Clin Physiol Funct Imaging (2018) doi: 10.1111/cpf.

12513

REVIEW ARTICLE

Tabata protocol: a review of its application, variations and

outcomes
~o Pedro Arau
Ricardo Borges Viana1 , Claudio Andre Barbosa de Lira1, Joa jo Naves1, Victor Silveira Coswig2,
 3
Fabrıcio Boscolo Del Vecchio and Paulo Gentil 1

1
Department of Physical Education, Faculty of Physical Education and Dance, Federal University of Goi
as, Goi^ania, 2Department of Physical Education, Federal
University of Par
a, Castanhal, and 3Department of Physical Education, Superior School of Physical Education, Federal University of Pelotas, Pelotas, Brazil

Summary

Correspondence
Paulo Gentil, FEFD – Faculdade de Educacß~ao
F
ısica e Dancßa, Universidade Federal de Goi
as –
UFG, Avenida Esperancßa s/n, Campus Samam-
baia – CEP: 74
690-900, Goi^ania – Goi
as,
Brazil
E-mail: paulogentil@hotmail.com
Accepted for publication
Received 5 December 2017;
accepted 6 February 2018
Key words
exercise performance; high-intensity interval
training; metabolism; physical education; sports
medicine
Purpose: The great popularity of the Tabata Protocol is accompanied by an uncom-
fortable lack of consistency and criteria in its use, which results in many contro-
versies in the results obtained from its utilization. The purpose of this study was
to analyse the studies that based their interventions on the Tabata Protocol and to
provide a critical analysis of its use.
Methods: A systematic literature search was conducted in PubMed and Scopus. All
articles published between 1996 and October 2017 that cited at least one of the
original studies of Tabata et al. were considered. Inclusion criteria were as fol-
lows: original articles, human trials and English language.
Results: Thirty studies were included for analysis. Almost 37% of the studies (n = 11)
used a variation of the Tabata Protocol on a cycle ergometer. Only five studies stated
the use of the original Tabata Protocol. Exercise intensity was controlled by percent-
age of iVO_ 2max (n = 8) or iVO_ 2peak (n = 3), number of bouts performed (n = 3),
all out (n = 10), rate of perceived exertion (n = 1), self-perception of paces (n = 1),
maximal power output (n = 1), aerobic power (n = 1) and other forms (n = 2).
Conclusion: Based on our results, variations of the Tabata Protocol seem to be indi-
cated to provide increases in aerobic power that are similar to traditional aerobic
training while being less time consuming. These adaptations seem to be mainly
due to peripheral adaptations. Moreover, the use of Tabata Protocols to promote
weight loss is not substantiated by the reviewed studies.

than 15% of the results for ‘interval training’. Additionally,

Introduction
More than two decades ago, Tabata et al. (1996) published a
study in which a high-intensity interval training protocol last-
ing 4 min was as or more efficient in increasing aerobic and
anaerobic fitness than 1 h of moderate-intensity activities .
One year after, the same group tested the acute effects of this
protocol and confirmed its high metabolic demand (Tabata
et al., 1997). The protocol introduced by Kouichi Irisawa, for-
mer head coach of the Japanese Speed Skating Team, was bap-
tized as the ‘Tabata Protocol’, in recognition of the lead
author of the pioneer study. Now it is used in scientific exper-
iments as well as by the lay public. An exploratory search for
the ‘Tabata Protocol’ in Google retrieved approximately
129 000 results (9 November 2017), ‘Tabata Interval Train-
ing’ retrieved more than 170 000, which represents more
the search for terms related to the Tabata Protocol in the Uni-
ted States Patent and Trademark Office retrieves dozens of
results among apps for mobiles and tablets, as well as exercise
programmes. The great popularity of the protocol is probably
associated with its astonishing results together with an attrac-
tive time efficiency. However, such popularity is accompanied
by a lack of consistency, supervision and criteria in the use of
the protocol, which results in some questionings and uncer-
tainties about the results obtained from its utilization (Coswig
et al., 2016; Gentil et al., 2016).
The 1996 study reported the use of a load that allowed a
participant to perform 7–8 bouts of 20 s at a minimum
cadence of 85 rpm on a cycle ergometer, with passive recov-
ery of 10 s (Tabata et al., 1996). In the 1997 study, the proto-
col involved 6–7 bouts at an intensity equivalent to 170% of

© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd 1
2 Tabata protocol: a review, R. B. Viana et al.
the intensity associated with the achievement maximal con-
sumption of oxygen during an incremental test (iVO _ 2max)
(Tabata et al., 1997). Apparently, both studies were merged to
form the ‘Tabata Protocol’, which consists in one set of 7–8
bouts of 20 s performed at 170% of iVO _ 2max, with passive
recovery of 10 s. However, the feasibility of this recommen-
dation has been challenged, as it would be unrealistic to accu-
mulate 160 s of work at 170% of iVO _ 2max using 20 s:10 s
intervals, especially in cycling (Coswig et al., 2016; Gentil
et al., 2016).
When analysing the literature, it is possible to find scientific
articles that used the original Tabata’s studies as references for
planning their interventions (Buchan et al., 2011a,b; McRae
et al., 2012; Invernizzi et al., 2014; Joanisse et al., 2015; Logan
et al., 2016). However, many methodological aspects were not
clearly described in the original studies (Gastin, 1997), and
they were revealed as being very difficult or even unrealistic
to reproduce in a real-world setting. These difficulties led to
the creation of a great variety of ‘Tabata-like protocols’ that
may have a big difference in the physiological response when
compared to the initial protocols, which may result in quite
different acute and chronic results. For example, the ‘Tabata
Protocol’ was adapted using other activities (i.e. calisthenics,
running and resistance exercises) (McRae et al., 2012; Logan
et al., 2016) and varying intensities (all out, based on ratings
of perceived exertion, heart rate and body mass percentage)
(Buchan et al., 2011a,b; Invernizzi et al., 2014; Joanisse et al.,
2015; Logan et al., 2016). Although this has brought interest-
ing results, the physiological responses to these variations
seem to be inconsistent and one may question in each case
the use of a ‘Tabata Protocol’ would be recommended.
Therefore, due to the lack of consistency and criteria in the
use of the Tabata Protocol and the potential for specific appli-
cation of the framework of Tabata Protocol to various training
methods, the present systematic review aimed to analyse the
studies that based their interventions on the Tabata Protocol
and to provide a critical analysis of its use. The discussion pre-
sented here will help coaches and researchers to adequately
choose whether to use the Tabata Protocol, which outcomes
should be pursued and which variations to choose when aim-
ing specific objectives.
Methods
Search strategy
This systematic review conforms to the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses (PRISMA)
statement (Moher et al., 2009).
A computerized systematic search was conducted in the
PubMed and Scopus databases. All articles published between
1996 and October 2017 that cited one or two of the original
studies and that used the Tabata Protocol (Tabata et al., 1996,
1997) were considered. Two independent researchers evalu-
ated the articles and extracted the data. Controversies on data
© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
extraction were solved by discussion between them and, in
case of disagreement, by the judgement of a third researcher.
Study selection
Studies that met the following criteria were included (i)
original articles; (ii) human trials; (iii) English language; and
(iv) applied any variation of the Tabata Protocol (Tabata
et al., 1996, 1997) in their procedures. Articles were
excluded if they (i) had no full text; (ii) cited Tabata et al.
(1996, 1997) only in the references but not in the text; and
(iii) did not use any variation of the original protocols in
the methods. A first screening process of the full text of the
articles was performed by two independent researchers. After
initial screening of titles, irrelevant studies were removed,
which include overlapping studies, abstracts and irrelevant
articles, such as editorials and discussion papers that did not
match the inclusion criteria. Duplicate papers were then
identified and removed. We excluded further studies due to
insufficient data for analyses and non-exercise interventions
(Fig. 1). In addition, quality of included studies was evalu-
ated by Physiotherapy Evidence Database Scale (PEDro) and
is shown in Table S1.
Outcome measures
The following data were extracted: study authors, year of
publication, number of participants, sex, age, training status,
maximal consumption of oxygen (VO _ 2max), details of the
variations of the Tabata Protocol used (exercise modality,
number of bouts, duration of effort and pause, and intensity
control) and main outcomes of the study.
Results
Included studies
The search strategy retrieved 356 records. After deduplication
and language examination, 50 were excluded from the review
process; 306 full-text copies of the remaining studies were
obtained and subjected to further evaluation, 276 of these
studies were excluded and the reasons for exclusion were
annotated. At the end of the process, 30 publications meeting
the eligibility criteria were selected for review, of which 22
(Truijens et al., 2003; Amtmann et al., 2008; Ravier et al.,
2009; Buchan et al., 2011a,b; Farney et al., 2012; McRae et al.,
2012; Morifuji et al., 2012; Rebold et al., 2013; Fortner et al.,
2014; Invernizzi et al., 2014; Ma et al., 2014, 2015; Scribbans
et al., 2014a,b; Foster et al., 2015; Jabbour et al., 2015; Joanisse
et al., 2015; Garcıa-Hermoso et al., 2016; Harnish & Sabo,
2016; Laird et al., 2016; Logan et al., 2016; Mate-Mu~ noz et al.,
2017) cited the Tabata et al. (1996) and eight (Vuorimaa et al.,
2000; Truijens et al., 2003; Feriche et al., 2007; Farney et al.,
2012; Stanley et al., 2014; Nicol o et al., 2015; Williams &
Kraemer, 2015; Holmstrup et al., 2016) cited Tabata et al.

