of upper vs lower body isometric training were not significant for any of the four outcome measures (sBP, dBP, mBP and RHR).
There was significant heterogeneity between studies in the IET sBP (I2=78%), dBP (I2=52%) and mBP arms (I2=84%), as well as the dBP control group (I2=0.47). The Egger’s regression test was performed, which was only significant for dBP in the IET trials (p=0.04) (online supplemental figure S1).
High-intensity interval training
HIIT significantly reduced sBP and dBP by a MD of 2.86 mm Hg (95% CI 1.11 to 4.62, p=0.0014) and 2.48 mmHg (95% CI 1.49 to 3.48, p<0.001) respectively, with no significant sBP change observed in the control group, but a statistically significant increase in the control group dBP (MD: −0.50, 95% CI −0.961 to −0.041, p=0.03). mBP also significantly decreased following HIIT by 3.15 mm Hg (95% CI 0.96 to 5.34, p=0.0049) with no significant change in the control group. There were significant differences between groups for dBP (p<0.001) and mBP (p=0.038), with a (MD: 1.03, 95% CI −0.170 to 2.224, p=0.09). HIIT also had a significant effect on RHR by a MD of 3.17 bpm (95% CI 2.49 to 3.85, p<0.001) with no changes in the control group and a significant between-groups effect.
There was significant heterogeneity between studies in the HIIT sBP (I2=52%), dBP (I2=34%) and mBP arms (I2=47%), with no significant heterogeneity in the control group. Eggers test (1997) for publication bias was significant for asymmetry in sBP in the HIIT trials (p=0.02) (online supplemental figure S2).
Random-effects meta-analysis of the effects of IET versus HIIT in the reduction of SBP. HIIT, high-intensity interval training; IET, isometric exercise training; MD, mean difference; SBP, systolic blood pressure.
IET versus HIIT
There was a significant treatment difference between IET and HIIT trials for all variables. IET produced significantly greater reductions in resting BP compared with HIIT in sBP (MD: 5.29, 95% CI 3.97 to 6.61, Q=17.1, p<0.01, I2=86%; figure 2), dBP (MD: 3.25, 95% CI 2.53 to 3.96, Q=4.71, p=0.03, I2=70%; figure 3) and mBP (MD: 4.63, 95% CI 3.09 to 6.17, Q=4.51, p=0.04, I2=83%). However, HIIT reduced RHR to a significantly greater extent than IET (MD: 2.45, 95% CI 1.78 to 3.11, Q=7.63, p=0.006, I2=0%).
Random-effects meta-analysis of the effects of IET versus HIIT in the reduction of DBP. DBP, diastolic blood pressure; HIIT, high-intensity interval training; IET, isometric exercise training; MD, mean difference.
Meta-regression
There was no statistical significance for the moderator variables: hypertension diagnosis, medication status and intervention duration. As such, TESTEX score was the only statistically significant moderator, which explained a significant proportion of the dBP inter-study variance in the HIIT trials only (R2=75.41, p=0.014, 95% CI 0.137 to 1.205) (online supplemental figure S3).
Discussion
The aim of this study was to perform a meta-analytical comparison of randomised controlled trials to directly compare the efficacy of IET versus HIIT in the management of BP. While the findings of our analysis support both IET and HIIT as efficacious modalities in reducing resting sBP, dBP and mBP, this novel comparison reported a statistically significant between-treatment difference, with reductions of a greater magnitude in all BP measures following IET compared with HIIT. However, RHR was reduced to a significantly greater extent following HIIT compared with IET.
