- Email: [email protected]
Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training
G Model
ARTICLE IN PRESS
JSAMS-2159; No. of Pages 6
Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training Michael Del Giudice, Jacob T. Bonafiglia, Hashim Islam, Nicholas Preobrazenski, Alessandra Amato, Brendon J. Gurd ∗ School of Kinesiology and Health Studies, Queen’s University, Canada
a r t i c l e
i n f o
Article history: Received 14 June 2019 Received in revised form 3 September 2019 Accepted 10 September 2019 Available online xxx Keywords: VO2 max Individual responses Reproducibility High-intensity interval training (HIIT)
a b s t r a c t Objectives: To test the hypothesis that observed maximal oxygen uptake (VO2 max) and time to fatigue (TTF) responses to two identical periods of standardized high-intensity interval training are reproducible. Design: Fourteen recreationally active and healthy young males completed two identical four-week periods of high-intensity interval training (4 × 4-min intervals at 90–95% maximum heart rate [HRmax ] separated by 3-min periods of active recovery at 70–75% HRmax ). Training periods were separated by a three-month washout period. Methods: VO2 max and TTF were assessed via incremental tests with supramaximal verification before and after each training period. Pearson correlation coefficients (r), intraclass correlation coefficients (ICC), and within-subjects coefficients of variation (CV) were used to assess reproducibility of observed VO2 max and TTF responses. Results: VO2 max and TTF values before the second training period were not significantly higher than baseline values and there were no significant (p > 0.05) interaction effects (period 1: VO2 max: +4.04 ± 2.29 mL/kg/min, TTF: +70.75 ± 35.87 s; period 2: VO2 max: +2.83 ± 2.74 mL/kg/min, TTF: +83.46 ± 34.55 s). We found very weak-to-moderate correlations and poor reproducibility for observed VO2 max (mL/kg/min: r = 0.40, ICC = 0.369, CV = 74.4) and TTF (r = 0.11. ICC = 0.048, CV = 45.6) responses to training periods 1 and 2. Conclusions: Our ANOVA results confirmed that the three-month washout period returned VO2 max and TTF levels to baseline and prevented carryover effects. Contrary to our hypothesis, our results suggest that individual observed VO2 max and TTF responses to identical training stimuli are not reproducible. © 2019 Published by Elsevier Ltd on behalf of Sports Medicine Australia.
Practical implications
1. Introduction
• Two four-week periods of high-intensity interval training resulted in similar mean changes in maximal oxygen uptake (VO2 max) and time to fatigue (TTF). • However, individual changes in VO2 max and TTF following two identical periods of high-intensity interval training were not reproducible. • These findings highlight a potential challenge for determining whether an individual has benefitted from a given exercise prescription.
Improvements in maximal oxygen uptake (VO2 max) are associated with reductions in the risk of all-cause mortality/morbidity.1 Interestingly, there is considerable heterogeneity in observed changes in VO2 max following standardized aerobic training.2–4 However, it remains unclear whether individuals demonstrate similar observed responses if they are repeatedly exposed to the same exercise training intervention.8,9 Findings from selective-breeding rodent studies10,11 and human twin,12 family,2 and genome-wide association13 studies provide evidence that genetics contributes to the variance in observed VO2 max responses to exercise training. This apparent influence of genetics raises the possibility that observed changes in VO2 max are reproducible if a group of participants are repeatedly exposed to the same exercise training intervention. Although one other study measured changes in VO2 max after repeatedly exposing participants
∗ Corresponding author. E-mail address: [email protected] (B.J. Gurd). https://doi.org/10.1016/j.jsams.2019.09.007 1440-2440/© 2019 Published by Elsevier Ltd on behalf of Sports Medicine Australia.
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
2
to the same exercise training intervention,14 this study did not assess intraindividual variability in observed VO2 max responses. Moreover, a recent consensus statement acknowledged that determining the reproducibility of observed VO2 max responses to training remains a major knowledge gap in the individual response literature.8 Therefore, the purpose of this study was to test the hypothesis that observed VO2 max responses to two identical periods of standardized exercise training are reproducible. Specifically, given the apparent role of genetics, we hypothesized that there would be high reproducibility between individuals’ observed VO2 max responses to two identical training periods. To test this hypothesis, we exposed a group of participants to two identical periods of standardized high-intensity interval training separated by a threemonth wash-out period. 2. Methods Forty-four potential participants were screened. Twenty-two enrolled and 14 completed all aspects of the study. Participants were only enrolled in the study if they met the following inclusion criteria: between 18 and 30 years of age, non-smokers, not taking any prescription medication, free of cardiometabolic disease, self-reported less than three hours of physical activity per week, and not involved in a systematic training program at the time of enrollment. All participants were instructed to maintain their regular physical activity and nutritional habits throughout the duration of the study. All experimental procedures were approved by the Health Sciences Human Research Ethics Board at Queen’s University (approval number: 6021938). Verbal and written explanation of the experimental protocol and associated risks were provided to all participants prior to obtaining written informed consent during a preliminary screening session. An a priori sample size was calculated for the primary outcome, the relationship between observed VO2 max responses following training periods 1 and 2. Given that ∼50% of the variance in observed VO2 max responses following training can be explained by genetics,2 we expected a Pearson correlation coefficient of r = 0.71 (i.e. an r2 value of 0.5). Using the sample size formula relevant to correlations,15 we determined that a sample size of 13 was needed (Z␣ = 1.96, Z = 0.84, C = 0.89, r = 0.71) to detect significance for a correlation with an expected coefficient of r = 0.71 with 80% power. The current study exposed the same individuals to two identical four-week training periods separated by a three-month washout period during which participants were instructed to return to their pre-study levels of physical activity. Considering the evidence supporting a potential learning effect associated with VO2 max testing,16 participants completed a familiarization test prior to each training period of the study. Hereafter, participants reported to the lab on three separate occasions separated by 24–48 h in the weeks preceding (PRE) and following (POST) training. During the first visit of PRE and POST testing, a resting muscle biopsy was taken from the vastus lateralis (biopsy data is not presented in this manuscript). Twenty-four hours after their biopsy visit, participants reported to the lab to complete a VO2 max test (day 2), and 24–48 h later participants returned to complete a second VO2 max test (day 3). Following training, the order of the three visits was identical to PRE and the first POST-training visit occurred 72–96 h after the final training session. All participants were asked to refrain from alcohol and caffeine 12 h before, and nutritional supplements and exercise 24 h before all physiological testing. All physiological testing and training were performed on a Treadmill (Sports Art Fitness 6300HR, Taiwan, R.O.C.). Participants consumed a standardized dinner the night before each VO2 max test (Stauffer’s Sauté Sensations [520 kcal; 74 g carbohydrate, 10 g fat, 32 g protein]) and arrived at the laboratory in
