ORIGINAL RESEARCH

published: 04 September 2018

doi: 10.3389/fphys.2018.01170

Tendinous Tissue Adaptation to

Explosive- vs. Sustained-Contraction
Strength Training
Garry J. Massey 1,2 , Thomas G. Balshaw 1,2 , Thomas M. Maden-Wilkinson 2,3 ,
Neale A. Tillin 4 and Jonathan P. Folland 1,2*
1
Arthritis Research UK Centre for Sport, Exercise and Osteoarthritis, Loughborough University, Loughborough,
United Kingdom, 2 School of Sport, Exercise, and Health Sciences, Loughborough University, Loughborough,
United Kingdom, 3 Faculty of Health and Wellbeing, Sheffield Hallam University, Sheffield, United Kingdom, 4 Department
of Life Sciences, University of Roehampton, London, United Kingdom

The effect of different strength training regimes, and in particular training utilizing
brief explosive contractions, on tendinous tissue properties is poorly understood. This
study compared the efficacy of 12 weeks of knee extensor explosive-contraction
(ECT; n = 14) vs. sustained-contraction (SCT; n = 15) strength training vs. a non-
training control (n = 13) to induce changes in patellar tendon and knee extensor
tendon–aponeurosis stiffness and size (patellar tendon, vastus-lateralis aponeurosis,
quadriceps femoris muscle) in healthy young men. Training involved 40 isometric knee
extension contractions (three times/week): gradually increasing to 75% of maximum
voluntary torque (MVT) before holding for 3 s (SCT), or briefly contracting as fast
Edited by:
Gary Iwamoto,
as possible to ∼80% MVT (ECT). Changes in patellar tendon stiffness and Young’s
University of Illinois modulus, tendon–aponeurosis complex stiffness, as well as quadriceps femoris muscle
at Urbana-Champaign, United States
volume, vastus-lateralis aponeurosis area and patellar tendon cross-sectional area were
Reviewed by:
quantified with ultrasonography, dynamometry, and magnetic resonance imaging. ECT
Sylvain Dorel,
University of Nantes, France and SCT similarly increased patellar tendon stiffness (20% vs. 16%, both p < 0.05
Bruce M. Damon, vs. control) and Young’s modulus (22% vs. 16%, both p < 0.05 vs. control). Tendon–
Vanderbilt University Medical Center,
United States
aponeurosis complex high-force stiffness increased only after SCT (21%; p < 0.02),
*Correspondence:
while ECT resulted in greater overall elongation of the tendon–aponeurosis complex.
Jonathan P. Folland Quadriceps muscle volume only increased after sustained-contraction training (8%;
[email protected]
p = 0.001), with unclear effects of strength training on aponeurosis area. The changes in
Specialty section:
patellar tendon cross-sectional area after strength training were not appreciably different
This article was submitted to to control. Our results suggest brief high force muscle contractions can induce increased
Exercise Physiology,
free tendon stiffness, though SCT is needed to increase tendon–aponeurosis complex
a section of the journal
Frontiers in Physiology stiffness and muscle hypertrophy.
Received: 28 December 2017 Keywords: tendon, aponeurosis, stiffness, young’s modulus, muscle, strength training, hypertrophy
Accepted: 03 August 2018
Published: 04 September 2018
Citation: INTRODUCTION
Massey GJ, Balshaw TG,
Maden-Wilkinson TM, Tillin NA and The mechanical stiffness (resistance to deformation) of muscle tendinous tissues (aponeurosis and
Folland JP (2018) Tendinous Tissue
extramuscular free tendon) is integral to the effectiveness of these tissues to transmit skeletal muscle
Adaptation to Explosive- vs.
Sustained-Contraction Strength
force to the bone and thus generate movement. Stiffer tissues may be protective in injury-related
Training. Front. Physiol. 9:1170. situations, for instance maintaining balance in response to mechanical perturbation (Karamanidis
doi: 10.3389/fphys.2018.01170 et al., 2008). Moreover, stiffer tendons undergo less strain in response to stress, which reduces their

Frontiers in Physiology | www.frontiersin.org 1 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

susceptibility to damage (Buchanan and Marsh, 2002). Likewise, to more conventional sustained-contraction strength training
stiffer tissues may limit injury risk by providing greater joint (SCT).
stability and by perhaps reducing the loading imposed on Changes in tendon–aponeurosis complex and free tendon
passive joint tissue structures (meniscus, cartilage, ligaments), stiffness after strength training may depend upon the increase in
(Lipps et al., 2014). A particular concern is that traumatic the size of these tissues. Muscle hypertrophy is a well-recognized
joint injuries predispose to degenerative disease (e.g., anterior characteristic response to conventional strength training regimes
cruciate ligament) and the increased risk of knee osteoarthritis, (Folland and Williams, 2007) that is suggested to be coincident
which contributes to a reduced quality of life (Salaffi et al., with an increase in aponeurosis size (Wakahara et al., 2015), but
2005). Therefore, increased tendinous tissue stiffness could longitudinal changes in aponeurosis size are largely unknown.
have functional and clinical implications, thus identifying A solitary report documented a 1.9% increase in vastus lateralis
effective interventions to stimulate tendinous tissue adaptations aponeurosis width to accompany a 10.7% increase in quadriceps
is warranted. muscle size after 12 weeks of SCT (Wakahara et al., 2015). Free
In vivo tendinous tissue stiffness is typically determined from tendon hypertrophy after SCT has received much more attention,
force–elongation relationships acquired by combining tissue but the evidence remains equivocal. While some studies utilizing
elongation visualized via ultrasonography with estimates of magnetic resonance imaging (MRI) have reported modest
tendon force during ramp isometric contractions. In response increases in free tendon cross-sectional area (CSA) (∼3–6%:
to a constant rate of increase in contractile force, elongation Arampatzis et al., 2007; Kongsgaard et al., 2007; Seynnes et al.,
of the free tendon [between proximal and distal osteotendon 2009; Bohm et al., 2014) that may be region specific, others
junction’s (Kongsgaard et al., 2007; Seynnes et al., 2009)] found no change (Arampatzis et al., 2010; Kubo et al., 2012;
and elongation of the distal tendon–aponeurosis complex (i.e., Bloomquist et al., 2013). The responses of muscle, aponeurosis
aponeurosis and free tendon) via the displacement of a muscle- and tendon size to ECT are largely unknown. Given the marginal
fascicle aponeurosis intersection (Kubo et al., 2001, 2006c; changes in free tendon size after SCT, the increases in free
Arampatzis et al., 2007) can be used to determine stiffness tendon stiffness (e.g., 15–65%: Reeves et al., 2003; Kongsgaard
of both these structures. During muscle contraction the free et al., 2007; Seynnes et al., 2009; Malliaras et al., 2013; McMahon
tendon experiences tensile loading and positive longitudinal et al., 2013) have predominantly been attributed to the nearly
strain, whereas the radial expansion of muscle fascicles during parallel increases in free tendon Young’s modulus (stiffness
force-generation and shortening causes the aponeurosis to also relative to tendon dimensions, i.e., material stiffness), although
undergo transverse elongation and positive strain (Azizi and the changes in free tendon modulus after ECT have yet to be
Roberts, 2009; Raiteri et al., 2016). The alternative strain behavior documented.
of the free tendon and aponeurosis may lead to differential The aim of the present study was to comprehensively compare
adaptations in the separate free tendon and combined tendon– the mechanical and morphological adaptations of the tendinous
aponeurosis complex in response to training. However, very few tissues, both the patellar tendon and tendon–aponeurosis
studies have made simultaneous measurements of the mechanical complex, to 12 weeks ECT vs. SCT vs. a non-training control
properties of both structures (Kubo et al., 2006a,c, 2009), group. The mechanical properties examined were patellar tendon
therefore the comparative changes in free tendon and tendon– stiffness and Young’s modulus, as well as tendon–aponeurosis
aponeurosis complex stiffness after exercise training remains complex stiffness. Morphological measures investigated were
opaque. quadriceps femoris muscle volume, vastus lateralis aponeurosis
The mechanical stiffness of the tendon–aponeurosis complex area and patellar tendon CSA. As both training regimes involved
has been repeatedly found to increase following strength training high force production, we hypothesized that ECT and SCT would
with sustained contractions at high loads (≥2 s duration with be similarly effective training interventions to increase tendinous
loads of >70% maximum: Bohm et al., 2015; Wiesinger et al., tissue stiffness.
2015), e.g., 16–54% after 12–14 weeks (Kubo et al., 2001,
2006b; Arampatzis et al., 2007). Interestingly, two recent studies
reported that strength training with brief explosive-contractions MATERIALS AND METHODS
(<1 s) characterized by maximum/near maximum rate of force
development up to a high level of force produced increases in Participants and Ethical Approval
stiffness after merely 4 (34%; Tillin et al., 2012) and 6 weeks (62%; Forty-two young, healthy, asymptomatic, males who had
Burgess et al., 2007) of training. These preliminary results suggest not completed lower body-strength training for >18 months
that explosive-contraction strength training (ECT) may provide and were not involved in systematic physical training were
a potent stimulus for increasing tendon–aponeurosis complex randomly assigned to ECT (n = 14), SCT (n = 15) or control
stiffness. Furthermore due to the brief nature of the contractions (CON, n = 13) groups. Baseline recreational physical activity
(Balshaw et al., 2016), ECT is a relatively non-fatiguing training level was assessed with the International Physical Activity
regime that may be preferable for older adults and patient groups Questionnaire (IPAQ, short format). Each participant provided
(e.g., mobility, limited, osteoarthritis, tendinopathy: Reid et al., written informed consent prior to completing this study, which
2015) and thus facilitate higher levels of adherence. However, a was approved by the Loughborough University Ethical advisory
comprehensive longer-term investigation is required to validate committee and conformed to the principles of the Declaration of
the efficacy of ECT to increase tissue stiffness in comparison Helsinki.

