See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/349460630

The dose-response relationship between interval-training and VO 2max in

well-trained endurance runners: A systematic review

Article  in  Journal of Sports Sciences · February 2021

DOI: 10.1080/02640414.2021.1876313

CITATION READS

1 349

3 authors:

Arran Parmar Thomas W. Jones

Northumbria University Northumbria University
4 PUBLICATIONS   2 CITATIONS    40 PUBLICATIONS   343 CITATIONS   

SEE PROFILE SEE PROFILE

Philip R Hayes
Northumbria University
68 PUBLICATIONS   1,485 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Investigating the complexity of the role of the cricket coach, in developing psychophysiological athlete outcomes / examining the relationships between cricket
coaching practice and psychophysiological athlete outcomes View project

Energy expenditure during exercise View project

All content following this page was uploaded by Philip R Hayes on 28 March 2021.

The user has requested enhancement of the downloaded file.

 Journal of Sports Sciences

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rjsp20

The dose-response relationship between interval-

training and VO2max in well-trained endurance
runners: A systematic review

Arran Parmar , Thomas W. Jones & Philip, R. Hayes

To cite this article: Arran Parmar , Thomas W. Jones & Philip, R. Hayes (2021): The dose-
response relationship between interval-training and VO2max in well-trained endurance runners: A
systematic review, Journal of Sports Sciences, DOI: 10.1080/02640414.2021.1876313

To link to this article: https://doi.org/10.1080/02640414.2021.1876313

Published online: 19 Feb 2021.

Submit your article to this journal

Article views: 24

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rjsp20
JOURNAL OF SPORTS SCIENCES
https://doi.org/10.1080/02640414.2021.1876313

PHYSIOLOGY AND NUTRITION

The dose-response relationship between interval-training and VO2max in well-trained

endurance runners: A systematic review
Arran Parmar , Thomas W. Jones and Philip, R. Hayes
Department of Sport, Exercise and Rehabilitation, Faculty of Health and Life Sciences, Northumbria University, Newcastle upon Tyne, UK

ABSTRACT ARTICLE HISTORY

Success in endurance running is primarily determined by maximal aerobic power (VO2max), fractional Accepted 11 January 2021
utilization, and running economy (RE). Within the literature, two training modalities have been identified
KEYWORDS
to improve VO2max; continuous training (CT) and interval-training (IT). The efficacy of IT to improve VO2max Running; endurance;
in well-trained runners remains equivocal, as does whether a dose-response relationship exists between training-impulse; training-
the IT training load performed and changes in VO2max. A keyword search was performed in five electronic load; high-intensity
databases. Seven studies met the inclusion criteria for this systematic review. The training impulse
(TRIMP) was calculated to analyse relationships between training load and changes in VO2max, by
calculating the time accumulated in certain intensity domains throughout a training intervention. Non-
significant (P>0.05) improvements in VO2max were reported in six studies, with only one study reporting
a significant (P<0.05) improvement following the IT interventions. A relationship between the training
session impulse of the interval-training performed (IT STRIMP) and VO2max improvements were observed.
The efficacy of IT to improve VO2max in well-trained runners remains equivocal, nevertheless, the novel
method of training-load analysis demonstrates a relationship between the IT STRIMP and VO2max improve­
ments. This provides practical application for the periodization of IT within the training regime of well-
trained distance runners.

Introduction longer duration work intervals have been shown to elicit

greater increases in VO2max (Bacon et al., 2013; Milanović
Maximal aerobic power (VO2max) is an established determinant
et al., 2015; Wen et al., 2019), as such work intervals maximally
of endurance performance (Blagrove et al., 2018; Foster, 1983;
stress cardiorespiratory parameters by increasing the time
JDR Bassett & Howley, 1997; Joyner, 1991; Pollock et al., 1980).
spent at, or close to VO2max (T@VO2max) to a greater extent
Training methods to improve VO2max are characterized in two
than shorter work intervals, leading to greater adaptations
modes: continuous training (CT) and interval training (IT) meth­
(ÅStrand et al., 1960; Bacon et al., 2013; Buchheit & Laursen,
ods (Laursen & Jenkins, 2002; KS Seiler & Kjerland, 2006). CT
2013; Helgerud et al., 2007; Milanović et al., 2015; Seiler et al.,
methods consist of long durations of sub-maximal intensity
2013; S Seiler & Sjursen, 2004). This suggests a dose-response
exercise typically eliciting adaptations associated with oxygen
relationship exists between T@VO2max and improvements in
utilization (F Daussin et al., 2007). By contrast, IT methods
VO2max, hence IT methods maximizing T@VO2max might elicit
consist of repeated higher intensity work durations (above or
the greatest improvements in VO2max. The vast majority of
equal to the maximal steady state) interspersed with periods of
evidence supporting this has primarily been reported in lesser-
recovery (light exercise or rest) typically eliciting adaptations
trained populations, meaning the efficacy of IT methods in
associated with oxygen delivery (Astorino et al., 2017; Billat,
well-trained populations is unclear.
2001; FN Daussin et al., 2008; Milanović et al., 2015).
A lack of conclusive evidence exists to support the efficacy
The use of IT methods to improve VO2max has increased in
of IT methods to improve VO2max in well-trained endurance
popularity with evidence showing the same if not greater
athletes. Midgley et al. (Midgley et al., 2007) reviewed 23 stu­
improvements compared to CT methods (Bacon et al., 2013;
dies for improvements in VO2max, RE, and lactate threshold (LT)
Helgerud et al., 2007; Milanović et al., 2015). Training intensities
in response to training interventions consisting of plyometric
at or close to VO2max have been suggested to be optimal in
training, interval training, resistance training, and continuous
improving VO2max (Midgley et al., 2006), with the total time
training, of which, 14 included well-trained runners (VO2max >
spent at this intensity proportional to the increase in VO2max.
60 mL·kg−1·min−1). Only one study included in the review uti­
Further supporting this, numerous studies and reviews have
lized IT in well-trained runners (n = 8; VO2max = 71.2 ± 5 mL·kg−1­
shown greater improvements in VO2max utilizing IT methods
·min−1), with runners performing 1 IT session per week for
close to, or at an intensity eliciting VO2max compared to CT
4 weeks increasing to 3 IT sessions per week for another
methods matched for load (Bacon et al., 2013; Ní Chéilleachair
4 weeks, consisting of 5 repetitions at 100% vVO2max for 3 min­
et al., 2017; Driller et al., 2009; Helgerud et al., 2007; Koral et al.,
utes separated by 3 minutes of rest at 50% vVO2max, however
2018; Mallol et al., 2018; Milanović et al., 2015). Interestingly,

CONTACT Arran Parmar, Email arran.parmar@northumbria.ac.uk Department of Sport, Exercise and Rehabilitation, Faculty of Health and Life Sciences,
Northumbria University, Newcastle upon Tyne, UK
© 2021 Informa UK Limited, trading as Taylor & Francis Group
2 A. PARMAR ET AL.

no improvements in VO2max were reported (Billat et al., 1999). In
contrast, a study utilizing a near identical IT intervention in
well-trained runners (VO2max = 61.5 ± 2.9 mL·kg−1·min−1)
reported significant increases in VO2max, however, the sample
size was low (n = 5) and therefore potentially underpowered
(Smith et al., 1999). In a follow up study utilizing this near
identical IT intervention of 2 IT sessions per week for 4 weeks
consisting of 5–6 repetitions at 100% vVO2max for approxi­
mately 2–3 minutes at a work: rest ratio of 1:2 in a larger sample
size of well-trained runners (two groups: n = 9 in each group;
VO2max ≥ 60.1 mL·kg−1·min−1), no significant improvements in
VO2max were reported (Smith et al., 2003). The IT interventions
used within the abovementioned studies might not have pro­
vided the T@VO2max required to stimulate improvements in
VO2max, with previous work in lesser-trained populations sug­
gesting >15 mins of T@VO2max accumulated per session to be
effective in maximizing VO2max improvements (Bacon et al.,
2013; Wen et al., 2019), supporting the notion of a dose-
response relationship. In well-trained endurance athletes
a minimum training dose combining volume and intensity
might need to be exceeded to elicit chronic adaptations in
the cardiorespiratory parameters mediating VO2max (Midgley
et al., 2007). In support of this, a 10-week IT intervention similar
to that used previously (Billat et al., 1999; Smith et al., 2003)
resulted in improvements in VO2max, albeit in lesser-trained
runners (VO2max = 51.6 ± 2.7 mL·kg−1·min−1) (Esfarjani &
Laursen, 2007). The longer training intervention (10-weeks
(Esfarjani & Laursen, 2007) vs 4-weeks (Smith et al., 2003) vs
8-weeks (Billat et al., 1999)) perhaps provided the accumulated
T@VO2max and training dose required to stimulate improve­
ments in VO2max. Had the previous studies with similar IT pro­
tocols used longer interventions (Billat et al., 1999; Smith et al.,
2003), improvements in VO2max could have been observed due
to the greater accumulated training load exceeding the mini­
mum training dose required for improvements in VO2max in
these well-trained runners, however this remains speculative.
The efficacy of IT methods is well-established in lesser-
trained populations (Bacon et al., 2013; Milanović et al., 2015),
however, inconclusive evidence exists to support the effective­
ness of IT methods to improve cardiorespiratory and metabolic
factors in well-trained runners (Laursen & Jenkins, 2002;
Midgley et al., 2007). The apparent dose-response relationship
between training load and improvements in VO2max, perhaps
offers a method of training load analysis that could be effective
for well-trained runners aiming to improve VO2max. Therefore,
this review aimed to analyse the volume and quality of the
current evidence pertaining to the chronic effects of IT meth­
ods in well-trained runners on improving VO2max. A further aim
was to analyse the dose-response relationship associated with
the total load of IT and changes in VO2max.
Materials and methods
This systematic review was conducted according to the
‘Preferred Reporting Items for Systematic Reviews and Meta-
Analyses’ (PRISMA) guidelines (Moher et al., 2009). All the fol­
lowing steps were implemented by three independent raters
(AP, PH, TJ), with discrepancies and conflicts resolved by
discussion.
Literature search strategy
Electronic database searches were carried out in PubMed,
MEDLine, SPORTDiscus, CINAHL, and Web of Science. All
searches were conducted between 24 January 2019 and
12 April 2019. Searches were limited to papers published in
English and from 1 January 1960 to 12 April 2019. Further
searches on authors known by the investigators to have pub­
lished papers using interval-training interventions in well-
trained distance runners were conducted. Additionally, refer­
ence lists of all eligible studies were reviewed to identify poten­
tially eligible studies that may have been missed. The following
strategy using Boolean search terms and operators was used in
each electronic database:
(“Runners” OR “Endurance Runners” OR “Running” OR
“Middle Distance” OR “Long Distance” OR “Marathon” OR
“Well Trained” OR “Competitive” OR “Athlete” OR “Endurance
Athlete” OR “Elite” OR “High Level”) AND (“Interval Training” OR
“High Intensity Interval Training” OR “High Intensity Interval
Exercise” OR “HIIT” OR “HIIE” OR “Sprint Interval Training” OR
“SIT” OR “Aerobic Interval Training” OR “Maximum Intensity
Interval” OR Intermittent”) AND (“VO2” OR “VO2 Max” OR
“VO2 Peak” OR “Maximal Oxygen Uptake” OR “Aerobic
Fitness” OR “Maximum Aerobic Capacity” OR “Maximum
Oxygen Consumption” OR “Peak Oxygen Uptake” OR “Peak
Oxygen Consumption” OR “Maximum Aerobic Power”) NOT
(“Cycling” OR “Swimming” OR “Rowing” OR “Skiing” OR
“Soccer” OR “Football” OR “Basketball” OR “Untrained” OR
“Recreational” OR “Clinical” OR “Obese” OR “Youth” OR
“Adolescent” OR “Older”).
Inclusion and exclusion criteria
To be eligible for inclusion, studies met each of the following
inclusion criteria:
● Participants were adult well-trained runners
- Well trained was defined as runners exhibiting a VO2max greater
than 60 mL kg−1 min−1 or a performance score greater than 600 in
the Mercier Scoring Tables (Mercier & Rioux, 1999). (Triathletes were
included if they constituted a small proportion [less than 20%] of the
population sample).
● The interval-training intervention lasted 4 weeks or
longer, with a minimum of two interval sessions per­
formed per week.
● Interval-training interventions consisted of running only.
● Interval-training interventions reported the intensity,
volume and duration of the work and relief periods used
throughout the intervention.
● The intensity of the work interval was greater than the
lactate turn point/second ventilatory threshold (VT2)/max­
imal lactate steady state (MLSS).
● Continuous training during the intervention was below
the LT/first ventilatory threshold (VT1) with the intensity,
volume and duration reported.
● Data on VO2max values pre- and post-training intervention
were reported in addition to one or more of the following
physiological variables: speed at VO2max (sVO2max) speed

