Skip to main content
NCBI home page
European Journal of Sport Science logoLink to European Journal of Sport Science
. 2024 Oct 9;24(11):1583–1596. doi: 10.1002/ejsc.12202

The higher the fraction of maximal oxygen uptake is during interval training, the greater is the cycling performance gain

Ingvill Odden 1,✉, Lars Nymoen 2, Tomas Urianstad 1, Morten Kristoffersen 3, Daniel Hammarström 1, Joar Hansen 1, Knut Sindre Mølmen 1, Bent R Rønnestad 1
PMCID: PMC11534653  PMID: 39385317

Abstract

It has been suggested that time at a high fraction (%) of maximal oxygen uptake (VO2max) plays a decisive role for adaptations to interval training. Yet, no study has, to date, measured the % of VO2max during all interval sessions throughout a prolonged training intervention and subsequently related it to the magnitude of training adaptations. Thus, the present study aimed to investigate the relationship between % of VO2max achieved during an interval training intervention and changes in endurance performance and its physiological determinants in well‐trained cyclists. Twenty‐two cyclists (VO2max 67.1 (6.4) mL·min−1 ·kg−1; males, n = 19; females, n = 3) underwent a 9‐week interval training intervention, consisting 21 sessions of 5 × 8‐min intervals conducted at their 40‐min highest sustainable mean power output (PO). Oxygen uptake was measured during all interval sessions, and the relationship between % of VO2max during work intervals and training adaptations were investigated using linear regression. A performance index was calculated from several performance measures. With higher % of VO2max during work intervals, greater improvements were observed for maximal PO during the VO2max test (R2 adjusted = 0.44, p = 0.009), PO at 4 mmol·L−1 [blood lactate] (R2 adjusted = 0.25, p = 0.035), the performance index (R2 adjusted = 0.36, p = 0.013), and VO2max (R2 adjusted = 0.54, p = 0.029). Other measures, such as % of maximal heart rate, were related to fewer outcome variables and exhibited poorer session‐to‐session repeatability compared to % of VO2max. In conclusion, improvements in endurance measures were positively related to the % of VO2max achieved during interval training. Percentage of VO2max was the measure that best reflected the magnitude of training adaptations.

Keywords: cycling, endurance training, high‐intensity interval training, performance, training intensity

Highlights

  • In the present study, we measured well‐trained cyclists' oxygen uptake during 21 power output‐matched interval sessions throughout a 9‐week training intervention and related it to the magnitude of training adaptations.

  • For the first time, we demonstrate that gains in endurance measures following an interval training intervention are positively related to the fraction of maximal oxygen uptake (V̇O2max) and time spent ≥90% of V̇O2max during interval sessions.

  • Fraction of V̇O2max during sessions demonstrated better session‐to‐session repeatability and displayed stronger associations with improvements in indicators of endurance performance compared to the percentage of maximal heart rate.

1. INTRODUCTION

Exercising at intensities close to maximal oxygen uptake (V̇O2max) is argued to largely stress the body's oxygen delivery and utilization system (Billat et al., 1996; Buchheit et al., 2013; Midgley et al., 2006a, 2006b). For that reason, it has been claimed that both the fraction (%) of V̇O2max achieved during exercise and the time spent ≥90% of V̇O2max are valid indicators of the training session's aerobic stimulus (Thevenet et al., 2007). Consequently, a growing number of studies investigating how to optimize interval training sessions, particularly by maximizing time ≥90% V̇O2max, have been conducted (Almquist et al., 2020; Bossi et al., 2020; Held et al., 2023; Rønnestad et al., 2016; Thevenet et al., 2007). As frequently highlighted in these studies, it is important to note that acute responses to exercise do not necessarily accurately reflect the magnitude of adaptations following a training intervention. Despite that discussions on this topic dates back to at least 1984 (Daniels et al., 1984), only a limited number of studies have investigated the specific importance of % of V̇O2max on adaptations to prolonged endurance training, or examined if this metric is better than other adaptive potential measures in predicting training adaptations. Surrogate measures of oxygen uptake (V̇O2), such as heart rate (HR), power output (PO), and velocities associated with a high % of V̇O2max, have commonly been used as indicators of exercise intensity in training studies, but their associations with training adaptations may not necessarily align with direct measures of V̇O2.

To our knowledge, only two studies have measured time at V̇O2max or time ≥90% of V̇O2max during interval sessions in a training intervention and subsequently investigated how it complies with the magnitude of training adaptations. Turnes, et al. (Turnes et al., 2016) observed that during a 4‐week training intervention consisting of two types of interval training sessions, the session associated with the longest time at V̇O2max also induced the largest gains in V̇O2max. Noteworthy, V̇O2 was only measured during two out of 12 interval sessions, and they did not observe a correlation between time spent at V̇O2max and V̇O2max improvement (Turnes et al., 2016). In the other study, cross‐country skiers performed five interval sessions within a 1‐week interval block, with V̇O2 measurements during three of the sessions. Here, a tendency toward a positive association between time ≥90% of V̇O2max and V̇O2max improvement was observed (Rønnestad, Bjerkrheim, et al., 2022). Of note, maximal velocity achieved during an incremental test and velocity corresponding to 4 mmol·L−1 blood lactate concentration ([La−]) were improved, but the improvements did not correlate with time ≥90% of V̇O2max during interval sessions (Rønnestad, Bjerkrheim, et al., 2022). Consequently, scientific evidence supporting the importance of exercising at a high % of V̇O2max for superior training adaptations are ambiguous. Considering that current training advice are made based on the assumption that sustaining a high % of V̇O2max during exercise translates into greater training adaptations (Buchheit et al., 2013; Midgley et al., 2006a, 2006b; Wenger et al., 1986), studies investigating whether it actually does are warranted.

Hence, the objective of the present study was to explore the relationship between the average % of V̇O2max during PO‐matched interval sessions and changes in endurance performance and its physiological determinants among well‐trained cyclists. We hypothesized that improvements in endurance performance and its determinants would be positively related to both % of V̇O2max achieved and the time spent ≥90% of V̇O2max during interval sessions, and that these measures would outperform other metrics in accurately reflecting training adaptations.

2. MATERIALS & METHODS

2.1. Participants

Thirty participants were initially recruited for the study. Inclusion criteria were V̇O2max > 60 mL·min−1 ·kg−1 for males and >55 mL·min−1 ·kg−1 for females. Due to sickness (n = 5), injury (n = 1), and personal reasons unrelated to the study (n = 2), eight participants withdrew from the study, leaving a total of 22 participants (♂, 19 and ♀, 3). The participants had a history of 3.7 (2.9) years of competitive cycling and were categorized as performance level 3 (n = 5), 4 (n = 9), and 5 (n = 8) (Decroix et al., 2016; Pauw et al., 2013).

2.2. Experimental design

The participants completed a 9‐week training intervention divided into three different 3‐week training periods (period 1, 2, and 3). A battery of tests was performed before (test day 1 and 2) and after the intervention period (test day 1 only; see “Exercise testing procedures” and “Hemoglobin mass measurements”), and a standalone V̇O2max test was performed after period 1 and 2. On an average, each participant performed 20.6 (0.8) interval sessions with 5 × 8‐min work intervals separated by 3‐min active recovery periods between work intervals. The mean PO for each participant during the 8‐min work intervals corresponded to their highest sustainable PO during a 40‐min cycling trial (PO40min) performed at test day 2 before the intervention. For training variation purposes, the 8‐min work intervals varied slightly in each of the three 3‐week training periods: 1) alternating between 30‐s at 118% of PO40min and 15‐s at 60% of PO40min, 2) alternating between 60‐s at 110% and 90% of PO40min, and 3) constant at 100% of PO40min. The PO during the 3‐min active recovery periods corresponded to 35% of PO40min. The participants were assigned to perform the three training periods with different 5 × 8‐min interval protocols in six different orders using stratified randomization. For each interval session, V̇O2 was measured during all 8‐min work intervals and subsequently presented as the % of the V̇O2max value obtained during the most recently performed V̇O2max test. In addition to the primary analysis using multiple linear regression models (see “Statistics”), secondary analyses were performed by dividing participants into two groups based on whether their average % of V̇O2max during the interval sessions was in the highest or lowest half of the group (HIGH%V̇O2max and LOW%V̇O2max, respectively; Table 1).

