Reliability of acute training responses elicited by exhaustive work intervals prescribed with the delta concept

Abstract

Reliability of acute training responses elicited by exhaustive work intervals prescribed with the delta concept

Arthur H. Bossi¹, Wouter P. Timmerman¹, Louis Passfield^1,2, James G. Hopker¹

¹ School of Sport and Exercise Sciences, University of Kent, Chatham, Kent, England

² Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada

Correspondence:

Arthur Henrique Bossi

School of Sport and Exercise Sciences

University of Kent at Medway

Medway Building

Chatham Maritime

Chatham, Kent

ME4 4AG

England

asnb3@kent.ac.uk

+447398944056

Background: Biomolecular research has suggested that the chronic adaptation to an exercise programme is modulated by the extent individual training sessions produce homeostatic stress. Hence, training intensity must be carefully prescribed to ensure the expected stimulus for adaptation is provided. To this end, the delta concept (%Δ) has been proposed as an intensity prescription method to minimising inter-individual variability of physiological and perceptual responses—expressed by:

Ẇ_prescribed = Ẇ_GET + [(Ẇ_V̇O2max – Ẇ_GET) · %Δ]

where Ẇ_prescribed is the set work rate, Ẇ_GET is the work rate associated with the gas exchange threshold, Ẇ_V̇O2max is the work rate associated with the maximal oxygen uptake and %Δ is the targeted intensity. Surprisingly, the inter-individual variability of acute training responses has not been investigated during high-intensity interval training (HIIT), despite HIIT being commonly performed in the laboratory and field settings. Moreover, the intra-individual variability of acute HIIT responses has not been established in cycling, which is vital to understand whether the anticipated training stimulus is achieved whenever a session is performed.

Purpose: We explored the levels of inter- and intra-individual variability of acute training responses elicited by exhaustive work intervals prescribed with %Δ.

Methods: Eighteen male and four female cyclists [age: 36 ± 12 years, height: 178 ± 10 cm, body mass: 75.2 ± 13.7 kg, V̇O_2max: 52 ± 5 ml·kg^-1·min^-1, peak power output (PPO): 4.72 ± 0.48 W·kg^-1] volunteered for this study. They performed a ramp test in the first visit to determine V̇O_2max, PPO, Ẇ_GET and Ẇ_V̇O2max. The next four visits consisted of a standardised 21-min warm-up and a HIIT session performed to exhaustion [i.e. 4-min work intervals at 70%Δ (Ẇ_prescribed = 4.00 ± 0.43 W·kg^-1, 84.7 ± 0.4 %PPO), interspersed with 2-min active recovery at 0.2·Ẇ_prescribed]. Breath-by-breath gas exchanges and heart rate (HR) were continuously measured, with ratings of perceived exertion (RPE) and blood lactate concentration ([La]) obtained after each work interval and at exhaustion. Time at >90%V̇O_2max was quantified as absolute values and as a percentage of the time to exhaustion. One-way repeated measures analysis of variance was used to test for systematic changes between HIIT sessions. Statistical significance was set at P ≤ 0.1. Reliability estimates [typical error (TE), coefficient of variation (CV), and intraclass correlation coefficient (ICC)] were obtained through Hopkins spreadsheet and are reported with 90% confidence limits. Based on the acquired data, we used G*Power software to perform sample size estimations for a two-tailed matched paired t-test, with alpha error probability set at 0.05 and power at 0.80.

Results: Time to exhaustion, absolute and relative time at >90%V̇O_2max, peak HR, peak RPE, and peak [La] were not different between HIIT sessions (all F ≤ 2.10, P ≥ 0.13, η²_p ≤ 0.09). Dependent variables and their reliability estimates are reported in Table 1. Sample size estimations are reported in Table 2.

Table 1. Reliability data [with 90% confidence limits] for the dependent variables.

mean ± SD

TE

ICC

intra-individual CV (%)

inter-individual CV (%)

Time to exhaustion (s)

1219 ± 618

244

[210-294]

0.86

[0.76-0.92]

31.0

67.0

Absolute time at >90%V̇O_2max (s)

502 ± 366

137

[118-165]

0.87

[0.79-0.93]

67.0

139.3

Relative time at >90%V̇O_2max (%)

57.0 ± 22.0

14.2

[12.2-17.1]

0.61

[0.43-0.77]

63.7

109.2

Peak HR (b·min^-1)

179 ± 11

2

[1.7-2.4]

0.97

[0.94-0.98]

1.2

6.2

Peak RPE

19.6 ± 0.8

0.3

[0.3-0.4]

0.85

[0.75-0.92]

1.9

4.7

Peak [La] (mmol·L^-1)

14.3 ± 2.6

2.0

[1.7-2.4]

0.45

[0.25-0.65]

15.0

20.4

SD, standard deviation; TE, typical error; ICC, intraclass correlation coefficient; CV, coefficient of variation; HR, heart rate; RPE, ratings of perceived exertion; [La], blood lactate concentration

Table 2. Required sample size to detect baseline changes for a given variable analysed with two-tailed matched paired t-test, with alpha error probability set at 0.05 and power at 0.80.

2%

5%

10%

20%

30%

Time to exhaustion (s)

1411

228

59

17

9

Absolute time at >90%V̇O_2max (s)

2854

458

116

31

15

Relative time at >90%V̇O_2max (%)

2421

389

99

27

13

Peak HR (b·min^-1)

8

4

3

3

2

Peak RPE

14

5

3

3

2

Peak [La] (mmol·L^-1)

724

119

32

10

6

HR, heart rate; RPE, ratings of perceived exertion; [La], blood lactate concentration

Discussion: Although participants consistently achieved peak values of HR, RPE, and to a lesser extent [La], there was substantial inter-individual variability in time to exhaustion, and both absolute and relative time at >90%V̇O_2max. Importantly, inter-individual variability was much higher than intra-individual, suggesting a greater day-to-day consistency would still produce marked heterogeneity between participants. This raises questions over the validity of %Δ to normalise acute HIIT responses. The levels of intra-individual variability also cast doubt on the assumption that a similar stimulus for adaptation is triggered every time a standard HIIT session is performed. Besides the effect on sample size estimations, achievable only if studies aim to detect large changes, this result also suggests athletes may not need to overly adhere to the prescribed power output during HIIT.

Conclusions: In contrast to previous suggestions based on continuous exercise, %Δ does not produce consistent inter- and intra-individual acute HIIT responses. Future studies should consider alternative methods for training intensity normalisation. Whether day-to-day consistency in training stimulus is a pre-requisite for optimal adaptation following HIIT is another question that merits investigation.

Acknowledgements: A.H.B. is a CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) scholarship holder [200700/2015-4].