at LT (sLT), running economy (RE), peak treadmill speed,
time-trial performance, time to exhaustion (TTE).
● Published in an indexed peer-reviewed journal.
Studies were excluded if any of the following criteria applied:
● Not published in English
● Participants were non-runners (e.g., students, cyclists,
recreationally trained, team-sport athletes, etc.)
● Participants were reported to be in poor health and/or
suffering from any kind of acute or chronic diseases.
● Strength training and/or continuous training above
LT/VT1 were included.
● Interval-training interventions were performed using an
incline and/or hypoxic conditions.
● The characteristics of the interval-training interventions
implemented (e.g., repetition number, intensity used,
relief durations, etc.) were not reported in enough detail
to calculate the training load for comparisons between
studies to be made.
● Ergogenic aids were used as part of the intervention.
Study selection
Figure 1 provides a schematic overview of the study identifi­
cation, screening approach and selection process. Search
results were imported into a published software for systematic
Figure 1. Schematic overview of search, screening approach and selection
process for suitable studies.
JOURNAL OF SPORTS SCIENCES 3
reviews (Ouzzani et al., 2016), allowing a blinded screening
process to be performed by the three independent reviewers
(AP, PH and TJ). Conflicted decisions were resolved through
discussion of the full-text until a consensus was reached. The
initial search yielded 1588 results which increased to 1595
following addition of records identified through authors and
reviewing references known to be relevant. The publication
titles and abstracts remaining after the removal of duplicates
(n = 1018) were screened independently by the reviewers for
eligibility [inter-rater reliability (IRR): 98.4%, Fleiss’ k = 0.48].
Following this screening, 29 potentially eligible studies were
given full consideration, with the full-texts of each reviewed
for inclusion. Of the 29 potentially eligible studies (Figure 1),
a total of 7 studies meeting the criteria remained for further
analysis.
Analysis of results
A quality checklist for Randomized Controlled Trials (RCT) and
Observational Studies modified by Kennelly (Kennelly, 2011)
based on the Downs and Black Methodological Quality check­
list (Downs & Black, 1998) was used to assess quality of the 7
remaining studies. This modified version was used due to the
inclusion of non-randomized and observational studies, as
many of the items relating to the blinding of studies within
the original Downs and Black checklist were therefore not
appropriate. When scoring the quality of the study, the check­
list is split into five sections with a total score provided from
each: Reporting (items 1–12), External Validity (items 13–16),
Internal Validity – bias (items 17–25), Internal Validity – con­
founding (items 26–32), and Power (item 33). Studies are able
to attain a total score of 32 with the total score indicating the
quality (≥20 = good, 15–19 = fair, ≤14 = poor) (Kennelly, 2011).
Data extraction
Data Extraction was performed by one investigator (AP) using
a standardized form to allow the extraction of relevant study
characteristics. The data extracted from each study were:
● Study characteristics (author(s), title, year of publication).
● Participant characteristics (number of participants, age,
stature, body mass, training level). Training intervention
characteristics (duration, frequency, total training time, IT
work intensity, IT work duration, IT relief intensity, IT relief
duration, total number of intervals, CT intensity, CT
duration.
● Total time spent in intensity domains and as a percentage
of the total training time (below VT1/2 mmol·L−1 BLa−,
between VT1/2 mmol·L−1 and VT2/4 mmol·L1 BLa−,
between VT2/4 mmol·L−1 BLa− and sVO2max, and above
sVO2max).
● Training load characteristics (Total intervention training
impulse overall [TOTTRIMP], per week [WTRIMP] and
per session [STRIMP]. Interval-training impulse overall [IT
TOTTRIMP], per week [IT WTRIMP] and per session [IT
STRIMP]).
● Average intervention training intensity scaled to the
sVO2max, interval-training intensity scaled to the sVO2max.
4 A. PARMAR ET AL.

● Training response (initial VO2max, post-training VO2max,

change in VO2max, total TRIMP units to change VO2max
by 1 mL·kg−1·min−1, IT TRIMP units to change VO2max by
1 mL·kg−1·min−1, significance of change).

VO2max values were reported as mL·kg−1·min−1 in all cases,

along with the standard deviation of all variables reported.
Where standard errors were reported, these were converted
into standard deviation. The total training time and time spent
in each intensity domain were reported in hours, minutes and
seconds (h:min:s).

Training load quantification

Training load in each intervention was estimated by calculating
a modified version of the training impulse (TRIMP), commonly
referred to as Lucia’s’ TRIMP (Foster et al., 2001; Lucia et al.,
Figure 2. Forest plot of the pre- to post-VO2max scores in response to the
2003). This method of calculating TRIMP has also been pre­ training intervention implemented in each study, displayed as the bias corrected
viously used to estimate training load in well-trained endur­ hedge’s effect size with 95% confidence intervals.
ance runners (Esteve-Lanao et al., 2007, 2005). The TRIMP score
was calculated by multiplying the accumulated training dura­
square of the pre- and post-group standard deviations. This
tion spent in each intensity domain by an intensity-weighted
version does not specifically include the sample size (n), pre­
multiplier. For example; 1 min in the first intensity domain
venting any complications that could arise from inflating
[<VT1] is given a score of 1 arbitrary unit (AU), 1 min in
n when both group’s means are from the same sample. This
the second intensity domain [>VT1 < VT2] is given a score of 2
statistical approach was chosen due to data being from the
AU, and 1 min in the third intensity domain [>VT2] is given
same sample rather than a separate intervention and control
a score of 3 AU. The total TRIMP score is then obtained by
group, thus making a traditional weighted effects meta-analysis
summing the results of the three intensity domains. The total
pooling inappropriate. Traditional meta-analysis assumes two
TRIMP was calculated for the total duration of a training inter­
different sets of individuals in each group (Higgins, 2011)
vention (TOTTRIMP), the weekly TRIMP of a training intervention
meaning a violation of underlying assumptions would have
(WTRIMP), and the TRIMP of a training session (STRIMP). To further
occurred if applied to this review. 95% confidence intervals
investigate the effects of only the interval-training interven­
were calculated for individual Hedges’ g effect sizes. A forest
tions, the TRIMP of only the interval-training performed in
plot of the individual effect sizes with 95% confidence intervals
each study was calculated for the total training intervention
was created to display the pre-post VO2max responses to the
duration (IT TOTTRIMP), the weekly interval-training performed
training intervention (Figure 2). The qualitative inferences asso­
(WTRIMP), and the TRIMP of an interval-training session (IT STRIMP)
ciated with the calculated effect sizes were defined as trivial
. The calculated TRIMP scores were then expressed relative to
(<0.2), small (0.2–0.6), moderate (0.6–1.2), large (1.2–2.0), very
changes in VO2max to examine if a dose-response relationship
large (2.0–4.0), and nearly perfect (>4.0) (Batterham & Hopkins,
exists between training load and changes in VO2max for all
2006). As the data extracted from the included studies were
training performed throughout an intervention and the effects
from the same sample rather than a separate intervention and
of only the interval-training performed throughout an
control group, it was deemed unsuitable to amalgamate the
intervention.
results for a meta-analysis. The results in this review were
To determine the training intensity in each study and to
therefore analysed narratively.
allow comparisons to be made between studies, intensity was
scaled to the reported average sVO2max, with sVO2max being 1.
For example, if the reported training intensity for the interval Results
work duration was 90% of sVO2max, then the intensity was
Participant characteristics and quality assessment
calculated as 0.9. These scaled intensities were subsequently
used to examine relationships between the training intensities A summary of the participant characteristics and the quality
used in each study and changes in VO2max. assessment results for the studies included in this review are
displayed in Table 1. Seven studies with a total of 62 partici­
pants met the inclusion criteria for this review. One study
Main analysis
included runners with an average VO2max and SD marginally
Effect sizes (Hedges’ g) were calculated for pre- and post- lower than 60 ml·kg−1·min−1 (Ferley et al., 2013) as this group
training VO2max values for each study individually without an (VO2max = 59.4 ± 8.9 ml·kg−1·min−1) displayed no significant
overall pooled effect. Hedges’ g was used to bias correct for the difference to the other training group with a VO2max value
typically small sample sizes, as observational studies were pri­ greater than 60 ml·kg−1·min−1 (63.3 ± 8.0 ml·kg−1·min−1).
marily included with no control groups. The pooled standard Differences in the quality assessment scores were mainly in
deviation for Hedges’ g was calculated using the root mean the external validity and confounding sections of the
Table 1. Participant characteristics (Mean ± SD) and quality assessment results of each study.
Study Participant Characteristics Quality Assessment
Total
Age Stature External Internal Internal Validity – Quality Qualitative
n Sex (years) (cm) Mass (kg) Training level Reporting Validity Validity – Bias Confounding Power Score Inference
Demarle 6 Not 27 ± 5.1 174.2 ± 2.9 68.5 ± 5.4 Endurance trained, middle- and long- 9 1 7 2 0 19 Fair
et al. 2001 Reported distance runners
Slawinski 6 Not 27 ± 4 175 ± 6 72 ± 9 Well-trained runners 9 1 7 2 0 19 Fair
et al. 2001 Reported
Ferley et al. 12 M&F 27.4 ± 3.8 173.6 ± 6.4 66 ± 7.1 Well-trained club runners 12 4 5 5 0 26 Good
2013
Garcin et al. 8 M 20.1 ± 4 176 ± 5.1 62.5 ± 5.1 High-level endurance-trained runners 11 1 7 3 0 22 Good
2002
Bickham 7 M 27 ± 6.9 Not Reported 77.1 ± 9.8 Endurance 10 2 7 2
et al. 2004 trained
distance
runners
0 21 Good
Smith et al. 5 M 22.8 ± 4.5 181 ± 4.7 74.1 ± 3.2
Well-trained middle-distance state-level 9 1 7 2 0 19 Fair
1999 runners
Smith et al. 60% TTE: 9 Not 25.2 ± 0.2 178.8 ± 0.3 72.2 ± 0.3 Well-trained competitive middle-distance, 12 2 5 4 0 23 Good
2003 70% TTE: 9 Reported triathlete, 10 km runners
TTE Time to exhaustion
JOURNAL OF SPORTS SCIENCES
5
6 A. PARMAR ET AL.

assessment. All studies scored 0 as it relates to the calculation sVO2max was prescribed as the work intensity in three studies
of statistical power. (Ferley et al., 2013; Smith et al., 2003, 1999). The highest work
**Table 1** intensity was prescribed as 90–100% maximum effort sprints in
1 study (Bickham & Rossignol, 2004) (Table 3).