TABLE 1.

Baseline characteristics for all participants, as well as divided into groups eliciting the highest and lowest average fraction of maximal oxygen uptake during interval sessions (HIGH%V̇O2max and LOW%V̇O2max, respectively).

All participants (n = 22) HIGH%V̇O2max (♀ = 2, ♂ = 9) LOW%V̇O2max (♀ = 1, ♂ = 10)
Age (years) 22.6 (6.0) 21.4 (6.7) 22.6 (5.5)
Body mass (kg) 71.9 (9.3) 69.2 (9.8) 74.7 (8.2)
Body height (cm) 180 (8) 178 (9) 182 (6)
HRmax (bpm) 192 (8) 197 (8) * 188 (5)
[La−]max (mmol·L−1) 11.56 (1.99) 11.14 (1.29) 11.99 (2.50)
W max (W·kg−1) 5.8 (0.6) 5.7 (0.6) 5.8 (0.7)
PO4mmol (W·kg−1) 4.0 (0.5) 4.0 (0.6) 4.0 (0.5)
PO15min (W·kg−1) 4.2 (0.5) 4.3 (0.6) 4.2 (0.5)
PO40min (W·kg−1) 3.8 (0.4) 3.8 (0.5) 3.7 (0.4)
Performance index (arbitrary value, 0–1) 0.749 (0.086) 0.751 (0.092) 0.748 (0.085)
V̇O2max (mL·min−1·kg−1) 67.1 (6.4) 65.1 (5.3) 69.1 (7.1)
% of V̇O2max@4mmol (%) 82.4 (5.3) 84.1 (5.2) 80.8 (5.2)
GE175w (%) 18.9 (0.8) 19.2 (0.7) # 18.6 (0.9)
Peak leg press power (W·kg−1) 18.0 (2.7) 17.5 (3.1) 18.5 (2.2)
Hemoglobin mass (g·kg−1) 12.9 (1.3) 12.8 (1.5) 13.1 (1.1)
Open in a new tab

Note: HRmax, maximal heart rate during the maximal oxygen uptake test (V̇O2max test); [La−]max, blood lactate concentration measured 1‐min after the V̇O2max test; W max, maximal 1‐min incremental power output during the V̇O2max test; PO4mmol, power output at 4 mmol·L−1 blood lactate concentration ([La−]); PO15min, maximal average power output during the 15‐min cycling trial; PO40min, maximal average power output during the 40‐min cycling trial; % of V̇O2max@4mmol, fraction of V̇O2max at 4 mmol·L−1 [La−]; GE175, gross efficiency measured at 175 W. Values are mean (SD).

#

Tendency to different from LOW%V̇O2max (p < 0.1 and >0.05).

*Significantly different from LOW%V̇O2max (p ≤ 0.05).

2.3. Training intervention

At the onset of the 9‐week training intervention, the participants had undertaken their usual off‐season training (08:49 (02:55) h:m·week−1 during the 4 weeks preceding pre‐testing). Those who performed heavy strength training were instructed to simply continue with their regular strength training routines to avoid any potential changes in strength training that could affect the results. No one was allowed to start with strength training during the intervention period. During the training intervention, endurance training was reported according to a five‐zone intensity scale based on % of PO40min for endurance training performed as cycling and % of average HR during the 40‐min cycling trial for other types of endurance training (Table 2; Hunter et al., 2010). Of the total training, 85.7 (8.4) % was performed as cycling, 7.3 (3.4) % as running, and 5.5 (3.1) % as other types of endurance training. To evaluate how the intervention affected the perceived well‐being in the legs, this was reported after all training sessions using a nine‐point scale as described in Rønnestad, Hansen and Ellefsen (Rønnestad, Hansen, & Ellefsen, 2014; Table 2).

TABLE 2.

Average weekly training data during the 9‐week training intervention for groups eliciting the highest and lowest average fraction of maximal oxygen uptake during interval sessions (HIGH%V̇O2max and LOW%V̇O2max, respectively).

All participants (n = 22) HIGH%V̇O2max (n = 11) LOW%V̇O2max (n = 11)
Zone 1 (<55% of PO40min; h:m) 02:32 (01:25) 03:12 (01:29) * 01:48 (00:48)
Zone 2 (56%–75% of PO40min; h:m) 02:54 (01:20) 03:19 (01:22) 02:28 (01:13)
Zone 3 (76–90 of PO40min; h:m) 01:09 (00:25) 01:09 (00:26) 01:10 (00:24)
Zone 4 (91%–105% of PO40min; h:m) 01:22 (00:20) 01:20 (00:20) 01:25 (00:21)
Zone 5 (>106% of PO40min; h:m) 00:30 (00:21) 00:25 (00:16) 00:34 (00:25)
Heavy resistance training (h:m) 00:10 (00:23) 00:02 (00:07) 00:17 (00:31)
Core training (h:m) 00:17 (00:19) 00:15 (00:17) 00:18 (00:22)
Total training (h:m) 08:41 (02:36) 09:43 (02:24) # 07:40 (02:28)
Feeling legs (1–9) 5.1 (0.3) 5.1 (0.3) 5.2 (0.3)
Open in a new tab

Note: Feeling legs, perceived well‐being in the legs where 1 is very very good and nine is very very bad. Values are mean (SD).

#

Tendency to different from LOW%V̇O2max (p < 0.1 and >0.05).

*Significantly different from LOW%V̇O2max (p ≤ 0.05).

Each participant performed all interval sessions on their own bike connected to a stationary trainer device (Tacx Neo 2T Smart T2875/Tacx Neo 2 Smart T2850, Wassenaar). All interval sessions were performed on the same trainer device, and were supervised by a test leader who controlled the PO using the Tacx Training app (version 4.26.2, Garmin Ltd., the Netherlands) connected to the trainer device. The first two steps of the Lamberts and Lambert Submaximal Cycle Test (LSCT) was used as warm‐up (Lamberts et al., 2011) and after 3 min of recovery, the perceived well‐being in the legs was recorded. After another 5–10 min of recovery, the 5 × 8‐min interval protocol started.

If the participant was not able to complete all five 8‐min work intervals at 100% of PO40min, the PO was lowered to the highest achievable % of PO40min during the given work interval, as well as during the following work intervals if necessary. Rate of perceived exertion (RPE) on the Borg 6–20 scale (RPE) (Borg, 1998) was assessed immediately after each work interval, and if a RPE of ≤15 was reported following the fifth work interval, % of PO40min at the following session was increased by 2%‐points. During the 8‐min work intervals, V̇O2 was measured continuously (10‐s sampling time) using a computerized metabolic system with a mixing chamber (Oxycon Pro, Erich Jaeger,; Figure 1). A standardized calibration of the metabolic system was performed before each interval session. The flow turbine (Triple V, Erich Jaeger) was calibrated with a 3 L, 5530 series, calibration syringe (Hans Rudolph). Due to equipment malfunction, a total of 3.8% of the 10‐s V̇O2 measurements were imputed. This was done by copying the corresponding V̇O2 measurements from the previous 8‐min work interval. All participants performed the 8‐min work intervals at an average % of V̇O2max in the range of 74.2%–90.8%, with the mean values of HIGH%V̇O2max and LOW%V̇O2max being 86.2 (3.8)% and 79.9 (4.0)%, respectively (Figure 1). PO and HR were also measured continuously during each work interval. At the second, fifth, and seventh interval session in each 3‐week period, [La−] was measured from the fingertip immediately after all work intervals (Biosen C‐line Lactate Analyzer, EKF Diagnostic GmbH). 10 min after each interval session, the participants reported their session RPE using a 10‐point scale (sRPE; Table 3; Foster et al., 2001).

FIGURE 1.