IET demonstrated significant sBP, dBP and mBP reductions of −8.5 to –4.07 and −6.46 mm Hg, respectively. Importantly, these reductions are clinically significant and of a similar, or greater magnitude to the reductions commonly observed following standard antihypertensive pharmacotherapy.21 Although the lower limit CIs were below what is generally accepted as clinically significant, HIIT produced significant reductions of −2.86 to –2.48 and −3.15 in sBP, dBP and mBP respectively. While not to the same magnitude as IET, these HIIT reductions are certainly comparable to the effects typically seen following traditional moderate intensity aerobic exercise22 which is currently recognised as the recommended modality for the management of BP in international guidelines.23 Indeed, a previous meta-analysis22 reported reductions in resting sBP and dBP by −3.84 and −2.58 mm Hg following traditional aerobic exercise, thus demonstrating the comparative efficacy of these modes and highlighting IET as unequivocally superior. Ultimately, the presented antihypertensive effects of both IET and HIIT are associated with a reduced risk of cardiovascular disease and all-cause mortality, therefore, providing imperative clinical implications regarding the optimal control of BP.24 25
The mechanistic underpinnings behind the antihypertensive effects of IET and HIIT are highly complex and still largely unknown, but are likely to share some homogenous physiological processes.26 27 Despite the significant reductions in RHR observed in our analysis, both interventions have often been shown to reduce BP with only small or no changes in CO.8 10 This suggests that BP is reduced following either intervention primarily via changes in total peripheral resistance. As such, both modes have been linked with enhanced shear stress as a mechanical stimulus to facilitate increases in endothelial intracellular calcium via potassium channel activation, ultimately promoting endothelial nitric oxide synthase and thus nitric oxide bioavailability.28 While this mechanism is conceptually consistent across both modes, the stimulating process of flow mediated dilation (FMD) differs between IET and HIIT within an acute setting,27 29 30 which although speculative, may be implicit in determining the observed differences in magnitude of BP reduction. Specifically, IET acutely produces such a mechanical response via contraction-induced occlusion of the relevant vasculature, thus causing reactive hyperaemia on relaxation to produce such mechanism, which has been demonstrated to translate into chronic FMD changes adjacent to significant improvements in BP.31 Conversely, HIIT simply invokes standard exercise hyperaemia, and despite the considerable stress placed on the relevant skeletal muscles with such workload intensity, current evidence indicates that the shear rate following HIIT is no greater than that of traditional aerobic exercise, thus potentially limiting the antihypertensive effects.28 32 However, the mechanistic rationale behind such differences in BP reducing effects between both modes is not understood and requires future investigation.
Interestingly, we found no significant differences for any of the measured parameters between upper (handgrip) and lower (wall-squat and leg-extension) body IET modes. While handgrip remains the traditionally investigated IET mode, more recent data has demonstrated lower-body IET to achieve reductions to a much larger magnitude.8 Such differences have been mechanistically attributed to the larger muscle mass involvement in lower body IET, generating a greater occlusion-response stimulus and thus enhancing the metabolic and autonomic regulatory adaptations.8 29 To what extent the lack of differences observed in the present analysis is owing to the disparity in published trials between IET modes is unknown, with future lower body IET research warranted before definitive conclusions on the optimal application of IET can be made. Furthermore, we also found no significant effects of medication and hypertension status on BP or RHR changes following IET or HIIT. Similarly, this finding may also be somewhat attributable to methodological limitations, with many studies failing to report exact participant characteristic data on medication or hypertension status, as well as the role of different medications and varying severities of hypertension increasing the complexity of this analysis.
Separately, both modes also show significant reductions in RHR with a significantly greater magnitude of change following HIIT compared with IET, perhaps indicating a more prominent role of RHR in the observed BP changes following HIIT than IET. Given the strong association between RHR and cardiovascular and all-cause mortality,33 these findings may be of clinical importance. Both IET and HIIT have been shown to produce significant improvements in cardiac autonomic regulation and baroreceptor sensitivity,8 10 which as well as mechanistically contributing to the observed BP changes, are likely to be implicit in these RHR reductions. Additionally, previous investigations from our lab in HIIT10 and acute IET34 have demonstrated significant improvements in cardiac function and mechanics following both interventions, providing potential mechanistic implications for such RHR changes. In speculating the differences in the magnitude of RHR change, it is well established that, despite being an anaerobic modality in nature, HIIT may induce significant peripheral and haematological adaptations that are generally associated with traditional aerobic exercise, such as increased skeletal muscle mitochondrial content,35 36 capillary density37 and plasma and blood volume,38 thus contributing to enhanced arteriovenous oxygen difference and stroke volume, resulting in vagally mediated reductions in resting RHR to a greater extent.27 Conversely, such physiological changes are not generally associated with IET.