Training intervention
A summary of the training characteristics implemented in the
seven studies is displayed in Table 3. In all studies, interval-
training interventions lasted between 4- and 8-weeks, with
sessions performed two to three times per week (Bickham &
Rossignol, 2004; Demarle et al., 2001; Ferley et al., 2013; Garcin
et al., 2002; Slawinski et al., 2001; Smith et al., 2003, 1999).
Supplementary continuous training sessions were included in
six studies 1 to 3 times per week (Demarle et al., 2001; Ferley
et al., 2013; Garcin et al., 2002; Slawinski et al., 2001; Smith et al.,
2003, 1999), with one study not including any continuous
training (Bickham & Rossignol, 2004).
Maximal aerobic power
Only one study (Smith et al., 1999) reported a large, significant
increase of 4.9% in VO2max in response to the training interven­
tion (ES: 1.86, P = 0.007). Non-significant, trivial changes in
VO2max of less than 0.7% (ES: 0.02–0.06) were reported in four
studies (Bickham & Rossignol, 2004; Demarle et al., 2001; Ferley
et al., 2013; Slawinski et al., 2001). Smith et al. (Smith et al., 2003)
reported nearly perfect, non-significant increases in VO2max of
5.9% (ES: 5.33) and 4.2% (ES: 7.02) in the 60% of the time to
exhaustion (TTE) and 70% TTE experimental groups, respectively.
Only one study (Garcin et al., 2002) reported a non-significant,
trivial decrease in VO2max of −0.9% (ES: 0.14) (Table 2).
Continuous training
In all studies including supplementary continuous training,
sessions were performed at 60–75% sVO2max for 30:00–1:00:00,
1 to 3 times per week (Table 3).
Work duration
The longest work durations were prescribed as 50% TTE at s∆50
(Demarle et al., 2001; Garcin et al., 2002; Slawinski et al., 2001).
Work durations were prescribed as 60–75% TTE at sVO2max in
studies using sVO2max as the intensity (Ferley et al., 2013; Smith
et al., 2003, 1999). The shortest work durations were prescribed
as 5–15 s, 90–100% maximum effort sprints over 40–100 m
(Bickham & Rossignol, 2004) (Table 3).
Relief intensity
Static rest was prescribed as the relief intensity in three studies
using sVO2max as the work intensity (Ferley et al., 2013; Smith
et al., 2003, 1999). Active relief periods were prescribed as 50%
sVO2max in the 3 studies using s∆50 as the work intensity
(Demarle et al., 2001; Garcin et al., 2002; Slawinski et al., 2001),
and walking or jogging back to the start of a sprint effort in one
study (Bickham & Rossignol, 2004) (Table 3).
Relief duration
The shortest relief durations of 25% TTE at s∆50 were prescribed
in 3 studies using s∆50 as the work intensity (Demarle et al.,
2001; Garcin et al., 2002; Slawinski et al., 2001). Relief durations
prescribed as the time taken to recover to 65% HRmax were
almost equal to the work durations in one study using sVO2max
as the work intensity (Ferley et al., 2013). The relief duration was
prescribed as a 1:2 work: rest ratio in two studies using sVO2max
as the work intensity (Smith et al., 2003, 1999). In the one study
using sprint efforts (Bickham & Rossignol, 2004), relief durations
were prescribed as progressively declining work: rest ratios
throughout the training intervention of 1:5, 1:4, and 1:3 (Table 3).

Interval training characteristics

Interval-training duration
Work intensity
Repetitions
The lowest intensity for the work period was prescribed as the
median speed between the sLT and sVO2max (s∆50) (Demarle The higher the work intensity the greater number of repeti­
et al., 2001; Garcin et al., 2002; Slawinski et al., 2001). The tions performed per training session and throughout the

Table 2. The VO2max response to the training intervention used along with statistical interpretation is displayed (Mean ± SD).
Study Training response
Initial VO2max Post VO2max % Effect Size 95% CI for Effect Qualitative
(ml∙kg−1∙min−1) (ml∙kg−1∙min−1) change Sig (Hedges’ g) Size inference
Demarle et al. 2001 61.2 ± 6.6 61.6 ± 5.4 +0.65% NS 0.05 −1.08 to 1.18 Trivial
Slawinski et al. 2001 61.2 ± 6 61.6 ± 4.9 +0.65% NS 0.06 −1.07 to 1.19 Trivial
Ferley et al. 2013 59.4 ± 8.9 59.6 ± 7.6 +0.34% NS 0.02 −0.78 to 0.82 Trivial
Garcin et al. 2002 64.8 ± 3.6 64.2 ± 3.8 −0.92% NS −0.14 −1.12 to 0.84 Trivial
Bickham et al. 2004 60.1 ± 3.2 60.3 ± 5.3 +0.33% NS 0.04 −1.01 to 1.09 Trivial
Smith et al. 1999 61.46 ± 1.3 64.45 ± 0.9 +4.86% 0.007 1.86 0.38 to 3.35 Large
Smith et al. 2003 60% TTE 60.5 ± 0.6 64.1 ± 0.6 +5.95% NS 5.33 3.36 to 7.3 Nearly Perfect
group
Smith et al. 2003 70% TTE 60.1 ± 0.2 62.6 ± 0.4 +4.16% NS 7.02 4.55 to 9.49 Nearly Perfect
group
Sig Significance; NS Not Significant (P > 0.05); 95% CI 95% Confidence Interval
Table 3. Training intervention characteristics of each study (Mean ± SD).
Study Training Intervention Characteristics
Total Training
Characteristics Interval Training characteristics Continuous Training characteristics
Work Intensity Relief Intensity Total Duration
Duration Frequency (km∙h−1) Work Duration (s) (km∙h−1) Relief Duration (s) Intervals Intensity (km∙h−1) (mins)
Demarle et al. 2001 8 weeks 2 IT/wk 3 s∆50: 17 ± 0.9 50% TTE s∆50: 294 ± 49 50% sVO2max: 25% TTE s∆50: 147 ± 24 37.2 ± 7.8 60–70% sVO2max: 45–60
CT/wk 9.1 ± 0.2 11.8 ± 0.5
Slawinski et al. 2001 8 weeks 2 IT/wk 3 s∆50: 17 ± 0.7 50% TTE s∆50: 294 ± 45.5 50% sVO2max: 25% TTE s∆50: 147 ± 22.7 53.2 ± 7.9 60–70% sVO2max: 60
CT/wk 9.1 ± 0.3 11.8 ± 0.4
Ferley et al. 2013 6 weeks 2 IT/wk 2 100% sVO2max: 60% TTE sVO2max: 136 ± 36.7 Rest HRR to 65% HRmax: 142 ± 45.4 60 75% sVO2max: 45–60
CT/wk 17.3 ± 2.1 12.9 ± 1.6
Garcin et al. 2002 8 weeks 2 IT/wk 3 s∆50: 19.5 ± 1.0 50% TTE s∆50: 162.5 ± 52 50% sVO2max: 25% TTE s∆50: 81.2 ± 26 64 65–75% sVO2max: 60
CT/wk 10.3 ± 1.0 14.4 ± 1.0
Bickham et al. 2004 6 weeks 3 IT/wk 90–100% max effort 5–15s, 40–100 m Walk or jog Work:Rest – 1:5, 1:4, 1:3 5 mins between 412 None performed None
sprints sets
performed
Smith et al. 1999 4 weeks 2 IT/wk 1 100% sVO2max: 60–75% TTE sVO2max: 135.3 ± 9.96 to Rest Work:Rest – 1:2 60–75% TTE: 270.6 ± 19.92 48 60% sVO2max: 30
CT/wk 20.5 ± 0.39 169.1 ± 12.45 to 338.2 ± 24.9 12.3 ± 0.23
Smith et al. 2003 60% 4 weeks 2 IT/wk 1 100% sVO2max: 60% TTE sVO2max: 133.4 ± 1.4 Rest Work:Rest – 1:2 266.8 ± 2.8 48 60% sVO2max: 30
TTE group CT/wk 19.1 ± 0.13 11.46 ± 0.13
Smith et al. 2003 70% 4 weeks 2 IT/wk 1 100% sVO2max: 70% TTE sVO2max: 154 ± 4.5 Rest Work:Rest = 1:2 308 ± 9 40 60% sVO2max: 30
TTE group CT/wk 19.9 ± 0.4 11.94 ± 0.4
wk week; IT Interval training; CT Continuous training; sVO2max speed at VO2max; s∆50 median speed between the sLT and sVO2max; TTE Time to exhaustion; HRR Heart rate recovery; HRmax age predicted maximum heart rate.
JOURNAL OF SPORTS SCIENCES
7
8 A. PARMAR ET AL.

Figure 3. Panel A: Total duration of training performed in the interventions implemented by each study and the distribution of training within each intensity domain.
Total training duration is displayed at the end of each stacked column in h, min and s (h:min:s). The time spent in each intensity domain is displayed above each
respective section of the stacked column (h:min:s). <VT1: Domain 1 – Below the first ventilatory threshold; > VT2 < sVO2max: Domain 3 – between the second ventilatory
threshold and the velocity at maximal aerobic power; > sVO2max: Domain 4 – above the velocity at maximal aerobic power. Panel B: Total training impulse per week
(WTRIMP) of the training interventions implemented by each study with the total WTRIMP split into the interval training (IT) TRIMP and continuous training (CT) TRIMP per
week. Total WTRIMP is displayed at the end of each stacked column. The IT WTRIMP and CT WTRIMP is displayed above each respective section of the stacked column.

study intervention. In the only study using sprint efforts
(Bickham & Rossignol, 2004) the highest repetitions per
training session (n = 22) and in total (n = 412) were per­
formed over the 6-week intervention. The lowest repetitions
per training session (n = 2–4) and in total (n = 37–64) were
performed in the 3 studies using the lowest work intensity
of s∆50 (Demarle et al., 2001; Garcin et al., 2002; Slawinski
et al., 2001). In the three studies using sVO2max as the work
intensity (Ferley et al., 2013; Smith et al., 2003, 1999), repe­
titions performed per training session (n = 5–6) and in total
(n = 40–60) were similar (Table 3).
Duration
The average training session durations were longest (33:00–­
40:00) in the two 4-week studies using sVO2max as the work
intensity (Smith et al., 2003, 1999), along with the highest
average time spent at the work intensity per session (13:00–­
15:00). Two studies using s∆50 (Garcin et al., 2002; Slawinski
et al., 2001) and the 6-week study using sVO2max as the work
intensities (Ferley et al., 2013) had the shortest average training
session durations (12:00–20:00), with equal average time spent
at the work intensity per session (11:00). The shortest average
time spent at the work intensity per session (4:00) was in the
only study using sprint efforts as the work intensity (Bickham &
Rossignol, 2004), despite a relatively long average training ses­
sion duration of 33:00 (Figure 3A).
The total time at the work intensity increased as the pre­
scribed intensity decreased over the duration of the training
intervention. In the 8-week studies using the lowest work
intensity of s∆50 (Demarle et al., 2001; Garcin et al., 2002;
Slawinski et al., 2001) the total time at the work intensity was
highest (3:00:00–4:20:00). The total time spent at the work
intensity was lowest (1:10:00) in the 1 study using the highest
work intensity of sprint efforts for 6-weeks (Bickham &
Rossignol, 2004) (Figure 3A).
The total training volume was highest (25:30:00–30:30:00) in
the 8-week interventions using s∆50 as the work intensity
(Demarle et al., 2001; Garcin et al., 2002; Slawinski et al., 2001)
and lowest (3:40:00–4:00:00) in the 4-week studies using sVO2max
as the work intensity (Smith et al., 2003, 1999) (Figure 3A).
Training load
Total training impulse per week (WTRIMP)
The total WTRIMP increased as the work intensity prescribed
decreased, with the three studies using the lowest work inten­
sity of s∆50 over the 8-week intervention (Demarle et al., 2001;
Garcin et al., 2002; Slawinski et al., 2001) having the highest
total WTRIMP (237–294 AU·wk−1). In contrast, the 6-week study
with the highest work intensity of sprint efforts (Bickham &
Rossignol, 2004) had the lowest total WTRIMP of 123 AU·wk−1
(Figure 3B).
 JOURNAL OF SPORTS SCIENCES 9