Frontiers in Physiology | www.frontiersin.org 2 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

Experimental Design
Participants visited the laboratory for a familiarization session
that included measurement of muscle strength and body mass
to facilitate group allocation, as well as practice isometric ramp
contractions. Thereafter, two duplicate laboratory measurement
sessions were conducted both pre (sessions 7–10 days apart
prior to the first training session) and post (2–3 and 4–6 days
after the last training session). MRI scans of the thigh and
knee were conducted pre (5 days prior to the start of the first
training session) and post (2–3 days after the final training
session) to measure knee extensor tissue size (quadriceps muscle
volume, vastus lateralis aponeurosis area, patellar tendon CSA)
and patellar tendon moment arm. All measurement and training
sessions were performed with the same isometric apparatus
and the same joint angle configuration [knee and hip angles
of 115◦ and 126◦ (180◦ = full extension)]. Training for ECT
and SCT group’s involved unilateral isometric contractions of
both legs three times a week for 12 weeks (36 sessions in
total), whereas CON participants attended only the measurement
sessions and maintained their habitual activity. All participants
were instructed to maintain their habitual physical activity and
diet throughout the study, which was verified by informal
questioning during post measurement. Measurement sessions
involved a series of contractions of the dominant (preferred
kicking) leg in the following order: maximum voluntary
contraction [MVCs to establish maximum voluntary torque
(MVT)]; ramp voluntary contractions of the knee extensors to
establish tendinous tissue properties, and knee flexor MVCs.
Knee joint torque was recorded throughout contractions. Knee
flexor surface electromyography was recorded during knee flexor
MVCs, as well as during knee extensor ramp contractions to
account for antagonist co-activation in the estimate of tendon
force in knee extensor ramp contractions. Ultrasound images of
the vastus lateralis muscle and patellar tendon were recorded to
assess tissue elongation during the ramp contractions in order
to derive force–elongation relationships (to determine stiffness) FIGURE 1 | Example isometric knee extension torque–time traces performed
of the distal tendon–aponeurosis complex and patellar tendon, during (A) sustained-contraction strength training (SCT), and (B)
as well as stress–strain relationships for the patellar tendon (to explosive-contraction strength training (ECT). MVT, maximum voluntary
torque.
determine Young’s modulus). Measurement sessions were at a
consistent time of day and started between 12:00 h and 19:00 h.

When performing ECT the focus was on maximizing RTD,

Training which means participants cannot precisely control the peak
After a brief warm-up of sub-maximum contractions of both torque achieved. Therefore participants were instructed to
legs, participants completed four sets of 10 unilateral isometric simply achieve ∼80% MVT as quickly as possible to ensure
knee-extensor contractions of each leg with sets alternating that peak torque was at least practically equivalent to SCT.
between legs. Each set took 60 s with 2 min between successive A computer monitor displayed RTD (10 ms time epoch) to
sets on the same leg. SCT involved sustained contractions at provide biofeedback of explosive performance, with a cursor
75% MVT, with 2 s rest between contractions. In order to control indicating the highest peak RTD achieved throughout the session.
the rate of torque development (RTD) these participants were Participants were encouraged to achieve a higher peak RTD
presented with a target torque trace 2 s before every contraction with each subsequent contraction. The torque–time curve was
and instructed to match this target, which gradually increased also shown: with a horizontal cursor at 80%MVT to encourage
torque linearly from rest to 75%MVT over 1 s before holding a sufficiently forceful contractions, and on a sensitive scale baseline
plateau at 75%MVT for a further 3 s (Figure 1A). ECT involved torque was highlighted in order to observe and provide feedback
maximum/near maximum RTD contractions with participants to participants to correctly perform the contractions by avoiding
instructed to perform each contraction “as fast and hard as any pre-tension or countermovement. All training participants
possible” then relax for 5 s between repetitions (Figure 1B). (ECT and SCT) performed three isometric knee extensor MVCs

Frontiers in Physiology | www.frontiersin.org 3 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

at the start of each training week in order to re-establish MVT sampled at 2000 Hz via the same A/D converter and computer
and prescribe training torques. Torque data from each repetition software as the force signal, to enable data synchronization. In
of all training participants in the first session of weeks 1, 6, and offline analysis, EMG signals were corrected for the 48 ms delay
12 was analyzed and loading indices were averaged across the inherent to the Trigno EMG system. During knee flexor MVCs
three sessions: SCT vs. ECT, peak loading magnitude (75 vs 81% EMG amplitude was calculated as the root mean square (RMS) of
MVT), peak loading rate (1.4 vs. 8.9% MVT.s−1 ), impulse (28212 the filtered EMG signal of the biceps femoris and semitendinosus
vs. 3025 Nm.s). over a 500 ms epoch at knee flexion MVT (250 ms either side of
instantaneous peak torque) and averaged across the two muscles
Knee Extension and Flexion Maximum to give knee flexor EMGMAX .
Voluntary Contractions
Following a brief warm-up [3 s contractions at 50% (x3), 75% MRI Measurement of Muscle Tendon
(205 x3) and 90% (x1) of perceived maximum], participants
performed three to four MVCs and were instructed to either
Unit Morphology and Moment Arm
‘push as hard as possible’ (knee extension) or ‘pull as hard as Participants reported to the MRI scanner (1.5 T Signa HDxt,
possible’ (knee flexion) for 3–5 s and rest ≥30 s. A horizontal GE) having not engaged in strenuous activity in the prior 36 h,
cursor indicating the greatest torque obtained within the session and were instructed to arrive in a relaxed state having eaten
was displayed for biofeedback and verbal encouragement was and drunk normally, and sat quietly for 15 min prior to their
provided during all MVCs. The highest instantaneous torque MRI scans. T1-weighted MR images of the dominant leg (thigh
recorded during any MVC was defined as MVT. and knee) were acquired in the supine position at a knee
angle of 163◦ due to constraints in knee coil size (180◦ = full
extension) and analyzed using OsiriX software (Version 6.0,
Torque Measurement Pixmeo, Geneva, Switzerland). Using a receiver 8-channel whole
Measurement and training sessions were completed in the same body coil, axial images (image matrix 512 × 512, field of
custom-made isometric strength-testing chair with knee and hip view 260 mm × 260 mm, pixel size 0.508 mm × 0.508 mm,
angles of 115◦ and 126◦ (180◦ = full extension), respectively. slice thickness 5 mm, inter-slice gap 0 mm) were acquired
Adjustable straps were tightly fastened across the pelvis and from the anterior superior iliac spine to the knee joint space
shoulders to prevent extraneous movement. An ankle strap in two overlapping blocks. Oil filled capsules placed on the
(35 mm width reinforced canvas webbing) was placed ∼15% lateral side of the thigh allowed alignment of the blocks during
of tibial length (distance from lateral malleolus to knee joint analysis. The anatomical CSA of each of the four constituent
space) above the medial malleolus, and positioned perpendicular quadriceps femoris muscles (vastus lateralis, vastus intermedius,
to the tibia and in series with a calibrated S-Beam strain vastus medialis, and rectus femoris) was manually outlined
gauge (Force Logic, Berkshire, United Kingdom). The analog in every third image (i.e., every 1.5 cm) starting from the
force signal was amplified (×370; A50 amplifier, Force Logic, most proximal image in which the muscle was visible. A cubic
Berkshire, United Kingdom) and sampled at 2,000 Hz using an spline curve was fitted to the plot of anatomical CSA vs.
A/D converter (Micro 1401; CED, Cambridge, United Kingdom) femur length for each constituent muscle, and the muscle
and recorded with Spike 2 computer software (CED). In offline volume calculated as the area under the spline curve (GraphPad
analysis, force signals were low-pass filtered at 500 Hz using Prism 6, GraphPad Software, Inc.) Total quadriceps femoris
a fourth order zero-lag Butterworth filter, gravity corrected by muscle volume was given by the sum of the constituent muscle
subtracting baseline force, and multiplied by lever length, the volumes.
distance from the knee joint space to the center of the ankle strap, As previously described (Wakahara et al., 2015), the deep
to calculate torque values. aponeurosis of the vastus lateralis muscle was defined as the
visible dark black segment between the vastus lateralis and vastus
Knee Flexor Electromyography (EMG) intermedius muscles in the axial thigh MRI images (Figure 2).
Surface EMG recordings over the biceps femoris and The transverse length (cm) of the black segment was defined
semitendinosus muscles were made with a wireless EMG as vastus lateralis aponeurosis width, and was traced manually
system (Trigno; Delsys Inc., Boston, MA, United States) during on every third image (i.e., every 1.5 cm), starting in the most
knee flexor MVCs and knee extensor ramp contractions. distal image where the aponeurosis was visible. From the images
Following preparation of the skin (shaving, abrading, and analyzed, the measures of aponeurosis width were plotted against
cleansing with alcohol) single differential Trigno standard EMG femur length. A cubic spline curve was fitted to the plot of
sensors (1 cm inter electrode distance; Delsys Inc., Boston, MA, VL aponeurosis width vs. femur length and the vastus lateralis
United States) were attached over each muscle using adhesive aponeurosis area was calculated as the area under the spline curve
interfaces. Sensors were positioned parallel to the presumed (Figure 2).
frontal plane orientation of the underlying muscle fibers at Immediately after thigh imaging, a lower extremity
45% of thigh length (distance from the greater trochanter to the knee coil was used to acquire axial (image matrix
lateral knee joint space) measured from the popliteal crease. EMG 512 × 512, field of view 160 mm × 160 mm, pixel size
signals were amplified at source (×300; 20–450 Hz bandwidth) 0.313 mm × 0.313 mm, slice thickness 2 mm, inter-slice gap
before further amplification (overall effective gain × 909) and 0 mm) and sagittal images (image matrix 512 × 512, field of