the morning following a 12-h overnight fast. Upon arrival, participants were fed a standardized breakfast (bagel [181 kcal] with 15 g of cream cheese [44 kcal]). Thirty minutes after breakfast consumption, participants completed a VO2 max test on a motorized treadmill following an incremental test protocol with a supramaximal verification phase. Gas exchange and heart rate were collected throughout the incremental test and verification phase using a metabolic cart (Moxus AEI Technologies, Pittsburgh, PA) and heart rate monitor (Polar Team2 Pro, Kempele, Finland), respectively. The incremental test protocol consisted of three minutes of resting data collection (participants were asked to stand on the treadmill and breathe normally) followed by a five-minute warm-up with the treadmill set to 2.5 mph at an incline of 2 and subsequent increases of either incline or speed every two minutes until volitional fatigue (see Supplemental Table 1 for details). Following the incremental test protocol, participants were provided with a minimum of 10 min of rest prior to commencing a supramaximal verification phase. The metabolic cart was not re-calibrated in between phases. During the supramaximal verification phase, participants ran until volitional fatigue at a speed that was 0.5 mph faster than the final stage attempted during the incremental test protocol. For each VO2 max test, VO2 was collected breath-by-breath, sampled from a mixing chamber, and averaged into 10-s bins. VO2 max was calculated as the highest 30-s average during each part of the protocol (incremental test and supramaximal verification) resulting in four values (two incremental test values and two supramaximal verification values) for each participant at each time point (PRE1 and 2 and POST1 and 2). The four values at each time point were averaged together to provide each participant with a single PRE and POST VO2 max value for each training period. Time to fatigue (TTF) was recorded as the duration (seconds) of the incremental test. Five VO2 max tests were missed due to time constraints (n = 4) and illness (n = 1), resulting in 107 of the expected 112 VO2 max tests being completed. Of the 107 completed tests, two supramaximal verification phases were not completed due to time constraints. In an attempt to improve confidence in our reported VO2 max values, we implemented three novel statistical procedures to remove outlier test values prior to averaging participant VO2 max values from each time point (details in Supplementary material). While these procedures removed outlying tests from being included in the calculations of average VO2 max or TTF values, we did not completely omit data from any participant (i.e. all 14 participants were included in final analysis). Following removal of outlying tests, all remaining values for each time point were averaged together to provide a single VO2 max value for each individual at each time point of both training period one (PRE1 and POST1) and training period two (PRE2 and POST2). Maximal HR (HRmax ) was determined as the highest 30 s average HR value from any of the PRE tests from each training protocol. The highest HRmax value observed from familiarization or PRE testing was used to prescribe exercise training intensity for that period. Training consisted of two, four-week periods separated by a three-month wash-out period. During each training period, participants trained four times per week with each training session consisting of four, four-minute intervals at 90–95% HRmax interspersed with three minutes of active recovery at 70–75% HRmax on a motorized treadmill. Each training session commenced with a 10min warm-up at 70–75% HRmax and ended with a five-minute cool down at 70–75% HRmax , requiring a total of 40 min to complete.17 If the target HR was not attained by the two-minute mark during each four-minute interval, speed or incline (based on participant preference) were adjusted to achieve the target HR. HR, speed and incline were recorded 30-s before the end of each interval. Statistical analyses were performed using GraphPad Prism Version 7.0 and IBM SPSS Statistics Version 25. Unpaired t-tests
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
were used to compare PRE1 characteristics between participants who completed all aspects of the study (completers) and participants who dropped out (non-completers). Separate two-way repeated measures ANOVAs (period × time) were used to compare changes in VO2 max and TTF between training periods one and two. Two-way repeated measures ANOVAs were also used to compare speed/incline values between training periods as well as HR responses, which were characterized by averaging HR values for each week of training (i.e. average of HR values obtained at the end of every interval within a given week). Bonferroni post-hoc tests were used to compare VO2 max and TTF at PRE1 and PRE2 to determine whether pre-training values returned to baseline following the washout period. Simple linear regressions were used to determine the relationship between VO2 max and TTF at PRE1 and PRE2. Reproducibility of responses to both training periods was determined using Pearson correlation coefficients (r), intraclass correlation coefficients (ICC: two-way random effects, absolute agreement, single rater/measurement) with 95% confidence intervals (CIs),18,19 and within-subject coefficients of variation (CV).20 Pearson correlation coefficients were classified as very weak (<0.19), weak (0.20–0.39), moderate (0.40–0.59), strong (0.60–0.79) or very strong (>0.80). ICC values were interpreted as indicators of poor (<0.50), moderate (0.50–0.75), good (0.75–0.9) or excellent (>0.90) reproducibility.21 Data are presented as mean ± SD. Statistical significance was accepted at p < 0.05. We also followed Swinton et al.’s22 methods to classify observed VO2 max and TTF responses using 50% confidence intervals (CIs) based on the typical error (TE) for each variable. Following recent recommendations,23 TEs were calculated using the SD of prepost differences from a no-exercise control group that was a part of a separate unpublished study (TEs: relative VO2 max: 1.77 mL/kg/min, absolute VO2 max: 0.111 L/min, TTF: 35.47 s; CVs: 3.90%, 3.78%, and 3.03%, respectively). Importantly, this separate study used the same equipment, separated pre-post measurements by the same duration (four weeks), and used the same inclusion/exclusion criteria compared to the present study. As recommended by Swinton et al.,22 we used a clinically-relevant response threshold of 1.75 mL/kg/min for relative VO2 max.1 The response thresholds for absolute VO2 max and TTF were calculated as 0.2 times the standard deviation of PRE1 values (0.114 L/min and 24.75 s, respectively), which is a recommended approach when no clinically-relevant threshold exists for a given outcome.22 Responders, uncertain responders, and non-responders were identified as participants with 50% CIs that lay above, crossed, or fell below these response thresholds, respectively.22
3. Results In total, eight participants dropped out after PRE1 (noncompleters), and fourteen participants completed all physiological testing (completers). Table 1 presents the PRE1 characteristics of completers and non-completers, and no significant differences were observed except for TTF, which was significantly lower in noncompleters than completers (p < 0.05). All average HRs during the high-intensity portions of training sessions were within the prescribed 90–95% HRmax range (Fig. 1A), and there were no significant differences for average training HR (Fig. 1) or speed/incline between training periods (data not shown). Strong correlations between PRE1 and PRE2 values were observed for VO2 max (absolute and relative) and TTF (Fig. 2A–C). ANOVA results are presented in Fig. 2 (panels D–F). Significant main effects of time (p < 0.01) and condition were observed for VO2 max in mL/kg/min (Fig. 2D; period 1: +4.04 ± 2.29, period 2: +2.83 ± 2.74) and in L/min (Fig. 2E period 1: +0.27 ± 0.16, period 2: +0.21 ± 0.18).
3
Table 1 Participant characteristics completers vs. non-completersa .
Age (years) Height (cm) Body weight (lbs) VO2 max (mL/kg/min) VO2 max (L/min) TTF (s)
Completers (n = 14) PRE1
Non-completers (n = 8) PRE1
21.5 ± 2.1 180.1 ± 5.9 162.8 ± 16.3 60.2 ± 6.9 4.5 ± 0.5 1463.3 ± 123.3
20.9 ± 2.0 177.4 ± 7.8 165.2 ± 24.7 54.7 ± 4.2 4.1 ± 0.8 1326.0 ± 111.4*
Values are means ± standard deviation. * Significantly different from completers, p < 0.05. a Reasons for non-completion: injury related to training (n = 5), injury not related to study (n = 2), and scheduling conflicts (n = 1).