Interval training impulse per week (IT WTRIMP)
The training impulse score per week for only the interval-training
performed (IT WTRIMP) was lowest (67–80 AU·wk−1) in two of the
8-week intervention studies using s∆50 for the work intensity
(Demarle et al., 2001; Garcin et al., 2002), accounting for 30% of
the total WTRIMP. In the two 4-week studies using sVO2max for the
work intensity (Smith et al., 2003, 1999) the IT WTRIMP (118–140
AU·wk−1) was highest, accounting for 80% of the total WTRIMP. The
only study to use sprint efforts for the work intensity (Bickham &
Rossignol, 2004) had an IT WTRIMP of 123 AU·wk−1 accounting for
100% of the total WTRIMP as no continuous training was prescribed
as part of the training intervention (Figure 3B).
Continuous training impulse per week (CT WTRIMP)
The training impulse score per week for the continuous training
only (CT WTRIMP) was lowest and identical (30 AU·wk−1) in the
two 4-week studies using sVO2max for the work intensity (Smith
et al., 2003, 1999), with a CT TRIMP: IT TRIMP ratio of approxi­
mately 1:4. The 8-week studies using s∆50 for the work intensity
(Demarle et al., 2001; Garcin et al., 2002; Slawinski et al., 2001)
had the highest CT WTRIMP (157–180 AU·wk−1), with a CT TRIMP:
IT TRIMP ratio of approximately 2:1 (Figure 3B).
Dose-response relationship
Training distribution and change in VO2max
Improvements in VO2max decreased as the percentage of total
training time spent <VT1 increased (Figure 4A). In contrast,
improvements in VO2max increased as the percentage of total
training time spent >VT2 increased (Figure 4B). The greatest
improvements in VO2max (2.5–3.6 mL·kg−1·min−1) were reported
in 2 studies (Smith et al., 2003, 1999) with the highest percen­
tage of total training time >VT2 (46.1% – 49.9%). By contrast,
a decrease in VO2max of −0.6 mL·kg−1·min−1 was reported in one
study (Garcin et al., 2002) with the lowest percentage of total
training time >VT2 (10.2%).
Training intensity and change in VO2max
Improvements in VO2max decreased as the average total train­
ing intensity increased (Figure 5 A1); however, if the sprint
interval study is removed (Bickham & Rossignol, 2004) (the
only study to prescribe a work intensity above sVO2max),
improvements in VO2max increased as the average total training
intensity increased (Figure 5 A2). Similarly, improvements in
VO2max decreased as the work intensity prescribed increased
(Figure 5 B1). After removing the sprint interval study (Bickham
& Rossignol, 2004), improvements in VO2max increased as the
work intensity increased (Figure 5 B2). When using aerobic
intensities ≤sVO2max, greater improvements in VO2max are
shown with higher average total training intensities and higher
interval-training intensities up to sVO2max (Figures 5A2 & 5B2).
Total training load and change in VO2max
As the TOTTRIMP and total WTRIMP performed during the training
interventions increased, improvements in VO2max decreased
(Figure 6A & 6C). By contrast, when displayed as the total
TRIMP per training session (STRIMP), improvements in VO2max
increased as the total STRIMP increased (Figure 6D). The total
TRIMP required to change VO2max by 1 mL·kg−1·min−1
(TRIMP·mL·kg−1·min−1) showed no clear relationship with the
average total training intensity used throughout the training
interventions (Figure 6B).
Interval-training load and change in VO2max
As the IT TRIMP performed during the training interventions
increased, improvements in VO2max decreased (Figure 7A).
When displayed as the IT WTRIMP and IT TRIMP per training
session (IT STRIMP) however, improvements in VO2max increased
as the IT WTRIMP and IT STRIMP increased (Figure 7C & 7D). The IT
TRIMP required to change VO2max by 1 mL·kg−1·min−1 (IT
TRIMP·mL·kg−1·min−1) increased with higher interval-training
intensities (Figure 7B).

Figure 4. Panel A: The relationship between the percentage of total training time spent below VT1 and the change in VO2max (∆ VO2max) in the training intervention
implemented in each study. Linear regression line, regression equation and R2 value displayed for the relationship between the percentage of training time below VT1
and the ∆ VO2max. *denotes statistically significant change in VO2max. Panel B: The relationship between the percentage of total training time spent above VT2 and the
change in VO2max (∆ VO2max) in the training intervention implemented in each study. Linear regression line, regression equation and R2 value displayed for the
relationship between the percentage of training time above VT2. *denotes statistically significant change in VO2max. Legend with coloured boxes relates to all panels to
signify each study.
10 A. PARMAR ET AL.

Figure 5. Panel A1: The relationship between the average intensity of the total training intervention implemented in each study and the change in VO2max (∆ VO2max).
Average intervention intensity is calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1. Linear regression line, regression equation and R2 value
displayed for the relationship between the average intervention intensity and the ∆ VO2max. *denotes statistically significant change in VO2max. Panel B1: The
relationship between the prescribed intensity of the interval training work period implemented in each study and the change in VO2max (∆ VO2max). Interval intensity is
calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1. Linear regression line, regression equation and R2 value displayed for the relationship
between the interval intensity and the ∆ VO2max. *denotes statistically significant change in VO2max. Panel A2: The relationship between the average intensity of the
total training intervention implemented in each study and the change in VO2max (∆ VO2max) excluding data from Bickham & Le Rossignol (2004). Average intervention
intensity is calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1. Linear regression line, regression equation and R2 value displayed for the
relationship between the average intervention intensity and the ∆ VO2max. *denotes statistically significant change in VO2max. Panel B2: The relationship between the
prescribed intensity of the interval training work period implemented in each study and the change in VO2max (∆ VO2max) excluding data from Bickham & Le Rossignol
(2004). Interval intensity is calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1. Linear regression line, regression equation and R2 value displayed
for the relationship between the interval intensity and the ∆ VO2max. *denotes statistically significant change in VO2max. Legend with coloured boxes relates to all panels
to signify each study.

Discussion difficulties associated with overcoming the reluctance of this

This systematic review aimed to provide an updated evaluation of
current evidence investigating the efficacy of interval-training
interventions to improve VO2max in well-trained, middle- and long-
distance runners. A further aim was to examine if a dose-response
relationship existed between the interval-training load and
changes in VO2max in a well-trained population. Empirical evidence
to support the efficacy of interval-training to improve VO2max in
well-trained runners remains in the same inconclusive state as the
review previously conducted by Midgley (Midgley et al., 2007).
A dose-response relationship might be evident, with a higher
total load and individual session load correlating with greater
increases in VO2max. Furthermore, intensities up to, but not exceed­
ing sVO2max, performed for long durations (>2 min) provide the
greatest stimulus for improvements in VO2max. These findings,
however, should be interpreted with caution due to the limited
number of included studies with small sample sizes and relatively
short training interventions (4–8 weeks). This might be due to the
population to alter their training regime for extended periods
(Hawley et al., 1997). The quality of the studies included within
this review is considered acceptable based on the quality checklist
for RCT and Observational studies modified by Kennelly (Kennelly,
2011), with all studies displaying fair to good quality assessment
scores (n = 19–26).
Training intensity distribution
The data analysed here indicate that interval-training interventions
have little effect on changes in VO2max in well-trained runners,
contrary to previous findings reporting significant improvements
in VO2max, albeit in lesser-trained populations (Milanović et al.,
2015; Bacon et al., 2013; LinkManagerBM_REF_9mpUW2jeWen
et al., 2019). Nevertheless, the available data suggests increasing
the percentage of total training time performed above VT2 in
comparison to the percentage of training time performed below
 JOURNAL OF SPORTS SCIENCES 11

Figure 6. Panel A: The relationship between the total training impulse (TRIMP) of the training intervention implemented in each study and the change in VO2max (∆
VO2max). The total TRIMP is the accumulated TRIMP of both interval and continuous training sessions completed throughout the training intervention. Linear regression
line, regression equation and R2 value displayed for the relationship between the TRIMP and the change in VO2max. *denotes statistically significant change in VO2max.
Panel B: The relationship between the average intensity of the total training intervention implemented in each study and the total training impulse units required to
change VO2max by 1 ml−1·kg−1·min−1 (TRIMP·mL·kg−1·min−1). Average intervention intensity is calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1.
Linear regression line, regression equation and R2 value displayed for the relationship between the average intervention intensity and the TRIMP·ml−1·kg−1·min−1.
*denotes statistically significant change in VO2max. Panel C: The relationship between the total training impulse per week (Total WTRIMP) and the change in VO2max (∆
VO2max). Total WTRIMP is the accumulated TRIMP of both interval and continuous training sessions completed per week. Linear regression line, regression equation and
R2 value displayed for the relationship between the WTRIMP and the change in VO2max. *denotes statistically significant change in VO2max. Panel D: The relationship
between the total training impulse per session (Total STRIMP) and the change in VO2max (∆ VO2max). Linear regression line, regression equation and R2 value displayed for
the relationship between the STRIMP and the change in VO2max. *denotes statistically significant change in VO2max. Legend with coloured boxes relates to all panels to
signify each study.

VT1 elicits greater improvements in VO2max, but only in short-term
training interventions (<8 weeks). Supporting the inclusion of high
proportions of training performed at high intensities within
a micro-cycle, top-class marathon runners have been shown to
train at relatively higher velocities for more total kilometres per
week than lower-level competitors, and exhibit significantly higher
VO2max values (Billat et al., 2001). Recent analyses have also
reported the volume of training performed above VT2 to be
a predictor of world-class endurance running performance, with
well-trained endurance runners performing high volumes of long
intervals slightly above VT2 (Casado et al., 2019; Tjelta, 2019). This
conflicts the training intensity distribution (TID) of well-trained
endurance athletes who reportedly perform the majority of their
training below VT1 (approximately 80%), in comparison to the
training performed above VT2 (Billat, 2001; Billat et al., 2001;
Seiler, 2009; KS Seiler & Kjerland, 2006; Stöggl & Sperlich, 2015,
2019; Tjelta, 2016). Previous observations of TID have been
reported over longer training durations (8 weeks to a season)
(Seiler, 2009; KS Seiler & Kjerland, 2006; Stöggl & Sperlich, 2015,
2019; Tjelta, 2016) than the study interventions presented here
(4–8 weeks) (Demarle et al., 2001; Ferley et al., 2013; Garcin et al.,
2002; Slawinski et al., 2001; Smith et al., 2003, 1999) therefore,
increased training time above VT2 might be indicative of greater
VO2max improvements, but only in short-term training interven­
tions (<8 weeks). It is likely sustained durations of increased train­
ing intensity (>8 weeks) are not conducive to performance
improvements, particularly due to the high mechanical and neu­
romuscular demands associated with running at high velocities
(Papacosta & Gleeson, 2013; Stöggl & Sperlich, 2015).
Appropriately periodizing training intensity is therefore necessary,
with polarized TID models shown to elicit greater performance
improvements without inducing signs of overtraining, even with
an equal TRIMP, than other TID models (Angeli et al., 2004; Billat
et al., 1999; Esteve-Lanao et al., 2007; Seiler, 2010; Stöggl & Sperlich,
2015; T Stöggl & Sperlich, 2014).
Interval intensity and duration
The results of this review suggest short-term training interven­
tions (4–6 weeks) with work intervals at sVO2max, sustained for
12 A. PARMAR ET AL.