Frontiers in Physiology | www.frontiersin.org 4 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 2 | Example axial magnetic resonance images: (A) most proximal, (B) middle, and (C) most distal, showing the transverse length of the vastus lateralis (VL)
deep aponeurosis which was traced manually in order to measure aponeurosis width. (D) A cubic spline curve was fitted through the aponeurosis width data points
measured at 1.5 cm intervals from the most proximal and distal image where the aponeurosis was visible (aponeurosis length) and the area under the curve was
defined as vastus lateralis aponeurosis area.

view 160 mm × 160 mm, pixel size 0.313 mm × 0.313 mm, Ramp Contractions for Determination of
slice thickness 2 mm, inter-slice gap = 0 mm) of the knee Tendinous Tissue Stiffness
joint. Contiguous axial images spanned patellar tendon length,
Tendinous tissue stiffness was derived from synchronous
which prior to analysis were reconstructed with an orientation
recordings of torque and tissue elongation (corrected for passive
perpendicular to the patellar tendon via the mutli-plane view
feature of OsiriX. Images spanned from 2 cm superior to the tissue displacement via video recording of knee joint changes;
patella apex to 2 cm inferior to the tendon tibial insertion. Patellar see below) during isometric knee extension ramp contractions
tendon CSA was measured on each contiguous image along the (experimental set-up: Figure 4). Participants completed two sub-
tendon’s length (first image where the patellar was no longer maximum practice ramp contractions prior to five maximum
visible to the last image before the tibial insertion). Images, attempts with 90 s of rest between contractions. Prior to each
viewed in grayscale, were sharpened and the perimeter manually ramp contraction participants were shown a target torque–time
outlined (Figure 3). Mean tendon CSA (mm2 ) was defined by trace on a computer monitor that increased at a constant gradient
the average of all measured analyzed images. Patellar tendon (50 Nm.s−1 loading rate) from zero up to MVT. They were
moment arm length was estimated from sagittal plane images, instructed to match the target trace as closely as possible for
as the perpendicular distance from the patellar tendon to the as long as possible (i.e., up to MVT), and then relax promptly.
midpoint of the distance between the tibio-femoral contact points Real-time torque was displayed over the target rising torque–
in the lateral and medial femoral condyles (Blazevich et al., 2009; time trace for feedback. The preceding knee extensor MVCs and
Seynnes et al., 2009). sub-maximum contractions were considered sufficient to elicit

Frontiers in Physiology | www.frontiersin.org 5 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 3 | Example magnetic resonance images of the knee: (A) proximal; just distal to the apex of the patella, (B) mid-length; 50% distance between the
patella-tibia attachments, and (C) distal; just proximal to the tendon tibial insertion. (i) Sagittal images show the position along the tendon length, of where the
example axial images shown (ii) were acquired perpendicular to the tendon line of action. (iii) The perimeter of the patellar tendon (PT) was manually traced to
determined PT cross-sectional area (CSA), with the average of the measures from each contiguous 2 mm image spanning tendon length being defined as mean
patellar tendon CSA.

tissue preconditioning (Seynnes et al., 2014). The three most thigh with the mid-point of the probe positioned at ∼50% thigh
suitable ramp contractions, according to highest peak torque, length. The probe was aligned so the fascicles inserting into the
the closeness to the target loading rate, as well the clarity of the vastus lateralis muscle deep aponeurosis could be visualized at
ultrasound images of both the patellar tendon and vastus lateralis rest and during contraction. An echo absorptive marker (multiple
muscle (clearly visible osteotendon attachments and fascicle- layers of transpore medical tape) was placed beneath the
aponeurosis intersection), were analyzed and measurements ultrasound probe to provide a reference for any probe movement
averaged across these three contractions. over the skin. Vastus lateralis muscle fascicle deep aponeurosis
cross-point displacement relative to the skin marker provided
a measure of distal tendon–aponeurosis complex elongation
Measurement of Tendinous Tissue (Figure 4). To enable correction of aponeurosis displacement
Elongation due to joint angle changes during ramp contractions, individual
Two ultrasound machines and a camera were interfaced with the ratios of aponeurosis displacement relative to joint angular
computer collecting torque data in Spike 2, and video images displacement (mm/◦ ) were obtained from passive movements
were synchronously recorded with torque (and EMG) using Spike (i.e., plotting the aponeurosis displacement-knee joint angle
2 video capture at 25 Hz. Video images were captured to obtain relationship). The mean ± standard deviation for this ratio was
tissue (tendon–aponeurosis and patellar tendon) and knee joint 0.37 ± 0.09 mm/◦ . Passive movements were conducted prior to
displacements during ramp contractions, which were measured the ramp contractions. Participants were instructed to completely
in off-line analysis by tracking specific anatomical landmarks relax as their knee was moved through 90◦ –130◦ . During passive
frame-by-frame in public domain semi-automatic video analysis movements and ramp contractions, knee joint angle (angle
software: Tracker, version 4.86 (www.physlets.org/tracker/). between visible markers placed on the greater trochanter, lateral
An ultrasound linear array probe (60 mm, B-mode, 7.5 MHz knee joint space and lateral malleolus) was derived from sagittal
scanning frequency, 39 Hz sampling frequency, Toshiba Power plane video recorded using a camera mounted on a tripod
Vision 6000, SSA-370A) was fitted into a custom made high- positioned (1.5 m) perpendicular to the strength-testing chair.
density foam cast that was strapped to the lateral aspect of the During ramp contractions knee angle changes were 3.1 ± 1.2◦ .

Frontiers in Physiology | www.frontiersin.org 6 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 4 | The experimental set-up and ultrasound images during the ramp contractions. Participants were tightly fastened to a rigid isometric strength-testing
chair with resting knee and hip angles of 115◦ and 126◦ , respectively (A). Unilateral knee extension torque, video of the knee joint angle, antagonist muscle [biceps
femoris (BF), semitendinosus (ST)] surface electromyography (EMG) and ultrasound video images were synchronously recorded during constant-loading rate
isometric ramp knee extensor contractions (example in B). Ultrasound images are of the patellar tendon (C) and vastus lateralis muscle (D) at rest (top) and at peak
ramp torque (bottom) and indicate the measurement of patellar tendon (tibia-patella apex displacement, 1T + 1P) and tendon–aponeurosis complex (vastus lateralis
muscle fascicle-deep aponeurosis cross point proximal displacement, 1M) elongation.

A second ultrasound linear array probe (92 mm EUP-L53L, tibial insertion (Figure 4). Under passive conditions, patellar
B-mode, 10 MHz scanning frequency, 32 Hz sampling frequency; tendon elongation was deemed negligible.
Hitachi EUB-8500) was fitted into a custom made high-density
foam cast that was held firmly over the anterior aspect of the Calculation of Patellar Tendon Force
knee with the probe aligned longitudinal to the patellar tendon Patellar tendon force was calculated by dividing total knee
such that the patella apex and insertion of the posterior tendon extensor torque by the patellar tendon moment arm length.
fibers at the tibia could be visualized at rest and throughout the Direct measures of moment arm were acquired at rest from
contraction. Patellar tendon elongation was determined by the MRI images as indicated above (MRI measurement). Due to
longitudinal displacement of both the patella apex and the tendon constraints in the size of the knee coil, sagittal images were