FIGURE 1

Open in a new tab

Average fraction of maximal oxygen uptake (% of V̇O2max) elicited during the 8‐min work intervals during all interval sessions for all participants (dashed lines), as well as divided into groups eliciting the highest and lowest % of V̇O2max during intervals (HIGH%V̇O2max, black solid line and LOW%V̇O2max, gray solid line, respectively). The light gray areas represent 95% confidence intervals.

TABLE 3.

Interval session data for all participants and groups eliciting the highest and lowest average fraction of maximal oxygen uptake during interval sessions (HIGH%V̇O2max and LOW%V̇O2max, respectively), presented as averages of all collected measurements per interval session during period 1, 2 and 3.

All participants (n = 22) HIGH%V̇O2max (n = 11) LOW%V̇O2max (n = 11)
Period 1 Period 2 Period 3 Period 1 Period 2 Period 3 Period 1 Period 2 Period 3
Time ≥90% of V̇O2max (m:s) 08:50 (10:03) 09:03 (10:08) 10:23 (09:55) 15:32 (10:03)* 15:05 (10:47)* 14:52 (09:43)* 02:14 (03:38) 03:06 (04:18) 05:52 (07:53) a
% of V̇O2max 82.6 (5.1) 83.0 (5.1) 83.5 (4.8) 86.3 (3.6)* 86.2 (4.2)* 85.9 (3.8)* 79.0 (3.6) 79.7 (3.6) 81.0 (4.5) a
V̇O2 (mL·min−1·kg−1) 55.4 (5.6) 57.3 (5.8) a 58.6 (6.1) a b 56.3 (5.6) 58.2 (6.0) a 59.2 (6.0) a , b 54.6 (5.4) 56.3 (5.5) a 58.0 (6.2) a , b
Time ≥90% of HRmax (m:s) 12:19 (10:23) 15:01 (10:44) a 18:36 (10:16) a , b 12:57 (10:03) 18:53 (09:46) a 21:17 (10:12) a 11:41 (10:44) 11:11 (10:19) 15:53 (09:40) a
% of HRmax 87.2 (2.2) 87.6 (2.4) 88.3 (2.2) a , b 87.6 (2.0) 88.4 (2.4) 88.9 (2.4) a 86.8 (2.3) 86.9 (2.3) 87.7 (1.7)
HR (bpm) 168 (8) 168 (9) 169 (8) a 172 (8)* 173 (10)* 174 (7)* 163 (5) 164 (5) 165 (5)
PO (W·kg−1) 3.7 (0.4) 3.9 (0.5) a 3.9 (0.5) a , b 3.7 (0.5) 3.9 (0.5) a 3.9 (0.6) a , b 3.7 (0.4) 3.9 (0.4) a 3.9 (0.5) a , b
[La−] (mmol·L−1) 5.24 (1.76) 5.85 (1.65) a 6.32 (2.34) a 5.15 (1.94) 5.88 (1.39) 6.19 (2.53) 5.31 (1.62) 5.83 (1.90) 6.45 (2.17) a
RPE (6–20) 16.0 (1.0) 16.1 (1.1) 16.3 (1.2) a , b 16.2 (1.0) 16.4 (1.2) 16.6 (1.1) a 15.8 (1.0) 15.9 (1.1) 16.0 (1.2)
Feeling legs (1–9) 5.3 (0.9) 5.4 (0.8) 5.4 (0.7) 5.2 (0.9) 5.5 (0.9) 5.4 (0.7) 5.4 (0.9) 5.2 (0.7) 5.4 (0.7)
sRPE (0–10) 6.4 (1.6) 6.5 (1.5) 6.8 (1.5) a 6.9 (1.7) 7.0 (1.6) 7.4 (1.4) 5.9 (1.4) 6.1 (1.3) 6.3 (1.4)
Completed sessions (n) 6.9 (0.3) 7.0 (0.2) 6.8 (0.5) a , b 6.8 (0.4) 6.9 (0.3) 6.8 (0.6) 7.0 (0.0) 7.0 (0.0) a 6.7 (0.6) a , b
Open in a new tab

Note: Time ≥90% of V̇O2max, average time spent at or above 90% of maximal oxygen uptake (V̇O2max) during work intervals; % of V̇O2max, average fraction of V̇O2max during work intervals; V̇O2, average oxygen uptake during work intervals; time ≥90% of HRmax, average time spent at or above 90% of maximal heart rate elicited during the V̇O2max test (HRmax) during work intervals; % of HRmax, average fraction of HRmax during work intervals; HR, average heart rate during work intervals; PO, average power output during work intervals; [La−], average blood lactate concentration measured 1‐min after each work interval at the second, fifth, and seventh interval session in each 3‐week period; RPE, average rating of perceived exhaustion reported after each work interval at each interval session; feeling legs, perceived well‐being in the legs reported after the warm‐up prior to each interval session; sRPE, session rate of perceived exhaustion reported 10 min after each interval session. Values are mean (SD).

a

Significantly different from period 1.

b

Significantly different from period 2.

*Significantly different from LOW%V̇O2max within period (p ≤ 0.05).

2.4. Exercise testing procedures

The participants were instructed to refrain from exercise the day before test day 1, and their last three meals, fluid intake, and caffeine intake prior to testing and nutritional energy intake during pre‐test were noted and subsequently repeated at the following tests. The cycle ergometer (Lode Excalibur Sport, Groningen, The Netherlands) was adjusted according to each participant's preference and testing was performed under similar environmental conditions (17.8 (1.7) °C). Each participant had the same test leader during all tests, and strong verbal encouragement was given to ensure maximal effort. The tests were performed at the same time of the day (±2 h) for each participant. The endurance test at test day 2 was performed on the participants' own bike connected to the same individual stationary trainer device as used during interval sessions.

Test day 1 started with a 7‐min standardized cycling warm‐up, including 2 min at an intensity corresponding to an RPE of 11, 2 min at 13, 1 min at 15, and 2 min at 12. Then, the participants performed a peak power (Ppeak) leg press test (Keiser AIR300 Leg Press, Keiser Corp.), as described in detail elsewhere (Rønnestad, Urianstad, et al., 2022). Following 5 min of recovery, a shorter version of the long protocol described in Rønnestad, et al. (Rønnestad, Urianstad, et al., 2022) was initiated. Briefly, the participants started with an incremental blood lactate profile test with 50 W increases in PO every 5th minute until the [La−] measured from the fingertip reached ≥4 mmol∙L−1. V̇O2 and respiratory exchange ratio (RER) were measured (30‐s sampling time) from 2.5 to 4.5 min during each 5‐min bout, using the same metabolic system as during interval sessions with the same calibration routines. From the blood lactate profile test, PO corresponding to 4 mmol·L−1 [La−] (PO4mmol) and fractional utilization of V̇O2max at 4 mmol·L−1 [La−] (% of V̇O2max@4mmol) were calculated by plotting [La−] as a function of PO and % of V̇O2max at each 5‐min bout. Steady‐state V̇O2 and RER were used to calculate exercise efficiency measured as gross efficiency at 175 W (GE175W), where GE was defined as the ratio between mechanical PO and metabolic power input (PI; GE = PI·PO−1·100). PI was calculated using the V̇O2 and RER measured at the given PO and the energetic equivalent of O2 (Péronnet et al., 1991) according to the equation of Noordhof, Skiba and de Koning (Noordhof et al., 2013; PI = V̇O2 L·s−1 · (4840 J·L−1 RER + 16,890 J·L−1). Following 5 min of active recovery, the participants performed an incremental V̇O2max test with 25 W increases every minute until exhaustion, defined as cadence dropping <60 revolutions per minute. V̇O2max was defined as the mean of the two highest consecutive 30‐s V̇O2 measurements, maximal 1‐min incremental PO (W max) was defined as the mean PO during the last min of the test, and maximal HR (HRmax) was defined as the highest value achieved during the test. [La−] was measured 1 min after the test was terminated. Following 10 min of active recovery, a 15‐min maximal cycling trial was initiated with the instruction of aiming for the highest possible mean PO (PO15min). The participants adjusted the PO themselves using an external control unit. At test day 2 after completing the LSCT warm‐up, the participants performed a 40‐min maximal cycling trial for determination of PO40min. PO and HR were measured continuously throughout the test (Garmin Edge 530 Cycle Computer, Garmin Ltd.).