Aside from RHR, these physiological adaptations raise an important point in distinguishing the appropriate application of either modality. In particular, these HIIT-induced adaptations may directly contribute to functional capacity and therefore have significant implications for clinical prognosis as one of the strongest predictors of outcome.39 Thus, it may be speculated that, while IET is superior in the management of resting BP, especially in hypertensive populations, HIIT may provide wider physiological advantages and greater clinical implications in individuals without hypertension, making it the preferred modality for the maintenance of general health. Therefore, since both modes may have different physiological advantages, future research should explore the application of both IET and HIIT in combination versus single-mode exercise training.
Limitations
As a primary limitation, we found significant heterogeneity for all BP outcomes in both interventions. Attributing this variance to methodological differences, our meta-regression accounted for hypertension diagnosis, medication status and intervention duration, but found no statistical significance for these moderators. Risk of bias TESTEX score explained a significant proportion of the HIIT dBP interstudy variance only. Nonetheless, random-effects models were applied to account for such heterogeneity. The Egger’s plots showed significant publication bias for sBP in the IET trials and dBP in the HIIT trials, suggesting studies that found no significant reductions in BP following either intervention may not have been published. Furthermore, the participants and investigators would have been aware of group allocation, as assessed in the TEXTEX scale. Thus, future IET and HIIT research should attempt to employ a sham-design methodology to blind the participants to their group allocation. It should also be considered that the present within-group analyses are subject to various biases, such as regression to the mean. There were also search limitations as we applied an English language filter, which carries the risk of introducing selection bias, and we did not search Scopus or Web of Science databases. Finally, this analysis did not compare IET or HIIT against MICT, which remains the most commonly applied mode, thus limiting the wider interpretations of the present results.
Clinical implications
Hypertension remains the leading risk factor for cardiovascular disease and all-cause mortality worldwide.1 Given the widely documented limitations that come with medication usage, it is crucial that effective non-pharmacological/lifestyle treatment options are established. IET and HIIT are short duration, highly time-efficient modes which have been shown to produce clinically significant reductions in resting BP at a magnitude superior to that of traditional exercise training modalities.7 9 This study demonstrates the efficacy of both interventions in the management of BP, highlighting IET as the superior mode in reducing BP, while HIIT remains more efficacious in reducing RHR. Specifically, at a magnitude of 8.5 mm Hg sBP and 4 mm Hg dBP, IET may produce reductions in BP to a greater extent than that typically achieved with standard dose antihypertensive monotherapy.40 While less effective for BP, HIIT may induce wider cardiovascular adaptations, making it the preferred mode in the general maintenance of health.
Conclusion
Both IET and HIIT are efficacious modes in the management of resting BP. IET appears the superior mode, producing a greater magnitude of BP reduction when compared with HIIT. However, HIIT remains a more effective mode in reducing RHR. Future research, ideally employing a sham-design protocol, is needed to establish the mechanistic underpinnings behind such differences in BP reducing effects and to explore the efficacy of both modes applied in combination.
What is already known?
Antihypertensive pharmacotherapy carries limitations and the current exercise guidelines for the control of resting blood pressure are limited by poor compliance.
Isometric exercise training and high-intensity interval training have been shown to improve markers of cardiovascular health.
Both modalities have promising future clinical utility in the management of resting blood pressure.
What are the new findings?
Compared with high-intensity interval training, isometric exercise training is the superior modality in the management of resting blood pressure, with reductions similar to, or greater than that of antihypertensive medication.
While still efficacious in reducing resting blood pressure, high-intensity interval training may achieve wider physiological benefits, with greater reductions in resting heart rate.
As such, the clinical implementation of isometric exercise training for blood pressure control should be considered, with high-intensity interval training likely preferred for the general maintenance of optimal health.
Ethics statements
Patient consent for publication
Ethics approval
This study does not involve human participants.
Acknowledgments
The authors would like to thank Michelle Crowther who is Canterbury Christ Church Universities learning and research librarian. Michelle assisted in developing the search strategy for this meta-analysis.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @JODriscoll9
Contributors JE, ADC and JO’D contributed to the conception and design of the study. JE and ADC contributed equally to this work. JE, ADC and JO’D contributed to the development of the search strategy. JE and ADC conducted the systematic review. JE, ADC and JO’D completed the acquisition of data. JE, ADC, JD and JO’D performed the data analysis. All authors assisted with the interpretation. JE and JO’D were the principal writers of the manuscript. All authors contributed to the drafting and revision of the final article. All authors approved the final submitted version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.