Frontiers in Physiology | www.frontiersin.org 7 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

acquired in an extended knee position (∼163◦ : 180◦ = full were averaged to produce criterion pre and post values for
extension). Moment arm length for any specific knee angle statistical analysis. Data are reported as mean ± standard
measured at rest or during ramp contraction was estimated from deviation (SD). Statistical significance tests were conducted using
previously published data fitted with a quadratic function (Kellis SPSS Version 20.0 (IBM Corp., Armonk, NY, United States),
and Baltzopoulos, 1999) scaled to each participant’s measured and significance was accepted at p < 0.05. 0.05 < p < 0.1
moment arm length at 163◦ . Total knee extensor torque was was considered a tendency. One-way analysis of variance
given by summing external net knee extension torque and (ANOVA) tests were conducted on all pre-training variables to
the estimated knee flexor co-contraction torque. Antagonist determine whether baseline differences existed between groups.
knee flexor torque was estimated by expressing the average The primary comparison of training effects involved between
knee flexor EMG amplitude (RMS 50 ms moving window) group comparisons to the intervention, and assessment of
during ramp contractions relative to the knee flexor EMGMAX , repeated measures analysis of variance [ANCOVA; group (ECT
and then multiplying by the knee flexor MVT (assuming a vs. SCT vs. CON) × time (pre vs. post)] with corresponding
linear relationship between EMG amplitude and torque). During pre-training values used as covariates. When group × time
analysis, torque and EMG amplitude were down sampled to interaction effects displayed p < 0.05, least significant difference
25 Hz to match the ultrasound video recording. (LSD) post hoc pairwise comparisons [with Holm–Bonferroni
adjustment applied to the p-values (LSDHB )] of absolute changes
Calculation of Tendinous Tissue Stiffness (pre to post) between groups (i.e., ECT vs. SCT, ECT vs.
CON, SCT vs. CON) were performed to delineate specific
and Patellar Tendon Young’s Modulus between-group differences. In addition to the between group
For each of the three best ramp contractions analyzed, comparisons, secondary within-group changes (absolute values)
both patellar tendon and distal tendon–aponeurosis complex were evaluated with paired t-tests. Effect size (ES: specifically
(corrected for passive tissue displacement due to knee joint Hedges g, incorporating correction for small sample bias; Lakens,
angle displacement) and during elongation contraction were 2013) was calculated for between-group comparisons and within
separately plotted against total tendon force (corrected for group changes.
antagonist force). Patellar tendon and tendon–aponeurosis
complex and force–elongation plots were fitted with a second-
order polynomial. To standardize the tendon force level, both
pre and post-training, tendon–aponeurosis complex and patellar
RESULTS
tendon stiffness for each individual was calculated as the
slope of the respective force–elongation curve over an absolute Group Characteristics at Baseline
tendon force range that equated to 70–80% of pre-training At baseline, no differences (p ≥ 0.579) were observed between
MVT. 70–80% pre-training MVT corresponded to the highest groups for age (ECT 25 ± 2; SCT 25 ± 2; CON 25 ± 3 years),
common torque range that all participants could individually height (ECT 174 ± 7; SCT 175 ± 8; CON 176 ± 6 cm), body
achieve during pre-training measurements sessions Patellar mass (ECT 71 ± 10; SCT 70 ± 8; CON 72 ± 7 kg) or habitual
tendon Young’s modulus was calculated for each individual physical activity level (ECT 1971 ± 1077; SCT 2084 ± 1256;
as the slope of the stress–strain curve derived over a stress CON 2179 ± 1588 metabolic equivalent minutes per week).
range that corresponded to 70–80% of pre-training MVT. Likewise, there were no differences in MVT (p = 0.304), tendon–
Stiffness/modulus measures derived over the highest attainable aponeurosis complex stiffness (p = 0.328), patellar tendon
force/stress range are recommended and deemed suitably reliable stiffness (p = 0.215), Young’s modulus (p = 0.184), quadriceps
(Hansen et al., 2006; Kösters et al., 2014; Seynnes et al., muscle volume (p = 0.508), and vastus lateralis aponeurosis area
2014). Tendon stress was obtained by dividing tendon force (p = 0.815), though a tendency existed for patellar tendon mean
by mean patellar tendon CSA. Patellar tendon strain was the CSA (p = 0.073).
percentage tendon displacement relative to the resting tendon
length. Resting patellar tendon length was defined as the distance Reproducibility of Measurements
between the patella apex and tibial insertion as measured prior The reproducibility of pre and post measures for the CON group
to the ramp contractions. The measures of patellar tendon over the 12-week intervention period was excellent for MVT
and tendon–aponeurosis complex stiffness, and the patellar (CVw 2.9%) and tendon–aponeurosis complex stiffness (3.9%),
tendon modulus derived from each of the three analyzed and very good for patellar tendon stiffness (7.2%) and Young’s
ramps were averaged to give a representative value for each modulus (6.8%). Excellent reproducibility was also observed for
individual. quadriceps muscle volume (1.7%), vastus lateralis aponeurosis
area (2.7%) and patellar tendon mean CSA (2.9%).
Statistical Analysis
The reproducibility of measurements (all muscle and tendinous Strength and Muscle–Tendon
tissue variables) over the 12-week intervention period was Morphology (Tables 1, 2 and Figure 5)
calculated for CON (pre vs. post) as within-participant coefficient Considering within-group changes, MVT increased after ECT
of variation [CVw, %; (SD/mean) × 100]. Muscle and tendon (paired t-test p < 0.001, ES = 1.15) and SCT (p < 0.001, ES = 1.11)
variables measured during the duplicate laboratory sessions but not following CON (p = 0.868, ES = 0.01). Between group

Frontiers in Physiology | www.frontiersin.org 8 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

comparisons showed the absolute increase in MVT was greater ES = 0.29), but was unchanged following SCT (p = 0.746,
than CON for both ECT (LSDHB p < 0.001, ES = 1.90) and SCT ES = 0.03). However, there was no group × time effect (Table 1
(LSDHB p < 0.001, ES = 2.64), and 45% larger after SCT than ECT and Figure 5C).
(LSDHB p = 0.032, ES = 0.75).
Quadriceps muscle volume increased after SCT (paired t-test
p = 0.001, ES = 0.47) but not following ECT (p = 0.195, ES = 0.17) Tendinous Tissue Mechanical Properties
or CON (p = 0.661, ES = 0.04). There was a group × time (Tables 1, 2)
effect for quadriceps muscle volume (Table 1), with the absolute Patellar tendon elongation at 80% pre-training MVT was less
change (Figure 5A) after SCT being greater than CON (LSDHB after ECT (paired t-test p = 0.011, ES = 0.75, but was unchanged
p = 0.021, ES = 1.12), and a tendency to be different to ECT after SCT (p = 0.246, ES = 0.24) or CON (p = 0.331, ES = 0.15),
(p = 0.074, ES = 0.72). Absolute changes in quadriceps muscle (Figure 6), and no group × time effect was observed (Table 1).
volume after ECT were not greater than CON (LSDHB p = 0.479, Patellar tendon strain (relative elongation) at 80% pre-training
ES = 0.31). MVT was also less after ECT (paired t-test p = 0.010, ES = 0.54),
Vastus lateralis aponeurosis area increased after SCT (paired but was unchanged after SCT (p = 0.542, ES = 0.11) or CON
t-test p = 0.015, ES = 0.32), and also tended to increase after (p = 0.263, ES = 0.15), (Figure 6), and there was no group × time
ECT (p = 0.060, ES = 0.35), while remaining unchanged in CON effect (Table 1).
(p = 0.408, ES = 0.11). However, there was no group × time effect Patellar tendon stiffness increased after both ECT (paired
(Table 1 and Figure 5B). t-test p = 0.002, ES = 0.88) and SCT (p = 0.019, ES = 0.74),
Patellar tendon mean CSA showed a small decrease in CON but was unchanged in CON (p = 0.711, ES = 0.07). There was a
(paired t-test p = 0.028, ES = 0.27), and after ECT (p = 0.012, group × time effect (Table 1), and absolute changes (Figure 7)

TABLE 1 | Strength, muscle–tendon unit size, patellar tendon moment arm, and patellar tendon and tendon–aponeurosis complex mechanical properties.

Explosive-contraction strength Sustained-contraction strength Non-training Two-way ANCOVA

training (ECT) training (SCT) control (CON) Group ×
time (p value)
Pre Post Pre Post Pre Post

Strength and
morphology
Maximum voluntary 234 ± 27 273 ± 36∗∗∗ L 237 ± 49 293 ± 47∗∗∗ L 255 ± 50 256 ± 58 <0.001
torque (MVT), Nm
Quadriceps muscle 1778 ± 244 1827 ± 277 1820 ± 273 1967 ± 316∗∗∗ S 1897 ± 282 1909 ± 271 0.018
volume, cm3
Vastus lateralis 137.1 ± 16.4 143.1 ± 15.2∼ S 136.3 ± 26.1 144.3 ± 21.2∗ S 138.8 ± 13.7 140.5 ± 15.7 0.242
aponeurosis area, cm2
Patellar tendon mean 98.7 ± 10.0 95.9 ± 8.3∗ S 97.3 ± 12.9 97.7 ± 13.0 106.5 ± 9.0 103.6 ± 10.7∗ S 0.129
CSA, mm2
Patellar tendon length, 47.5 ± 5.7 47.2 ± 5.7 45.4 ± 5.5 45.1 ± 5.5 47.1 ± 5.7 46.6 ± 6.8 0.829
mm
Patellar tendon 40.6 ± 2.4 40.7 ± 2.3 42.4 ± 2.9 42.5 ± 2.9 41.2 ± 2.9 41.3 ± 2.9 0.902
moment arm, mm

Patellar tendon
properties
Elongation at 80% 3.17 ± 0.52 2.82 ± 0.42∗∗ M 3.23 ± 0.54 3.07 ± 0.64 3.12 ± 0.62 3.02 ± 0.63 0.270
pre-MVT, mm
Stiffness, N.mm−1 2605 ± 446 3122 ± 632∗∗ L 2835 ± 444 3239 ± 575∗ M 2534 ± 501 2569 ± 413 0.018
Strain at 80% pre-MVT, 6.8 ± 1.7 6.0 ± 1.1∗∗ M 7.2 ± 1.4 6.9 ± 1.7 6.6 ± 1.1 6.4 ± 1.1 0.093
%
Young’s modulus, GPa 1.23 ± 0.18 1.49 ± 0.27∗∗∗ L 1.32 ± 0.27 1.51 ± 0.36∗ M 1.14 ± 0.27 1.16 ± 0.20 0.012

Tendon–aponeurosis
complex properties
Elongation at 80% 15.0 ± 2.6 17.4 ± 2.2∗∗ L 16.9 ± 4.6 16.4 ± 5.3 16.3 ± 5.7 16.6 ± 4.4 0.020
pre-MVT, mm
Stiffness, N.mm−1 592 ± 118 595 ± 101 560 ± 177 687 ± 285∗∗ M 507 ± 130 511 ± 116 0.007

Data are mean ± SD. ECT, n = 13; SCT, n = 15 (size and strength), n = 14/15 (tendon–aponeurosis complex/patellar tendon); CON, n = 13 (size and strength) and
n = 13/12 (tendon–aponeurosis/patellar tendon). ∗∗∗ Different to pre, p ≤ 0.001, ∗∗ p < 0.01, ∗ p < 0.05. ∼0.05 < p < 0.1. Within-group effect size: S, small (0.2–0.5);
M, moderate (>0.5–0.8), L, large (>0.8).