Fig. 1. Average heart rate (HR) responses expressed as a percentage of maximal heart rate (HRMAX ) or in beats per minute (bpm) during two separate periods of identical high-intensity interval training. HR values were collected within the last 30 s of each interval and were averaged across each interval within a given week. Circles and error bars represent mean and standard deviation of HR responses, respectively.
Mean changes in TTF revealed a significant (p < 0.01) main effect of time (Fig. 2F; period 1: +70.75 ± 35.87 s, period 2: +83.46 ± 34.55 s) but not condition (p > 0.05). There were no significant interaction effects (period × time) observed for any variable (Fig. 2D–F). Posthoc analyses revealed that that absolute VO2 max (L/min; Fig. 2E) and TTF (Fig. 2F) were significantly (p < 0.05) lower at PRE2 than PRE1, but relative VO2 max (mL/kg/min) was not different between PRE1 and PRE2 (Fig. 2D). Fig. 3 presents the reproducibility statistics for observed training responses between training periods. A moderate non-significant regression with poor reproducibility was observed for changes in relative VO2 max (r = 0.40, p = 0.15; Fig. 3A). Weak non-significant
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6 4
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
Fig. 2. Correlations of individual PRE1 and PRE2 relative (A) and absolute (B) VO2 max and TTF (C) values, as well as group responses (D–F) following exposure to the same 4-week high-intensity interval training intervention separated by a 3-month washout period. Error bars represent standard deviation. NS, not significant. * Significantly (p < 0.05) different than PRE1.
regressions with poor reproducibility were observed for changes in absolute VO2 max (r = 0.31, p = 0.28; Fig. 3B) and TTF (r = 0.11, p = 0.69; Fig. 3C). Following the completion of both training periods, three, six, and eight participants were consistently classified as responders for relative VO2 max, absolute VO2 max, and TTF, respectively (green circles in Fig. 3). Some participants were consistently classified as an uncertain responder in a given variable (orange circles; Fig. 3), and one participant was consistently classified as a non-responder in relative VO2 max (red circle; Fig. 3A). Interestingly, many participants were classified differently for the same variable following training periods 1 and 2 (black and blue circles; Fig. 3).
4. Discussion The observation that VO2 max and TTF values at PRE2 were not significantly higher than PRE1 suggests that the washout period returned these outcomes to baseline levels. Both training periods increased mean VO2 max and TTF, and the magnitude of these changes were similar to previous four-week training studies.6,24,25 Importantly, the mean changes in VO2 max and TTF were similar following both periods, suggesting that potential carryover effects from period one did not impact responses to period two, and that mean changes in VO2 max and TTF appeared to be reproducible at the group level. However, the linear regressions, ICCs, and CVs
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
5
(Fig. 3) suggest that observed VO2 max and TTF responses to two identical periods of standardized training are not reproducible at the individual level. Consistent with our results demonstrating poor reproducibility in observed VO2 max responses to training, Lindholm et al.26 reported poor correlations for individual changes in single-leg exercise performance and markers of skeletal muscle adaptation following two identical training protocols separated by a washout period. Further, we recently found poor reproducibility for acute changes in skeletal muscle mRNA expression following identical exercise bouts in a similar population.27 Because changes in mRNA expression following acute exercise are thought to represent the initiation of adaptive processes that contribute to chronic changes in aerobic capacity,28 the lack of reproducibility in mRNA responses may partly explain the lack of reproducibility in observed VO2 max responses to training. The lack of reproducibility in observed VO2 max and TTF responses poses a potential challenge for personalized exercise prescription in clinical/applied settings. Specifically, our observation that not all participants were consistently classified as either a responder, uncertain responder, or non-responder for a given variable following two identical training periods suggests that it may be difficult to determine whether an individual will benefit from a given exercise prescription. Indeed, these possible challenges to personalized exercise prescription are specific to VO2 max and TTF, and future work is needed to determine the reproducibility of other outcomes. It remains unclear whether a prior bout of training has a similar influence on adaptations to subsequent training periods across a group of individuals. Recent evidence suggests that adaptations to resistance training are larger following a second training period, and these augmented responses are accompanied by changes in skeletal muscle myonuclei content and DNA methylation.29 These findings highlight potential mechanisms of “muscle memory” whereby training-induced molecular changes prime an individual to experience larger adaptations following subsequent training periods. Therefore, individual differences in responsiveness to repeat training periods may be explained by variability in the mechanisms of “muscle memory”. However, future studies are needed to test this speculation. There are several limitations with the current study that should be acknowledged. First, the training period was quite intensive (four HIIT sessions per week), which likely contributed to the high dropout rate (8/22 or 36% of enrolled participants; Table 1). Second, given that our training periods only captured initial responses to training (i.e. 4 weeks), it remains unknown whether individual responses are reproducible if participants are exposed to longer training periods. It also remains unknown whether our observed lack of reproducibility is attributable to interindividual baseline physiological or genetic differences, and/or whether these findings are specific to treadmill exercise. Third, despite being powered to detect the effect size associated with the heritability estimate of observed VO2 max responses (r = 0.71),2 we obtained much smaller effect sizes (r = 0.31–0.40; Fig. 3A–B), thus raising the possibility that these non-significant correlations reflect
Fig. 3. Correlations of individual relative (A) and absolute (B) VO2 max and TTF (C) responses following exposure to the same four-week high-intensity interval training intervention separated by a 3-month washout period. Dashed lines represent lines of best fit. The Pearson correlation coefficient (r values) and p values are presented with for each correlation. Additional reproducibility statistics (intraclass correlation coefficients [ICCs] with 95% confidence intervals and coefficients of variation [CVs]) are also presented in each panel. Green, orange, and red circles represent participants that were consistently classified as a responder, uncertain responder, or non-responder for a given outcome following both training periods. Conversely,
black and blue circles represent participants that were classified differently for a given outcome following both training periods. Specifically, black circles represent participants classified as a responder and an uncertain responder, and blue circles represent participants classified as an uncertain responder and a non-responder. The dashed lines represent the thresholds used to differentiate responders and uncertain responders (furthest right and top lines) and to differentiate uncertain responders and non-responders (furthest left and bottom lines) (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model
ARTICLE IN PRESS
JSAMS-2159; No. of Pages 6
M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
6
type II errors. Our smaller effect sizes may be explained by differences in participant characteristics between our study and the HERITAGE Family Study2 (e.g. physical activity levels, baseline VO2 max, and sex). Future replication studies are needed to verify the effect size associated with reproducibility of observed VO2 max responses in young recreationally-active males. Finally, the current study did not include a control group or measure/control for behavioural/environmental factors such as diet, sleep, or habitual physical activity, and thus we were unable to determine the degree of random within-subjects variability in observed VO2 max responses to training.30 Although not included in the present study, future studies should consider recording diet, sleep, and physical activity levels to measure changes in these behavioural factors over the course of an exercise training intervention. 5. Conclusion The current study assessed the reproducibility of observed VO2 max and TTF responses to two identical exercise training periods. While both training periods elicited significant group-level VO2 max and TTF improvements, there was poor reproducibility in the observed individual responses. These results highlight a potential challenge in the application of a personalized exercise training approach based on an individual’s observed VO2 max or TTF response to a single training period. Acknowledgement This project was supported by an operating grant from the Natural Science and Engineering Research Council of Canada (NSERC; grant number: 402635. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jsams.2019.09. 007. References 1. Ross R, Blair SN, Arena R et al. Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign: a scientific statement from the American Heart Association. Circulation 2016; 134(24):e653–e699. http://dx.doi.org/10.1161/CIR.0000000000000461. 2. Bouchard C, An P, Rice T et al. Familial aggregation of VO2 max response to exercise training: results from the HERITAGE Family Study. J Appl Physiol 1999; 87(3):1003–1008. 3. Bonafiglia JT, Rotundo MP, Whittall JP et al. Inter-individual variability in the adaptive responses to endurance and sprint interval training: a randomized crossover study. PLoS One 2016; 11(12):e0167790. http://dx.doi.org/10.1371/ journal.pone.0167790. 4. Gurd BJ, Giles MD, Bonafiglia JT et al. Incidence of nonresponse and individual patterns of response following sprint interval training. Appl Physiol Nutr Metab 2016; 41(3):229–234. http://dx.doi.org/10.1139/apnm-2015-0449. 6. Raleigh JP, Giles MD, Islam H et al. Contribution of central and peripheral adaptations to changes in maximal oxygen uptake following 4 weeks of sprint interval training. Appl Physiol Nutr Metab 2018; 43(10):1059–1068. http://dx.doi.org/10. 1139/apnm-2017-0864.