Figure 7. Panel A: The relationship between the interval training impulse (IT TRIMP) implemented in each study and the change in VO2max (∆ VO2max). The IT TRIMP is
the TRIMP of only the interval training sessions completed throughout the training intervention. Linear regression line, regression equation and R2 value displayed for
the relationship between the IT TRIMP and the change in VO2max. *denotes statistically significant change in VO2max. Panel B: The relationship between the prescribed
intensity of the interval training work period implemented in each study and the interval training impulse units required to change VO2max by 1 ml−1·kg−1·min−1 (IT
TRIMP·mL·kg−1·min−1). Interval intensity is calculated scaled to the speed at VO2max (sVO2max) with sVO2max being 1. Linear regression line, regression equation and R2
value displayed for the relationship between the interval intensity and the IT TRIMP·mL·kg−1·min−1. *denotes statistically significant change in VO2max. Panel C: The
relationship between the interval training impulse per week (IT WTRIMP) and the change in VO2max (∆ VO2max). IT WTRIMP is the TRIMP of only the interval training sessions
completed per week. Linear regression line, regression equation and R2 value displayed for the relationship between the IT WTRIMP and the change in VO2max. *denotes
statistically significant change in VO2max. Panel D: The relationship between the interval training impulse per session (IT STRIMP) and the change in VO2max (∆ VO2max).
Linear regression line, regression equation and R2 value displayed for the relationship between the IT STRIMP and the change in VO2max. *denotes statistically significant
change in VO2max. Legend with coloured boxes relates to all panels to signify each study.

long durations (>2 min) are the most effective in improving
VO2max in well-trained runners, so long as the intensity allows
a sustained T@VO2max to create a strong enough stimulus to
elicit VO2max improvements. Accumulating T@VO2max in well-
trained runners appears necessary as the studies that accumu­
lated 13–16 min of running at sVO2max per training session
reported the greatest VO2max improvements (Smith et al.,
2003, 1999). In contrast, lesser VO2max improvements were
reported in the studies that accumulated running times
per session of 10–16 min at s∆50 (Demarle et al., 2001; Garcin
et al., 2002; Slawinski et al., 2001), 9–12 min at sVO2max (Ferley
et al., 2013), and 4-min as 90–100% maximum sprint efforts
(Bickham & Rossignol, 2004). This supports previous recom­
mendations that interval-training protocols accumulating
≥15 min of T@VO2max per session are optimal to maximally
stress cardiorespiratory parameters for VO2max improvements
in well-trained runners (Buchheit & Laursen, 2013; Midgley &
McNaughton, 2006; Wen et al., 2019). Selecting appropriate
intensities and durations to maximize T@VO2max is highlighted
by the contrasting improvements in VO2max shown in the stu­
dies using sVO2max as the work intensity (Ferley et al., 2013;
Smith et al., 2003, 1999). Greater improvements in VO2max were
reported in the two studies that accumulated more time (13–­
16 min) running at sVO2max per training session (Smith et al.,
2003, 1999) than the study accumulating less time per training
session (9–12 min) (Ferley et al., 2013), despite using the same
work intensity of sVO2max.
The lack of improvements in VO2max reported in the studies
using s∆50 as the work intensity (Demarle et al., 2001; Garcin
et al., 2002; Slawinski et al., 2001) suggests the duration and
intensity might have been too low to evoke a maximal VO2
response and in turn stimulate improvements in VO2max. The
use of s∆50 to elicit a maximal VO2 response relies on the
presence of a VO2 slow component (Billat et al., 1998), with
Billat et al. (Billat et al., 2000) previously showing the time to
reach VO2max to be approximately 5-min when running at s∆50.
Moreover, well-trained runners have been shown to maintain
a plateau in VO2 below VO2max during exhaustive runs at s∆50
(Billat et al., 1998). These data indicate s∆50 is insufficient to
provide the T@VO2max required to maximally stress cardiore­
spiratory parameters in well-trained runners and in turn
improve VO2max (Buchheit & Laursen, 2013; Midgley &
McNaughton, 2006; Midgley et al., 2007). Conversely, the one
study using maximal sprint efforts reported a trivial increase in

VO2max (Bickham & Rossignol, 2004), indicating this to be too
high an intensity to allow a maximal VO2 response to be
sustained to stimulate improvements in VO2max. Such high
intensities above sVO2max limit the duration of the interval
and in turn the ability to attain and sustain a maximal VO2
response, due to the greater contribution from anaerobic meta­
bolism increasing the intramuscular accumulation of lactate
and hydrogen ion production associated with fatigue
(Buchheit & Laursen, 2013; Midgley et al., 2006).
The use of long duration interval repetitions (≥2 min) per­
formed at 100% sVO2max, accumulating ≥15 min T@VO2max per
training session is therefore suggested to be optimal to maxi­
mally stress cardiorespiratory parameters in well-trained endur­
ance runners. It should be noted that despite the data
suggesting this to be optimal, VO2max improvements were non-
significant in the included studies; perhaps due to the small
sample sizes increasing the probability of type 2 errors when
using probability-based statistics (Nevill et al., 2007).
Nevertheless, based on current evidence the efficacy of inter­
val-training to improve VO2max in well-trained runners remains
equivocal.
Dose – response relationship
Total training load and changes in VO2max
In this study, data from the original studies have been used to
calculate a TRIMP to evaluate the relationship between the
TOTTRIMP and changes in VO2max. This novel approach provides
a different interpretation of this relationship that has not been
presented in previous reviews. The present data show dimin­
ishing increases in VO2max as the TOTTRIMP increases, both
throughout the duration of an intervention and when
expressed as the total TRIMP per week (WTRIMP). However,
when the TOTTRIMP is expressed per training session (STRIMP),
greater STRIMP elicited greater increases in VO2max.
The diminishing increases in VO2max with increasing
TOTTRIMP and WTRIMP follow the relationship displayed between
the total training time <VT1 and changes in VO2max. This sug­
gests the high TOTTRIMP values and decreasing VO2max improve­
ments displayed are a result of the inclusion of high
proportions of CT <VT1. Performing such high proportions of CT
<VT1 reduces the average intervention intensity due to
a potentially excessive accumulation of fatigue affecting the
ability to perform and recover from IT sessions for improve­
ments in cardiorespiratory parameters mediating VO2max to
occur. This contrasts previous observations and reports on TID
in well-trained endurance athletes showing greater volumes of
training performed <VT1, with very little performed >VT2; how­
ever, these observations have been reported over much longer
training durations (Esteve-Lanao et al., 2007; Kenneally et al.,
2018; Seiler, 2009; KS Seiler & Kjerland, 2006; Stöggl & Sperlich,
2015, 2019).
The greater increases in VO2max with increasing STRIMP dis­
played here is in line with the reportedly rapid effects of interval
training on physiology and performance, albeit with rapid pla­
teau effects as well (Seiler, 2009). This perhaps indicates
improvements in VO2max are elicited in the initial 4–8 weeks
of interval-training interventions but only with high STRIMP in
JOURNAL OF SPORTS SCIENCES 13
well-trained runners, supporting suggestions that intensities
greater than s∆50 are required to improve VO2max in well-
trained populations (Buchheit & Laursen, 2013; Laursen &
Jenkins, 2002; Londeree, 1997; Midgley et al., 2007). Extending
to longer training interventions might require lower STRIMP due
to the accumulated load of repeatedly performing high-
intensity interval-training sessions, in addition to the potential
de-training of other aspects of performance (Mujika & Padilla,
2000a; I Mujika & SPadilla, 2000b), especially in well-trained
runners performing high volumes of training (Billat et al.,
2001; Casado et al., 2019; Kenneally et al., 2018). In support of
this, the inclusion of 10% of the weekly running volume con­
stituting intense interval training at 3 km and 10 km race pace
over an 8-week period significantly improved VO2max in well-
trained marathon runners (Billat et al., 2002). The inclusion of
this interval-training however was accompanied by a decrease
in the total weekly running volume to 90% of that performed
previously. This indicates that in well-trained runners, increases
in training intensity can potentially elicit improvements in
VO2max using short-term interventions. The total volume of
training might need to be reduced to allow high intensities to
be reached and to recover from the demands of these sessions,
thereby reducing the TOTTRIMP but maintaining or even increas­
ing the STRIMP, as indicated by the present findings. Periodizing