Frontiers in Physiology | www.frontiersin.org 9 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

TABLE 2 | Summary of within-group changes and between-group differences from pre to post training in strength, muscle–tendon unit morphology and tendinous tissue
stiffness indices.

Within-group changes Between-group differences

Explosive-contraction Sustained-contraction Non-training

strength training (ECT) strength training (SCT) control (CON)

Strength and morphology

Maximum voluntary torque (MVT), Nm ↑ +17% ↑ +24% ↔ ECT and SCT↑ > CON
Quadriceps muscle volume, cm3 ↔ ↑ +8% ↔ SCT↑ > CON
Vastus lateralis aponeurosis area, cm2 ↔ ↑ +7% ↔ −
Patellar tendon mean CSA, mm2 ↓ −3% ↔ ↓ −3% −

Tendinous tissue stiffness indices

Patellar tendon
Elongation at 80% pre-MVT, mm ↓ −10% ↔ ↔ −
Strain at 80% pre-MVT, % ↓ −11% ↔ ↔ −
Stiffness, N.mm−1 ↑ +20% ↑ +16% ↔ ECT and SCT↑ > CON
Young’s modulus, GPa ↑ +22% ↑ +16% ↔ ECT and SCT↑ > CON

Tendon–aponeurosis complex
Elongation at 80% pre-MVT, mm ↑ +17% ↔ ↔ ECT↑ > SCT
Stiffness, N.mm−1 ↔ ↑ +21% ↔ SCT↑ > ECT and CON

The directions of the group changes are shown by ↑ or ↓ with the percentage change in the group mean also shown. Non-significant within-/between group changes are
indicated by ↔/−.

FIGURE 5 | Pre to post absolute changes (1) in (A) quadriceps femoris muscle volume (B) vastus lateralis aponeurosis area and (C) patellar tendon mean
cross-sectional area (CSA) in response to explosive-contraction (ECT, n = 13) or sustained-contraction strength training (SCT, n = 14) interventions and in a
non-training control group (CON, n = 13). Symbols indicate between-group differences: ∗ SCT vs. CON, p<0.05; † ECT vs. SCT, trend 0.05 < p < 0.09. Letter
denotes effect size magnitude: M, moderate (0.5–0.8); L, large (>0.8). Data are group mean ± SD.

in both ECT (LSDHB p = 0.030, ES = 1.18) and SCT (LSDHB Patellar tendon Young’s modulus increased after ECT (paired
p = 0.034, ES = 0.73) were greater than CON. ECT and SCT had t-test p = 0.004, ES = 1.05), and SCT (p = 0.017, ES = 0.57),
a similar effect on patellar tendon stiffness (LSDHB p = 0.500, and was unchanged in CON (p = 0.637, ES = 0.05), resulting in
ES = 0.29). a group × time effect (Table 1). Absolute changes (Figure 7)

Frontiers in Physiology | www.frontiersin.org 10 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 6 | Patellar tendon force- elongation (A–C) and stress–strain (D–F) relationships pre (black diamonds) and post (gray squares) 12 weeks of
explosive-contraction [ECT, n = 13 (A,D)] or sustained-contraction [SCT, n = 15 (B,E)] strength training interventions and in an untrained control group [CON, n = 12
(C,F)]. Data are group mean ± SD. Data points are plotted at the elongation or strain corresponding to tendon force or stress at 10% increments of pre-training
maximum voluntary torque (MVT). Symbols indicate within-group difference ∗∗ p<0.01. Letter denotes effect size magnitude: M, medium (0.5–0.8).

Frontiers in Physiology | www.frontiersin.org 11 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 7 | Pre to post absolute changes (1) in (A) Patellar tendon elongation at 80% of pre-training maximum voluntary torque (MVT), (B) patellar tendon stiffness,
(C) patellar tendon Young’s modulus, in response to explosive-contraction (ECT, n = 13) or sustained-contraction (SCT, n = 15) strength training interventions and in
a non-training control group (CON, n = 12). Symbols indicate between-group differences: § ECT vs. CON p < 0.05; ∗ SCT vs. CON, p < 0.05; Letter denotes effect
size magnitude: M, moderate (>0.5–0.8); L, large (>0.8). Data are mean ± SD.

were greater in both ECT (LSDHB p = 0.012, ES = 1.38) and SCT increased tendon–aponeurosis complex stiffness (21%), and
SCT (LSDHB p = 0.042, ES = 0.75) than CON. Positive effects of quadriceps muscle volume (8%). There was a marginal effect of
ECT and SCT on tendon Young’s modulus were similar (LSDHB SCT on aponeurosis area (within-group increase, but no between
p = 0.830, ES = 0.21). group differences), while patellar tendon hypertrophy was not
Tendon–aponeurosis complex elongation at 80% pre-training clearly apparent after either SCT or ECT.
MVT increased after ECT (paired t-test p = 0.003, ES = 0.89) Sustained-contraction training increased high-force free
but was unchanged after SCT (p = 0.428, ES = 0.09) and CON tendon stiffness, as has been commonly reported in response
(p = 0.637, ES = 0.06), (Figure 8). There was a group × time effect to strength training regimes utilizing sustained (>2 s) high
(Table 1), with increases in ECT being greater than SCT (LSDHB force (>70% maximum) dynamic and/or isometric muscle
p = 0.021, ES = 1.23) and tended to be greater than CON (LSDHB contractions (e.g., Seynnes et al., 2009; Malliaras et al., 2013;
p = 0.098, ES = 0.80) (Figure 9). McMahon et al., 2013). A more original finding was the increase
Tendon–aponeurosis complex stiffness increased after SCT in free tendon stiffness after ECT, as this had not been investigated
(paired t-test p = 0.005, ES = 0.50) but was unchanged after ECT in previous studies (Burgess et al., 2007; Tillin et al., 2012).
(p = 0.938, ES = 0.02) and CON (p = 0.695, ES = 0.03,), with Intriguingly, ECT (+20%) was similarly effective as SCT (+16%)
a group × time effect (Table 1). Absolute changes in tendon– for stimulating increases in free tendon high-force stiffness, and
aponeurosis complex stiffness (Figure 9) following SCT were both increased by more than CON. The greater patellar tendon
greater than ECT (LSDHB p = 0.015, ES = 0.94) and CON (LSDHB stiffness after ECT and SCT can be explained by the parallel
p = 0.016, ES = 1.12), while ECT vs. CON changes were alike increase in patellar tendon Young’s modulus in response to
(LSDHB p = 0.846 ES = 0.02). training. This adaptation to SCT is consistent with multiple
previous studies (Seynnes et al., 2009; Malliaras et al., 2013;
McMahon et al., 2013) although the similar effect of ECT
DISCUSSION on free tendon Young’s modulus we have observed has not
been investigated before. Our findings support the view that
The present randomized controlled study compared the efficacy the changes in free tendon Young’s modulus is the primary
of 12 weeks of explosive- (ECT) vs. sustained- (SCT) contraction mechanism for the increased in tendon stiffness during the initial
strength training to increase patellar tendon stiffness and Young’s months of strength training (Wiesinger et al., 2015). Increased
modulus, knee extensor tendon–aponeurosis complex stiffness Young’s modulus after SCT and ECT may be due to changes
as well as elicit tissue (muscle, aponeurosis, free tendon) to the patellar tendon intrinsic collagenous structure and/or
hypertrophy. ECT and SCT similarly increased patellar tendon biochemical composition, e.g., increased collagen content, cross-
stiffness and modulus (20 and 22% vs. 16 and 16%), whereas only link density, fibril size (Buchanan and Marsh, 2002; Kjaer et al.,

Frontiers in Physiology | www.frontiersin.org 12 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

FIGURE 9 | Pre to post absolute changes (1) in (A) tendon–aponeurosis

complex elongation at 80% pre-training MVT and (B) tendon–aponeurosis
complex stiffness, in response to explosive-contraction (ECT, n = 13) or
sustained-contraction (SCT, n = 14) strength training interventions and in a
non-training control group (CON, n = 13). Symbols indicate between-group
differences: ∗ SCT vs. CON, p < 0.05; † ECT vs. SCT p < 0.05. Letter denotes
effect size magnitude: L, large (>0.8). Data are mean ± SD.