2.5. Hemoglobin mass measurement

After the exercise tests at test day 1, measurement of hemoglobin mass was performed with an automated blood volume analyzer based on carbon monoxide (CO) rebreathing (Detalo Performance, Detalo Health), using a modified version of procedures described in detail elsewhere (Rønnestad, Lid, et al., 2022). Briefly, blood was sampled and analyzed in triplicate for %HbCO using a blood gas analyzer (ABL830 FLEX CO‐OX analyzer, Radiometer). The participants were then connected to a breathing circuit, breathing 100% O2 for 15 s before a bolus of 1.0 (males) or 0.8 (females) mL·kg−1 body mass of 99.997% chemically pure CO was administrated through the system. After rebreathing the O2–CO gas mixture for 6 min and breathing normally for 4 min, capillary blood was sampled and analyzed in triplicate for %HbCO. Following 10 min of rest, the rebreathing procedure was repeated. The means of duplicate measures were used in data analyses. Due to gas leakage, hematological data is only presented for 11 and 9 participants in HIGH%V̇O2max and LOW%V̇O2max, respectively.

2.6. Statistics

Descriptive data are presented as means with standard deviations (mean (SD)). Based on the main performance indicators from test day 1 (W max, PO4mmol, and PO15min), a performance index was calculated as the average of the given indicators after normalization (x i max(x)−1, where x i is a single observation for one performance indicator and max(x)−1 is the maximum value observed across all participants for the given indicator). The performance index was calculated to increase the statistical power for performance‐related measures for which small differences may be very difficult to detect but still of relevance for elite endurance performance. Relationships between the different adaptive potential measures and training adaptations were investigated using multiple linear regression models with baseline value, change in body mass, and sex as covariates. Relationships between baseline variables and % of V̇O2max during interval sessions were investigated using multiple linear regression models with sex as a covariate. For the regression models, R2 adjusted and estimated scores with 95% confidence intervals (CI) are reported. Reliability for adaptive potential measures across the 21 interval sessions were determined using the intraclass correlation coefficient (ICC) and their 95% CI using the psych package written for R. A two‐way random effects model was used, and the single rater value reported. ICC values < 0.5 indicate poor reliability, 0.5–0.75 moderate reliability, 0.75–0.9 good reliability, and >0.90 excellent reliability (Koo et al., 2016). Differences between groups at baseline and in total training during the intervention were investigated using one‐way analyses of variance (ANOVA). Differences in interval session measures between periods within groups and for all participants combined, as well as between groups within periods, were investigated using linear mixed models using the lme4 package written for R. Differences in absolute changes in outcome variables were investigated using analyses of covariance (ANCOVA) with baseline value as a covariate. Results were considered statistically significant if p ≤ 0.05 and described as tendencies if p < 0.10 and >0.05. All data analyses were performed in R (R Core Team, 2018).

3. RESULTS

3.1. The effect of % of V̇O2max on changes in endurance measures

There was a positive relationship between % of V̇O2max achieved during all interval sessions and changes in both W max (p = 0.009; R2 adjusted = 0.44; estimate = 0.04 W·kg−1 [0.01, 0.06]; Figure 2A), PO4mmol (p = 0.035, R2 adjusted = 0.25; estimate = 0.02 W·kg−1 [0.00, 0.04]; Figure 2B), and the performance index (p = 0.013; R2 adjusted = 0.36; estimate = 0.004 AU [0.001, 0.007]; Figure 2D), whereas no such association was applicable for change in PO15min (p = 0.215; R2 adjusted = 0.15; estimate = 0.02 W·kg−1 [−0.01, 0.04]; Figure 2C). Additionally, there was a positive relationship between % of V̇O2max achieved during the intervals and change in V̇O2max (p = 0.029; R2 adjusted = 0.54; estimate = 0.25 mL·min−1 kg−1 [0.03, 0.48]; Figure 2E) and a tendency toward a positive relationship between % of V̇O2max and % of V̇O2max@4mmol (p = 0.085; R2 adjusted = 0.41; estimate = 0.39%‐points [−0.06, 0.83]; Figure 2F), whereas no such relationship was displayed for change in GE175W (p = 0.706; R2 adjusted = 0.06; estimate = −0.02%‐points [−0.12, 0.09]). No significant relationships were observed between % of V̇O2max during the intervals and change in hemoglobin mass (p = 0.853; data not shown).

FIGURE 2.

FIGURE 2

Open in a new tab

Multiple linear regression of the average fraction of maximal oxygen uptake (% of V̇O2max) elicited during the intervals related to changes in (A) maximal 1‐min incremental power output during the V̇O2max test (W max), (B) power output at 4 mmol·L−1 lactate concentration (PO4mmol), (C) maximal average power output during a 15‐min cycling trial (PO15min), (D) the performance index, (E) V̇O2max, and (F) fractional utilization of V̇O2max at 4 mmol·L−1 lactate concentration (%V̇O2max@4mmol) when controlling for baseline values, change in body mass, and sex. Individual data points for groups eliciting the highest (HIGH%V̇O2max; black dots) and lowest (LOW%V̇O2max; white dots) average % of V̇O2max during intervals in addition to pooled regression slopes (solid lines) with 95% confidence intervals (light gray areas) are shown. “80%–90%” in the panels with significant or tendencies toward relationships represent the theoretical increase in the given outcome variable if % of V̇O2max during intervals is increased from 80% to 90%.

3.2. Comparisons of participants with the highest and lowest % of V̇O2max during intervals

Compared to LOW%V̇O2max, HIGH%V̇O2max displayed larger absolute increases in W max (0.51 (0.33) versus 0.23 (0.14) W·kg−1; p = 0.022; Figure 3A), PO4mmol (0.31 (0.15) versus 0.12 (0.18) W·kg−1; p = 0.015; Figure 3B), the performance index (0.059 (0.028) versus 0.027 (0.021) AU; p = 0.005; Figure 3D), and V̇O2max (5.23 (2.76) versus 1.78 (1.72) mL·min−1 kg−1; p = 0.003; Figure 3E). Moreover, the absolute change in PO15min tended to be larger in HIGH%V̇O2max compared to LOW%V̇O2max (0.30 (0.20) versus 0.15 (0.20) W·kg−1; p = 0.096; Figure 3C), whereas the %‐point change in GE175W tended to be larger in LOW%V̇O2max compared to HIGH%V̇O2max (0.71 (0.98) versus 0.06 (0.47) %‐points; p = 0.065). No differences were observed between HIGH%V̇O2max and LOW%V̇O2max for %‐point change in % of V̇O2max@4mmol (−1.13 (4.29) versus −1.31 (4.44) %‐points; p = 0.914; Figure 3F) and absolute change in hemoglobin mass (0.13 (0.44) versus 0.13 (0.44) g·kg−1; p = 0.359). PO4mmol, PO15min, and % of V̇O2max@4mmol at baseline were all positively related to % of V̇O2max during intervals (Table 4).

FIGURE 3.

FIGURE 3

Open in a new tab

Individual data points (dashed lines) and mean values (solid lines) for (A) maximal 1‐min incremental power output during the maximal oxygen uptake (V̇O2max) test (W max), (B) power output at 4 mmol·L−1 lactate concentration (PO4mmol), (C) maximal average power output during a 15‐min cycling trial (PO15min), (D) the performance index, (E) V̇O2max, and (F) fractional utilization of V̇O2max at 4 mmol·L−1 lactate concentration (% of V̇O2max@4mmol), before (pre) and after (post) the training intervention for groups eliciting the highest and lowest average fraction of V̇O2max during intervals (HIGH%V̇O2max and LOW%V̇O2max, respectively). The values presented in each panel represent the mean (SD) percentage change in the variable of interest for HIGH%V̇O2max and LOW%V̇O2max, respectively. * Absolute change significantly different from LOW%V̇O2max (p ≤ 0.05). # Absolute change tends to be different from LOW%V̇O2max (p < 0.1 and >0.05).

TABLE 4.