Figure 1 PRISMA schematic summarizing
the process of data collection. PRISMA, Pre-
ferred Reporting Items for Systematic Reviews
and Meta-Analyses.
(1997). Among them, two cited both studies (Truijens et al.,
2003; Farney et al., 2012).
Date of publication of the included studies
Although the Tabata et al. (1996, 1997) studies were published
in 1996 and 1997, most of the references that cited them were
published in the last 7 years, especially in the years 2014 and
2016 (Fig. 2). Two studies (Scribbans et al., 2014a; Joanisse
et al., 2015) conducted two surveys each on different samples.
Another two studies (Buchan et al., 2011a,b) used the same
sample, however, evaluated different variables.
Participant characteristics
The number of participants in the studies varied from four
(Amtmann et al., 2008) to 88 (Ma et al., 2015) with an average
© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
Tabata protocol: a review, R. B. Viana et al. 3
of approximately 25 participants. Thirteen studies examined
exclusively males (Vuorimaa et al., 2000; Feriche et al., 2007;
Amtmann et al., 2008; Ravier et al., 2009; Farney et al., 2012;
Morifuji et al., 2012; Scribbans et al., 2014a,b; Stanley et al.,
2014; Nicol o et al., 2015; Williams & Kraemer, 2015; Logan
et al., 2016; Mate-Mu~ noz et al., 2017), three exclusively
females (McRae et al., 2012; Holmstrup et al., 2016; Laird et al.,
2016), whilst the remaining study assessed a mixed sample
(Truijens et al., 2003; Buchan et al., 2011a,b; Rebold et al.,
2013; Fortner et al., 2014; Ma et al., 2014, 2015; Scribbans
et al., 2014a; Foster et al., 2015; Jabbour et al., 2015; Joanisse
et al., 2015; Harnish & Sabo, 2016). In total, 431 men and
217 women participated in the studies (two studies used the
same sample). The mean age of study participants ranged from
9 to 29 years. The training status of the participants ranged
from sedentary (Jabbour et al., 2015; Joanisse et al., 2015) to
high-level athletes (Vuorimaa et al., 2000; Truijens et al., 2003;
4 Tabata protocol: a review, R. B. Viana et al.
Ravier et al., 2009; Invernizzi et al., 2014; Stanley et al., 2014;
Nicolo et al., 2015). In addition, in those studies that measured
_ 2max (n = 21), the mean of VO
VO _ 2max ranged from
34
0  6
5 to 69
4  5
1 ml kg 1min 1. Details of the par-
ticipant characteristics of the included studies are detailed in
Tables S2 and S3.
Exercise training parameters
Regarding training protocols, almost 37% of the studies
(n = 11) used a variation of the Tabata Protocol in a cycle
ergometer (Fig. 3). Only five studies (Amtmann et al., 2008;
Scribbans et al., 2014a,b; Foster et al., 2015; Joanisse et al.,
2015) stated the use of the original Tabata Protocol. Exercise
_ 2max (n = 8) or
intensity was prescribed by percentage of iVO
intensity of peak consumption of oxygen (VO _ 2peak) (n = 3),
number of repetitions performed in fixed time (n = 3), all-
out efforts (n = 10), rate of perceived exertion (n = 1), self-
perception of paces (n = 1), maximal power output (n = 1),
© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
aerobic power (n = 1) and other forms (n = 2). Details of the
acute and chronic interventions of the included studies are
showed in Tables S2 and S3, respectively.
Discussion
In general, a high variability and low consistency with the
original Tabata Protocols (Tabata et al., 1996, 1997) were
found. Considering the high heterogeneity within studies, we
chose to discuss the influence of exercise modality (cycling,
running, calisthenics, etc.), prescribed intensity and study
design (acute and chronic) on main outcomes by topics.
Exercise modality
Only 11 used a cycle ergometer and most used other forms of
exercise (Fig. 3). It is important to note that different types of
exercise (cycling exercise and whole-body exercise) evoke dif-
ferent physiological demands. For example, at the same
Figure 2 Number of studies that cited
Tabata et al. (1996, 1997).
Figure 3 Type of training modality used in
the studies that cited Tabata et al. (1996,
1997).