short-term, higher intensity training blocks into the training
regime are therefore recommended to ensure continued per­
formance improvements without compromising recovery and
performance in the long-term.
Interestingly, the TRIMP units required to change VO2max
(TRIMP·mL·kg−1·min−1) displayed little relationship with the
average intervention intensity. This might be due to methodo­
logical differences of the interventions implemented in the
included studies (total duration and inclusion of continuous
training); with longer interventions displaying higher
TOTTRIMP, along with high volumes of CT increasing TOTTRIMP
further without improvements in VO2max. Despite this, it
appears increases in average intervention training intensity
reduce TRIMP·mL·kg−1·min−1, but only up to sVO2max. This sup­
ports previous recommendations regarding increasing
T@VO2max by using 100% sVO2max as the interval intensity for
long durations (≥2-min) as this also maximizes the total STRIMP
(Buchheit & Laursen, 2013; Midgley & McNaughton, 2006; Wen
et al., 2019). In addition, decreasing the total volume of training
performed as CT below VT1 reduces the TOTTRIMP and the
TRIMP·mL·kg−1·min−1 in the short-term, perhaps providing
greater recovery from IT sessions. This might therefore allow
higher intensities to be sustained for longer durations during IT
sessions, increasing the total average intervention intensity,
T@VO2max, and in turn improvements in VO2max.
Interval-training load and changes in VO2max
The IT TIRIMP (IT TOTTRIMP) performed throughout the study
interventions shows diminishing improvements in VO2max as
the IT TOTTRIMP increased. This is likely the result of the shorter
interventions being more intense with less CT, whereas the
longer interventions were less intense with more CT. To over­
come this, the IT TRIMP per week (IT WTRIMP) and the IT TRIMP
per session (IT STRIMP) were calculated, both showing greater
14 A. PARMAR ET AL.
improvements in VO2max as the IT WTRIMP and IT STRIMP
increased. This further supports the notion that high training
loads accumulated through high-intensity interval-training ses­
sions are required to improve VO2max in well-trained runners.
This is due to the ability to increase T@VO2max and thereby
maximally stress cardiorespiratory parameters (Buchheit &
Laursen, 2013; Laursen & Jenkins, 2002; Midgley &
McNaughton, 2006; Midgley et al., 2006). In line with this, the
present data indicate the IT WTRIMP and IT STRIMP are of greater
importance in eliciting VO2max improvements than the total
WTRIMP and STRIMP. This suggests the intensity during interval
sessions to be the likely key driver of VO2max adaptations, more
so than the overall training load in short-term training inter­
ventions. It is recognized that VO2max is predominantly limited
by maximal cardiac output (Qmax) (DR Bassett & Howley, 2000;
JDR Bassett & Howley, 1997). Changes in Qmax are primarily due
to increases in stroke volume (SV), as HRmax displays little
change in response to training, along with previous observa­
tions even reporting small decreases in HRmax (Clausen, 1977;
Cooper, 1997; Støren et al., 2017; Zavorsky, 2000). Increases in
SV occur due to a mechanical overload of the ventricles
through the filling and ejection of blood stimulating hypertro­
phy of the myocardium, with maximum filling pressure, and
therefore maximum mechanical overload, occurring at VO2max
(Clausen, 1977; Cooper, 1997; Vega et al., 2017). This further
supports the use of intensities at, or close to, VO2max to provide
the mechanical overload necessary to elicit improvements in
Qmax, and in turn VO2max.
Greater improvements in VO2max with greater IT STRIMP were
observed here, in well-trained runners. Studies using 100%
sVO2max as the interval intensity displayed greater IT STRIMP
and improvements in VO2max than studies using s∆50 as the
intensity. This suggests short-term training interventions aim­
ing to improve VO2max should increase the IT TOTTRIMP by
increasing the time spent running at 100% sVO2max during IT
sessions to provide the optimal stimulus for adaptation. This
should be incorporated into the training regime however, by
increasing both IT STRIMP rather than increasing the weekly
volume of IT, and reducing the total running volume to allow
recovery and reduce the risk of overtraining. While the impor­
tance of high total running volumes in well-trained distance
runners is not to be dismissed (Casado et al., 2019; Kenneally
et al., 2018), when the focus of training is to improve VO2max,
the intensity of training is the predominant factor regulating
improvements that can be optimized through interval-training.
Supporting this, in well-trained endurance athletes, maximal
(and near maximal) intensities (90–100% VO2max) have been
suggested to be optimal in eliciting maximal SV values, increas­
ing the time spent at Qmax (T@Qmax) and therefore the cardio­
pulmonary stress (Buchheit & Laursen, 2013; Midgley &
McNaughton, 2006). This is in line with the optimal intensity
suggested to increase T@VO2max during interval-training ses­
sions; however, Seiler et al. (Seiler et al., 2013) showed that an
interval protocol enabling 90% HRmax to be sustained for 32-
min induced greater increases in cardiorespiratory parameters
than an interval protocol enabling 95% HRmax to be sustained
for 16-min. While SV and Q were not measured, this raises the
question as to whether slightly lower intensities may be opti­
mal in increasing T@Qmax, rather than higher intensities that
increase T@VO2max. These findings were reported in recrea­
tional cyclists, therefore the intensity and mode of exercise
might differ in well-trained runners (Seiler et al., 2013). By
contrast, maximal SV and Q values were elicited following an
exhaustive exercise bout at 100% power at VO2max (pVO2max),
whereas an exhaustive bout at the power at ∆50 (p∆50) was
unable to elicit maximal SV and Q values in well-trained triath­
letes (VO2max = 64 mL·kg−1·min−1) (Lepretre et al., 2004). Even
though this was shown during exhaustive cycling, it is specu­
lated that higher intensities are able to elicit maximal SV and
Q responses in well-trained runners; thereby increasing T@Qmax
and in turn T@VO2max, optimizing the overall stimulus for
VO2max improvements. The use of IT methods appears to be
effective in enabling intensities eliciting Qmax to be attained
and sustained; thus providing the stimulus required to further
improve VO2max in well-trained runners.
In the studies using work intensities at 100% sVO2max per­
formed for long durations (>2 mins), increases in VO2max per IT
TRIMP unit are greater than in studies using work intensities
above (sprint efforts) and below (∆50) sVO2max. This suggests
interval intensities of 100% sVO2max performed for long dura­
tions (>2 mins) to be an optimal protocol to increase VO2max in
well-trained runners, with the lowest IT TRIMP units required to
improve VO2max (IT TRIMP·mL·kg−1·min−1). In contrast, the cur­
rent data show work intensities below (s∆50) and above (sprint
efforts) sVO2max would require more IT TRIMP·mL·kg−1·min−1 for
the same increase in VO2max. This indicates that higher volumes
of interval-training would need to be performed either during
individual training sessions or by including a greater frequency
of interval-training sessions. Increasing the IT TRIMP by increas­
ing the volume of running, however, increases the risk of over­
training and injury, particularly due to high frequencies of
running at high velocities limiting the recovery time available
between training bouts (Billat et al., 1999; T Stöggl & Sperlich,
2014). Furthermore, as evidence suggests s∆50 to be an inten­
sity unable to evoke maximal VO2 and Q responses in well-
trained runners (Billat et al., 1998; Lepretre et al., 2004), it is
speculated these sessions may be more characteristic of
‘threshold’ training where long durations of running are per­
formed close to, or slightly above the maximal steady state,
accumulating blood lactate concentrations of approximately
2–4 mmol·L−1 (KS Seiler & Kjerland, 2006; Tjelta, 2016). Well-
trained runners typically exhibit high fractional utilization in
addition to a high VO2max allowing high but sub-maximal velo­
cities, such as s∆50, to be sustained (Billat et al., 1998, 2001,
2000; Demarie et al., 2000); with well-trained endurance run­
ners running at s∆50 reportedly sustaining blood lactate con­
centrations comparable to that achieved during a 10-km race
(Billat, 2001; Billat et al., 1998, 2000). Greater volumes of run­
ning at s∆50 to increase the IT TRIMP would lead to well-trained
runners performing higher volumes of “threshold” training;
however, this exerts a greater demand on the autonomic ner­
vous system (Seiler et al., 2007; T Stöggl & Sperlich, 2014),
endocrine system (Kenneally et al., 2018; Muñoz et al., 2014),
and carbohydrate fuelling (Beneke et al., 2011; Stöggl &
Sperlich, 2015). This in turn restricts training time due to an
increased recovery time and limited glycogen storing, increas­
ing the risk of overtraining without offering further VO2max
improvements. This further supports the need to maximize