8. Ross R, Goodpaster BH, Koch LG et al. Precision exercise medicine: understanding exercise response variability. Br J Sport Med 2019; 0:1–13. http://dx.doi.org/ 10.1136/bjsports-2018-100328. 9. Pickering C, Kiely J. Do non-responders to exercise exist—and if so, what should we do about them? Sport Med 2019; 49(1):1–7. http://dx.doi.org/10.1007/ s40279-018-01041-1. 10. Koch LG, Pollott GE, Britton SL. Selectively bred rat model system for low and high response to exercise training. Physiol Genomics 2013; 45(14):606–614. http://dx.doi.org/10.1152/physiolgenomics.00021.2013. 11. Marton O, Koltai E, Takeda M et al. Mitochondrial biogenesis-associated factors underlie the magnitude of response to aerobic endurance training in rats. Pflugers Arch 2015; 467(4):779–788. http://dx.doi.org/10.1007/s00424014-1554-7. 12. Prud’homme D, Bouchard C, Leblanc C et al. Sensitivity of maximal aerobic power to training is genotype-dependent. Med Sci Sport Exerc 1984:489–493. http://dx. doi.org/10.1249/00005768-198410000-00012. 13. Bouchard C, Sarzynski MA, Rice TK et al. Genomic predictors of the maximal O2 uptake response to standardized exercise training programs. J Appl Physiol 2011; 110(5):1160–1170. http://dx.doi.org/10.1152/japplphysiol.00973.2010. 14. Simoneau JA, Lortie G, Boulay MR et al. Effects of two high-intensity intermittent training programs interspaced by detraining on human skeletal muscle and performance. Eur J Appl Physiol 1987; 56(5):516–521. 15. Hulley SB. Designing clinical research, Lippincott Williams & Wilkins, 2007. 16. Edgett BA, Bonafiglia JT, Raleigh JP et al. Reproducibility of peak oxygen consumption and the impact of test variability on classification of individual training responses in young recreationally active adults. Clin Physiol Funct Imaging 2018; 38(4):630–638. http://dx.doi.org/10.1111/cpf.12459. 17. Tjønna AE, Leinan IM, Bartnes AT et al. Low- and high-volume of intensive endurance training significantly improves maximal oxygen uptake after 10weeks of training in healthy men. PLoS One 2013; 8(5):e65382. http://dx.doi. org/10.1371/journal.pone.0065382. 18. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979; 86(2):420–428. 19. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychol Methods 1996; 1(1):30–46. http://dx.doi.org/10.1037/1082989X.1.1.30. 20. Hopkins WG. Measures of reliability in sports medicine and science. Sport Med 2000:15. 21. Watkins MP, Portney L. Foundations of clinical research: applications to practice, 3rd ed. Upper Saddle River, NJ, Pearson/Prentice Hall, 2009. 22. Swinton PA, Hemingway BS, Saunders B et al. A statistical framework to interpret individual response to intervention: paving the way for personalised nutrition and exercise prescription. Front Nutr 2018; 5:41. http://dx.doi.org/10.3389/ FNUT.2018.00041. 23. Williamson PJ, Atkinson G, Batterham AM. Inter-individual responses of maximal oxygen uptake to exercise training: a critical review. Sport Med 2017; 47(8):1501–1513. http://dx.doi.org/10.1007/s40279-017-0680-8. 24. Preobrazenski N, Bonafiglia JT, Nelms MW et al. Does blood lactate predict the chronic adaptive response to training: a comparison of traditional and talk test prescription methods. Appl Physiol Nutr Metab 2019; 44(2):179–186. http://dx. doi.org/10.1139/apnm-2018-0343. 25. McKie GL, Islam H, Townsend LK et al. Modified sprint interval training protocols: physiological and psychological responses to 4 weeks of training. Appl Physiol Nutr Metab 2018; 43(6):595–601. http://dx.doi.org/10.1139/apnm2017-0595. 26. Lindholm ME, Giacomello S, Werne Solnestam B et al. The impact of endurance training on human skeletal muscle memory, global isoform expression and novel transcripts. PLoS Genet 2016; 12(9):1–24. http://dx.doi.org/10.1371/journal. pgen.1006294. 27. Islam H, Edgett BA, Bonafiglia JT et al. Repeatability of exercise-induced changes in mRNA expression and technical considerations for qPCR analysis in human skeletal muscle. Exp Physiol 2019; 104(3):407–420. http://dx.doi.org/10.1113/ EP087401. 28. Egan B, Zierath JR. Exercise metabolism and the molecular regulation of skeletal muscle adaptation. Cell Metab 2013; 17(2):162–184. http://dx.doi.org/10.1016/ j.cmet.2012.12.012. 29. Gundersen K, Bruusgaard JC, Egner IM et al. Muscle memory: virtues of your youth? J Physiol 2018; 596(18):4289–4290. http://dx.doi.org/10.1113/JP276354. 30. Hecksteden A, Kraushaar J, Scharhag-Rosenberger F et al. Individual response to exercise training—a statistical perspective. J Appl Physiol 2015; 118:1450–1459. http://dx.doi.org/10.1152/japplphysiol.00714.2014.