2015). At present evidence for specific alterations in free tendon

intrinsic structure/composition after strength training in healthy
individuals are lacking, and therefore further investigations to
uncover the mechanism(s) for the increases in Young’s modulus
are required.
The similar increases in patellar tendon Young’s modulus
after ECT and SCT may be attributable to their similar
loading magnitude (%MVT). It is recognized that in vitro
mechanotransduction responses of tenocytes (resident tendon
cells responsible for extracellular matrix remodeling) are highly
dependent on strain magnitude (Lavagnino et al., 2008) as
reflected by in vivo studies showing increased free tendon
stiffness and modulus only after high vs. low force strength
training (Kongsgaard et al., 2007; Arampatzis et al., 2010). The
similar changes to free tendon Young’s modulus after ECT
and SCT despite the previously documented (Balshaw et al.,
2016) differences in time related loading parameters with these
training regimes (loading rate, ECT 6-fold > SCT; loading
duration SCT 13-fold > ECT), strongly suggests that loading
magnitude, irrespective of duration or rate, is the primary
mechanostimulatory parameter for the free tendon.
In the present study, the increases in patellar tendon
stiffness in ECT and SCT were independent of free tendon
hypertrophy. Whilst it is curious there was a small within-
group decrease in mean patellar tendon CSA in CON, this
possible negative bias in post-training measures had only a
FIGURE 8 | Tendon force-tendon–aponeurosis complex elongation small effect size (0.27). Moreover, the primary between group
relationships pre (black diamonds) and post (gray squares) 12 weeks
explosive-contraction [ECT, n = 13 (A)] or sustained-contraction [SCT, n = 15
comparisons, that is the most robust indicator of training
(B)] strength training interventions and in a non-training control group [CON, effects in comparison to CON, revealed no between group
n = 13 (C)]. Data are group mean ± SD. Data points are plotted at the differences. Several earlier studies have similarly reported no
elongation corresponding to tendon forces at 10% increments of pre-training change in free tendon CSA after a comparable period of SCT
maximum voluntary torque (MVT). Within-group effect, tendon–aponeurosis
(Arampatzis et al., 2010; Kubo et al., 2012; Bloomquist et al.,
complex elongation at 80% pre-training MVT, post different to pre ∗∗ p < 0.01.
Letter denotes effect size magnitude: L, large (>0.8).
2013). However, others have reported small increases in free
tendon CSA following similar SCT regimes (∼3–6%: Arampatzis

Frontiers in Physiology | www.frontiersin.org 13 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

et al., 2007; Kongsgaard et al., 2007; Seynnes et al., 2009; Bohm complex stiffness (Farcy et al., 2013). Tillin et al. (2012) trained
et al., 2014). With regards to our patellar tendon mean CSA data their participants at a longer muscle length (knee joint angle 85◦
it is unlikely that our measurements simply failed to detect a vs. 115◦ in the current study), which has been shown to result
change. Pre and post free tendon CSA analysis was performed by in greater increases in knee extensor tendon–aponeurosis which
a single investigator blinded to the group allocation, and involved has been shown to result in greater increases in knee extensor
precise measurements of tendon CSA along the full length of tendon–aponeurosis complex stiffness (Kubo et al., 2006b) in
the tendon from high resolution MRI (2 mm slice thickness, accordance with high force development in conditions of higher
pixel size 0.313 mm × 0.313 mm), with excellent reproducibility tissue strain magnitude (McMahon et al., 2013), and this could
even over the duration of the intervention (∼3% pre–post CVw explain their contrasting findings of increased knee extensor
in CON). It is possible the magnitude of tendon hypertrophy tendon–aponeurosis complex stiffness.
after relatively short-term resistance training is small, and on the An interesting observation was that the force–elongation
borderline of what can be detected. Importantly, however, we relationship post ECT was actually shifted to the right (greater
recently found no evidence for free tendon hypertrophy in long- elongation at specific forces). The increase in elongation in
term (4 years) resistance trained men, despite their substantially response to the same high force after ECT was greater than
greater muscle volume (56%) and strength (58%) compared to after SCT and tended to be greater than the CON group.
untrained controls (Massey et al., 2017). Based on those findings The rightward shift in the force–elongation curves after ECT
and the current results it seems unlikely that high-load resistance appears to result from a change in elongation at the initial
training causes tendon hypertrophy even after months and years level (10%MVT), that persists throughout the rise in tendon
of training. force, as after 10%MVT the gradients of the force–elongation
Moreover, the lack of free tendon hypertrophy after strength relationships pre–post ECT are equivalent. Consistent with
training in the current study is consistent with some evidence our data, there is some evidence that sprint trained athletes
that resistance exercise/training may not noticeably stimulate (who inherently utilize explosive contractions) display greater
increased in vivo collagen protein synthesis. For instance, an knee extensor tendon–aponeurosis complex elongation at the
acute bout of high load dynamic knee extensor contractions lowest levels of force (<20%MVT), with resultantly greater
(3 × 10 repetitions, 70% 1 repetition maximum) had no effect on elongation throughout the measured force range (Kubo et al.,
patellar tendon collagen type I messenger RNA expression 24 h 2000, 2011). It is possible that a reduction in low force tendon
post exercise (Sullivan et al., 2009). Also, 12 weeks of isoinertial aponeurosis complex stiffness (i.e., 0–10% MVT) after ECT
squat training failed to increase the concentration of procollagen with no changes at higher forces indicates changes in tissue
type 1 N-propeptide (biomarker of collagen synthesis) in patellar collagenous structure/composition that specifically influence the
tendon peritendinous tissue [Bloomquist et al., 2013; this study lower region of the force–elongation relationship. In contrast,
also observed no change in patellar tendon CSA (via MRI)]. whilst SCT increased high force stiffness there was no clear
Contrarily there is some evidence that mechanical loading of leftward shift in the force–elongation curve. Indeed, some
free tendon tissue can induce an increased collagen synthesis previous studies have concordantly reported an increase in
(Miller et al., 2005) although it is not a consistent finding high force tendon–aponeurosis complex stiffness, along with
(Dideriksen et al., 2013). Therefore mechanical loading in vivo no apparent effect on the elongation at lower force levels
may not necessarily stimulate a sufficiently robust induction of (Kubo et al., 2001, 2010). These results perhaps imply that SCT
the appropriate biochemical response needed to elicit free tendon may induce tissue collagenous structure/composition changes
hypertrophy. that specifically impact the high stiffness region of the force–
In contrast to the free tendon, the tendon–aponeurosis elongation relationship (e.g., collagen cross-links: Kjaer et al.,
complex stiffness measured at high force levels (i.e., 70–80% 2015). Further work is needed to fervently elucidate whether
pre-training MVT) increased only after SCT, but not ECT. force level specific changes in stiffness are likely to occur with
The increased tendon–aponeurosis complex high force stiffness different interventions, and identify any possible mechanistic
after SCT is consistent with previous findings (Kubo et al., basis for this supposition.
2001; Arampatzis et al., 2007, 2010; Bohm et al., 2014) and the Collectively our findings show that in comparison to a
greater increase after SCT than ECT may be attributable to the control intervention patellar tendon stiffness but not tendon–
substantially longer loading duration in SCT. Previous work has aponeurosis complex stiffness increased after ECT, whereas
shown greater increases in tendon–aponeurosis complex stiffness SCT increased both patellar tendon and tendon–aponeurosis
after strength training with long vs. short duration contractions complex stiffness, indicating a differential adaptive response of
(Kubo et al., 2001; Arampatzis et al., 2007). The absence of change these tendinous tissues according to the training regime. The
in tendon–aponeurosis complex stiffness for ECT in the current contrasting patellar tendon and tendon–aponeurosis complex
study contrasts with earlier studies examining the triceps surae stiffness changes after ECT demonstrates the independence
(Burgess et al., 2007) and knee extensors (Tillin et al., 2012). It of these adaptations. The simple observation that only a
is possible that our results diverge from Burgess et al. (2007), small proportion of tendon–aponeurosis complex elongation
because an increase in free tendon stiffness as we have observed is due to the patellar tendon elongation (19%) further
after ECT, may be of greater consequence to the triceps surae highlights the distinction of these measures. From our study
tendon–aponeurosis complex, as the Achilles tendon accounts we cannot discount a contribution of the quadriceps tendon
for a larger proportion of the triceps surae tendon–aponeurosis and vastus lateralis extramuscular tendon to tendon–aponeurosis