Multiple linear regression of baseline measures related to % of V̇O2max during intervals, all when controlling for sex.

Independent variables Dependent variable
% of V̇O2max during intervals
Estimate a p 95% CI R2 adjusted
Baseline W max (W·kg−1) 2.66 0.172 [−1.26, 6.57] 0.00
Baseline PO4mmol (W·kg−1) 4.70 0.028* [0.57, 8.83] 0.15
Baseline PO15min (W·kg−1) 4.76 0.013* [1.13, 8.38] 0.21
Baseline V̇O2max (mL·min−1·kg−1) −0.06 0.757 [−0.44, 0.33] −0.10
Baseline % of V̇O2max@4mmol (%‐points) 0.42 0.011* [0.11, 0.73] 0.22
Baseline GE175W (%‐points) 2.36 0.052 # [−0.02, 4.73] 0.11
Open in a new tab

Note: % of V̇O2max during intervals; average fraction of maximal oxygen uptake (V̇O2max) elicited during intervals; W max, maximal 1‐min incremental power output during the V̇O2max test; PO4mmol, power output at 4 mmol·L−1 blood lactate concentration ([La−]); PO15min, maximal power output during the 15‐min cycling trial; % of V̇O2max@4mmol, fractional utilization of V̇O2max at 4 mmol·L−1 [La−]; GE175W, gross efficiency at 175 W.

aFor each unit higher baseline value of the independent variable, the estimate indicates how much higher % of V̇O2max (dependent variable) theoretically would be.

#

Tendency to relationship (p < 0.1 and >0.05).

*Significant relationship (p ≤ 0.05).

3.3. Other adaptive potential measures, their associations with training adaptations, and their reproducibility

There was a positive relationship between time ≥90% of V̇O2max during intervals and change in W max, PO4mmol, and the performance index, and a tendency for change in V̇O2max and % of V̇O2max@4mmol (Table 5). Percentage of HRmax during intervals only displayed a positive relationship with change in the performance index, and a tendency for change in PO4mmol (Table 5). There was a positive relationship between time ≥90% of HRmax during intervals and changes in PO4mmol and the performance index, as well as a tendency for change in PO15min (Table 5). The 21 interval sessions had an ICC of 0.61 [0.47, 0.76] for % of V̇O2max, 0.56 [0.42, 0.73] for time ≥90% of V̇O2max, 0.31 [0.20, 0.49] for % of HRmax, and 0.24 [0.14, 0.42] for time ≥90% of HRmax.

TABLE 5.

Multiple linear regression of the adaptive potential measures time ≥90% of V̇O2max, % of HRmax, and time ≥90% of HRmax during intervals related to training adaptations when controlling for baseline values, change in body mass, and sex.

Dependent variables Independent variable
Time ≥90% of V̇O2max during intervals (minutes)
Estimate a p 95% CI R2 adjusted
∆ W max (W·kg−1) 0.02 0.026* [0.00, 0.03] 0.37
∆ PO4mmol (W·kg−1) 0.01 0.045* [0.00, 0.02] 0.23
∆ PO15min (W·kg−1) 0.01 0.165 [−0.00, 0.02] 0.17
∆ performance index (AU) 0.00 0.021* [0.00, 0.00] 0.33
∆ V̇O2max (mL·min−1·kg−1) 0.11 0.094 # [−0.02, 0.23] 0.48
∆ % of V̇O2max@4mmol (%‐points) 0.20 0.071 # [−0.02, 0.42] 0.42
∆ GE175W (%‐points) −0.02 0.562 [−0.07, 0.04] 0.08
% of HRmax during intervals
∆ W max (W·kg−1) 0.07 0.134 [−0.02, 0.16] 0.25
∆ PO4mmol (W·kg−1) 0.06 0.060 # [−0.00, 0.13] 0.21
∆ PO15min (W·kg−1) 0.05 0.172 [−0.02, 0.12] 0.16
∆ performance index (AU) 0.01 0.036* [0.00, 0.02] 0.29
∆ V̇O2max (mL·min−1·kg−1) 0.48 0.226 [−0.32, 1.27] 0.43
∆ % of V̇O2max@4mmol (%‐points) 0.86 0.300 [−0.84, 2.57] 0.34
∆ GE175W (%‐points) −0.08 0.574 [−0.40, 0.23] 0.07
Time ≥90% of HRmax during intervals (minutes)
∆ W max (W·kg−1) 0.01 0.266 [−0.01, 0.04] 0.21
∆ PO4mmol (W·kg−1) 0.02 0.043* [0.00, 0.03] 0.24
∆ PO15min (W·kg−1) 0.01 0.095 # [−0.00, 0.03] 0.21
∆ performance index (AU) 0.00 0.037* [0.00, 0.00] 0.29
∆ V̇O2max (mL·min−1·kg−1) 0.10 0.292 [−0.09, 0.30] 0.42
∆ % of V̇O2max@4mmol (%‐points) 0.27 0.139 [−0.10, 0.64] 0.38
∆ GE175W (%‐points) 0.02 0.515 [−0.10, 0.05] 0.08
Open in a new tab

Note: V̇O2max, maximal oxygen uptake; HRmax, maximal heart rate during the incremental test for determination of V̇O2max (V̇O2max‐test); W max, maximal 1‐min incremental power output during the V̇O2max test; PO4mmol, power output at 4 mmol·L−1 blood lactate concentration ([La−]); PO15min, maximal power output during the 15‐min cycling trial; % of V̇O2max@4mmol, fractional utilization of V̇O2max at 4 mmol·L−1 [La−]; GE175W, gross efficiency at 175 W.

#Tendency to relationship (p < 0.1 and >0.05).

aFor each 1‐min increase in time ≥90% of V̇O2max and HRmax, and each %‐point increase in % of HRmax during intervals, the dependent variables change according to the estimates.

*Significant relationship (p ≤ 0.05).

4. DISCUSSION

The main findings of the present study support our initial hypothesis that gains in endurance measures after an interval training intervention are positively related to the % of V̇O2max and time spent ≥90% of V̇O2max during the sessions, and that these measures are more reliable expressions of the adaptive stimulus elicited by interval training and better predictors of endurance performance gains. Compared to % of V̇O2max and time ≥90% of V̇O2max during intervals, % of HRmax and time ≥90% of HRmax showed weaker and fewer positive associations with changes in outcome variables, in addition to weaker reliability across sessions.

4.1. Effects of % of V̇O2max on changes in endurance performance measures

The efficacy of exercising at intensities near V̇O2max for adaptations to prolonged endurance training have been repeatedly discussed for at least 40 years (Billat et al., 1996; Buchheit et al., 2013; Daniels et al., 1984; Held et al., 2023; Midgley, McNaughton, & Wilkinson, 2006; Rønnestad, Hansen, et al., 2014; Thevenet et al., 2007; Wenger et al., 1986). In recent years, this topic has been actualized through several acute studies comparing the V̇O2 response to different types of interval sessions (Almquist et al., 2020; Bossi et al., 2020; Held et al., 2023; Rønnestad et al., 2016; Thevenet et al., 2007). In these studies, the interval sessions eliciting the longest accumulated time ≥90% of V̇O2max have been argued to be the most efficient for enhancing V̇O2max and endurance performance. However, this relationship has not yet been directly measured and investigated. Consequently, the present study is the first to provide scientific evidence for the statement that longer accumulated time ≥90% of V̇O2max during intervals, as well as higher % of V̇O2max, provides better training adaptations. The current study shows that both measures are positively associated with improvements in W max, PO4mmol, and the calculated performance index. Such findings may have significant implications for competitive cyclists, as W max is considered a good predictor of cycling performance (Faria et al., 2005) and has previously been observed to distinguish well‐trained cyclists from elite cyclists (Lucia et al., 1998). Similarly, PO4mmol, frequently referred to as “lactate threshold PO,” is known to play a key role in endurance performance (Joyner et al., 2008). Furthermore, the performance index in the current study is considered to reflect the performance ability of the participants more accurately since it, in a weighted fashion, combines several indices of cycling performance (W max, PO4mmol, and PO15min; Rønnestad, Urianstad, et al., 2022). Notably, the observed associations were slightly more robust for % of V̇O2max than time ≥90% of V̇O2max. This could be because % of V̇O2max utilizes all collected data across the work intervals, while time ≥90% of V̇O2max is a binary measure (above/under 90% of V̇O2max), which therefore potentially are less accurate in reflecting the overall training stimulus. Both % of V̇O2max and time ≥90% of V̇O2max showed moderate repeatability across sessions and can, therefore, be recommended as accurate expressions of adaptive potential. Based on the numerically larger ICC of % of V̇O2max compared to time ≥90% of V̇O2max (0.61 vs. 0.56, respectively) and the more robust association between % of V̇O2max and V̇O2max improvement, it can be suggested that the former is the preferred adaptive potential measure.