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
ARTICLE IN PRESS
JSAMS-2159; No. of Pages 6
Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training Michael Del Giudice, Jacob T. Bonafiglia, Hashim Islam, Nicholas Preobrazenski, Alessandra Amato, Brendon J. Gurd ∗ School of Kinesiology and Health Studies, Queen’s University, Canada
a r t i c l e
i n f o
Article history: Received 14 June 2019 Received in revised form 3 September 2019 Accepted 10 September 2019 Available online xxx Keywords: VO2 max Individual responses Reproducibility High-intensity interval training (HIIT)
a b s t r a c t Objectives: To test the hypothesis that observed maximal oxygen uptake (VO2 max) and time to fatigue (TTF) responses to two identical periods of standardized high-intensity interval training are reproducible. Design: Fourteen recreationally active and healthy young males completed two identical four-week periods of high-intensity interval training (4 × 4-min intervals at 90–95% maximum heart rate [HRmax ] separated by 3-min periods of active recovery at 70–75% HRmax ). Training periods were separated by a three-month washout period. Methods: VO2 max and TTF were assessed via incremental tests with supramaximal verification before and after each training period. Pearson correlation coefficients (r), intraclass correlation coefficients (ICC), and within-subjects coefficients of variation (CV) were used to assess reproducibility of observed VO2 max and TTF responses. Results: VO2 max and TTF values before the second training period were not significantly higher than baseline values and there were no significant (p > 0.05) interaction effects (period 1: VO2 max: +4.04 ± 2.29 mL/kg/min, TTF: +70.75 ± 35.87 s; period 2: VO2 max: +2.83 ± 2.74 mL/kg/min, TTF: +83.46 ± 34.55 s). We found very weak-to-moderate correlations and poor reproducibility for observed VO2 max (mL/kg/min: r = 0.40, ICC = 0.369, CV = 74.4) and TTF (r = 0.11. ICC = 0.048, CV = 45.6) responses to training periods 1 and 2. Conclusions: Our ANOVA results confirmed that the three-month washout period returned VO2 max and TTF levels to baseline and prevented carryover effects. Contrary to our hypothesis, our results suggest that individual observed VO2 max and TTF responses to identical training stimuli are not reproducible. © 2019 Published by Elsevier Ltd on behalf of Sports Medicine Australia.
Practical implications
1. Introduction
• Two four-week periods of high-intensity interval training resulted in similar mean changes in maximal oxygen uptake (VO2 max) and time to fatigue (TTF). • However, individual changes in VO2 max and TTF following two identical periods of high-intensity interval training were not reproducible. • These findings highlight a potential challenge for determining whether an individual has benefitted from a given exercise prescription.
Improvements in maximal oxygen uptake (VO2 max) are associated with reductions in the risk of all-cause mortality/morbidity.1 Interestingly, there is considerable heterogeneity in observed changes in VO2 max following standardized aerobic training.2–4 However, it remains unclear whether individuals demonstrate similar observed responses if they are repeatedly exposed to the same exercise training intervention.8,9 Findings from selective-breeding rodent studies10,11 and human twin,12 family,2 and genome-wide association13 studies provide evidence that genetics contributes to the variance in observed VO2 max responses to exercise training. This apparent influence of genetics raises the possibility that observed changes in VO2 max are reproducible if a group of participants are repeatedly exposed to the same exercise training intervention. Although one other study measured changes in VO2 max after repeatedly exposing participants
∗ Corresponding author. E-mail address: [email protected] (B.J. Gurd). https://doi.org/10.1016/j.jsams.2019.09.007 1440-2440/© 2019 Published by Elsevier Ltd on behalf of Sports Medicine Australia.
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
2
to the same exercise training intervention,14 this study did not assess intraindividual variability in observed VO2 max responses. Moreover, a recent consensus statement acknowledged that determining the reproducibility of observed VO2 max responses to training remains a major knowledge gap in the individual response literature.8 Therefore, the purpose of this study was to test the hypothesis that observed VO2 max responses to two identical periods of standardized exercise training are reproducible. Specifically, given the apparent role of genetics, we hypothesized that there would be high reproducibility between individuals’ observed VO2 max responses to two identical training periods. To test this hypothesis, we exposed a group of participants to two identical periods of standardized high-intensity interval training separated by a threemonth wash-out period. 2. Methods Forty-four potential participants were screened. Twenty-two enrolled and 14 completed all aspects of the study. Participants were only enrolled in the study if they met the following inclusion criteria: between 18 and 30 years of age, non-smokers, not taking any prescription medication, free of cardiometabolic disease, self-reported less than three hours of physical activity per week, and not involved in a systematic training program at the time of enrollment. All participants were instructed to maintain their regular physical activity and nutritional habits throughout the duration of the study. All experimental procedures were approved by the Health Sciences Human Research Ethics Board at Queen’s University (approval number: 6021938). Verbal and written explanation of the experimental protocol and associated risks were provided to all participants prior to obtaining written informed consent during a preliminary screening session. An a priori sample size was calculated for the primary outcome, the relationship between observed VO2 max responses following training periods 1 and 2. Given that ∼50% of the variance in observed VO2 max responses following training can be explained by genetics,2 we expected a Pearson correlation coefficient of r = 0.71 (i.e. an r2 value of 0.5). Using the sample size formula relevant to correlations,15 we determined that a sample size of 13 was needed (Z␣ = 1.96, Z = 0.84, C = 0.89, r = 0.71) to detect significance for a correlation with an expected coefficient of r = 0.71 with 80% power. The current study exposed the same individuals to two identical four-week training periods separated by a three-month washout period during which participants were instructed to return to their pre-study levels of physical activity. Considering the evidence supporting a potential learning effect associated with VO2 max testing,16 participants completed a familiarization test prior to each training period of the study. Hereafter, participants reported to the lab on three separate occasions separated by 24–48 h in the weeks preceding (PRE) and following (POST) training. During the first visit of PRE and POST testing, a resting muscle biopsy was taken from the vastus lateralis (biopsy data is not presented in this manuscript). Twenty-four hours after their biopsy visit, participants reported to the lab to complete a VO2 max test (day 2), and 24–48 h later participants returned to complete a second VO2 max test (day 3). Following training, the order of the three visits was identical to PRE and the first POST-training visit occurred 72–96 h after the final training session. All participants were asked to refrain from alcohol and caffeine 12 h before, and nutritional supplements and exercise 24 h before all physiological testing. All physiological testing and training were performed on a Treadmill (Sports Art Fitness 6300HR, Taiwan, R.O.C.). Participants consumed a standardized dinner the night before each VO2 max test (Stauffer’s Sauté Sensations [520 kcal; 74 g carbohydrate, 10 g fat, 32 g protein]) and arrived at the laboratory in
