The time course of physiological adaptations to high‐intensity interval training in older adults

Abstract Objective High‐intensity interval training (HIIT) has been shown to be more effective than moderate continuous aerobic exercise for improving cardiorespiratory fitness (CRF) in a limited time frame. However, the length of time required for HIIT to elicit clinically significant improvements in the CRF of older adults is currently unknown. The aim of this study was to compare changes in the CRF of older adults completing identical HIIT protocols of varying durations. Methods Forty healthy, community‐dwelling older adults completed a cardiopulmonary exercise test (CPET) before and after 2, 4, or 6 weeks of fully supervised HIIT on a cycle ergometer, or a no‐intervention control period. Results Anaerobic threshold (AT) was increased only after 4 (+1.9 [SD 1.1] mL/kg/min) and 6 weeks (+1.9 [SD 1.8] mL/kg/min) of HIIT (both P < 0.001), with 6‐week HIIT required to elicit improvements in VO2 peak (+3.0 [SD 6] mL/kg/min; P = 0.04). Exercise tolerance increased after 2 (+15 [SD 15] W), 4 (+17 [SD 11] W), and 6 weeks (+16 [SD 11] W) of HIIT (all P < 0.001), with no difference in increase between the groups. There were no changes in any parameter in the control group. Conclusion Improvements in exercise tolerance from HIIT precede changes in CRF. Just 4 weeks of a well‐tolerated, reduced‐exertion HIIT protocol are required to produce significant changes in AT, with a further 2 weeks of training also eliciting improvements in VO2 peak.

are in young individuals, often utilizing supramaximal protocols (ie, Wingate) that would not be tolerable for the majority of older adults. 14 Additionally, robust data on the time course of physiological adaptions to HIIT in older adults is lacking. One recent meta-analysis did evaluate the effect of HIIT over different durations and demonstrated greater benefit from longer (>4 weeks) training durations; however, that analysis only included studies in younger adults, including athletic populations, and considered a variety of different HIIT regimes. 12 Similarly, individual studies that have evaluated the time course of adaptations to a single HIIT protocol have so far only been conducted in younger population groups 15 and we are unaware of any studies of this nature in older adults.
Based on a previously published HIIT protocol that was both well-tolerated and effective at improving the CRF of older adults over a 4-week period, 16 the primary aim of this study was to determine the effect of either shortening or lengthening this training program on CRF adaptations in healthy older adults. We hypothesized that longer durations would lead to larger improvements, which may have clinical relevance given the mandated time constraints for interventions in certain age-associated clinical conditions. 16

| Pre-and post-intervention testing
Testing began with a whole-body dual-energy X-ray absorptiom-

| Interventions
Subjects assigned to a HIIT program attended our exercise laboratory three times each week for 2, 4, or 6 weeks to perform a previously published HIIT protocol. 16 In brief, each training session lasted 16.5 minutes, comprising a 2-minute warm-up of unloaded cycling followed by five 1-minute intervals at 90%-110% of the peak power output achieved during the pre-intervention CPET. Working intervals were interspersed with 90 seconds of active recovery cycling, with a final 3.5-minute active recovery and monitoring period. Each training session was fully supervised with continuous 12-lead electrocardiogram monitoring throughout. Subjects in the control group attended only for pre-and postintervention testing, 6 weeks apart. All subjects were asked to maintain their habitual diet and levels of physical activity for the duration of the study.

| Statistics
All calculations were performed using GraphPad Prism Version 7.02. Data are presented as mean (SD). Data were tested for normality and analyzed appropriately. Participant demographics and outcome variables at baseline were compared using one-way analysis of variance (ANOVA), whilst outcome data were compared using two-way ANOVA for time (pre-/post-intervention) and group with Sidak's post hoc testing. Significance was taken as an alpha of P < 0.05.

| Subject characteristics
Forty subjects, mean age (SD) 71 (5) years, mean BMI (SD) 25.8 (2.6) kg/m 2 , were recruited into this study. There were no significant differences in sex, age, baseline height or weight (Table 1), or in any outcome variable (all P > 0.1) between the groups at baseline. All 40 participants completed the study with none lost to follow-up.
Training compliance was 100% with no adverse events reported.

| Anaerobic threshold
There was a main effect of time across groups (P < 0.001) and a

| VO 2 peak
There was a main effect of time across groups (P < 0.002), but no sig-

| Exercise tolerance
For exercise tolerance, determined by peak power at termination of CPET (which is directly related to CPET duration), there was a main effect of time across groups (P < 0.001) and a significant Group ×  Figure 3).

| Body composition
There was no significant change in any of our body composition parameters (whole body lean mass, body fat percentage, and lean leg mass) in any group (Table 2), with no main effect of time across groups, nor significant Group × Time interaction.

| D ISCUSS I ON
This study has demonstrated that a time-efficient, well-tolerated HIIT program elicits appreciable improvements in exercise tolerance before any measurable changes in CRF in a cohort of older adults.
Two weeks of HIIT appears to be insufficient to improve AT, but this can be achieved with 4 or 6 weeks of training. Six weeks of HIIT is seemingly required to improve VO 2 peak. To our knowledge, this is the first study to attempt to plot the time course of CRF changes resulting from a single HIIT protocol in healthy older adults.
Supporting the efficacy of our specific HIIT protocol, other stud- Note: Data are presented as mean (SD). There were no significant differences between the groups.
improvements of ~4 mL/kg/min at 12 weeks. 21 However, these data are drawn from different HIIT protocols and not different durations of a single HIIT protocol and as such cannot be used to truly determine the temporal nature of adaptations to HIIT in older adults.
The physiological basis behind our observation that significant improvements in AT occur before increases in VO 2 peak may reside at the mitochondrial level. Mitochondrial biogenesis pathways, arguably the main determinant of AT, are seen to be upregulated after just a single HIIT session 23 and therefore could be important early in a training regime. Conversely, central cardiovascular adaptions (ie, blood volume and cardiac output), which are thought to be the main mechanisms through which HIIT may influence VO 2 peak, tend to occur later in a training programme. 24 Although we were unable to find any previous studies evaluating HIIT over durations as short as 2 weeks in older adults, those conducted in young men showed similar findings to ours: an increase in peak power F I G U R E 1 Anaerobic threshold before (pre) and after (post) 2-wk, 4-wk, or 6-wk high-intensity interval training (HIIT), or a no-intervention control period. N = 10 subjects per group. Analysis via twoway analysis of variance with Sidak's multiple comparison post hoc analysis. ***P < 0.001 vs pre-HIIT.