relative intensity, running may induce greater oxygen con-
sumption (VO _ 2) than cycling, with lower lactate levels
(Matsui et al., 1978). These differences might explain the vari-
ations in the outcomes reported in the reviewed studies. For
example, McRae et al. (2012) used calisthenic exercises and
found no differences in anaerobic capacity between high-
intensity interval training (HIIT) and continuous activity at
moderate intensity, contradicting the results of Tabata et al.
(1996) on the cycle ergometer.
Indeed, running and calisthenics create less peripheral fati-
gue when compared with cycling (Fitzsimmons et al., 1993),
which might result in a lower glycogen depletion and this can
influence the results in power outputs (Hulston et al., 2010).
Fortner et al. (2014) conducted a study with 14 non-obese
young participants (body mass index: 25
7  0
8 kg m 2) to
investigate cardiovascular and metabolic demands of a ‘Tabata
Protocol’ performed with kettlebell swing exercises. The
results revealed a higher rate of perceived exertion, average
_ 2 value, percentage of VO
VO _ 2peak achieved, maximal heart
rate and post-exercise blood lactate concentration in the
Tabata Protocol than in traditional resistance protocols.
Although the post-exercise blood lactate concentration was
greater in the Tabata Protocol than in the traditional protocol,
the value (6
4  1
1 mmol l 1) was lower when compared
to the value reported by Foster et al. (2015) when using a
cycle ergometer (~12 mmol l 1). On the other hand, using
the Tabata Protocol with mixed martial arts resulted in higher
blood lactate concentration (range 8
1–19
7 mmol l 1) (Amt-
mann et al., 2008). This can be a consequence of the use of
different exercise modalities and methods of controlling the
intensity of effort, resulting in different releases of lactate into
the bloodstream. Therefore, if the purpose is to provide
higher stress on the anaerobic system, the use of cycling
might be considered. However, to provide higher VO _ 2 with
less peripheral fatigue running, calisthenics or kettlebell exer-
cise might be a better choice. In addition, besides McRae et al.
(2012) suggested that exercise modality choice and variability
could enhance motivation and long-term adherence, this issue
was not yet tested and should be considered in future investi-
gations.
Prescribed intensity
About 47% of the selected studies (Amtmann et al., 2008;
McRae et al., 2012; Rebold et al., 2013; Fortner et al., 2014;
Ma et al., 2014, 2015; Scribbans et al., 2014a,b; Foster et al.,
2015; Joanisse et al., 2015; Williams & Kraemer, 2015; Holm-
strup et al., 2016; Laird et al., 2016; Mate-Mu~
noz et al., 2017)
used the same number of bouts, time of effort and time to
recovery between the bouts that were used by Tabata et al.
(1996). Regarding intensity, only four studies (Scribbans et al.,
2014a,b; Foster et al., 2015; Joanisse et al., 2015) supposedly
applied the original intensity used in the Tabata Protocol (i.e.
170% of iVO _ 2 max) (Tabata et al., 1996, 1997). However, the
feasibility of such a protocol was recently questioned (Coswig
© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
Tabata protocol: a review, R. B. Viana et al. 5
et al., 2016; Gentil et al., 2016) due to the difficulties of the
subjects in performing 7–8 bouts at an intensity correspond-
ing to 170% of iVO _ 2max on a cycle ergometer. Unpublished
data from our research group reveal that most participants
interrupt the exercise by the third bout when cycling at an
intensity equivalent to 170% of the iVO _ 2max and a plausible
intensity for 7–8 bouts should be around 115% of the
_ 2max. Therefore, it appears that using 170% of iVO
iVO _ 2max
is unfeasible in a real-world setting, especially if we consider
a clinical population (such as overweight and obese subjects).
The difficulty in using the intensity equivalent to 170% of
_ 2max might have led researchers to change the form of
iVO
intensity control (e.g. ‘all-out’, rate of perceived exertion and
number of bouts performed) in order to accomplish the 7–8
bouts proclaimed by Tabata et al. (1996). For example, Laird
et al. (2016) used 110%, 115% and 120% of iVO _ 2max. We
speculate that inconsistencies in Tabata’s Protocol intensity
prescription could be related to poor pretesting description on