the morning following a 12-h overnight fast. Upon arrival, participants were fed a standardized breakfast (bagel [181 kcal] with 15 g of cream cheese [44 kcal]). Thirty minutes after breakfast consumption, participants completed a VO2 max test on a motorized treadmill following an incremental test protocol with a supramaximal verification phase. Gas exchange and heart rate were collected throughout the incremental test and verification phase using a metabolic cart (Moxus AEI Technologies, Pittsburgh, PA) and heart rate monitor (Polar Team2 Pro, Kempele, Finland), respectively. The incremental test protocol consisted of three minutes of resting data collection (participants were asked to stand on the treadmill and breathe normally) followed by a five-minute warm-up with the treadmill set to 2.5 mph at an incline of 2 and subsequent increases of either incline or speed every two minutes until volitional fatigue (see Supplemental Table 1 for details). Following the incremental test protocol, participants were provided with a minimum of 10 min of rest prior to commencing a supramaximal verification phase. The metabolic cart was not re-calibrated in between phases. During the supramaximal verification phase, participants ran until volitional fatigue at a speed that was 0.5 mph faster than the final stage attempted during the incremental test protocol. For each VO2 max test, VO2 was collected breath-by-breath, sampled from a mixing chamber, and averaged into 10-s bins. VO2 max was calculated as the highest 30-s average during each part of the protocol (incremental test and supramaximal verification) resulting in four values (two incremental test values and two supramaximal verification values) for each participant at each time point (PRE1 and 2 and POST1 and 2). The four values at each time point were averaged together to provide each participant with a single PRE and POST VO2 max value for each training period. Time to fatigue (TTF) was recorded as the duration (seconds) of the incremental test. Five VO2 max tests were missed due to time constraints (n = 4) and illness (n = 1), resulting in 107 of the expected 112 VO2 max tests being completed. Of the 107 completed tests, two supramaximal verification phases were not completed due to time constraints. In an attempt to improve confidence in our reported VO2 max values, we implemented three novel statistical procedures to remove outlier test values prior to averaging participant VO2 max values from each time point (details in Supplementary material). While these procedures removed outlying tests from being included in the calculations of average VO2 max or TTF values, we did not completely omit data from any participant (i.e. all 14 participants were included in final analysis). Following removal of outlying tests, all remaining values for each time point were averaged together to provide a single VO2 max value for each individual at each time point of both training period one (PRE1 and POST1) and training period two (PRE2 and POST2). Maximal HR (HRmax ) was determined as the highest 30 s average HR value from any of the PRE tests from each training protocol. The highest HRmax value observed from familiarization or PRE testing was used to prescribe exercise training intensity for that period. Training consisted of two, four-week periods separated by a three-month wash-out period. During each training period, participants trained four times per week with each training session consisting of four, four-minute intervals at 90–95% HRmax interspersed with three minutes of active recovery at 70–75% HRmax on a motorized treadmill. Each training session commenced with a 10min warm-up at 70–75% HRmax and ended with a five-minute cool down at 70–75% HRmax , requiring a total of 40 min to complete.17 If the target HR was not attained by the two-minute mark during each four-minute interval, speed or incline (based on participant preference) were adjusted to achieve the target HR. HR, speed and incline were recorded 30-s before the end of each interval. Statistical analyses were performed using GraphPad Prism Version 7.0 and IBM SPSS Statistics Version 25. Unpaired t-tests
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
were used to compare PRE1 characteristics between participants who completed all aspects of the study (completers) and participants who dropped out (non-completers). Separate two-way repeated measures ANOVAs (period × time) were used to compare changes in VO2 max and TTF between training periods one and two. Two-way repeated measures ANOVAs were also used to compare speed/incline values between training periods as well as HR responses, which were characterized by averaging HR values for each week of training (i.e. average of HR values obtained at the end of every interval within a given week). Bonferroni post-hoc tests were used to compare VO2 max and TTF at PRE1 and PRE2 to determine whether pre-training values returned to baseline following the washout period. Simple linear regressions were used to determine the relationship between VO2 max and TTF at PRE1 and PRE2. Reproducibility of responses to both training periods was determined using Pearson correlation coefficients (r), intraclass correlation coefficients (ICC: two-way random effects, absolute agreement, single rater/measurement) with 95% confidence intervals (CIs),18,19 and within-subject coefficients of variation (CV).20 Pearson correlation coefficients were classified as very weak (<0.19), weak (0.20–0.39), moderate (0.40–0.59), strong (0.60–0.79) or very strong (>0.80). ICC values were interpreted as indicators of poor (<0.50), moderate (0.50–0.75), good (0.75–0.9) or excellent (>0.90) reproducibility.21 Data are presented as mean ± SD. Statistical significance was accepted at p < 0.05. We also followed Swinton et al.’s22 methods to classify observed VO2 max and TTF responses using 50% confidence intervals (CIs) based on the typical error (TE) for each variable. Following recent recommendations,23 TEs were calculated using the SD of prepost differences from a no-exercise control group that was a part of a separate unpublished study (TEs: relative VO2 max: 1.77 mL/kg/min, absolute VO2 max: 0.111 L/min, TTF: 35.47 s; CVs: 3.90%, 3.78%, and 3.03%, respectively). Importantly, this separate study used the same equipment, separated pre-post measurements by the same duration (four weeks), and used the same inclusion/exclusion criteria compared to the present study. As recommended by Swinton et al.,22 we used a clinically-relevant response threshold of 1.75 mL/kg/min for relative VO2 max.1 The response thresholds for absolute VO2 max and TTF were calculated as 0.2 times the standard deviation of PRE1 values (0.114 L/min and 24.75 s, respectively), which is a recommended approach when no clinically-relevant threshold exists for a given outcome.22 Responders, uncertain responders, and non-responders were identified as participants with 50% CIs that lay above, crossed, or fell below these response thresholds, respectively.22
3. Results In total, eight participants dropped out after PRE1 (noncompleters), and fourteen participants completed all physiological testing (completers). Table 1 presents the PRE1 characteristics of completers and non-completers, and no significant differences were observed except for TTF, which was significantly lower in noncompleters than completers (p < 0.05). All average HRs during the high-intensity portions of training sessions were within the prescribed 90–95% HRmax range (Fig. 1A), and there were no significant differences for average training HR (Fig. 1) or speed/incline between training periods (data not shown). Strong correlations between PRE1 and PRE2 values were observed for VO2 max (absolute and relative) and TTF (Fig. 2A–C). ANOVA results are presented in Fig. 2 (panels D–F). Significant main effects of time (p < 0.01) and condition were observed for VO2 max in mL/kg/min (Fig. 2D; period 1: +4.04 ± 2.29, period 2: +2.83 ± 2.74) and in L/min (Fig. 2E period 1: +0.27 ± 0.16, period 2: +0.21 ± 0.18).
3
Table 1 Participant characteristics completers vs. non-completersa .
Age (years) Height (cm) Body weight (lbs) VO2 max (mL/kg/min) VO2 max (L/min) TTF (s)
Completers (n = 14) PRE1
Non-completers (n = 8) PRE1
21.5 ± 2.1 180.1 ± 5.9 162.8 ± 16.3 60.2 ± 6.9 4.5 ± 0.5 1463.3 ± 123.3
20.9 ± 2.0 177.4 ± 7.8 165.2 ± 24.7 54.7 ± 4.2 4.1 ± 0.8 1326.0 ± 111.4*
Values are means ± standard deviation. * Significantly different from completers, p < 0.05. a Reasons for non-completion: injury related to training (n = 5), injury not related to study (n = 2), and scheduling conflicts (n = 1).