F I G U R E 2
Peak oxygen consumption before (pre) and after (post) 2-wk, 4-wk, or 6-wk high-intensity interval training (HIIT), or a no-intervention control period. N = 10 subjects per group. Analysis via two-way analysis of variance with Sidak's multiple comparison post hoc analysis. *P < 0.05 vs pre-HIIT. without significant changes in CRF. 25 It may therefore be that the peak power increases seen very early in training are a consequence of improved cycling efficiency and improved exercise tolerance, 25 after which changes in AT related to mitochondrial upregulation become apparent before, finally, central cardiovascular changes manifest.
Despite our HIIT program eliciting CRF and blood pressure improvements in just 4 weeks and physical function gains in just 2 weeks, it did not lead to any significant changes in body composition over 2, 4, or 6 weeks. This is despite previous studies in older adults demonstrating gains in lean mass 26 and reductions in body fat 27 with HIIT. This disparity may be due to differences in participant demographics and/or training modalities. For example, Herbert and colleagues 26 employed 30-second Wingate sprints in an exclusively male cohort of participants who were a decade younger than our cohort, including some masters' athletes, and who were able to demonstrate a ~3% increase in lean mass with 6 weeks of HIIT. Søgaard and colleagues 27 also studied a cohort of participants who were a decade younger than our participants and who were overweight or obese and failed to significantly increase in lean mass, but were able to significantly reduce their body fat percentage by ~0.6%. In support of the notion of exercise modality having a role to play in disparate F I G U R E 3 Peak power output before (pre) and after (post) 2-wk, 4-wk, or 6-wk high-intensity interval training (HIIT), or a no intervention control period. N = 10 subjects per group. Analysis via twoway analysis of variance with Sidak's multiple comparison post hoc analysis. ***P < 0.001 vs pre-HIIT. adults) has demonstrated cycling-based HIIT to be far less effective than running HIIT for reducing body fat percentage, 28 likely due to the recruitment of more muscle groups when running.

Group
One final consideration as to the reason for the lack of body composition changes must be given to diet, which was not controlled or measured in this study, although participants were instructed to maintain their habitual pre-study dietary behavior.
There is a possibility that our HIIT may have led to an increase in appetite and subsequent energy intake that may have diminished body fat reductions.
This study has also demonstrated significant reductions in resting systolic blood pressure with just 4 weeks of HIIT. This reduction of 8 mmHg systolic blood pressure is greater than that previously observed in trials of both moderate AET and resistance exercise training in older adults, 9 but is similar to that observed in the few previous studies of HIIT in older adults that have measured this. 22 Of interest, the 4 weeks that our HIIT protocol required to produce significant changes in blood pressure is much shorter than that described in the majority of exercise trials in older adults, with the majority lasting for 3 months or longer. 9 One important feature of the training program used herein is its tolerability, as was demonstrated by the 100% training compliance in a cohort with a mean age over 70 years. This demonstrates its suitability for use in older adults per se and in older patient groups, where adherence to previous exercise prehabilitation has been reported to be as low as 16%. 29 That our HIIT protocol can increase AT in 4 weeks would allow this protocol to be used as exercise prehabilitation before, for example, major surgical intervention for cancer where there is a limitation on the preoperative time available: for example, the 31-day time-tofirst-treatment cancer target in the UK. 30 An increase in CRF prior to major surgery has been shown to reduce both perioperative morbidity and mortality, 5 with ongoing large randomized clinical trials of exercise prehabilitation (ie, NCT03509428) currently underway to fully determine the effects of exercise prehabilitation on direct clinical outcomes (eg, complication rates and length of hospital stay).
Although 6 weeks of HIIT appears to be needed to improve VO 2 peak, this measure is prone to a substantial degree of subjectivity with participant effort a significant contributing factor. As such, our data would not support a delay to any preexisting clinical treatment targets beyond 4 weeks to achieve this.
As with all studies, we must acknowledge some limitations to our study design. That we did not have a single intervention group with repeated measurements at baseline, 2, 4, and 6 weeks is our main limitation. However, incorporating repeated CPET assessments every 2 weeks would have impacted delivery of our HIIT protocol (as these sessions would have had to replace HIIT sessions) and based on patient and public involvement feedback from our participants, was not something many of them would have been willing to undertake. We also acknowledge that prescribed exercise interventions interact with habitual physical activity and nutritional intake, aspects that were not measured in this study. We did, however, ask all participants to maintain their normal diet and levels of physical activity for the duration of the study.
In conclusion, this study is the first to map the time course of physiological adaptions to a single, well-tolerated HIIT protocol in older adults, demonstrating that a 5-by-1-minute HIIT protocol performed 3-times each week can elicit significant improvements in the exercise tolerance of older adults in just 2 weeks, whilst 4 weeks of this training regime can lead to significant improvements in CRF and resting systolic blood pressure.