Previously, only two studies have measured time at V̇O2max or time ≥90% of V̇O2max during interval sessions and related this to changes in endurance measures. Both studies observed improved maximal velocity achieved during an incremental test and/or improved velocity/PO at lactate threshold, but in contrast to our observation, there were no positive relationships between these improvements and time ≥90% of V̇O2max (Rønnestad, Bjerkrheim, et al., 2022; Turnes et al., 2016). Rønnestad, et al. (Rønnestad, Bjerkrheim, et al., 2022) performed a 1‐week interval block in cross‐country skiers and it can be speculated that the lack of agreement with the current study can be due to the shorter duration of the training intervention, lack of V̇O2 measurement during all interval sessions, and/or different forms of exercise. Despite no correlations with change scores, Turnes, et al. (Turnes et al., 2016) observed that the interval session eliciting the longest time at V̇O2max also induced the largest improvement in PO at lactate threshold. Accordingly, in our secondary analyses, we observed larger improvements for the HIGH%VȮ2max group compared to the LOW%VȮ2max group for changes in W max, PO4mmol, and the performance index, as well as a tendency to a greater improvement in PO15min. Notably, we did not observe a positive relationship between % of V̇O2max during intervals and improvements in PO15min. The 15‐min cycling trial was performed in a fatigued state (after the leg press, the lactate profile and the V̇O2max test on test day 1), and considering that greater variability in professional cyclists' fatigued power profile has been observed when compared to the power profile conducted in a fresh state (Spragg et al., 2023), this could potentially have influenced the results.

4.2. Effects of % of V̇O2max on changes in endurance performance determinants

In their well‐cited review, Wenger and Bell (Wenger et al., 1986) argued that “the magnitude of change in V̇O2max increases as exercise intensity increases from 50% to 100% of V̇O2max.” Notably, this claim was based on converting % of HRmax or % of heart rate reserve to % of V̇O2max, and the statistics used were not presented. Still, their ∼40‐year‐old statement coincides well with our present finding that the increase in V̇O2max is positively related to the % of V̇O2max during interval sessions. This was further emphasized by the greater increase in V̇O2max for the HIGH%V̇O2max group compared to the LOW%V̇O2max group. For time ≥90% of V̇O2max, only a tendency toward a positive relationship with increase in V̇O2max was observed. This is consistent with the relationship observed by Rønnestad, et al. (Rønnestad, Bjerkrheim, et al., 2022), and indicates that % of V̇O2max is better than time ≥90% of V̇O2max in reflecting the magnitude of training adaptations following interval training.

Percentage of V̇O2max and time ≥90% of V̇O2max during intervals only tended to be related to changes in % of V̇O2max@4mmol. Moreover, there was no difference between HIGH%V̇O2max and LOW%V̇O2max for change in this variable. This is not surprising as improvements of fractional utilization of V̇O2max do not appear to be exercise intensity‐dependent (Helgerud et al., 2007). Moreover, it seems that endurance training interventions lasting eight to 12 weeks are not sufficient to increase the fractional utilization of V̇O2max in already trained cyclists (Rønnestad et al., 2014c, 2017; Sunde et al., 2010); years of systematic training are probably required (Rusko, 1987; Rønnestad et al., 2019). Training intensity and volume have both been observed to correlate with improvements in gross efficiency over the course of a competitive cycling season (Hopker et al., 2009), but is reported to remain unchanged following weeks of high‐intensity interval training (Rønnestad et al., 2014; Rønnestad et al., 2020; Skovereng et al., 2018). In line with this, both % of V̇O2max and time ≥90% of V̇O2max during intervals was not positively related to GE changes. Unexpectedly, we did observe a tendency to a greater change in GE175W in LOW%V̇O2max compared to HIGH%V̇O2max. However, previous findings suggest that gross efficiency may initially be decreased when V̇O2max is increased in competitive cyclists following an interval training intervention (Skovereng et al., 2018; Sylta et al., 2016). Furthermore, HIGH%V̇O2max had a higher GE175W compared to LOW%V̇O2max at the baseline, and a moderate negative correlation has previously been observed between initial gross efficiency and its change following a relatively short endurance training period (Skovereng et al., 2018).

4.3. Training variables and their associations with training adaptations

Among the training variables recorded during interval sessions, only time ≥90% of V̇O2max and % of V̇O2max differed significantly between the HIGH%V̇O2max and the LOW%V̇O2max groups. In contrast, measures such as % of HRmax and time ≥90% of HRmax were not different between groups. Importantly, there was no differences in PO during interval sessions as this was matched between participants (corresponding to PO40min). Furthermore, % of HRmax during intervals was only positively related to improvements in the performance index, and time ≥90% of HRmax only cohered with improvements in PO4mmol and the performance index. Additionally, % of HRmax and time ≥90% of HRmax had both poor session‐to‐session repeatability (0.31 and 0.24, respectively), contrasting to the moderate repeatability observed for % of V̇O2max and time ≥90% of V̇O2max (0.61 and 0.56, respectively). Interestingly, Rønnestad, et al. (Rønnestad, Bjerkrheim, et al., 2022) observed the same level of repeatability for time ≥90% of V̇O2max and time ≥90% of HRmax in cross‐country skiers (0.57 and 0.24, respectively), implying that this level of repeatability is consistent between exercise modes. Taken together, these findings indicate that % of V̇O2max is an accurate measure and a robust determinant of physical adaptations to prolonged interval training and that exercising at a higher % of V̇O2max during intervals induces a greater adaptational stimulus compared to exercising at lower fractions of V̇O2max. Adding to this, the other collected training variables, [La−], RPE, perceived well‐being in the legs, and sRPE were not different between the HIGH%V̇O2max and LOW%V̇O2max groups within all training periods. This further underlines the superiority of V̇O2 measurements to characterize the training stimulus.

4.4. Limitations

5 × 8‐min interval sessions performed at the PO corresponding to PO40min is not necessarily the optimal session design to elicit a high % of V̇O2max and long durations ≥90% of V̇O2max for all individuals. Shorter work intervals (<8‐min) at subsequently higher POs could potentially have elicited a greater V̇O2 for some participants. The HIGH%V̇O2max group performed significantly more zone 1 training and tended to have a greater total training volume compared to LOW%V̇O2max, which could have affected the results. However, LOW%V̇O2max had a numerically longer training duration in zone 4 and 5, and as high‐intensity training is considered a more time‐efficient form of training to improve endurance performance in endurance‐trained athletes (Buchheit et al., 2013; Laursen et al., 2002), it was most likely the higher % of V̇O2max and time ≥90% of V̇O2max during interval sessions that led to the greater training adaptations in HIGH%V̇O2max compared to LOW%V̇O2max.

4.5. Practical implications

In the present study, the % of V̇O2max during intervals ranged from 74% to 91% for the participants, and those who elicited the highest % of V̇O2max during the 21 interval sessions also got the largest training adaptations. Thus, for optimizing training adaptations, we recommend cyclists to perform interval sessions that elicits a V̇O2 close to their maximal. To verify what type of interval session protocol(s) that elicits the greatest V̇O2 response for the individual cyclist, regular screening visits to a physiological laboratory can be useful. Additionally, we observed that the cyclists' initial % of V̇O2max@4mmol was positively related to the % of V̇O2max they were able to sustain during the interval sessions. Therefore, it can be speculated whether cyclists with a relatively low % of V̇O2max@4mmol would benefit from performing shorter work intervals (e.g., <8 min) at subsequently higher POs to increase the likelihood of a higher % of V̇O2max during the interval session.