Frontiers in Physiology | www.frontiersin.org 14 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

complex stiffness because the fascicle-aponeurosis intersection of the aponeurosis (Zuurbier et al., 1994). Training-induced
displacement reflects elongation in all tendinous tissues distal changes in muscle morphology and architecture, as well as
to the tracked point (Stafilidis et al., 2005). However, from our neural recruitment strategy along the muscle length, may have
data and previous measures of vastus lateralis myotendinous influenced muscle–aponeurosis interaction and thus aponeurosis
junction and aponeurosis elongation (Stafilidis et al., 2005), the behavior during contraction, conceivably confounding the
muscle aponeurosis apparently comprises the most influential interpretation of differences in tendon–aponeurosis stiffness pre–
component of tendon–aponeurosis elongation and stiffness. post intervention. However, at present we are not aware of a
The tendon–aponeurosis complex stiffness changes after SCT better technique for investigating the mechanical behavior of the
could reflect adaptations (material properties and/or size) of tendon–aponeurosis complex.
the aponeurosis component of the tendon–aponeurosis complex,
and there was some indication of increased aponeurosis area
after SCT (+7% within-group change, but insufficient for a CONCLUSION
between group effect), that could conceivably have contributed
to the increased tendon–aponeurosis complex stiffness after In conclusion, ECT was equally effective as SCT for stimulating
SCT. Aponeurosis hypertrophy is thought to be necessary to an increase in patellar tendon stiffness and Young’s modulus,
provide an enlarged attachment area for an increased muscle demonstrating that in order to induce free tendon adaptation,
CSA (Wakahara et al., 2015), thus our finding is consistent strength training need only involve brief, high force muscle
with the similar hypertrophic response of the quadriceps contractions. However, brief high force muscle contractions
femoris muscle (+8%) after SCT and not ECT (or CON). are not solely sufficient to stimulate muscle and aponeurosis
The muscle hypertrophic response to SCT but not ECT is adaptations as only SCT increased tendon–aponeurosis complex
most likely a consequence of the greater total loading duration stiffness, muscle size, and aponeurosis size, while ECT was
with SCT. Following bouts of isoinertial knee extensions ineffective. Thus our results suggest muscle–aponeurosis
with equivalent load, a greater total loading duration was adaptations are specific to the loading regime and sensitive to
associated with increased acute amplitude of muscle myofibrillar loading duration.
protein synthesis (Burd et al., 2012). Therefore, the limited
total loading duration in ECT is perhaps an insufficient
stimulus for the necessary muscle protein synthesis, and likely AUTHOR CONTRIBUTIONS
accounts for the lack of muscle hypertrophy in response
to this training modality. Although it should be recognized GM, TB, TM-W, NT, and JF conceived and designed the study.
that overall muscle volume is a relatively gross measure that GM, TB, and TM-W performed the experiments. GM, TB,
may not capture regional remodeling or hypertrophy within TM-W, and NT analyzed the data. GM and JF interpreted the
specific regions of the muscle according to localized mechanical data and drafted the manuscript. TB, TM-W, and NT critically
stimuli. evaluated the manuscript. All authors are responsible for the final
A potential limitation of our study concerns the methodology content of the manuscript.
for determining tendon–aponeurosis mechanical properties,
even though it has been used very extensively (Kubo et al.,
2001, 2006b, 2009; Bojsen-Møller et al., 2005; Tillin et al., 2012). FUNDING
In addition to the patellar tendon, which we have assessed,
the contribution of other intermediary tendinous tissues (i.e., This study was financially supported by the Arthritis Research
quadriceps and vastus lateralis tendon), to tendon–aponeurosis UK Centre for Sport, Exercise and Osteoarthritis (Grant reference
complex elongation appears relatively small (Stafilidis et al., 20194).
2005), but has limited attention. The measurement of tendon–
aponeurosis complex elongation could also be influenced by the
active state of muscle fibers in parallel with the aponeurosis. ACKNOWLEDGMENTS
Aponeurosis stiffness is considered muscle-activation dependent
as muscle fibers anchor the aponeurosis during contraction The authors thank Clare Appleby, Antonio Morales, and Alex
(Lieber et al., 2000), and is also modulated by muscle McKeown for their assistance during laboratory measurement
deformation during contraction (Azizi and Roberts, 2009) and training sessions and participants for their time in taking part
as well as the relative force distribution along the length in the study.

REFERENCES Arampatzis, A., Peper, A., Bierbaum, S., and Albracht, K. (2010). Plasticity of
human Achilles tendon mechanical and morphological properties in response
Arampatzis, A., Karamanidis, K., and Albracht, K. (2007). Adaptational to cyclic strain. J. Biomech. 43, 3073–3079. doi: 10.1016/j.jbiomech.2010.
responses of the human Achilles tendon by modulation of the applied 08.014
cyclic strain magnitude. J. Exp. Biol. 210, 2743–2753. doi: 10.1242/jeb.00 Azizi, E., and Roberts, T. J. (2009). Biaxial strain and variable stiffness in
3814 aponeurosis. J. Physiol. 587, 4309–4318. doi: 10.1113/jphysiol.2009.173690

Frontiers in Physiology | www.frontiersin.org 15 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