the cardiorespiratory demand during training interventions to
provide the optimal stimulus required to improve VO2max in
well-trained runners, without increasing the risk of overtraining
and injury during such short-term interventions. It is therefore
suggested to utilize 100% sVO2max for repetition durations
>2-min, accumulating >15-min of total work at this intensity
to provide the greatest IT STRIMP, thereby maximizing T@VO2max
and T@Qmax whilst reducing the IT TRIMP·mL·kg−1·min−1 in
well-trained runners.
Limitations
The limited number of studies included in this review highlights
the lack of training studies in well-trained runners to establish
the efficacy of interval-training, making it difficult to provide
valid training recommendations in this population. This is likely
due to the reluctance of well-trained runners to modify their
training programmes for a significant period (Hawley et al.,
1997). Nevertheless, conclusions based on a systematic analysis
of the literature can be made, but interpretations should be
made within the limitations of this review. Furthermore, all
recommendations and analyses presented within this review
are only applicable to endurance running therefore caution is
advised if these results are attempted to be applied to other
endurance disciplines. For all studies included in the review,
interpretations of the data are based predominantly upon
probability values, however, these can be misleading due to
the low sample sizes and the heterogeneity in the pool of
participants studied. In addition, the recommendations and
conclusions presented on how to periodize interval-training
into the training regime are limited to the results and analyses
of the short-term interventions (4–8 weeks) of the studies
included in this review, with the absence of longer-term inter­
vention studies not allowing comparisons to be made as to
whether longer intervention durations elicit inferior or superior
adaptations. A further limitation is related to the calculation of
the training load within the included studies as no studies
reported HR data during training sessions. Previous work sup­
ports the approach used here to calculate training load (Sylta
et al., 2014), however it is acknowledged that the calculation of
training load and demarcation of training zones is based upon
the intended prescription of the training interventions, which
might differ from the actual physiological response evoked
during the sessions. Arguably, the greatest limitation when
interpreting the conclusions made is the lack of studies includ­
ing more than one experimental group. This makes evaluating
the efficacy of interval-training to improve VO2max difficult due
to the lack of a control group to compare training effects. It is
acknowledged however, that including control groups within
this population presents challenges due to well-trained runners
potentially being unwilling to participate in a reduced training
volume or intensity for a training block. Finally, the ecological
validity of the training interventions implemented within the
included studies is debatable, with evidence from a survey
under review showing the interval sessions completed by run­
ners differ from those implemented in study interventions
(Parmar et al., 2020). Consequently, the lack of research con­
ducted on well-trained runners along with the questionable
ecological validity, mean the conclusions drawn as it pertains
JOURNAL OF SPORTS SCIENCES 15
to the efficacy of interval-training to improve VO2max might be
misleading. This is due to the interval-training methods pre­
scribed in practice potentially resulting in different physiologi­
cal responses compared to those reported in research.
Practical applications
The novel TRIMP quantification presented in this review dis­
plays a relationship between a high IT TRIMP per session (IT
STRIMP) and VO2max improvements that coaches of well-trained
runners can implement to optimize training adaptations during
focussed interval-training blocks aiming to improve VO2max.
The data presented herein suggest performing 2 to 3 interval-
training sessions per week, at a work intensity of 100% sVO2max
for repetitions >2 min, accumulating >15 min of total work
per session at this intensity to optimally accumulate a high IT
STRIMP. Such protocols appear to maximize the T@VO2max and
T@Qmax, thereby reducing the total interval-training TRIMP
required to improve VO2max. The total running volume per­
formed below VT1 should be decreased in already well-
trained runners during such periods of training intensification
in the short-term (≤8 weeks) to appropriately balance training
stress, recovery and adaptation. When the aim is to improve
VO2max, the current data indicate short-term training blocks
(≤8 weeks) prioritizing the intensity of interval-training sessions
and reducing the total volume of running are an effective
means of improving VO2max. High-intensity training focussed
on improving VO2max should be periodized appropriately into
the training regime over prolonged periods (>8 weeks) due to
the stress associated with high-intensity training, with
a polarized TID model appearing to be effective in eliciting
training adaptations without inducing signs of overtraining.
Such short-term blocks of training intensification might not
display the typical polarized TID observed in the long term,
however, the reduction in running volume to accommodate
the increase in intensity appears to be optimal for interval-
training sessions to be performed at the necessary intensity,
whilst also allowing for appropriate recovery and adaptation.
Conclusions and future research
In conclusion, the available evidence is insufficient to unequi­
vocally support the efficacy of interval-training to improve
VO2max in well-trained runners. The novel method of analysis
used here for the first time has quantified the IT across different
studies and has shown that a dose-response relationship
appears to be evident, perhaps providing a guide as to the
minimal dose required to improve VO2max. The lack of research
supporting the use of interval-training in well-trained runners
warrants further research to elucidate the effectiveness of these
training methods, with particular reference to the training
intensity and total training load required to improve VO2max.
Furthermore, future research should aim to include interval-
training protocols performed by well-trained runners with
more than one experimental group to allow confident conclu­
sions to be drawn, and in turn provide worthwhile recommen­
dations to researchers and practitioners alike.
This review provides an updated systematic analysis of the
literature to date regarding the use of interval-training
16 A. PARMAR ET AL.
methods to improve VO2max in well-trained middle- to long-
distance runners. The dose-response relationship evident
between the IT STRIMP and changes in VO2max suggests this
novel method to be a useful metric coaches can implement
to optimize the interval-training protocols prescribed. Evidence
from this review suggests interval-training interventions per­
formed at 100% sVO2max for repetitions >2 min, accumulating
>15 min of total work per session at this intensity maximize the
T@VO2max and T@Qmax, reducing the interval-training load
required to improve VO2max in well-trained runners.
Acknowledgments
Not applicable
Disclosure statement
No potential conflict of interest was reported by the authors.
Disclosure of Interest
The authors report no conflicts of interest.
Funding
The authors have no funding to report.
ORCID
Arran Parmar http://orcid.org/0000-0001-6780-3922
Thomas W. Jones http://orcid.org/0000-0003-2263-5261
Philip, R. Hayes http://orcid.org/0000-0002-4197-2848
References
Angeli, A., Minetto, M., Dovio, A., & Paccotti, P. (2004). The overtraining
syndrome in athletes: A stress-related disorder. Journal of
Endocrinological Investigation, 27(6), 603–612. https://doi.org/10.1007/
BF03347487
Astorino, T. A., Edmunds, R. M., Clark, A., King, L., Gallant, R. A., Namm, S.,
FISCHER, A., & WOOD, K. M. (2017). High-intensity interval training
increases cardiac output and VO2max. Medicine & Science in Sports &
Exercise, 49(2), 265–273. https://doi.org/10.1249/MSS.
0000000000001099
ÅStrand, I., ÅStrand, P., Christensen, E. H., & Hedman, R. (1960). Intermittent
muscular work. Acta Physiologica Scandinavica, 48(3–4), 448–453.
https://doi.org/10.1111/j.1748-1716.1960.tb01879.x
Bacon, A. P., Carter, R. E., Ogle, E. A., & Joyner, M. J. (2013). VO2max
trainability and high intensity interval training in humans: A
meta-analysis. PLoS One, 8(9), e73182. https://doi.org/10.1371/journal.
pone.0073182
Bassett, D. R., & Howley, E. T. (2000). Limiting factors for maximum oxygen
uptake and determinants of endurance performance. Medicine & Science
in Sports & Exercise, 32(1), 70–84. https://doi.org/10.1097/00005768-
200001000-00012
Bassett, J. D. R., & Howley, E. T. (1997). Maximal oxygen uptake: “classical”
versus “contemporary” viewpoints. Medicine &amp Science in Sports
&amp Exercise, 29(5), 591–603. https://doi.org/10.1097/00005768-
199705000-00002
Batterham, A. M., & Hopkins, W. G. (2006, March). Making meaningful
inferences about magnitudes. International Journal of Sports Physiology
and Performance, 1(1), 50–57. https://doi.org/10.1123/ijspp.1.1.50
Beneke, R., Leithäuser, R. M., & Ochentel, O. (2011). Blood lactate diagnostics
in exercise testing and training. International Journal of Sports Physiology
and Performance, 6(1), 8–24. https://doi.org/10.1123/ijspp.6.1.8
Bickham, D., & Rossignol, P. (2004). Effects of high-intensity interval training
on the accumulated oxygen deficit of endurance-trained runners. The
Journal of Exercise Physiology Online, 7(1), 40–47. https://www.research­
gate.net/profile/Dale_Bickham/publication/
228500219_Effects_of_high-intensity_interval_training_on_the_accu­
mulated_oxygen_deficit_of_endurance-trained_runners/links/
0deec52c3b3594be7f000000.pdf
Billat, V., Binsse, V., Petit, B., & Koralsztein, J. J. P. (1998). High level runners
are able to maintain a VO2 steady-state below VO2max in an all-out run
over their critical velocity. Archives of Physiology and Biochemistry, 106(1),
38–45. https://doi.org/10.1076/apab.106.1.38.4396
Billat, V., Demarle, A., Paiva, M., & Koralsztein, J. P. (2002). Effect of training
on the physiological factors of performance in elite marathon runners
(males and females). International Journal of Sports Medicine, 23(5),
336–341. https://doi.org/10.1055/s-2002-33265
Billat, V. L. (2001). Interval training for performance: A scientific and empiri­
cal practice. Special recommendations for middle- and long-distance
running. Part I: Aerobic interval training. Sports Medicine, 31(1), 13–31.
https://doi.org/10.2165/00007256-200131010-00002
Billat, V. L., Demarle, A., Slawinski, J., Paiva, M., & Koralsztein, J. J. P. (2001).
Physical and training characteristics of top-class marathon runners.
Medicine & Science in Sports & Exercise, 33(12), 2089–2097. https://doi.
org/10.1097/00005768-200112000-00018
Billat, V. L., Flechet, B., Petit, B., Muriaux, G., & Koralsztein, J. J. P. (1999).
Interval training at VO2max: Effects on aerobic performance and over­
training markers. Medicine & Science in Sports & Exercise, 31(1), 156–163.
https://doi.org/10.1097/00005768-199901000-00024
Billat, V. L., Slawinski, J., Bocquet, V., Demarle, A., Lafitte, L., Chassaing, P., &
Koralsztein, J.-P. (2000, February). Intermittent runs at the velocity asso­
ciated with maximal oxygen uptake enables subjects to remain at max­
imal oxygen uptake for a longer time than intense but submaximal runs.
European Journal of Applied Physiology, 81(3), 188–196. https://doi.org/
10.1007/s004210050029
Blagrove, R. C., Howatson, G., & Hayes, P. R. (2018). Effects of strength
training on the physiological determinants of middle- and
long-distance running performance: A systematic review. Sports
Medicine, 48(5), 1117–1149. https://doi.org/10.1007/s40279-017-0835-7
Buchheit, M., & Laursen, P. B. (2013). High-intensity interval training, solu­
tions to the programming puzzle: Part 1: Cardiopulmonary emphasis.
Sports Medicine, 43(5), 313–338. https://doi.org/10.1007/s40279-013-
0029-x
Casado, A., Hanley, B., Santos-Concejero, J., & Ruiz-Pérez, L. M. (2019).
World-class long-distance running performances are best predicted by
volume of easy runs and deliberate practice of short-interval and tempo
runs. Journal of Strength and Conditioning Research, (29), 1. https://doi.
org/10.1519/JSC.0000000000003176
Clausen, J. P. (1977). Effect of physical training on cardiovascular adjust­
ments to exercise in man. Physiological Reviews, 57(4), 779–815. https://
doi.org/10.1152/physrev.1977.57.4.779
Cooper, G. (1997, February). Basic determinants of myocardial hypertrophy:
A review of molecular mechanisms. Annual Review of Medicine, 48(1),
13–23. https://doi.org/10.1146/annurev.med.48.1.13
Daussin, F., Ponsot, E., Dufour, S., Lonsdorfer-Wolf, E., Doutreleau, S.,
Geny, B., Piquard, F., & Richard, R. (2007). Improvement of VO2max by
cardiac output and oxygen extraction adaptation during intermittent
versus continuous endurance. European Journal of Applied Physiology,
101(3), 377–383. https://doi.org/10.1007/s00421-007-0499-3
Daussin, F. N., Zoll, J., Dufour, S. P., Ponsot, E., Lonsdorfer-Wolf, E.,
Doutreleau, S., Mettauer, B., Piquard, F., Geny, B., & Richard, R. (2008,
July). Effect of interval versus continuous training on cardiorespiratory
and mitochondrial functions: Relationship to aerobic performance
improvements in sedentary subjects. American Journal of Physiology-
Regulatory, Integrative and Comparative Physiology, 295(1), 264–272.
https://doi.org/10.1152/ajpregu.00875.2007
Demarie, S., Koralsztein, J. P., & Billat, V. (2000). Time limit and time at
VO2max during a continuous and an intermittent run. The Journal of
Sports Medicine and Physical Fitness, 40(2), 96–102. https://www.