Fig. 1. Average heart rate (HR) responses expressed as a percentage of maximal heart rate (HRMAX ) or in beats per minute (bpm) during two separate periods of identical high-intensity interval training. HR values were collected within the last 30 s of each interval and were averaged across each interval within a given week. Circles and error bars represent mean and standard deviation of HR responses, respectively.
Mean changes in TTF revealed a significant (p < 0.01) main effect of time (Fig. 2F; period 1: +70.75 ± 35.87 s, period 2: +83.46 ± 34.55 s) but not condition (p > 0.05). There were no significant interaction effects (period × time) observed for any variable (Fig. 2D–F). Posthoc analyses revealed that that absolute VO2 max (L/min; Fig. 2E) and TTF (Fig. 2F) were significantly (p < 0.05) lower at PRE2 than PRE1, but relative VO2 max (mL/kg/min) was not different between PRE1 and PRE2 (Fig. 2D). Fig. 3 presents the reproducibility statistics for observed training responses between training periods. A moderate non-significant regression with poor reproducibility was observed for changes in relative VO2 max (r = 0.40, p = 0.15; Fig. 3A). Weak non-significant
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6 4
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
Fig. 2. Correlations of individual PRE1 and PRE2 relative (A) and absolute (B) VO2 max and TTF (C) values, as well as group responses (D–F) following exposure to the same 4-week high-intensity interval training intervention separated by a 3-month washout period. Error bars represent standard deviation. NS, not significant. * Significantly (p < 0.05) different than PRE1.
regressions with poor reproducibility were observed for changes in absolute VO2 max (r = 0.31, p = 0.28; Fig. 3B) and TTF (r = 0.11, p = 0.69; Fig. 3C). Following the completion of both training periods, three, six, and eight participants were consistently classified as responders for relative VO2 max, absolute VO2 max, and TTF, respectively (green circles in Fig. 3). Some participants were consistently classified as an uncertain responder in a given variable (orange circles; Fig. 3), and one participant was consistently classified as a non-responder in relative VO2 max (red circle; Fig. 3A). Interestingly, many participants were classified differently for the same variable following training periods 1 and 2 (black and blue circles; Fig. 3).
4. Discussion The observation that VO2 max and TTF values at PRE2 were not significantly higher than PRE1 suggests that the washout period returned these outcomes to baseline levels. Both training periods increased mean VO2 max and TTF, and the magnitude of these changes were similar to previous four-week training studies.6,24,25 Importantly, the mean changes in VO2 max and TTF were similar following both periods, suggesting that potential carryover effects from period one did not impact responses to period two, and that mean changes in VO2 max and TTF appeared to be reproducible at the group level. However, the linear regressions, ICCs, and CVs
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model JSAMS-2159; No. of Pages 6
ARTICLE IN PRESS M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
5
(Fig. 3) suggest that observed VO2 max and TTF responses to two identical periods of standardized training are not reproducible at the individual level. Consistent with our results demonstrating poor reproducibility in observed VO2 max responses to training, Lindholm et al.26 reported poor correlations for individual changes in single-leg exercise performance and markers of skeletal muscle adaptation following two identical training protocols separated by a washout period. Further, we recently found poor reproducibility for acute changes in skeletal muscle mRNA expression following identical exercise bouts in a similar population.27 Because changes in mRNA expression following acute exercise are thought to represent the initiation of adaptive processes that contribute to chronic changes in aerobic capacity,28 the lack of reproducibility in mRNA responses may partly explain the lack of reproducibility in observed VO2 max responses to training. The lack of reproducibility in observed VO2 max and TTF responses poses a potential challenge for personalized exercise prescription in clinical/applied settings. Specifically, our observation that not all participants were consistently classified as either a responder, uncertain responder, or non-responder for a given variable following two identical training periods suggests that it may be difficult to determine whether an individual will benefit from a given exercise prescription. Indeed, these possible challenges to personalized exercise prescription are specific to VO2 max and TTF, and future work is needed to determine the reproducibility of other outcomes. It remains unclear whether a prior bout of training has a similar influence on adaptations to subsequent training periods across a group of individuals. Recent evidence suggests that adaptations to resistance training are larger following a second training period, and these augmented responses are accompanied by changes in skeletal muscle myonuclei content and DNA methylation.29 These findings highlight potential mechanisms of “muscle memory” whereby training-induced molecular changes prime an individual to experience larger adaptations following subsequent training periods. Therefore, individual differences in responsiveness to repeat training periods may be explained by variability in the mechanisms of “muscle memory”. However, future studies are needed to test this speculation. There are several limitations with the current study that should be acknowledged. First, the training period was quite intensive (four HIIT sessions per week), which likely contributed to the high dropout rate (8/22 or 36% of enrolled participants; Table 1). Second, given that our training periods only captured initial responses to training (i.e. 4 weeks), it remains unknown whether individual responses are reproducible if participants are exposed to longer training periods. It also remains unknown whether our observed lack of reproducibility is attributable to interindividual baseline physiological or genetic differences, and/or whether these findings are specific to treadmill exercise. Third, despite being powered to detect the effect size associated with the heritability estimate of observed VO2 max responses (r = 0.71),2 we obtained much smaller effect sizes (r = 0.31–0.40; Fig. 3A–B), thus raising the possibility that these non-significant correlations reflect
Fig. 3. Correlations of individual relative (A) and absolute (B) VO2 max and TTF (C) responses following exposure to the same four-week high-intensity interval training intervention separated by a 3-month washout period. Dashed lines represent lines of best fit. The Pearson correlation coefficient (r values) and p values are presented with for each correlation. Additional reproducibility statistics (intraclass correlation coefficients [ICCs] with 95% confidence intervals and coefficients of variation [CVs]) are also presented in each panel. Green, orange, and red circles represent participants that were consistently classified as a responder, uncertain responder, or non-responder for a given outcome following both training periods. Conversely,
black and blue circles represent participants that were classified differently for a given outcome following both training periods. Specifically, black circles represent participants classified as a responder and an uncertain responder, and blue circles represent participants classified as an uncertain responder and a non-responder. The dashed lines represent the thresholds used to differentiate responders and uncertain responders (furthest right and top lines) and to differentiate uncertain responders and non-responders (furthest left and bottom lines) (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007
G Model
ARTICLE IN PRESS
JSAMS-2159; No. of Pages 6
M. Del Giudice, J.T. Bonafiglia, H. Islam et al. / Journal of Science and Medicine in Sport xxx (2019) xxx–xxx
6
type II errors. Our smaller effect sizes may be explained by differences in participant characteristics between our study and the HERITAGE Family Study2 (e.g. physical activity levels, baseline VO2 max, and sex). Future replication studies are needed to verify the effect size associated with reproducibility of observed VO2 max responses in young recreationally-active males. Finally, the current study did not include a control group or measure/control for behavioural/environmental factors such as diet, sleep, or habitual physical activity, and thus we were unable to determine the degree of random within-subjects variability in observed VO2 max responses to training.30 Although not included in the present study, future studies should consider recording diet, sleep, and physical activity levels to measure changes in these behavioural factors over the course of an exercise training intervention. 5. Conclusion The current study assessed the reproducibility of observed VO2 max and TTF responses to two identical exercise training periods. While both training periods elicited significant group-level VO2 max and TTF improvements, there was poor reproducibility in the observed individual responses. These results highlight a potential challenge in the application of a personalized exercise training approach based on an individual’s observed VO2 max or TTF response to a single training period. Acknowledgement This project was supported by an operating grant from the Natural Science and Engineering Research Council of Canada (NSERC; grant number: 402635. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jsams.2019.09. 007. References 1. Ross R, Blair SN, Arena R et al. Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign: a scientific statement from the American Heart Association. Circulation 2016; 134(24):e653–e699. http://dx.doi.org/10.1161/CIR.0000000000000461. 2. Bouchard C, An P, Rice T et al. Familial aggregation of VO2 max response to exercise training: results from the HERITAGE Family Study. J Appl Physiol 1999; 87(3):1003–1008. 3. Bonafiglia JT, Rotundo MP, Whittall JP et al. Inter-individual variability in the adaptive responses to endurance and sprint interval training: a randomized crossover study. PLoS One 2016; 11(12):e0167790. http://dx.doi.org/10.1371/ journal.pone.0167790. 4. Gurd BJ, Giles MD, Bonafiglia JT et al. Incidence of nonresponse and individual patterns of response following sprint interval training. Appl Physiol Nutr Metab 2016; 41(3):229–234. http://dx.doi.org/10.1139/apnm-2015-0449. 6. Raleigh JP, Giles MD, Islam H et al. Contribution of central and peripheral adaptations to changes in maximal oxygen uptake following 4 weeks of sprint interval training. Appl Physiol Nutr Metab 2018; 43(10):1059–1068. http://dx.doi.org/10. 1139/apnm-2017-0864.