Noteworthy, the moderate and, thus, not perfect repeatability for both % of V̇O2max and time ≥90% of V̇O2max across the 21 interval sessions implies a certain session‐to‐session variation for these measures. Therefore, caution should be exerted when interpreting and generalizing the V̇O2 response of a single interval session. Moreover, as % of V̇O2max showed slightly better repeatability across sessions and was related to slightly more outcome variables than time ≥90% of V̇O2max in the present study, we suggest that % of V̇O2max is the best available adaptive potential measure. HR measurements are often used as a surrogate measure for V̇O2, but based on the presented findings we argue that these measures should not be used interchangeably.

5. CONCLUSION

In cyclists, the gains in endurance measures following 9 weeks of interval training are positively related to the % of V̇O2max achieved and time spent ≥90% of V̇O2max during the interval sessions. When compared to these metrics, % of HRmax and time ≥90% of HRmax were found to be less accurate indicators of the adaptive stimulus. Furthermore, % of V̇O2max and time spent ≥90% of V̇O2max displayed the best session‐to‐session reliability, which further emphasizes the superiority of these metrics.

CONFLICT OF INTEREST STATEMENT

No conflicts of interest, financial or otherwise, are declared by the authors.

PATIENT CONSENT STATEMENT

Before inclusion, the participants were informed of any potential risks and discomfort associated with the study, and they all provided their written informed consent to participate.

ACKNOWLEDGMENTS

The authors thank Torkil Rogneflåten, Anders Sørensen, Andreas Øhrn, Halvor T. Tjøntveit, Martine S. Hasle, Aida Besic, Maria G. Jacobsen, Malin Aannestad, and Simen Næss Berge for their help during data collection. We also thank the dedicated group of cyclists who made this study possible.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