minervamedica.it/en/journals/sports-med-physical-fitness/article.php?
cod=R40Y2000N02A0096
Demarle, A. P., Slawinski, J. J., Laffite, L. P., Bocquet, V. G., Koralsztein, J. P., &
Billat, V. L. (2001, March). Decrease of O2 deficit is a potential factor in
increased time to exhaustion after specific endurance training. Journal of
Applied Physiology, 90(3), 947–953. https://doi.org/10.1152/jappl.2001.
90.3.947
Downs, S. H., & Black, N. (1998, June). The feasibility of creating a checklist
for the assessment of the methodological quality both of randomised
and non-randomised studies of health care interventions. Journal of
Epidemiology & Community Health, 52(6), 377–384. https://doi.org/10.
1136/jech.52.6.377
Driller, M. W., Fell, J. W., Gregory, J. R., Shing, C. M., & Williams, A. D. (2009).
The effects of high-intensity interval training in well-trained rowers.
International Journal of Sports Physiology and Performance, 4(1),
110–121. https://doi.org/10.1123/ijspp.4.1.110
Esfarjani, F., & Laursen, P. B. (2007, February). Manipulating high-intensity
interval training: Effects on VO2max, the lactate threshold and 3000m
running performance in moderately trained males. Journal of Science and
Medicine in Sport, 10(1), 27–35. https://doi.org/10.1016/j.jsams.2006.05.014
Esteve-Lanao, J., Foster, C., Seiler, S., & Lucia, A. (2007). Impact of training
intensity distribution on performance in endurance athletes. Journal of
Strength and Conditioning Research, 21(3), 943–949. https://doi.org/10.
1519/R-19725.1
Esteve-Lanao, J., Juan, A. F. S., Earnest, C. P., Foater, C., & Lucia, A. (2005,
March). How do endurance runners actually train? Relationship with
competition performance. Medicine & Science in Sports & Exercise, 37(3),
496–504. https://doi.org/10.1249/01.MSS.0000155393.78744.86
Ferley, D. D., Osborn, R. W., & Vukovich, M. D. (2013, June). The effects of
uphill vs level-grade high-intensity interval training on VO2max, Vmax,
VLT, and Tmax in well-trained distance runners. Journal of Strength and
Conditioning Research, 27(6), 1549–1559. https://doi.org/10.1519/JSC.
0b013e3182736923
Foster, C. (1983). VO2max and training indices as determinants of compe­
titive running performance. Journal of Sports Sciences, 1(1), 13–22.
https://doi.org/10.1080/02640418308729657
Foster, C., Florhaug, J. A., Franklin, J., Gottschall, L., Hrovatin, L. A., Parker, S.,
Doleshal, P., & Dodge, Cet al. (2001). A new approach to monitoring
exercise training. Journal of Strength and Conditioning Research, 15(1),
109–115. https://doi.org/10.1519/1533-4287(2001)015<0109:anatme>2.
0.co;2
Garcin, M., Fleury, A., & Billat, V. (2002, January). The ratio HLa : RPE as a tool
to appreciate overreaching in young high-level middle-distance runners.
Medicine & Science in Sports & Exercise, 23(1), 16–21. https://doi.org/10.
1055/s-2002-19275
Hawley, J. A., Myburgh, K. H., Noakes, T. D., & Dennis, S. C. (1997). Training
techniques to improve fatigue resistance and enhance endurance
performance. Journal of Sports Sciences, 15(3), 325–333. https://doi.org/
10.1080/026404197367335
Helgerud, J., Høydal, K., Wang, E., Karlsen, T., Berg, P., Bjerkaas, M.,
SIMONSEN, T., HELGESEN, C., HJORTH, N., BACH, R., & HOFF, J. (2007).
Aerobic high-intensity intervals improve VO2max more than moderate
training. Medicine & Science in Sports & Exercise, 39(4), 665–671. https://
doi.org/10.1249/mss.0b013e3180304570
Higgins, J. Cochrane handbook for systematic reviews of interventions. The
Cochrane Collaboration. Wiley. Version 5.1.0 [updated March 2011].
www.cochrane-handbook.org
Joyner, M. J. (1991). Modeling: Optimal marathon performance on the basis
of physiological factors. Journal of Applied Physiology, 70(2), 683–687.
https://doi.org/10.1152/jappl.1991.70.2.683
Kenneally, M., Casado, A., & Santos-Concejero, J. (2018, October). The effect
of periodization and training intensity distribution on middle- and
long-distance running performance: A systematic review. International
Journal of Sports Physiology and Performance, 13(9), 1114–1121. https://
doi.org/10.1123/ijspp.2017-0327
Kennelly, J. (2011). Methodological approach to assessing the evidence. In
A. Handler, J. Kennelly, & N. Peacock (Eds.), Reducing racial/ethnic dispa­
rities in reproductive and perinatal outcomes: The evidence from popula­
tion-based interventions (pp. 7–19). Springer US.
JOURNAL OF SPORTS SCIENCES 17
Koral, J., Oranchuk, D. J., Herrera, R., & Millet, G. Y. (2018, March). Six sessions
of sprint interval training improves running performance in trained
athletes. Journal of Strength and Conditioning Research, 32(3), 617–623.
https://doi.org/10.1519/JSC.0000000000002286
Laursen, P. B., & Jenkins, D. G. (2002). The scientific basis for high-intensity
interval training: Optimising training programmes and maximising per­
formance in highly trained endurance athletes. Sports Medicine, 32(1),
53–73. https://doi.org/10.2165/00007256-200232010-00003
Lepretre, P. M., Koralsztein, J. P., & Billat, V. L. (2004). Effect of exercise
intensity on relationship between VO2max and cardiac output.
Medicine & Science in Sports & Exercise, 36(8), 1357–1363. https://doi.
org/10.1249/01.MSS.0000135977.12456.8F
Londeree, B. R. (1997). Effect of training on lactate/ventilatory thresholds: A
meta-analysis. Medicine &amp Science in Sports &amp Exercise, 29(6),
837–843. https://doi.org/10.1097/00005768-199706000-00016
Lucia, A., Hoyos, J., Santalla, A., Earnest, C., & Chicharro, J. L. (2003). Tour de
France versus Vuelta a España: Which is harder? Medicine & Science in
Sports & Exercise, 35(5), 872–878. https://doi.org/10.1249/01.MSS.
0000064999.82036.B4
Mallol, M., Bentley, D. J., Norton, L., Norton, K., Mejuto, G., & Yanci, J. (2018).
Comparison of reduced-volume high-intensity interval training and
high-volume training on endurance performance in triathletes.
International Journal of Sports Physiology and Performance, 14(2),
239–245. https://doi.org/10.1123/ijspp.2018-0359
Mercier, D., & Rioux, N. (1999). Mercier scoring tables. In Athletics Canada.
Midgley, A. W., & McNaughton, L. R. (2006). 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.
Midgley, A. W., McNaughton, L. R., & Jones, A. M. (2007). Training to enhance
the physiological determinants of long-distance running performance:
Can valid recommendations be given to runners and coaches based on
current scientific knowledge? Sports Medicine, 37(10), 857–880. https://
doi.org/10.2165/00007256-200737100-00003
Midgley, A. W., McNaughton, L. R., & Wilkinson, M. (2006). Is there an
optimal training intensity for enhancing the maximal oxygen uptake of
distance runners? Empirical research findings, current opinions, physio­
logical rationale and practical recommendations. Sports Medicine, 36(2),
117–132. https://doi.org/10.2165/00007256-200636020-00003
Milanović, Z., Sporiš, G., & Weston, M. (2015, October). Effectiveness of
high-intensity interval training (HIT) and continuous endurance training
for VO2max improvements: A systematic review and meta-analysis of
controlled trials. Sports Medicine, 45(10), 1469–1481. https://doi.org/10.
1007/s40279-015-0365-0
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., Altman, D., & PRISMA Group.
(2009, August 18). Preferred reporting items for systematic reviews and
meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151
(4), 264–269. https://doi.org/10.7326/0003-4819-151-4-200908180-
00135
Mujika, I., & Padilla, S. (2000a, September 23). Detraining: Loss of
training-induced physiological and performance adaptations. Part II:
Long term insufficient training stimulus. Sports Medicine, 30(3),
145–154. https://doi.org/10.2165/00007256-200030030-00001
Mujika, I., & Padilla, S. (2000b, September 23). Detraining: Loss of training
induced physiological and performance adaptation. Part I. Short term
insufficient training stimulus. Sports Medicine, 30(2), 79–87. https://doi.
org/10.2165/00007256-200030020-00002
Muñoz, I., Seiler, S., Bautista, J., España, J., Larumbe, E., & Esteve-Lanao, J.
(2014). Does polarized training improve performance in recreational
runners? International Journal of Sports Physiology and Performance, 9
(2), 265–272. https://doi.org/10.1123/ijspp.2012-0350
Nevill, A. M., Holder, R. L., & Cooper, S.-M. (2007, March). Statistics, truth, and
error reduction in sport and exercise sciences. European Journal of Sport
Science, 7(1), 9–14. https://doi.org/10.1080/17461390701197767
Ní Chéilleachair, N. J., Harrison, A. J., & Warrington, G. D. (2017, June). HIIT
enhances endurance performance and aerobic characteristics more than
high-volume training in trained rowers. Journal of Sports Sciences, 35(11),
1052–1058. https://doi.org/10.1080/02640414.2016.1209539
18 A. PARMAR ET AL.
Ouzzani, M., Hammady, H., Fedorowicz, Z., & Elmagarmid, A. (2016,
December). Rayyan-a web and mobile app for systematic reviews.
Systematic Reviews, 5(1), 1–10. https://doi.org/10.1186/s13643-016-0384-4
Papacosta, E., & Gleeson, M. (2013, March). Effects of intensified training and
taper on immune function. Rev Bras Educ Física E Esporte, 27(1), 159–176.
https://doi.org/10.1590/S1807-55092013005000001
Parmar, A., Jones, T. W., & Hayes, P. R. The use of interval-training methods
by coaches of well-trained middle- to long-distance runners (Under
review). 2020;
Pollock, M. L., Jackson, A. S., & Pate, R. R. (1980). Discriminant analysis of
physiological differences between good and elite distance runners.
Research Quarterly for Exercise and Sport, 51(3), 521–532. https://doi.
org/10.1080/02701367.1980.10608075
Seiler, K. S., & Kjerland, G. O. (2006, February). Quantifying training intensity
distribution in elite endurance athletes: Is there evidence for an “opti­
mal” distribution? Scandinavian Journal of Medicine and Science in Sports,
16(1), 49–56. https://doi.org/10.1111/j.1600-0838.2004.00418.x
Seiler, S. (2009). Intervals, thresholds, and long slow distance: The role of
intensity and duration in endurance training. Sportscience, 13(1), 32–53.
sportsci.org/2009/ss.htm
Seiler, S. (2010, September). What is best practice for training intensity and
duration distribution in endurance athletes? International Journal of
Sports Physiology and Performance, 5(3), 276–291. https://doi.org/10.
1123/ijspp.5.3.276
Seiler, S., Haugen, O., & Kuffel, E. (2007). Autonomic recovery after exercise
in trained athletes: Intensity and duration effects. Medicine & Science in
Sports & Exercise, 39(8), 1366–1373. https://doi.org/10.1249/mss.
0b013e318060f17d
Seiler, S., Jøranson, K., Olesen, B. V., & Hetlelid, K. J. (2013, February).
Adaptations to aerobic interval training: Interactive effects of exercise
intensity and total work duration. Scandinavian Journal of Medicine &
Science in Sports, 23(1), 74–83. https://doi.org/10.1111/j.1600-0838.2011.
01351.x
Seiler, S., & Sjursen, J. E. (2004, October). Effect of work duration on physio­
logical and rating scale of perceived exertion responses during
self-paced interval training. Scandinavian Journal of Medicine and
Science in Sports, 14(5), 318–325. https://doi.org/10.1046/j.1600-0838.
2003.00353.x
Slawinski, J., Demarle, A., Koralsztein, J. P., & Billat, V. L. (2001). Effect of
supra-lactate threshold training on the relationship between mechanical
stride descriptors and aerobic energy cost in trained runners. Archives of
Physiology and Biochemistry, 109(2), 110–116. https://doi.org/10.1076/
apab.109.2.110.4270
Smith, T. P., Coombes, J. S., & Geraghty, D. P. (2003). Optimising
high-intensity treadmill training using the running speed at maximal
O2 uptake and the time for which this can be maintained. European
View publication stats
Journal of Applied Physiology, 89(3–4), 337–343. https://doi.org/10.1007/
s00421-003-0806-6
Smith, T. P., McNaughton, L. R., & Marshall, K. J. (1999, June). Effects of 4-wk
training using Vmax/Tmax on VO2max and performance in athletes.
Medicine & Science in Sports & Exercise, 31(6), 892–896. https://doi.org/
10.1097/00005768-199906000-00019
Stöggl, T., & Sperlich, B. (2014). Polarized training has greater impact on key
endurance variables than threshold, high intensity, or high volume
training. Frontiers in Physiology, 5(33), 1–9. https://doi.org/10.3389/
fphys.2014.00033
Stöggl, T. L., & Sperlich, B. (2015, October). The training intensity distribution
among well-trained and elite endurance athletes. Frontiers in Physiology, 6
(295), 1–14. https://doi.org/10.3389/fphys.2015.00295
Stöggl, T. L., & Sperlich, B. (2019, May). Editorial: Training intensity, volume
and recovery distribution among elite and recreational endurance
athletes. Frontiers in Physiology, 10(592), 1–4. https://doi.org/10.3389/
fphys.2019.00592
Støren, Ø., Helgerud, J., Saebø, M., Støa, E. M., Bratland-Sanda, S.,
Unhjem, R. J., HOFF, J., & WANG, E. (2017). The effect of age on the
VO2max response to high-intensity interval training. Medicine & Science
in Sports & Exercise, 49(1), 78–85. https://doi.org/10.1249/MSS.
0000000000001070
Sylta, Ø., Tønnessen, E., & Seiler, S. (2014). From heart-rate data to training
quantification: A comparison of 3 methods of training-intensity analysis.
International Journal of Sports Physiology and Performance, 9(1), 100–107.
https://doi.org/10.1123/ijspp.2013-0298
Tjelta, L. I. (2016, February). The training of international level distance
runners. International Journal of Sports Science & Coaching, 11(1),
122–134. https://doi.org/10.1177/1747954115624813
Tjelta, L. I. (2019, October). Three Norwegian brothers all European
1500 m champions: What is the secret? International Journal of Sports
Science & Coaching, 14(5), 694–700. https://doi.org/10.1177/
1747954119872321
Vega, R. B., Konhilas, J. P., Kelly, D. P., & Leinwand, L. A. (2017, May).
Molecular mechanisms underlying cardiac adaptation to exercise. Cell
Metabolism, 25(5), 1012–1026. https://doi.org/10.1016/j.cmet.2017.04.
025
Wen, D., Utesch, T., Wu, J., Robertson, S., Liu, J., Hu, G., & Chen, H.. (2019,
August 1). Effects of different protocols of high intensity interval training
for VO2max improvements in adults: A meta-analysis of randomised
controlled trials. Journal of Science and Medicine in Sport, 22(8),
941–947. https://doi.org/10.1016/j.jsams.2019.01.013
Zavorsky, G. S. (2000). Evidence and possible mechanisms of altered
maximum heart rate with endurance training and tapering. Sports
Medicine, 29(1), 13–26. https://doi.org/10.2165/00007256-200029010-
00002