Balshaw, T. G., Massey, G. J., Maden-Wilkinson, T. M., Tillin, N. A., and Folland, Kubo, K., Ikebukuro, T., Yaeshima, K., Yata, H., Tsunoda, N., and Kaneshisa, H.
J. P. (2016). Training-specific functional, neural and hypertrophic adaptations (2009). Effects of static and dynamic training on the stiffness and blood volume
to explosive- vs. sustained-contraction strength training. J. Appl. Physiol. 120, of tendon in vivo. J. Appl. Physiol. 106, 412–417. doi: 10.1152/japplphysiol.
1364–1373. doi: 10.1152/japplphysiol.00091.2016 91381.2008
Blazevich, A. J., Coleman, D. R., Horne, S., and Cannavan, D. (2009). Anatomical Kubo, K., Ikebukuro, T., Yata, H., Tomita, M., and Okada, M. (2011).
predictors of maximum isometric and concentric knee extensor moment. Eur. Morphological and mechanical properties of muscle and tendon in highly
J. Appl. Physiol. 105, 869–878. doi: 10.1007/s00421-008-0972-7 trained sprinters. J. Appl. Biomech. 27, 336–344. doi: 10.1123/jab.27.4.336
Bloomquist, K., Langberg, H., Karlsen, S., Madsgaard, S., Boesen, M., and Kubo, K., Ikebukuro, T., Yata, H., Tsunoda, N., and Kaneshisa, H. (2010).
Raastad, T. (2013). Effect of range of motion in heavy load squatting on muscle Time course of changes in muscle and tendon properties during strength
and tendon adaptations. Eur. J. Appl. Physiol. 113, 2133–2142. doi: 10.1007/ training and detraining. J. Strength Cond. Res. 24, 322–331. doi: 10.1519/JSC.
s00421-013-2642-7 0b013e318c865e2
Bohm, S., Mersmann, F., and Arampatzis, A. (2015). Human tendon adaptation Kubo, K., Kanehisa, H., and Fukunaga, T. (2001). Effect of different duration
in response to mechanical loading: a systematic review and meta-analysis of isometric contractions on tendon elasticity in human quadriceps muscles.
exercise intervention studies on healthy adults. Sports Med. Open. 1:7. doi: J. Physiol. 536, 639–655. doi: 10.1111/j.1469-7793.2001.0649c.xd
10.1186/s40798-015-0009-9 Kubo, K., Kanehisa, H., Kawakami, Y., and Fukunaga, T. (2000). Elasticity of
Bohm, S., Mersmann, F., Tettke, M., Kraft, M., and Arampatzis, A. (2014). Human tendon structures of the lower limbs in sprinters. Acta Physiol. Scand. 168,
Achilles tendon plasticity in response to cyclic strain: effect of strain rate and 327–335. doi: 10.1046/j.1365-201x.2000.00653.x
duration. J. Exp. Biol. 217, 4010–4017. doi: 10.1242/jeb.112268 Kubo, K., Lomuro, T., Ishihuro, N., Tsunoda, N., Sato, Y., Ishii, N., et al.
Bojsen-Møller, J., Magnusson, S. P., Rasmussen, L. R., Kjaer, M., and (2006a). Effects of low-load resistance training with vascular occlusion on the
Aagaard, P. (2005). Muscle performance during maximal isometric and mechanical properties of muscle and tendon. J. Appl. Biomech. 22, 112–119.
dynamic contractions is enhanced by the stiffness of the tendinous structures. doi: 10.1123/jab.22.2.112
J. Appl. Physiol. 99, 986–994. doi: 10.1152/japplphysiol.01305.2004 Kubo, K., Ohgo, K., Takeishi, R., Yoshinaga, K., Tsunoda, N., Kanehisa, H., et al.
Buchanan, C. I., and Marsh, R. L. (2002). Effects of exercise on the biomechanical, (2006b). Effects of isometric training at different knee angles on the muscle-
biochemical and structural properties of tendons. Comp. Biochem. Physiol. tendon complex in vivo. Scand. J. Med. Sci. Sports 16, 159–167. doi: 10.1111/j.
A Mol. Integr. Physiol. 133, 1101–1107. doi: 10.1016/S1095-6433(02)00 1600-0838.2005.00450.x
139-3 Kubo, K., Yata, H., Kaneshisa, H., and Fukunaga, T. (2006c). Effect of isometric
Burd, N. A., Andrews, R. J., West, D. W. D., Little, J. P., Cochran, A. J. R., squat training on the tendon stiffness and jump performance. Eur. J. Appl.
Hector, A. J., et al. (2012). Muscle time under tension during resistance exercise Physiol. 96, 305–314. doi: 10.1007/s00421-005-0087-3
stimulates differential muscle protein sub-fractional synthetic responses in men. Lakens, D. (2013). Calculating and reporting effect sizes to facilitate cumulative
J. Physiol. 590, 351–362. doi: 10.1113/jphysiol.2011.221200 science: a practical primer for t-tests and ANOVAs. Front. Psychol. 4:863. doi:
Burgess, K. E., Connick, M. J., Graham-Smith, P., and Pearson, S. J. (2007). 10.3389/fpsyg.2013.00863
Plyometric vs. isometric training influences on tendon properties and muscle Lavagnino, M., Arnoczky, S. P., Kepich, E., Calballero, O., and Haut, R. C. (2008).
output. J. Strength Cond. Res. 21, 986–989. doi: 10.1519/R-20235.1 A finite element model predicts the mechanotransduction response of tendon
Dideriksen, K., Sindby, A. K., Krogsgaard, M., Schjerling, P., Holm, L., and cells to cyclic tensile loading. Biomech. Model. Mechanobiol. 7, 405–416. doi:
Langberg, H. (2013). Effect of acute exercise on patella tendon protein synthesis 10.1007/s10237-007-0104-z
and gene expression. Springerplus 2:109. doi: 10.1186/2193-1801-2-109 Lieber, R. L., Leonard, M. E., and Brown-Maupin, C. G. (2000). Effects of muscle
Farcy, S., Nordez, A., Dorel, S., Hauraix, H., Portero, P., and Rabita, G. (2013). contraction on the load-strain properties of frog aponeurosis and tendon. Cells
Interaction between gastrocnemius medialis fascicle and Achilles tendon Tissues Organs 166, 48–54. doi: 10.1159/000016708
compliance: a new insight on the quick-release method. J. Appl. Physiol. 116, Lipps, D. B., Oh, Y. K., Ashton-Miller, J. A., and Wojtys, E. M. (2014). Effect of
259–266. doi: 10.1152/japplphysiol.00309.2013 increases quadriceps tensile stiffness on peak anterior cruciate ligament strain
Folland, J. P., and Williams, A. G. (2007). The adaptations to strength training: during a simulated pivot landing. J. Orthop. Res. 32, 423–430. doi: 10.1002/jor.
morphological and neurological contributions to increased strength. Sports 22531
Med. 37, 145–168. doi: 10.2165/00007256-200737020-00004 Malliaras, P., Kamal, B., Nowell, A., Farley, T., Dhamu, H., Simpson, V., et al.
Hansen, P., Bojsen-Møller, J., Aagaard, P., Kjaer, M., and Magnusson, S. P. (2006). (2013). Patellar tendon adaptation in relation to load-intensity and contraction
Mechanical properties of the human patellar tendon, in vivo. Clin. Biomech. 21, type. J. Biomech. 46, 1893–1899. doi: 10.1016/j.jbiomech.2013.04.022
54–58. doi: 10.1016/j.clinbioech.2005.07.008 Massey, G. J., Balshaw, T. G., Maden-Wilkinson, T. M., and Folland, J. P. (2017).
Karamanidis, K., Arampatzis, A., and Mademli, L. (2008). Age-related deficit in Tendinous tissue properties after short and long-term functional overload:
dynamic stability control after forward falls is affected by muscle strength and differences between controls, 12 weeks and 4 years of resistance training. Acta
tendon stiffness. J. Electromyogr. Kinesiol. 18, 980–989. doi: 10.1016/j.jelekin. Physiol. 222:e13019. doi: 10.1111/apha.13019
2007.04.003 McMahon, G. E., Morse, C. I., Burden, A., Winwood, K., and Onambélé-Pearson,
Kellis, E., and Baltzopoulos, V. (1999). In vivo determination of the patella tendon G. L. (2013). The manipulation of strain, when stress is controlled, modulates
and hamstrings moment arms in adult males using videofluoroscopy during in vivo tendon mechanical properties but not systematic TGF-B1 levels. Physiol.
submaximal knee extension. Clin. Biomech. 14, 118–124. doi: 10.1016/S0268- Rep. 1:e00091. doi: 10.1002/phy2.91
0033(98)00055-2 Miller, B. F., Olesen, J. L., Hansen, M., Døssing, S., Crameri, R. M., Welling, R. J.,
Kjaer, M., Jørgensen, N. R., Heinemeier, K., and Magnusson, S. P. (2015). Exercise et al. (2005). Coordinated collagen and muscle protein synthesis in human
and regulation of bone and collagen tissue biology. Prog. Mol. Biol. Transl. Sci. patella tendon and quadriceps muscle after exercise. J. Physiol. 567, 102–133.
135, 259–291. doi: 10.1016/bs.pmbts.2015.07.008 doi: 10.1113/jphysiol.2005.093690
Kongsgaard, M., Reitelseder, S., Pedersen, T. G., Holm, L., Aagaard, P., Kjaer, M., Raiteri, B. J., Cresswell, A. G., and Lichtwark, G. A. (2016). Three-dimensional
et al. (2007). Region specific patellar tendon hypertrophy in humans following geometrical changes of the human tibialis anterior muscle and its central
resistance training. Acta Physiol. 191, 111–121. doi: 10.1111/j.1748-1716.2007. aponeurosis measured witch three-dimensional ultrasound during isometric
01714.x contractions. PeerJ 4:e2260. doi: 10.7717/peerj.2260
Kösters, A., Wiesinger, H. P., Bojsen-Møller, J., Müller, E., and Seynnes, O. (2014). Reeves, N. D., Maganaris, C. N., and Narici, M. V. (2003). Effect of strength training
Influence of loading rate on patellar tendon mechanical properties in vivo. Clin. on human patella tendon mechanical properties of older individuals. J. Physiol.
Biomech. 29, 323–329. doi: 10.1016/j.clinbiomech.2013.12.010 548, 971–981. doi: 10.1113/jphysiol.2002.035576
Kubo, K., Ikebukuro, T., Maki, A., Yata, H., and Tsunoda, N. (2012). Time Reid, K. F., Martin, K. I., Doros, G., Clark, D. J., Hau, C., Patten, C., et al. (2015).
course of changes in the human Achilles tendon properties and metabolism Comparative effects of light or heavy resistance power training for improving
during training and detraining in vivo. Eur. J. Appl. Physiol. 112, 2679–2691. lower extremity power and physical performance in mobility-limited older
doi: 10.1007/s00421-011-2248-x adults. J. Gerontol. A Biol. Sci. Med. Sci. 70, 374–380. doi: 10.1093/gerona/glu156

Frontiers in Physiology | www.frontiersin.org 16 September 2018 | Volume 9 | Article 1170

Massey et al. Tendon Adaptation to Strength Training

Salaffi, F., Carotti, M., Stancati, A., and Grassi, W. (2005). Health-related quality of Wakahara, T., Ema, R., Miyamoto, N., and Kawakami, Y. (2015). Increase in vastus
life in older adults with symptomatic hip and knee osteoarthritis: a comparison lateralis aponeurosis width induced by resistance training: implications for a
with matched healthy controls. Aging Clin. Exp. Res. 17, 255–263. doi: 10.1016/ hypertrophic model of pennate muscle. Eur. J. Appl. Physiol. 115, 309–316.
0021-9290(94)90020-5 doi: 10.1007/s00421-014-3012-9
Seynnes, O. R., Bojsen-Møller, J., Albracht, K., Arndt, A., Cronin, N. J., Finni, T., Wiesinger, H. P., Kösters, A., Müller, E., and Seynnes, O. R. (2015). Effects
et al. (2014). Ultrasound-based testing of tendon mechanical properties: a of increased loading on in vivo tendon properties: a systematic review.
critical evaluation. J. Appl. Physiol. 118, 133–141. doi: 10.1152/japplphysiol. Med. Sci. Sports Exerc. 47, 1885–1895. doi: 10.1249/MSS.000000000000
00849.2014 0603
Seynnes, O. R., Erskine, R. M., Maganaris, C. N., Longo, S., Simoneau, E. M., Zuurbier, C. J., Everard, A. J., Van Der Wees, P., and Huijing, P. A. (1994).
Grosset, J. F., et al. (2009). Training-induced changes in structural and Length-force characteristics of the aponeurosis in the passive and active muscle
mechanical properties of the patellar tendon are related to muscle hypertrophy condition and in the isolated condition. J. Biomech. 27, 445–453. doi: 10.1016/
but not to strength gains. J. Appl. Physiol. 107, 523–530. doi: 10.1152/ 0021-9290(94)90020-5
japplphysiol.00213.2009
Stafilidis, S., Karamanidis, K., Morey-Klapsing, G., DeMonte, G., Brüggemann, Conflict of Interest Statement: The authors declare that the research was
G. P., and Arampatzis, A. (2005). Strain and elongation of the vastus lateralis conducted in the absence of any commercial or financial relationships that could
aponeurosis and tendon in vivo during maximal isometric contraction. Eur. J. be construed as a potential conflict of interest.
Appl. Physiol. 94, 317–322. doi: 10.1007/s00421-004-1301-4
Sullivan, B. E., Carroll, C. C., Jemiolo, B., Trappe, S. W., Magnusson, S. P., Copyright © 2018 Massey, Balshaw, Maden-Wilkinson, Tillin and Folland. This is an
Døssing, S., et al. (2009). Effect of acute resistance exercise and sex on human open-access article distributed under the terms of the Creative Commons Attribution
patellar tendon structural and regulatory mRNA expression. J. Appl. Physiol. License (CC BY). The use, distribution or reproduction in other forums is permitted,
106, 468–475. doi: 10.1152/japplphysiol.91341.2008 provided the original author(s) and the copyright owner(s) are credited and that the
Tillin, N. A., Pain, M. T., and Folland, J. P. (2012). Short-term training for explosive original publication in this journal is cited, in accordance with accepted academic
strength causes neural and mechanical adaptations. Exp. Physiol. 97, 630–641. practice. No use, distribution or reproduction is permitted which does not comply
doi: 10.1113/expphysiol.2011.063040 with these terms.

Frontiers in Physiology | www.frontiersin.org 17 September 2018 | Volume 9 | Article 1170