the original articles. Indeed, the original work indicates con-
stant power testing for VO _ 2max, while recent papers applied
different progressive tests (Scribbans et al., 2014a,b; Foster
et al., 2015; Joanisse et al., 2015), which means that different
parameters have been used and VO _ 2max could have been
underestimated in Tabata’s work.
Another possible consequence of the difficulties in establish-
ing the adequate intensity and consequently replicate the orig-
inal protocol is the increase in the use of ‘Tabata-like
Protocols’ to perform generic forms of ‘all-out’ efforts (Amt-
mann et al., 2008; Rebold et al., 2013; Fortner et al., 2014; Ma
et al., 2014, 2015; Joanisse et al., 2015; Harnish & Sabo, 2016;
Logan et al., 2016; Mate-Mu~ noz et al., 2017). However, the
performance of all-out efforts for 20 s at a fixed load inter-
spaced by 10 s could be highly demanding, which makes its
feasibility questionable. It is important to note that recent
studies with 20 s of all-out efforts employed 2 min of rest
between bouts and restricted the number of bouts to 3–4
(Gillen & Gibala, 2014; Gillen et al., 2016). Based on this, it is
recommended to individually test the participants in order to
establish the adequate intensity for the purposed work param-
eters.
Study design
Regarding the type of the study, 16 (53
3%) of the reviewed
studies were acute and 14 (46
6%) were chronic studies
(Tables S2 and S3). The analyses involved VO _ 2max (Vuorimaa
et al., 2000; Feriche et al., 2007; Fortner et al., 2014; Stanley
et al., 2014; Nicolo et al., 2015; Williams & Kraemer, 2015),
lactate levels (Feriche et al., 2007; Amtmann et al., 2008; Far-
ney et al., 2012; Fortner et al., 2014; Invernizzi et al., 2014;
Stanley et al., 2014; Nicol o et al., 2015), perceived exertion
(Amtmann et al., 2008; Farney et al., 2012; Fortner et al.,
2014), heart rate (Farney et al., 2012; Fortner et al., 2014;
Stanley et al., 2014; Nicol o et al., 2015; Williams & Kraemer,
2015), total work (Farney et al., 2012; Morifuji et al., 2012;
6 Tabata protocol: a review, R. B. Viana et al.
Scribbans et al., 2014a; Nicol o et al., 2015; Williams &
Kraemer, 2015; Harnish & Sabo, 2016), caloric expenditure
(Williams & Kraemer, 2015), oxidative stress (Farney et al.,
2012), blood glucose, insulin and plasma concentration of
amino acids (Morifuji et al., 2012), blood pressure (Buchan
et al., 2011a,b; Fortner et al., 2014; Stanley et al., 2014; Foster
et al., 2015), fibre-type distribution, whole-muscle capillary
density (Scribbans et al., 2014a), power output (Foster et al.,
2015; Holmstrup et al., 2016; Laird et al., 2016; Mate-Mu~ noz
et al., 2017), inflammatory markers (cytokines, interleukin-6
and interleukin-10), tumour necrosis factor (TNF-a) and
insulin sensitivity (Harnish & Sabo, 2016).
The analysis of VO _ 2 in the acute studies (Fortner et al.,
2014; Williams & Kraemer, 2015) that used the same number
of bouts, time of effort and time to recovery between the
bouts that were used by Tabata et al. (1996) indicates that
mean percentage of VO _ 2max achieved ranged from approxi-
mately 43% (Williams & Kraemer, 2015) to 71% (Fortner
et al., 2014). However, Tabata et al. (1997) reported that sub-
jects achieved values corresponding to 96
5% of the VO _ 2max.
The blood lactate, perceived exertion and heart rate ranged
from 6
4 (Fortner et al., 2014) to 16
3 mmol l 1 (Amtmann
et al., 2008), 15
1 (Fortner et al., 2014) to 17 (Amtmann et al.,
2008) values on the Borg Scale (6–20) and approximately
149 (Williams & Kraemer, 2015) to 162 bpm (Fortner et al.,
2014), respectively. The high variability shows that different
modalities and/or intensity prescription affect VO _ 2, blood lac-
tate, heart rate and rate perceived exertion behaviour, which
means that benefits derived from the Tabata Protocol change
according to the modality, form of control and magnitude of
the intensity adopted. This wide range of applications does
not allow authors to suggest that the use of ‘Tabata Protocols’
tax the aerobic and anaerobic systems maximally, as previ-
ously suggested (Gentil et al., 2016).
In the chronic studies, VO _ 2max, anaerobic power and
capacity were the most common outcomes evaluated (Table
S2). Improvements in VO _ 2max were shown in six chronic
studies (Ravier et al., 2009; McRae et al., 2012; Scribbans et al.,
2014a,b; Foster et al., 2015; Logan et al., 2016), while four