  1. Almquist, Nicki W. , Nygaard Håvard, Vegge Geir, Hammarström Daniel, Ellefsen Stian, and Rønnestad Bent R.. 2020. “Systemic and Muscular Responses to Effort‐Matched Short Intervals and Long Intervals in Elite Cyclists.” Scandinavian Journal of Medicine & Science in Sports 30(7): 1140–1150. 10.1111/sms.13672. [DOI] [PubMed] [Google Scholar]
  2. Billat, L. Véronique , and Koralsztein J. Pierre. 1996. “Significance of the Velocity at VO2max and Time to Exhaustion at This Velocity.” Sports Medicine 22(2): 90–108. 10.2165/00007256-199622020-00004. [DOI] [PubMed] [Google Scholar]
  3. Borg, G. 1998. Borg's Perceived Exertion and Pain Scales, 104–viii. Champaign, IL, US: Human Kinetics: viii. [Google Scholar]
  4. Bossi, Arthur H. , Mesquida Cristian, Passfield Louis, Rønnestad Bent R., and Hopker James G.. 2020. “Optimizing Interval Training through Power‐Output Variation within the Work Intervals.” International Journal of Sports Physiology and Performance 15(7): 982–989. 10.1123/ijspp.2019-0260. [DOI] [PubMed] [Google Scholar]
  5. Buchheit, Martin , and Laursen Paul B.. 2013. “High‐Intensity Interval Training, Solutions to the Programming Puzzle: Part I.” Sports Medicine 43(5): 313–338. 10.1007/s40279-013-0029-x. [DOI] [PubMed] [Google Scholar]
  6. Daniels, Jack , and Scardina Nancy. 1984. “Interval Training and Performance.” Sports Medicine 1(4): 327–334. 10.2165/00007256-198401040-00006. [DOI] [PubMed] [Google Scholar]
  7. Decroix, Lieselot , De Pauw Kevin, Foster Carl, and Meeusen Romain. 2016. “Guidelines to Classify Female Subject Groups in Sport‐Science Research.” International Journal of Sports Physiology and Performance 11(2): 204–213. 10.1123/ijspp.2015-0153. [DOI] [PubMed] [Google Scholar]
  8. Faria, Erik W. , Parker Daryl L., and Faria Irvin E.. 2005. “The Science of Cycling: Physiology and Training ‐ Part 1.” Sports Medicine 35(4): 285–312. 10.2165/00007256-200535040-00002. [DOI] [PubMed] [Google Scholar]
  9. Foster, Carl , Florhaug Jessica A., Franklin Jodi, Gottschall Lori, Hrovatin Lauri A., Parker Suzanne, Doleshal Pamela, and Dodge Christopher. 2001. “A New Approach to Monitoring Exercise Training.” The Journal of Strength & Conditioning Research 15(1): 109–115. . [DOI] [PubMed] [Google Scholar]
  10. Held, Steffen , Rappelt Ludwig, Giesen René, Wiedenmann Tim, Deutsch J.‐Philip, Wicker Pamela, and Donath Lars. 2023. “Increased Oxygen Uptake in Well‐Trained Runners during Uphill High Intensity Running Intervals: A Randomized Crossover Testing.” Frontiers in Physiology 14. 10.3389/fphys.2023.1117314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Helgerud, Jan , Høydal Kjetill, Wang Eivind, Karlsen Trine, Berg Pålr, Bjerkaas Marius, Simonsen Thomas, et al. 2007. “Aerobic High‐Intensity Intervals Improve VO2max More Than Moderate Training.” Medicine & Science in Sports & Exercise 39(4): 665–671. 10.1249/mss.0b013e3180304570. [DOI] [PubMed] [Google Scholar]
  12. Hopker, James , Coleman Damian, and Passfield Louis. 2009. “Changes in Cycling Efficiency during a Competitive Season.” Medicine & Science in Sports & Exercise 41(4): 912–919. 10.1249/MSS.0b013e31818f2ab2. [DOI] [PubMed] [Google Scholar]
  13. Hunter, A. , and Coggan A.. 2010. Training and Racing with a Power Meter, 2nd ed., 36. ColoradoBiz: Velopress. [Google Scholar]
  14. Joyner, Michael J. , and Coyle Edward F.. 2008. “Endurance Exercise Performance: the Physiology of Champions.” The Journal of Physiology 586(1): 35–44. 10.1113/jphysiol.2007.143834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Koo, Terry K. , and Li Mae Y.. 2016. “A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research.” J Chiropr Med 15(2): 155–163. 10.1016/j.jcm.2016.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lamberts, R. P. , Swart J., Noakes T. D., and Lambert M. I.. 2011. “A Novel Submaximal Cycle Test to Monitor Fatigue and Predict Cycling Performance.” British Journal of Sports Medicine 45(10): 797–804. 10.1136/bjsm.2009.061325. [DOI] [PubMed] [Google Scholar]
  17. Laursen, Paul B. , and Jenkins David G.. 2002. “The Scientific Basis for High‐Intensity Interval Training.” Sports Medicine 32(1): 53–73. 10.2165/00007256-200232010-00003. [DOI] [PubMed] [Google Scholar]
  18. Lucia, A. , Pardo J., Durantez A., Hoyos J., and Chicharro J.. 1998. “Physiological Differences between Professional and Elite Road Cyclists.” Int J Sports Med 19(5): 342–348. 10.1055/s-2007-971928. [DOI] [PubMed] [Google Scholar]
  19. Midgley, A. W. , and Mc Naughton L. R.. 2006a. “Time at or Near VO2max during Continuous and Intermittent Running. A Review with Special Reference to Considerations for the Optimisation of Training Protocols to Elicit the Longest Time at or Near VO2max.” The Journal of Sports Medicine and Physical Fitness 46(1): 1–14. [PubMed] [Google Scholar]
  20. Midgley, Adrian W. , McNaughton Lars R., and Wilkinson Michael. 2006. “Is There an Optimal Training Intensity for Enhancing the Maximal Oxygen Uptake of Distance Runners?” Sports Medicine 36(2): 117–132. 10.2165/00007256-200636020-00003. [DOI] [PubMed] [Google Scholar]
  21. Noordhof, Dionne A. , Skiba Philip F., and de Koning Jos J.. 2013. “Determining Anaerobic Capacity in Sporting Activities.” International Journal of Sports Physiology and Performance 8(5): 475–482. 10.1123/ijspp.8.5.475. [DOI] [PubMed] [Google Scholar]
  22. Pauw, Kevin De , Roelands Bart, Cheung Stephen S., de Geus Bas, Rietjens Gerard, and Meeusen Romain. 2013. “Guidelines to Classify Subject Groups in Sport‐Science Research.” International Journal of Sports Physiology and Performance 8(2): 111–122. 10.1123/ijspp.8.2.111. [DOI] [PubMed] [Google Scholar]
  23. Péronnet, F. , and Massicotte D.. 1991. “Table of Nonprotein Respiratory Quotient: an Update.” Canadian Journal of Sport Sciences 16(1): 23–29. [PubMed] [Google Scholar]
  24. Rønnestad, Bent R. , Hansen Joar, Nygaard Håvard, and Lundby Carsten. 2020. “Superior Performance Improvements in Elite Cyclists Following Short‐Interval vs Effort‐Matched Long‐Interval Training.” Scandinavian Journal of Medicine & Science in Sports 30(5): 849–857. 10.1111/sms.13627. [DOI] [PubMed] [Google Scholar]
  25. Rønnestad, Bent R. , Bjerkrheim Kjetil Andre, Hansen Joar, and Mølmen Knut Sindre. 2022. “A 6‐day High‐Intensity Interval Microcycle Improves Indicators of Endurance Performance in Elite Cross‐Country Skiers.” Front Sports Act Living 4: 948127. 10.3389/fspor.2022.948127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rønnestad, B. R. , Ellefsen S., Nygaard H., Zacharoff E. E., Vikmoen O., Hansen J., and Hallén J.. 2014. “Effects of 12 Weeks of Block Periodization on Performance and Performance Indices in Well‐Trained Cyclists.” Scandinavian Journal of Medicine & Science in Sports 24(2): 327–335. 10.1111/sms.12016. [DOI] [PubMed] [Google Scholar]
  27. Rønnestad, Bent R. , and Hansen Joar. 2016. “Optimizing Interval Training at Power Output Associated with Peak Oxygen Uptake in Well‐Trained Cyclists.” The Journal of Strength & Conditioning Research 30(4): 999–1006. 10.1519/JSC.0b013e3182a73e8a. [DOI] [PubMed] [Google Scholar]
  28. Rønnestad, B. R. , Hansen J., and Ellefsen S.. 2014. “Block Periodization of High‐Intensity Aerobic Intervals Provides Superior Training Effects in Trained Cyclists.” Scandinavian Journal of Medicine & Science in Sports 24(1): 34–42. 10.1111/j.1600-0838.2012.01485.x. [DOI] [PubMed] [Google Scholar]
  29. Rønnestad, Bent R. , Hansen Joar, and Nygaard Håvard. 2017. “10 Weeks of Heavy Strength Training Improves Performance‐Related Measurements in Elite Cyclists.” Journal of Sports Sciences 35(14): 1435–1441. 10.1080/02640414.2016.1215499. [DOI] [PubMed] [Google Scholar]
  30. Rønnestad, Bent R. , Hansen Joar, Stensløkken Lars, Joyner Michael J., and Lundby Carsten. 2019. “Case Studies in Physiology: Temporal Changes in Determinants of Aerobic Performance in Individual Going from Alpine Skier to World Champion Time Trial Cyclist.” Journal of Applied Physiology 127(2): 306–311. 10.1152/japplphysiol.00798.2018. [DOI] [PubMed] [Google Scholar]
  31. Rønnestad, B. R. , Hansen J., Vegge G., Tønnessen E., and Slettaløkken G.. 2014. “Short Intervals Induce Superior Training Adaptations Compared with Long Intervals in Cyclists – an Effort‐Matched Approach.” Scandinavian Journal of Medicine & Science in Sports 25(2): 143–151. 10.1111/sms.12165. [DOI] [PubMed] [Google Scholar]
  32. Rønnestad, Bent R. , Lid Ole Martin, Hansen Joar, Hamarsland Håvard, Mølmen Knut Sindre, Nygaard Håvard, Ellefsen Stian, Hammarström Daniel, and Lundby Carsten. 2022. “Heat Suit Training Increases Hemoglobin Mass in Elite Cross‐Country Skiers.” Scandinavian Journal of Medicine & Science in Sports 32(7): 1089–1098. 10.1111/sms.14156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rønnestad, Bent R. , Urianstad Tomas, Hamarsland Håvard, Hansen Joar, Nygaard Håvard, Ellefsen Stian, Hammarström Daniel, and Lundby Carsten. 2022. “Heat Training Efficiently Increases and Maintains Haemoglobin Mass and Temperate Endurance Performance in Elite Cyclists.” Medicine & Science in Sports & Exercise 54(9): 1515–1526. 10.1249/mss.0000000000002928. [DOI] [PubMed] [Google Scholar]
  34. Rusko, Heikki . 1987. “The Effect of Training on Aerobic Power Characteristics of Young Cross‐country Skiers.” Journal of Sports Sciences 5(3): 273–286. 10.1080/02640418708729782. [DOI] [PubMed] [Google Scholar]
  35. Skovereng, Knut , Sylta Øystein, Tønnessen Espen, Hammarström Daniel, Danielsen Jørgen, Seiler Stephen, Rønnestad Bent R., et al. 2018. “Effects of Initial Performance, Gross Efficiency and V̇O2peak Characteristics on Subsequent Adaptations to Endurance Training in Competitive Cyclists.” Frontiers in Physiology 9. 10.3389/fphys.2018.00713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Spragg, James , Leo Peter, and Swart Jeroen. 2023. “The Relationship between Training Characteristics and Durability in Professional Cyclists across a Competitive Season.” European Journal of Sport Science 23(4): 489–498. 10.1080/17461391.2022.2049886. [DOI] [PubMed] [Google Scholar]
  37. Sunde, Arnstein , Støren Øyvind, Bjerkaas Marius, Larsen Morten H., Hoff Jan, and Helgerud Jan. 2010. “Maximal Strength Training Improves Cycling Economy in Competitive Cyclists.” The Journal of Strength & Conditioning Research 24(8): 2157–2165. 10.1519/JSC.0b013e3181aeb16a. [DOI] [PubMed] [Google Scholar]
  38. Sylta, Øystein , Tønnessen Espen, Hammarström Daniel, Danielsen Jørgen, Skovereng Knut, Ravn Troels, Rønnestad Bent R., et al. 2016. “The Effect of Different High‐Intensity Periodization Models on Endurance Adaptations.” Medicine & Science in Sports & Exercise 48(11): 2165–2174. 10.1249/mss.0000000000001007. [DOI] [PubMed] [Google Scholar]
  39. Thevenet, Delphine , Tardieu‐Berger Magaly, Berthoin Serge, and Prioux Jacques. 2007. “Influence of Recovery Mode (Passive vs. Active) on Time Spent at Maximal Oxygen Uptake during an Intermittent Session in Young and Endurance‐Trained Athletes.” European Journal of Applied Physiology 99(2): 133–142. 10.1007/s00421-006-0327-1. [DOI] [PubMed] [Google Scholar]
  40. Turnes, Tiago , de Aguiar Rafael Alves, Cruz Rogério Santos de Oliveira, and Caputo Fabrizio. 2016. “Interval Training in the Boundaries of Severe Domain: Effects on Aerobic Parameters.” European Journal of Applied Physiology 116(1): 161–169. 10.1007/s00421-015-3263-0. [DOI] [PubMed] [Google Scholar]
  41. Wenger, Howard A. , and Bell Gordon J.. 1986. “The Interactions of Intensity, Frequency and Duration of Exercise Training in Altering Cardiorespiratory Fitness.” Sports Medicine 3(5): 346–356. 10.2165/00007256-198603050-00004. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from European Journal of Sport Science are provided here courtesy of Wiley

RESOURCES