8. Ross R, Goodpaster BH, Koch LG et al. Precision exercise medicine: understanding exercise response variability. Br J Sport Med 2019; 0:1–13. http://dx.doi.org/ 10.1136/bjsports-2018-100328. 9. Pickering C, Kiely J. Do non-responders to exercise exist—and if so, what should we do about them? Sport Med 2019; 49(1):1–7. http://dx.doi.org/10.1007/ s40279-018-01041-1. 10. Koch LG, Pollott GE, Britton SL. Selectively bred rat model system for low and high response to exercise training. Physiol Genomics 2013; 45(14):606–614. http://dx.doi.org/10.1152/physiolgenomics.00021.2013. 11. Marton O, Koltai E, Takeda M et al. Mitochondrial biogenesis-associated factors underlie the magnitude of response to aerobic endurance training in rats. Pflugers Arch 2015; 467(4):779–788. http://dx.doi.org/10.1007/s00424014-1554-7. 12. Prud’homme D, Bouchard C, Leblanc C et al. Sensitivity of maximal aerobic power to training is genotype-dependent. Med Sci Sport Exerc 1984:489–493. http://dx. doi.org/10.1249/00005768-198410000-00012. 13. Bouchard C, Sarzynski MA, Rice TK et al. Genomic predictors of the maximal O2 uptake response to standardized exercise training programs. J Appl Physiol 2011; 110(5):1160–1170. http://dx.doi.org/10.1152/japplphysiol.00973.2010. 14. Simoneau JA, Lortie G, Boulay MR et al. Effects of two high-intensity intermittent training programs interspaced by detraining on human skeletal muscle and performance. Eur J Appl Physiol 1987; 56(5):516–521. 15. Hulley SB. Designing clinical research, Lippincott Williams & Wilkins, 2007. 16. Edgett BA, Bonafiglia JT, Raleigh JP et al. Reproducibility of peak oxygen consumption and the impact of test variability on classification of individual training responses in young recreationally active adults. Clin Physiol Funct Imaging 2018; 38(4):630–638. http://dx.doi.org/10.1111/cpf.12459. 17. Tjønna AE, Leinan IM, Bartnes AT et al. Low- and high-volume of intensive endurance training significantly improves maximal oxygen uptake after 10weeks of training in healthy men. PLoS One 2013; 8(5):e65382. http://dx.doi. org/10.1371/journal.pone.0065382. 18. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979; 86(2):420–428. 19. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychol Methods 1996; 1(1):30–46. http://dx.doi.org/10.1037/1082989X.1.1.30. 20. Hopkins WG. Measures of reliability in sports medicine and science. Sport Med 2000:15. 21. Watkins MP, Portney L. Foundations of clinical research: applications to practice, 3rd ed. Upper Saddle River, NJ, Pearson/Prentice Hall, 2009. 22. Swinton PA, Hemingway BS, Saunders B et al. A statistical framework to interpret individual response to intervention: paving the way for personalised nutrition and exercise prescription. Front Nutr 2018; 5:41. http://dx.doi.org/10.3389/ FNUT.2018.00041. 23. Williamson PJ, Atkinson G, Batterham AM. Inter-individual responses of maximal oxygen uptake to exercise training: a critical review. Sport Med 2017; 47(8):1501–1513. http://dx.doi.org/10.1007/s40279-017-0680-8. 24. Preobrazenski N, Bonafiglia JT, Nelms MW et al. Does blood lactate predict the chronic adaptive response to training: a comparison of traditional and talk test prescription methods. Appl Physiol Nutr Metab 2019; 44(2):179–186. http://dx. doi.org/10.1139/apnm-2018-0343. 25. McKie GL, Islam H, Townsend LK et al. Modified sprint interval training protocols: physiological and psychological responses to 4 weeks of training. Appl Physiol Nutr Metab 2018; 43(6):595–601. http://dx.doi.org/10.1139/apnm2017-0595. 26. Lindholm ME, Giacomello S, Werne Solnestam B et al. The impact of endurance training on human skeletal muscle memory, global isoform expression and novel transcripts. PLoS Genet 2016; 12(9):1–24. http://dx.doi.org/10.1371/journal. pgen.1006294. 27. Islam H, Edgett BA, Bonafiglia JT et al. Repeatability of exercise-induced changes in mRNA expression and technical considerations for qPCR analysis in human skeletal muscle. Exp Physiol 2019; 104(3):407–420. http://dx.doi.org/10.1113/ EP087401. 28. Egan B, Zierath JR. Exercise metabolism and the molecular regulation of skeletal muscle adaptation. Cell Metab 2013; 17(2):162–184. http://dx.doi.org/10.1016/ j.cmet.2012.12.012. 29. Gundersen K, Bruusgaard JC, Egner IM et al. Muscle memory: virtues of your youth? J Physiol 2018; 596(18):4289–4290. http://dx.doi.org/10.1113/JP276354. 30. Hecksteden A, Kraushaar J, Scharhag-Rosenberger F et al. Individual response to exercise training—a statistical perspective. J Appl Physiol 2015; 118:1450–1459. http://dx.doi.org/10.1152/japplphysiol.00714.2014.
Please cite this article in press as: M. Del Giudice, J.T. Bonafiglia, H. Islam, et al.. Investigating the reproducibility of maximal oxygen uptake responses to high-intensity interval training. J Sci Med Sport (2019), https://doi.org/10.1016/j.jsams.2019.09.007