studies (Rebold et al., 2013; Scribbans et al., 2014a,b; Foster
et al., 2015) reported increases in anaerobic power and only
two in anaerobic capacity (Scribbans et al., 2014a; Foster et al.,
2015). Tabata’s original studies showed aerobic and anaerobic
improvements, which indicate that very low-volume, high-
intensity interval training, which lasts only 4 min, could be
an interesting strategy for improving physical fitness. Scrib-
bans et al. (2014b) replicated Tabata’s Protocol (eight bouts of
20-second intervals at  170% of iVO _ 2peak separated by 10
seconds rest no load on a cycle ergometer at a cadence of
their choice) in order to investigate possible mechanisms for
those responses and found similar aerobic and anaerobic
improvements. Aerobic changes were accompanied by
increases in mitochondrial proteins and peroxisome prolifera-
tor-activated receptor gamma coactivator 1 alpha (PGC1-a)
activity. In agreement, Scribbans et al. (2014a) showed that
© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
Tabata’s Protocol induced similar changes in VO _ 2peak and
exercise performance as endurance training. The authors indi-
cated that fibre-type distribution, fibre-type specific oxidative
and glycolytic capacity, glycogen and intramuscular triglyc-
eride (IMTG) storage, and whole-muscle capillary density, in
addition to anaerobic exercise performance and whole-muscle
glycolytic capacity, could explain results for both training
modes; however, this remains unclear. The advantage of the
very low-volume HIIT could be related to time efficiency
(~10% of the time expended on endurance training). Among
the seven chronic studies that compared the Tabata Protocol
with other forms of exercise, two (McRae et al., 2012; Scrib-
bans et al., 2014a) reported better results and five (Buchan
et al., 2011a,b; Foster et al., 2015; Holmstrup et al., 2016;
Laird et al., 2016) reported equivalent results to others forms
of exercise. Consequently, we can infer that the merit of the
Tabata Protocol is in its time efficiency more than in its supe-
riority in comparison with traditional protocols.
Only three chronic studies evaluated body composition
(Buchan et al., 2011a,b; Rebold et al., 2013). Two studies
compared with a moderate-intensity group, and the results
showed that body fat percentage significantly decreased only
in the moderate group postintervention (Buchan et al., 2011a,
b). Another study showed significant results when compared
to a non-exercise group (Rebold et al., 2013). Taken these
studies together, it should be considered that (i) body compo-
sition was not the main outcome; (ii) we found duplicated
data (Buchan et al., 2011a,b); and (iii) subjects already have
low values of body fat percentage. Therefore, to infer about
the use of this form of exercise for weight/fat loss is not pos-
sible yet. On the other hand, it is a relevant topic and should
be considered by researchers groups.
Surprisingly, our review found only five studies that
approximate the original protocol replication (Amtmann et al.,
2008; Scribbans et al., 2014a,b; Foster et al., 2015; Joanisse
et al., 2015) and not even the group that published the origi-
nal studies replicated it in later studies. The term ‘Tabata Pro-
tocol’ is a generic classification for high-intensity intermittent
protocols using low duration and high-intensity efforts with
even lower duration rest intervals. However, the inconsisten-
cies between them do not allow us to predict the outcomes
obtained from such interventions.
To the best of our knowledge, this review is the first one
to analyse the use of the Tabata Protocol in scientific articles
and our results may help coaches and researchers to replicate
previously used protocols by providing them information that
would help to choose the variations to perform and define
adequate expectations from the outcomes.
The findings of this review should be interpreted taking
into consideration some limitations. The first one is that we
might have neglected articles that tried to replicate the original
protocols but did not cite them in the references. Moreover,
the heterogeneity of the studies did not allow for further
meta-analytical approach, which precludes us estimating the
results obtained from the variations.
 Tabata protocol: a review, R. B. Viana et al. 7

that are similar to traditional aerobic training while being

Conclusions less time consuming. These adaptations seem to be mainly
The findings from this systematic review indicated that less due to peripheral adaptations. Moreover, the use of Tabata
than half of the studies that cited the high-intensity interval Protocols to promote weight loss is not substantiated by the
training named Tabata Protocol in the methods stated the reviewed studies.
use of the original protocol. Most of the existing literature
has focused only on applying the same number of bouts,
Acknowledgment
time duration of bout and/or time to recovery between
bouts used in the Tabata Protocol. Our analysis revealed a None.
large inconsistency in the use of the ‘Tabata Protocols’,
which results in divergent acute and chronic responses.
Conflict of interest
Based on our results, variations of the Tabata Protocols
seem to be indicated to provide increases in aerobic power The authors have no conflict of interests.

References
Amtmann JA, Amtmann KA, Spath WK. Lac-
tate and rate of perceived exertion responses
of athletes training for and competing in a
mixed martial arts event. J Strength Cond Res
(2008); 22: 645–647.
Buchan DS, Ollis S, Thomas NE, et al. Physical
activity interventions: effects of duration
and intensity. Scand J Med Sci Sports (2011a);
21: 341–350.
Buchan DS, Ollis S, Young JD, et al. The effects
of time and intensity of exercise on novel
and established markers of CVD in adolescent
youth. Am J Hum Biol (2011b); 23: 517–526.
Coswig VS, Gentil P, Naves JPA, et al. Com-
mentary on: the effects of high intensity
interval training vs steady state training on
aerobic and anaerobic capacity. Front Physiol
(2016); 7: 495.
Farney TM, McCarthy CG, Canale RE, et al.
Absence of blood oxidative stress in trained
men after strenuous exercise. Med Sci Sports
Exerc (2012); 44: 1855–1863.
Feriche B, Delgado M, Calder on C, et al. The
effect of acute moderate hypoxia on accu-
mulated oxygen deficit during intermittent
exercise in nonacclimatized men. J Strength
Cond Res (2007); 21: 413–418.
Fitzsimmons M, Dawson-Hughes B, Ware D,
et al. Cycling and running tests of repeated
sprint ability. Aust J Sci Med Sport (1993); 25:
82–87.
Fortner HA, Salgado JM, Holmstrup AM,
et al. Cardiovascular and metabolic demads
of the kettlebell swing using Tabata interval
versus a traditional resistance protocol. Int J
Exerc Sci (2014); 7: 179–185.
Foster C, Farland CV, Guidotti F, et al. The
effects of high intensity interval training vs
steady state training on aerobic and anaero-
bic capacity. J Sports Sci Med (2015); 14:
747–755.
Garcıa-Hermoso A, Cerrillo-Urbina AJ, Her-
rera-Valenzuela T, et al. Is high-intensity
interval training more effective on improv-
ing cardiometabolic risk and aerobic capac-
ity than other forms of exercise in
overweight and obese youth? A meta-analy-
sis Obes Rev (2016); 17: 531–540.
Gastin PG. Metabolic profile of high intensity
intermittent exercises. Med Sci Sports Exerc
(1997); 29: 1274–1276.
Gentil P, Naves JPA, Viana RB, et al. Revis-
iting Tabata’s protocol: does it even exist?
Med Sci Sports Exerc (2016); 48: 2070–
2071.
Gillen JB, Gibala MJ. Is high-intensity interval
training a time-efficient exercise strategy to
improve health and fitness? Appl Physiol Nutr
Metab (2014); 39: 409–412.
Gillen JB, Martin BJ, MacInnis MJ, et al. Twelve
weeks of sprint interval training improves
indices of cardiometabolic health similar to
traditional endurance training despite a five-
fold lower exercise volume and time com-
mitment. PLoS One (2016); 11: 1–14.
Harnish CR, Sabo RT. Comparison of two dif-
ferent sprint interval training work-to-rest
ratios on acute inflammatory responses.
Sports Med Open (2016); 2: 20.
Holmstrup ME, Jensen BT, Evans WS, et al.
Eight weeks of kettlebell swing training
does not improve sprint performance in
recreationally active females. Int J Exerc Sci
(2016); 9: 437–444.
Hulston CJ, Venables MC, Mann CH, et al.
Training with low muscle glycogen
enhances fat metabolism in well-trained
cyclists. Med Sci Sports Exerc (2010); 42:
2046–2055.
Invernizzi PL, Longo S, Scurati R, et al. Inter-
pretation and perception of slow, moderate,
and fast swimming paces in distance and
sprint swimmers. Percept Mot Skills (2014);
118: 833–849.
Jabbour G, Iancu HD, Paulin A. Effects of
high-intensity training on anaerobic and
aerobic contributions to total energy release
during repeated supramaximal exercise in
obese adults. Sports Med Open (2015); 1: 36.
Joanisse S, McKay BR, Nederveen JP, et al.
Satellite cell activity, without expansion,
following non-hypertrophic stimuli. Am J
Physiol Regul Integr Comp Physiol (2015); 309:
R1101–R1111.
Laird RH, Elmer DJ, Barberio MD, et al. Eval-
uation of performance improvements after
either resistance training or sprint interval-
based concurrent training. J Strength Cond Res
(2016); 30: 3057–3065.
Logan GRM, Harris N, Duncan S, et al. Low-
active male adolescents: a dose response to
high-intensity interval training. Med Sci Sports
Exerc (2016); 48: 481–490.
Ma JK, Le Mare L, Gurd BJ. Classroom-based
high-intensity interval activity improves off-
task behaviour in primary school students.
Appl Physiol Nutr Metab (2014); 39: 1332–
1337.
Ma JK, LeMare L, Gurd BJ. Four minutes of
in-class high-intensity interval activity
improves selective attention in 9- to 11-
year olds. Appl Physiol Nutr Metab (2015); 40:
238–244.
Mate-Mu~ noz JL, Lougedo JH, Barba M, et al.
Muscular fatigue in response to different
modalities of CrossFit sessions. PLoS One
(2017); 12: 1–17.
Matsui H, Kitamura K, Miyamura M. Oxy-
gen uptake and blood flow of the lower
limb in maximal treadmill and bicycle
exercise. Eur J Appl Physiol Occup Physiol
(1978); 40: 57–62.
McRae G, Payne A, Zelt JGE, et al. Extremely
low volume, whole-body aerobic-resistance
training improves aerobic fitness and mus-
cular endurance in females. Appl Physiol Nutr
Metab (2012); 37: 1124–1131.
Moher D, Liberati A, Tetzlaff J, et al. Pre-
ferred reporting items for systematic

© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd
8 Tabata protocol: a review, R. B. Viana et al.

reviews and meta-analyses: the PRISMA
statement. BMJ (2009); 339: b2535.
Morifuji M, Aoyama T, Nakata A, et al. Post-
exercise ingestion of different amounts of
protein affects plasma insulin concentration
in humans. Eur J Sport Sci (2012); 12: 152–
160.
Nicolo A, Bazzucchi I, Lenti M, et al. Neuro-
muscular and metabolic responses to high
intensity intermittent cycling protocols with
different work to rest ratios. Int J Sports Phys-
iol Perform (2015); 9: 2014–2015.
Ravier G, Dugue B, Grappe F, et al.
Impressive anaerobic adaptations in elite
karate athletes due to few intensive inter-
mittent sessions added to regular karate
training. Scand J Med Sci Sports (2009); 19:
687–694.
Rebold MJ, Kobak MS, Otterstetter R. The
influence of a Tabata interval training pro-
gram using an aquatic underwater treadmill
on various performance variables. J Strength
Cond Res (2013); 27: 3419–3425.
Scribbans TD, Edgett BA, Vorobej K, et al.
Fibre-specific responses to endurance and
low volume high intensity interval training:
striking similarities in acute and chronic
adaptation. PLoS One (2014a); 9: e98119.
Scribbans TD, Ma JK, Edgett BA, et al. Resver-
atrol supplementation does not augment
performance adaptations or fibre-type-speci-
fic responses to high-intensity interval train-
ing in humans. Appl Physiol Nutr Metab
(2014b); 39: 1305–1313.
Stanley J, Peake JM, Coombes JS, et al. Central
and peripheral adjustments during high-
intensity exercise following cold water
immersion. Eur J Appl Physiol (2014); 114:
147–163.
Tabata I, Nishimura K, Kouzaki M, et al.
Effects of moderate-intensity endurance and
high-intensity intermittent training on
anaerobic capacity and VO2max. Med Sci
Sports Exerc (1996); 28: 1327–1330.
Tabata I, Irisawa K, Kouzaki M, et al. Meta-
bolic profile of high intensity intermittent
exercises. Med Sci Sports Exerc (1997); 29:
390–395.
Truijens MJ, Toussaint HM, Dow J, et al.
Effect of high-intensity hypoxic training on
sea-level swimming performances. J Appl
Physiol (2003); 94: 733–743.
Vuorimaa T, Vasankari T, Rusko H. Compar-
ison of physiological strain and muscular
performance of athletes during two inter-
mittent running exercises at the velocity
associated with VO2max. Int J Sports Med
(2000); 21: 96–101.
Williams BM, Kraemer RR. Comparison of
cardiorespiratory and metabolic responses
in kettlebell high-intensity interval training
versus sprint interval cycling. J Strength Cond
Res (2015); 29: 3317–3325.

Supporting Information
Additional Supporting Information may be found in the Table S2. Summary of trial characteristics from acute
online version of this article: studies.
Table S1. Description of studies quality based on Physiother- Table S3. Summary of trial characteristics from chronic/train-
apy Evidence Database Scale (PEDro). ing studies.

© 2018 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd