Sports Medicine

https://doi.org/10.1007/s40279-018-0956-7

SYSTEMATIC REVIEW

Are the Mechanical or Material Properties of the Achilles

and Patellar Tendons Altered in Tendinopathy? A Systematic Review
with Meta‑analysis
Steven J. Obst1   · Luke J. Heales2 · Benjamin L. Schrader1 · Scott A. Davis1 · Keely A. Dodd1 · Cory J. Holzberger1 ·
Louis B. Beavis1 · Rod S. Barrett3,4

© Springer International Publishing AG, part of Springer Nature 2018

Abstract
Background  Changes in the mechanical behaviour of the Achilles and patellar tendons in tendinopathy could affect muscle
performance, and have implications for injury prevention and rehabilitation strategies.
Objectives  To determine the effect of clinically diagnosed tendinopathy on the mechanical and material properties of the
Achilles tendon (AT) and patellar tendon (PT).
Design  Systematic review with meta-analysis.
Methods  A search of electronic databases (SPORTDiscus, CINAHL, PubMed, ScienceDirect and Google Scholar) was
conducted to identify research articles that reported local and global in vivo mechanical (e.g. strain, stiffness) and/or mate-
rial properties (e.g. modulus) of the AT and/or PT in people with and without tendinopathy. Effect sizes and corresponding
95% confidence intervals for individual studies were calculated for tendon strain, stiffness, modulus and cross-sectional area.
Results  Eighteen articles met the inclusion criteria (AT only = 11, PT only = 5, AT and PT = 2). There was consistent evidence
that the reported AT strain was higher in people with tendinopathy, compared to asymptomatic controls. People with Achilles
tendinopathy had a lower AT global stiffness, lower global modulus and lower local modulus, compared to asymptomatic
controls. In contrast, there was no clear and consistent evidence that the global or local mechanical or material properties
of the PT are altered in tendinopathy.
Conclusions  The in vivo mechanical and material properties of the Achilles tendon-aponeurosis are altered in tendinopathy,
compared to asymptomatic tendons. Despite a similar clinical presentation to Achilles tendinopathy, patellar tendinopathy
does not appear to alter the tensile behaviour of the PT in vivo.

Key Points 

The mechanical and material properties of the Achilles

tendon-aponeurosis and patellar tendon (PT) appear to
be differentially affected by tendinopathy.
Clinical interventions to restore tendon strength and
* Steven J. Obst mechanical function of the tendinopathic Achilles
[email protected] tendon-aponeurosis appear to be warranted.
1
School of Health, Medicine and Applied Science, Central The lack of evidence for a consistent effect of tendi-
Queensland University, Bundaberg, QLD, Australia nopathy on the mechanical and/or material properties
2
School of Health, Medicine and Applied Science, Central of the PT raises the possibility that the therapeutic goal
Queensland University, Rockhampton, QLD, Australia for treating PT tendinopathy may differ from that for the
3
School of Allied Health Sciences, Griffith University, Achilles tendon.
Gold Coast, QLD, Australia
4
Menzies Health Institute Queensland, Griffith University,
Gold Coast, QLD, Australia

Vol.:(0123456789)
 S. J. Obst et al.

1 Introduction compared to the AT and mechanically better suited for

effective transmission of muscle force rather than storage
Tendinopathy is an umbrella term used to describe the and release of elastic strain energy [28].
clinical syndrome of pain and dysfunction arising from While the mechanical and material properties of the
tendon injury [1, 2]. In the majority of cases, the tendon healthy AT and PT appear to respond similarly to changes
injury is attributed to repetitive mechanical loading that in tendon homeostasis during periods of increased (e.g.
causes damage accumulation within the tendon consist- strength training [29]) or decreased (e.g. immobilisation)
ent with degeneration (i.e. tendinosis) [2]. Tendinosis is loading, or with ageing [30], it is not known whether similar
characterised by structural changes that are observable changes occur in the presence of degenerative tendinopa-
with ultrasound imaging, including diffuse or focal ten- thy. Identifying whether the mechanical behaviour of the
don thickening or calcification, irregular fibre hypertrophy, Achilles and patellar tendons are altered in tendinopathy is
increased vascularity and blood flow [3], and increased important for several reasons. Firstly, from a muscle–ten-
cross-sectional area, thickness and volume [4–6], com- don unit performance perspective, altered tendon stiffness
pared to healthy tendons. At the microscopic level, ten- could change the force–length relationship of the muscle
dinopathic tendons contain abnormal collagen fibre in-series, and either positively or negatively impact on the
composition and structure, including increased concen- rate, magnitude and efficiency of muscle force production
tration of type III collagen, increased ground substance [31]. Lower tendon stiffness for a muscle that operates on
and glycosaminoglycan content, and increased number either the descending or ascending limb of the force–length
of abnormal tenocytes [3, 7–12]. It is therefore evident curve would result in either more or less contractile force,
that tendinosis negatively affects both the solid and fluid respectively. Similarly, lower tendon stiffness in a compli-
components of the tendon load-bearing matrix [6]. What ant tendon such as the Achilles tendon may in fact reflect a
remains unclear is how these changes affect the mechani- positive adaptation that optimises, or fine-tunes, the com-
cal behaviour of the involved tendon. It is often assumed, pliance of the whole muscle–tendon unit to accommodate
perhaps incorrectly, that structural changes in tendinosis high power output movements that rely on the stretch short
result in a mechanically ‘weaker’ tendon in vivo [13, 14], cycle, such as running and jumping [17, 32–34]. For these
independent of the type of tendon, and therefore interven- reasons, the impact that altered tendon compliance has on
tions that target tendon remodeling are effective because muscle performance should be viewed in the context of the
they ‘strengthen’ the tendon [7, 8]. Such assumptions may functional requirements of the muscle–tendon unit, and may
be not be appropriate for all tendinopathies, and exercise not necessarily reflect a negative adaptation. Secondly, from
interventions may lead to inconsistent treatment outcomes. an injury-risk perspective, a lower tendon stiffness could
The Achilles tendon (AT) is the largest and strongest result in a higher strain for a given stress, and increase the
human tendon, yet it is also the most commonly injured risk of strain-induced damage accumulation or rupture
in both athletic and sedentary populations [15]. The AT via a reduction in the safety factor of the injured tendon
is considered a compliant tendon as it undergoes large [35]. Lower stiffness could also theoretically increase the
changes in length for a given tensile load and therefore amount of energy lost as heat and contribute to thermal dam-
has an important role in storing and releasing elastic age accumulation during repeated cyclic loading [31, 36];
strain energy during locomotion [16–18]. Because of its however, research has yet to confirm a correlation between
function, the AT, in particular the free tendon portion, is tendon stiffness and indices of tendon energy dissipation
vulnerable to repetitive strain-induced changes [18–23]. [37]. Altered mechanical and/or material properties could,
Age-related reduction in AT stiffness and modulus [24], therefore, accelerate an already failed healing response and
combined with increased damage accumulation with run- perpetuate further tendon degeneration. Finally, from a reha-
ning, could explain the high incidence rates of Achilles bilitation perspective, this information could improve exer-
tendinopathy (up to 50%) reported in middle-aged male cise prescription for the prevention and/or treatment of ten-
long distance runners [25] and compared to other lower dinopathy, by better matching the exercise mode, dose and
limb tendons, such as the patellar tendon (PT). In con- timing with the mechanical state of the target tendon [29].
trast to the AT, tendinopathy of the PT is more common The aim of this systematic review and meta-analysis was
in younger athletes who engage in sports that involve to determine the effect of clinically diagnosed tendinopathy
repeated jumping, such as volleyball, netball and bas- on the in vivo mechanical and material properties of the
ketball [26]. The PT is designed to transmit large tensile Achilles and patellar tendons.
forces from the quadriceps muscle group, via the patel-
lar pulley mechanism, to control the tibio-femoral joint
motion [27]. Morphologically, the PT is shorter and stiffer
Mechanical and material properties of lower limb tendinopathy

2 Methods reliability and validity [39]. Each item was scored 1 (Yes),
0.5 (Partial, lacks specific details; or, only some defined) or
2.1 Search Strategy 0 (No). Overall quality of the included studies was assessed
by summing the response to each item, resulting in a total
The review was conducted according to the PRISMA possible score of 14. Any discrepancies between review-
guidelines for performing systematic reviews [38]. Data- ers were discussed and resolved via consensus. The level of
base searches were conducted by two reviewers (S.J.O. agreement between reviewers was determined using Cohen’s
and S.A.D.) in May 2017 with the final search conducted kappa statistic [40].
in November 2017. The following databases were searched
for articles that examined in vivo tendon mechanical and/ 2.4 Data Extraction and Analysis
or material properties in humans diagnosed with Achil-
les or patellar tendinopathy: SPORTDiscus via EBSCO Means, standard deviations and sample sizes for tendi-
(1984–current), CINAHL via EBSCO (2002–current), nopathy and asymptomatic control groups for each out-
PubMed (1950–current), ScienceDirect (1950–current) come measure were used to compute standardized effect
and Google Scholar. The search string used was as follows: size (ES) (Cohen’s d) and corresponding 95% confidence
(Tendon* or tendin*) AND (mechanical OR stiffness OR intervals (CIs) [41]. Standardized ES are dimensionless and
elasticity OR modulus OR strain OR plasticity OR cross- account for differences in measurement units and sample
sectional area) AND (patella* tendon OR quadriceps tendon sizes between included studies. A positive mean ES reflects
OR Achilles OR gastroc* OR triceps surae) AND (in vivo); an increase in the outcome for the tendinopathic group,
where * donates a truncation to ensure all variations of the compared to the asymptomatic control group. If the 95%
term are included. CI of the ES did not cross the zero line, the outcome was
considered statistically significant [42]. Effect size data were
grouped based on outcome measure (strain, stiffness and
2.2 Study Selection modulus) and measurement level (global and local), and
presented as forest plots to facilitate comparison between
Retrieved references were imported into Endnote (version tendinopathy and control groups. The terms ‘global’ and
X8.1, Thomson Reuters, Carlsbad, CA, USA) and duplicate ‘local’ are used to distinguish between mechanical measure-
articles were removed. Two reviewers (S. J. O. and C. J. H.) ments made at the level of the whole tendon (i.e. using ultra-
independently screened titles and abstracts. Where insuffi- sound and dynamometry), and measurements made within
cient information was available in the title and abstract, the small regions of the tendon (i.e. using elastography-based
full-text version was screened. Articles that did not meet methods). Pooled ES data were not calculated for any tendon
inclusion criteria during the initial screen were excluded. outcome measure, due to the considerable heterogeneity in
Reference lists of eligible articles were also searched and sample characteristics (e.g. sex, age, duration and severity
additional articles screened for eligibility. Full journal arti- of symptoms) and measurement methods reported in the
cles reporting global and/or local in vivo mechanical (e.g. included studies. All data analyses, statistical analyses and
strain, stiffness), material properties (e.g. modulus) and/or plotting were performed in Matlab (version R2017a, The
morphological (e.g. cross-sectional area, diameter) in human MathWorks, Natick, MA, USA).
participants with clinically diagnosed Achilles and/or patel-
lar tendinopathy in comparison to asymptomatic controls
(matched or contralateral limb) were included. Articles were 3 Results
excluded if they: (1) did not report baseline (i.e. pre-inter-
vention) data for intervention studies; (2) did not include a 3.1 Database Search
suitable asymptomatic control group (e.g. contralateral limb
or control group); or (3) included participants with a recent The initial database search identified 643 potentially eligible
history of tendon surgery. articles. After removal of duplicates (n = 234) and articles
that did not meet the inclusion criteria (n = 363) based on
title and abstract screening, 46 articles remained (Fig. 1).
2.3 Quality Assessment A further 35 articles were removed based on the exclusion
criterion, leaving 11 full text articles. An additional refer-
Included articles were assessed for methodological quality ence list screening of these articles yielded a further seven
by two reviewers (S. A. D. and C. J. H.) using a modified eligible articles, leaving a final yield of 18 articles for data
qualitative review tool derived from Galna et al. [39]. The extraction, 11 of which reported only AT mechanical and/
tool includes 14 items, which assess internal and external or material properties [6, 14, 43–51], five reported only PT
 S. J. Obst et al.

Fig. 1  Flow chart showing study inclusion. The data in the heading boxes represent the number of studies that reported each tendon prop-
of the ‘Extraction’ boxes represent the total number of studies that erty. Because some studies reported both tendons (n = 2) and multiple
reported at least one mechanical or material property of the Achil- tendon properties (n = 8), these data do not reflect the total number of
les or patellar tendon. The data in the bullet points of the ‘Extraction’ included studies (n = 18)

mechanical and/or material properties [52–56], two reported from 46.3 [49] to 79.4 [44, 48] out of a possible score of
both AT and PT mechanical and/or material properties [57, 100, where a higher score represents less pain and disabil-
58] and 11 reported tendon morphological properties [6, 14, ity. The mean duration of symptoms was reported in six out
44–46, 49, 51–54, 57]. of 13 studies and ranged from 10 months [46] to 3.7 years
[57]. All studies included participants with mid-portion
3.2 Characteristics of Included Studies Achilles tendinopathy, with the exception of Chimenti et al.
[46, 51], who included participants with insertional Achil-
3.2.1 Achilles Tendinopathy les tendinopathy. The control group was the asymptomatic
contralateral limb in three studies (n = 81) [6, 50, 58]. Four
Study characteristics are included in Table 1. The sample studies included both asymptomatic contralateral limb and
size of the included studies that reported on the AT mechani- age- and sex-matched asymptomatic control groups with
cal and/or material properties in tendinopathy ranged from the remaining studies using age- and sex-matched asymp-
nine [44, 48] to 204 [49]. The total number of participants tomatic control groups [43, 44, 46, 48]. Of the studies that
was 381 for Achilles tendinopathy and 260 for asymptomatic included an asymptomatic control group, six also controlled
controls. Five studies included male-only participants [6, 14, for physical activity and/or training history [44, 45, 48, 49,
45, 48, 50], six included male and female participants [43, 54, 57]. Three studies reported AT cross-sectional area [6,
49, 51, 56–58] and one study did not report sex [44]. The 14, 49] and five reported AT thickness [44–46, 51, 57] in
mean age of participants with Achilles tendinopathy ranged both tendinopathic and contralateral control groups. With
from 27.3 [50] to 58.6 [46] years. The Victorian Institute the exception of measurements made at the distal AT [45,
of Sport Assessment-Achilles questionnaire (VISA-A) was 57], all studies that reported cross-sectional area or diameter
reported in seven out of 12 studies, with mean scores ranging had positive ES with lower CIs greater than zero (Fig. 2),
Table 1  Characteristics of included studies
Reference Sample size and characteristics Tendinopathy Methodology Measurement

Arya and Kulig [14] Tendinopathy: n = 12, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis strain (%),
age = 47.3 ± 8.3 years, males only, stiffness (N mm−1), modulus (GPa)
VISA-A = NR, DoS = NR
Asymptomatic control: n = 12,
age = 44.8 ± 7.2 years, males only
Aubrey et al. [43] Tendinopathy: n = 80, age = 56 years Mid-portion Achilles tendinopathy Shear wave elastography: axial and Achilles free tendon mean velocity (m
(range 46–63), males: female = 1.3, sagittal ­sec−1)
VISA-A = NR, DoS = NR
Asymptomatic control: n = 25,
age = 50 years (range 31–57), male:
female = 0.85
Chang et al. [44] Tendinopathy: n = 9, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis stiffness
age = 46.8 ± 6.3 years, male: (N mm−1)
female = NR, VISA-A = 79.4 ± 22,
DoS = NR
Asymptomatic control: n = 10,
Mechanical and material properties of lower limb tendinopathy

age = 48.7 ± 4.4 years, VISA-
A = 99.5 ± 1.2
Asymptomatic control (contralateral
limb): n = 9, age = 46.8 ± 6.3 years
Child et al. [45] Tendinopathy: n = 14, age 40 ± 8 years, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis strain (%)
males only, VISA-A = 70 ± 8,
DoS = 20 ± 47 months
Asymptomatic control: n = 15,
age = 35 ± 9 years, males only
Chimenti et al. [46] Tendinopathy: n = 20, age = 58.6 ± 7.8, Insertional Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis strain (%) and
female (55%), VISA-A = 47.6 ± 26.8, (passive) stiffness (N mm−1)
DoS = 10 months (range 4–24)
Asymptomatic control: n = 20,
age = 58.2 ± 8.5 years, female (55%),
VISA-A = 100 ± 0
Asymptomatic control (contralateral
limb): n = 20, age = 58.6 ± 7.8, female
(55%)
Chimenti et al. [51] Tendinopathy: n = 10, Insertional Achilles tendinopathy Axial strain elastography Achilles free tendon strain (%) in deep and
age = 53 ± 8.5 years, male (40%), VISA- superficial regions of distal tendon in
A = 57 ± 24, DoS = NR standing and squat
Asymptomatic control: n = 10,
age = 43.7 ± 15.9 years, male (50%)
Intziegianni et al. [47] Tendinopathy: n = 10, age = 40 ± 7 years, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis strain (%) and
male (70%), VISA-A = NR, DoS = NR compliance (mm N−1)
Asymptomatic control: n = 10,
age = 37 ± 8 years, male (70%)


Table 1  (continued)
Reference Sample size and characteristics Tendinopathy Methodology Measurement

Kulig et al. [48] Tendinopathy: n = 9, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis stiffness
age = 48.7 ± 4.4 years, males only, (N mm−1) and modulus (MPa)
VISA-A = 79.4 ± 22, DoS = NR
Asymptomatic control: n = 10,
age = 46.8 ± 6.3 years, male only, VISA-
A = 99.5 ± 1.2
Nuri et al. [6] Tendinopathy: n = 10, age = 42.2 ± 11.5, Mid-portion Achilles tendinopathy 3D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis and free
males only, VISA-A = 54.2 ± 16.5, tendon strain (%)
DoS = 3.5 ± 2 years
Asymptomatic control (contralateral
limb): n = 10, age = 42.2 ± 11.5, male
only
Ooi et al. [49] Tendinopathy: n = 120, age = 45.2 ± 13.1, Mid-portion Achilles tendinopathy Axial strain elastography Achilles free tendon strain-ratio
male (75%), VISA-A = 46.3 ± 16.8,
DoS = 13.9 ± 12.1 months
Asymptomatic control: n = 120,
age = 44.9 ± 13.6, VISA-A = 99.1 ± 1.9
Wang et al. [50] Tendinopathy: n = 17, Mid-portion Achilles tendinopathy 2D B-mode ultrasound and dynamometry Achilles tendon-aponeurosis stiffness
age = 27.3 ± 2 years, males only, VISA- (N.mm−1)
A = 70.7 ± 7.8, DoS = 5.9 ± 1.3 months
Asymptomatic control (contralateral
limb): n = 17, age = 27.3 ± 2 years
Coombes et al. [57] Achilles tendinopathy: n = 22, Mid-portion Achilles tendinopathy Supersonic shearwave elastography: Achilles tendon mean velocity (m.s−1):
age = 47.5 ± 11.4, female (41%), VISA- Patellar tendinopathy sagittal mid-tendon and tendon insertion
A = 58.6 ± 16.5, DoS = 3.7 ± 6.9 years Patellar tendon mean velocity (m.s−1): mid-
Patellar tendinopathy: n = 17, tendon and proximal tendon
age = 29.4 ± 6.6, female (24%), VISA-
P = 57.1 ± 14.4, DoS = 3.8 ± 3.4 years
Asymptomatic control: n = 28,
age = 38.3 ± 16.7 years, female (61%)
Dirrichs et al. [58] Achilles tendinopathy: n = 48 (34 unilat- Mid-portion Achilles tendinopathy Supersonic shearwave elastography: Achilles and patellar tendon mean velocity
eral, 7 bilateral), age = 50 ± 15.4, male Patellar tendinopathy sagittal (m.s−1) and modulus (kPa): proximal,
(63%), VISA-A = 53.6 (range 10–87), mid- and distal tendon
DoS = NR
Patellar tendinopathy: n = 38 (25 unilat-
eral, 13 bilateral), age = 35 ± 9.4, males
(87%), VISA-P = 58.5 (range 23–89),
DoS = NR
Asymptomatic control (contralateral
limb): n = 34 (Achilles), n = 25 (Patel-
lar), age = NR
S. J. Obst et al.
Table 1  (continued)
Reference Sample size and characteristics Tendinopathy Methodology Measurement

Couppé et al. [52] Tendinopathy: n = 7, age = 26 ± 1 years, Patellar tendinopathy 2D B-mode ultrasound and dynamometry Patellar tendon strain (%), stiffness
male only, VISA-P = NR, (N.mm−1), modulus (GPa)
Do = 3–96 months
Asymptomatic control: n = 9,
age = 22 ± 1 years, male only
Helland et al. [53] Tendinopathy: n = 13, Patellar tendinopathy 2D B-mode ultrasound and dynamometry Patellar tendon strain (%), stiffness
age = 20 ± 1.8 years, VISA-P = 76 ± 21, (N.mm−1), modulus (GPa)
DoS = NR.
Asymptomatic control: n = 15,
age = 21 ± 4 years, males only, VISA-
P = 98 ± 4
Kongsgaard et al. [54] Tendinopathy: n = 8, age = 32.9 ± 3.5, Patellar tendinopathy 2D B-mode ultrasound and dynamometry Patellar tendon strain (%), stiffness
males only, VISA-P = 57 ± 3, (N.mm−1), modulus (GPa)
DoS = 15 ± 3 months
Asymptomatic control: n = 9,
age = 31.5 ± 3.3 years, males only
Mechanical and material properties of lower limb tendinopathy

Lee et al. [55] Tendinopathy: n = 34, Patellar tendinopathy 2D B-mode ultrasound and dynamometry Patellar tendon strain (%) and stiffness
age = 22.2 ± 3.7 years, males (N.mm−1)
only, VISA-P = 55.9 ± 10.4,
DoS = 32.6 ± 25 months
Asymptomatic control: n = 13,
age = 24.9 ± 6 years
Zhang et al. [56] Tendinopathy: n = 13, age = 22.9 ± 4.6, Patellar tendinopathy Supersonic shearwave elastography: Patellar tendon shear elastic modulus
males only, DoS = 1.7 ± 1.6 years sagittal (kPa): distal tendon
Asymptomatic control: n = 20,
age = 24.9 ± 4.4, males only
 S. J. Obst et al.

Fig. 2  Forest plot (standardised effect size and 95% CI) illustrating matched controls and participants with tendinopathy. Unless other-
the differences in tendon cross-sectional area and diameter of the wise stated, all effect sizes reflect data for measurements obtained at
Achilles and patellar tendon between asymptomatic age- and sex- the tendon mid-portion

indicating increased tendon size in the Achilles tendinopathy wave mean velocity (m.s−1) [43, 57], one reported in units
group compared to asymptomatic controls. of shear wave modulus (kPa) [56] and one reported both
Of the articles that reported global AT mechanical and/or [58]. Because shear wave elastic modulus (E) represents the
material properties in tendinopathic participants (n = 193), product of the velocity of the shear wave propagation and
four reported strain [6, 14, 45, 47], five reported stiffness a constant [i.e. E = 3ρv2, where ρ is the density (equal to
[14, 44, 46, 48, 50] and two reported modulus [14, 48]. 1000 kg/m3) and v is the velocity of the shear wave propaga-
With the exception of Nuri et al. [6], who reported strain tion], all datasets were grouped together as local modulus
of the Achilles free tendon and tendon-aponeurosis meas- measures. Two studies that reported local modulus measures
ured using 3D ultrasound, all global tendon measurements included multiple datasets. Aubrey et al. [43] reported sagit-
were of the Achilles tendon-aponeurosis and measured tal- and axial-plane shear wave velocity data of the Achilles
using 2D ultrasound. Of the studies that reported local AT mid-tendon, and Coombes et al. [57] reported sagittal-plane
material properties, two reported strain-related parameters shear wave velocity data for the Achilles mid-tendon and
using quasi-static elastography methods [49, 51] and four proximal tendon.
reported estimates of tendon elastic modulus using shear
wave elastography (SWE) [43, 56–58]. Chimenti et al. [51] 3.2.2 Patellar Tendinopathy
used a non-rigid image registration method to examine deep
and superficial tissue strains (transverse-compression and Study characteristics are included in Table  1. The total
axial-tensile) at the AT insertion during two loading condi- sample size of all included studies that reported on the PT
tions (standing and squat). To facilitate comparison with properties ranged from 17 [50] to 47 [55]. The total number
other studies only the axial-tensile strain data were included of patellar tendinopathy participants (n = 130) was greater
in the current review. Ooi et al. [49] examined local tissue than the total number of asymptomatic control participants
strain of the Achilles mid-tendon during external mechani- (n = 122). With the exception of Coombes et al. [57], all arti-
cal compression and data were reported as a strain-ratio (i.e. cles included male participants only. The mean age for par-
ratio between AT strain and Kager’s fat pad strain), whereby ticipants with patellar tendinopathy ranged from 22.2 [55] to
a higher value represents a softer and ‘less stiff’ tendon 35 [58] years. The Victorian Institute of Sport Assessment-
region. For the studies that used SWE, two reported shear Patellar questionnaire (VISA-P) was reported in five out
Mechanical and material properties of lower limb tendinopathy

of seven studies with mean scores ranging from 53.6 [58] reported multiple datasets, meaning that each comparison
to 76 [53] out of a possible score of 100, where a higher between those with and without tendinopathy was reported
score represents less pain and disability. The mean dura- separately. This comprehensive approach resulted in 29 ES
tion of symptoms was reported in four out of seven studies measures for the AT (Fig. 3). Twelve studies (23 datasets)
and ranged from 15 months [54] to 3.8 years [57]. With the reported statistically significant within-study findings [6, 14,
exception of Dirrichs et al. [58], who used the asymptomatic 43–46, 48–51, 57, 58].
contralateral limb, all studies used an asymptomatic age- and
sex-matched control group. Some studies also controlled for 3.4.1 Global and Local Tendon Strain
height, weight and physical activity levels [52, 54–57]. Only
one study reported both asymptomatic and contralateral limb The mean ES for each of the five AT global strain datasets
control data [52]. Of the studies that included an asympto- was greater than zero (range 0.6–2.8), and in three datasets
matic PT control group, four also controlled for physical the lower confidence intervals did not cross zero [6, 14, 45],
activity and/or training history [52, 55–57]. Three studies indicating an increased global strain of the Achilles free
reported PT cross-sectional area [52–54] and one reported tendon and tendon-aponeurosis for the Achilles tendinopa-
PT thickness [57] in both tendinopathic and contralateral thy group compared to asymptomatic control. For AT local
control groups. Three out of four studies reported either a strain, one study reported higher strain [mean ES 1.5 (95%
significant increase [53, 57], or no difference [54, 57], in PT CI 1.2–1.8) [49] ], the other reported lower strain across
cross-sectional area or diameter (Fig. 2). One study reported multiple conditions in participants with insertional tendi-
a significantly smaller PT cross-sectional area, measured at nopathy (mean ES range − 0.9 to − 1.8 [51]).
the distal insertion [52].
Of the four studies that reported global PT mechanical
3.4.2 Global Tendon Stiffness
and/or material properties, four reported strain and stiffness
[52–55] and three reported modulus [52–54]. All global PT
The mean ES for each of the nine datasets for Achilles ten-
mechanical and/or material properties were measured during
don-aponeurosis stiffness was less than zero (range − 5.0
isometric knee extension contractions using 2D ultrasound
to − 0.6) and in all but two datasets the upper confidence
combined with dynamometry. Two studies reported local PT
interval was less than zero. The exceptions were those of
properties consistent with modulus using SWE. Coombes
Chimenti et al. [46] and Intziegianni et al. [47], which had
et al. [57] reported separate data for the proximal and mid
a mean ES of − 0.6 (95% CI − 1.3 to 0.0) and − 0.03 (95%
PT, whereas Dirrichs et al. [58] reported the most ‘rigid’
CI − 0.9 to 0.8), respectively.
area of the tendon measured either at the proximal, mid- or
distal tendon parts.
3.4.3 Global and Local Tendon Modulus
3.3 Quality Assessment
For AT global modulus, all three datasets had a mean ES less
The mean (SD) score for the modified qualitative review than zero (range − 8.7 to − 3.4) and upper confidence inter-
tool was 11.2 (2.5) out of a possible score of 14 (range vals greater than zero [14, 48]. For AT local modulus, three
9.1 [43]–12.7 [49, 57]) (Table  2). Mean scores greater studies (datasets) had negative ES with upper confidence
than 0.75 out of 1 were recorded for nine out of 14 criteria intervals less than zero [43, 57, 58], indicating decreased
used, with ‘recruitment and sampling method descriptions’ local modulus of the AT for tendinopathy compared to
[mean (SD) = 0.56 (0.31)], ‘controlled covariates’ [mean asymptomatic controls.
(SD) = 0.67 (0.31)], ‘reliability of the methodology stated’
[mean (SD) = 0.56 (0.51)], ‘internal validity of the method- 3.5 Tendon Properties in Patellar Tendinopathy
ology stated’ [mean (SD) = 0.06 (0.25)] and ‘clinical impli-
cations stated’ [mean (SD) = 0.64 (0.43)] each scoring below Seven studies incorporating 18 datasets reported mechanical
0.75. There was an acceptable level of agreement between and/or material properties of the PT in tendinopathy [52–58]
the two reviewers (Cohen’s kappa = 0.90; perfect agreement (Fig. 4). Four studies (seven datasets) reported statistically
on 501/558 single item scores). No studies were excluded significant within-study findings [53, 55, 57, 58].
from the review on the basis of the quality assessment score.
3.5.1 Global Tendon Strain
3.4 Tendon Properties in Achilles Tendinopathy
Three out of four data sets had a positive mean ES for PT
Thirteen studies reported data on AT mechanical and/or global strain (range 0.7–1.5) [52, 54, 58], but only one had
material properties [6, 14, 43–51, 57, 58]. Eight studies a lower confidence interval greater than zero [54]. Lee et al.


Table 2  Methodological quality review of included articles

Question Scoring Arya Aubry Chang Child Chi- Chi- Coombes Couppé Dirrichs Hel- Intzie- Kongsgaard Kulig Lee Nuri Ooi Wang Zhang Mean SD
criteria and et al. et al. et al. menti menti et al. [57] et al. et al. [58] land gianni et al. [54] et al. et al. et al. et al. et al. et al.
Kulig [43] [44] [45] et al. et al. [52] et al. et al. [48] [55] [6] [49] [50] [56]
[14] [46] [51] [53] [47]

1. Research 1—yes; 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00

aims or 0.5—
questions yes,
stated lacking
clearly detail;
0—no
2. Par- 1—yes;
ticipants 0—no
detailed
 Number 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Age 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Sex 1 1 0 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0.89 0.34
 Population 1 1 0 1 0 1 1 1 1 1 1 0 0 1 0 1 1 1 0.72 0.48
 Subtotal 1.00 1.00 0.50 1.00 0.75 1.00 1.00 1.00 1.00 1.00 1.00 0.75 0.50 1.00 0.75 1.00 1.00 1.00 0.90 0.18
3. Recruit- 1—yes; 0.5 0.5 0.5 0.5 0.5 0.5 1 0.5 0.5 1 0.5 0.5 0 1 0 1 1 0 0.56 0.31
ment and 0.5—
sampling yes,
methods lacking
described detail;
0—no
4. Inclu- 1—yes; 1 1 1 1 1 1 1 0.5 1 0.5 0.5 1 1 1 1 1 1 0.5 0.89 0.20
sion and 0.5—
exclusion yes,
criteria lacking
described detail;
0—no
5. Con- 1—yes;
trolled 0—no
covariates
 Age 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 0.89 0.34
 Sex 1 0 0 1 1 1 1 1 0 1 1 1 0 1 1 1 1 0 0.72 0.45
 Tendon 1 1 0 1 0 0 0 1 0 1 1 0 0 1 0 0 0 0 0.39 0.51
pre-
condi-
tioning
 Subtotal 1.00 0.33 0.33 1.00 0.67 0.67 0.67 1.00 0.00 1.00 1.00 0.67 0.33 1.00 0.67 0.67 0.67 0.33 0.67 0.31
S. J. Obst et al.
Table 2  (continued)
Question Scoring Arya Aubry Chang Child Chi- Chi- Coombes Couppé Dirrichs Hel- Intzie- Kongsgaard Kulig Lee Nuri Ooi Wang Zhang Mean SD
criteria and et al. et al. et al. menti menti et al. [57] et al. et al. [58] land gianni et al. [54] et al. et al. et al. et al. et al. et al.
Kulig [43] [44] [45] et al. et al. [52] et al. et al. [48] [55] [6] [49] [50] [56]
[14] [46] [51] [53] [47]

6. Key 1—yes; 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00

outcome 0.5—
variable only
clearly some
described defined;
0.5—
yes,
lacking
detail;
0—no
7. Adequate 1—yes;
methodol- 0—no
ogy able
to repeat
study
 Participant 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.94 0.25
Mechanical and material properties of lower limb tendinopathy

sam-
pling
 Equipment 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Procedure 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Data pro- 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
cessing
 Statistical 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
analysis
 Sub total 1.00 0.80 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.99 0.05
8. Method- 1—yes;
ology able 0—no
to answer
research
question
 Participant 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
sam-
pling
 Equipment 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Procedure 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
 Data pro- 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
cessing
 Statistical 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
analysis
 Sub total 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.00
9. Reliabil- 1—yes; 1 0 1 1 1 0 1 0 0 0 0 0 1 1 1 1 0 1 0.56 0.51
ity of the 0—no
methodol-
ogy stated


Table 2  (continued)
Question Scoring Arya Aubry Chang Child Chi- Chi- Coombes Couppé Dirrichs Hel- Intzie- Kongsgaard Kulig Lee Nuri Ooi Wang Zhang Mean SD
criteria and et al. et al. et al. menti menti et al. [57] et al. et al. [58] land gianni et al. [54] et al. et al. et al. et al. et al. et al.
Kulig [43] [44] [45] et al. et al. [52] et al. et al. [48] [55] [6] [49] [50] [56]
[14] [46] [51] [53] [47]

10. Internal 1—yes; 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0.06 0.25

validity of 0—no
the meth-
odology
stated
11. 1—yes; 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
Research 0—no
questions
answered
ade-
quately
in the
discus-
sion
12. Key 1—yes; 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.00 0.00
findings 0—no
supported
by the
results
13. Key 1—yes; 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.94 0.25
findings 0—no
inter-
preted in
a logical
manner
which is
supported
by refer-
ence
14. Clinical 1—yes; 0 0.5 0 1 1 1 1 0 1 1 1 0.5 0.5 0.5 0 1 0.5 1 0.64 0.43
implica- 0.5—
tions yes,
stated lacking
detail;
0—no
Total (/14) 11.5 9.1 10.3 12.5 11.9 11.2 12.7 10.0 11.5 11.5 11.0 10.4 10.3 12.5 10.4 12.7 11.2 10.8 11.2 2.5
S. J. Obst et al.
Mechanical and material properties of lower limb tendinopathy

Fig. 3  Forest plot (standardised effect size and 95% CI) illustrating property (global and local) and listed alphabetically. Unless otherwise
the effect of tendinopathy on the mechanical properties of the Achil- stated, all effect sizes reflect differences between tendinopathy and
les tendon. Numerical values adjacent to error bars indicate the per- asymptomatic control group. Additional information is provided for
centage change. Bold error bars indicate results reported as signifi- studies that report multiple datasets
cant in the original study. Articles are grouped by tendon mechanical

[55] had the only study with a negative mean ES for PT modulus, two studies [56, 57] both reported significantly
global strain [mean − 1.0 (95% CI − 1.7 to − 0.3)]. higher mean velocity in tendinopathy compared to age-
matched asymptomatic controls, whereas Dirrichs et al. [58]
3.5.2 Global Tendon Stiffness reported significantly lower mean velocity [mean ES − 8.2
(95% CI − 9.5 to − 6.8)] in the symptomatic compared to
Three studies (three datasets) had a negative mean ES for the asymptomatic side in participants with unilateral patellar
PT global stiffness [52–54] and of these one had an upper tendinopathy.
confidence interval less than zero [53] [mean ES − 1.2 (95%
CI − 1.9 to − 0.5)].
4 Discussion
3.5.3 Global and Local Tendon Modulus
Eighteen articles published between 2010 and 2017 that
For PT global modulus, the mean ES ranged from − 0.8 [53] assessed global and/or local tendon mechanical and/
to 1.5 [52] and two datasets had ES confidence intervals that or material properties in individuals with tendinopathy
did not cross zero [52, 53]. Couppé et al. [52] and Helland compared to asymptomatic controls were included in the
et al. [53] reported a 56% higher and 15% lower PT global review. In general, the methodological quality of included
modulus, respectively, in athletes with patellar tendinopathy articles was acceptable; however, there were notable
compared to an asymptomatic control group. For PT local inconsistencies in the use of tendon preconditioning
 S. J. Obst et al.

Fig. 4  Forest plot (standardised effect size and 95% CI) illustrating property (global and local) and listed alphabetically. Unless otherwise
the effect of tendinopathy on the mechanical properties of the patel- stated, all effect sizes reflect differences between tendinopathy and
lar tendon. Numerical values adjacent to error bars indicate the per- asymptomatic control group. Additional information is provided for
centage change. Bold error bars indicate results reported as signifi- studies that report multiple datasets
cant in the original study. Articles are grouped by tendon mechanical

protocols and lack of detail of recruitment and sampling such as therapeutic exercise, which target adaptation of
procedures. There was also considerable heterogeneity in tendon mechanical and/or material properties.
participant characteristics (e.g. age, sex, type of tendinopa-
thy, duration and severity of symptoms) and measurement 4.1 Tendon Properties in Achilles Tendinopathy
techniques, which must be considered when interpreting
the results of this review. Irrespective of these limitations, There was consistent evidence that the AT (free tendon
the review identified consistent evidence of a moderate and tendon-aponeurosis) strained more for a similar tendon
to large effect of tendinopathy on the mechanical and force, and had a lower stiffness (tendon-aponeurosis) and
material properties of the Achilles tendon-aponeurosis, modulus (tendon-aponeurosis) in people with tendinopathy,
compared to an asymptomatic control group. More spe- compared to an asymptomatic control group. These find-
cifically, the Achilles tendon-aponeurosis in tendinopathy ings are broadly consistent with the studies that examined
was reported to strain more and had a lower stiffness and tendon modulus in localised regions of the AT using SWE
modulus compared to an asymptomatic control group. [43, 57, 58]. Lower global and local tendon stiffness and
Despite the tendinopathic free AT being larger in cross- modulus in the tendinopathic AT would result in greater
sectional area and thickness, compared to an asymptomatic local and global length changes for a given force or stress,
control tendon, these morphological changes do not appear compared to a healthy tendon. Such changes could lead to
to offset the reduction in mechanical stiffness and suggest non-uniform strains and stresses and earlier mechanical
that tendinopathic AT has both inferior mechanical and fatigue in response to cyclic loading that could increase the
material properties. In contrast, there is insufficient evi- risk of cumulative or single-event local and global tissue
dence in the literature to date to support a consistent effect damage [18, 19, 23, 35]. It is therefore not surprising that
of tendinopathy on the mechanical and/or material proper- mid-portion Achilles tendinopathy is considered a strong
ties of the PT. Although the histopathology and clinical risk factor for AT ruptures [11], and that complete failure
presentation may be similar in Achilles and patellar ten- can occur in low intensity and non-sports-related activities
dinopathy [59], our findings based on the literature to date [60]. Similarly, exercise interventions shown to elicit global
suggest there is a differential effect of tendinopathy on the tendon strains considered optimal for tendon adaptation in
mechanical behaviour of the AT and PT. These findings non-injured tendons (e.g. 3–6% [61]), may not be generalis-
could have implications for the design of interventions, able to the injured tendon, and could have catabolic rather
Mechanical and material properties of lower limb tendinopathy

than anabolic effects on AT remodeling. Whether reduced stress (stress = force/cross-sectional area) to estimate modu-
mechanical and material stiffness precipitate the clinical lus. Helland et al. [53] reported a 19% larger (measured at
presentation of tendinopathy is unknown; however, such mid-tendon) and Couppé et al. [52] reported a 17% smaller
changes would undoubtedly increase the risk of further ten- (measured at distal tendon) tendon cross-sectional area in
don degeneration in the absence of appropriate load man- tendinopathic compared to asymptomatic control tendons.
agement [2]. Prospective studies that evaluate the temporal As a result, although Couppé et al. [52] reported higher
progression of morphological and mechanical properties in global tendon strain in the tendinopathic group (p < 0.05),
individuals who develop and recover from Achilles tendi- the small tendon cross-sectional area resulted in significantly
nopathy are required. higher tendon stress for a given tendon force and therefore
Contrasting findings were reported in the two studies higher tendon modulus. It should be noted, however, that the
that examined local tendon strain [49, 51]. Chimenti et al. variation in tendon cross-sectional area between these two
[51] reported lower local tendon strain of both the deep studies could reflect differences in the site of tendon pathol-
and superficial regions of the distal AT in participants with ogy between individuals, and, therefore, reflect differences
insertional tendinopathy, whereas Ooi et al. [49] reported in the measurement location used in the two studies. It will
higher local tendon strain (measured as strain ratio) of the be important for future studies to either standardise the loca-
mid-tendon in participants with mid-portion tendinopathy. tion of tendon cross-sectional area measurements or provide
The disparity in these findings may be due to the different a composite (e.g. average) or minimum measurement based
regions of AT tendinopathy investigated between these stud- imaging of the entire tendon [5, 67]. Although based on a
ies (insertional vs. mid-tendon), which experience different small number of studies, our results do not indicate a clear
loading and exhibit distinct histological alterations in tendi- and consistent effect of tendinopathy on PT global stiffness
nopathy. Whereas the principal loads in the AT mid-portion or modulus.
are tensile and parallel to the direction of the tendon fibres, The inconsistency in studies of PT global measures was
the distal AT experiences non-uniform strains and compres- also reflected in the local modulus results wherein two stud-
sive loads at the calcaneal insertion [62, 63]. The findings ies reported significantly higher [56, 57] and one signifi-
of Ooi et al. [49] could be consistent with changes in degen- cantly lower [58] shear wave modulus/mean velocity data
erative mid-portion tendinopathy, whereby the histological in tendinopathy compared to asymptomatic controls. These
changes (i.e. increased type III collagen and abnormal teno- discrepancies could reflect differences in the sample popula-
cytes) result in tendon tissue that is less able to resist tensile tion, methodologies used to measure tendon properties, and/
forces. Conversely, the histology of insertional tendinopa- or the choice of control group. Dirrichs et al. [58] recorded
thy may include higher amounts of cartilage matrix proteins the elastic modulus using a very small (1 mm diameter)
(i.e. type II collagen and aggrecan) and cartilage cells (i.e. region of interest in the ‘most conspicuous (or pathologi-
chondrocytes) [64] that could result in a more rigid tissue cal) area’ of the tendon as determined from B-mode images,
structure [65, 66]. whereas Coombes et al. [57] and Zhang et al. [56] recorded
a larger region of interest (> 5 mm diameter) at standard-
4.2 Tendon Properties in Patellar Tendinopathy ised locations at the proximal [56, 57] and distal tendon
[57]. While the former approach would include exclusively
In contrast to the AT, there was inconsistent and conflicting pathological tissue and may contribute to measurement
evidence that the global and local mechanical and material bias, the latter approach would likely include both healthy
properties of the PT are altered in tendinopathy, compared and pathological tendon tissue and provide a more com-
to asymptomatic controls. Four studies reported global PT prehensive assessment of tendon mechanics. Differences
strain, two of which had ES confidence intervals that did not in the measurement region could explain small variations
cross zero [54, 55] and only one reported significant within- in data between studies, but would be unlikely to account
study findings [55]. Similar inconsistencies were reported for a clear discrepancy in the direction of the effect size
for PT global stiffness and global modulus. Helland et al. data. While difficult to determine, these results could reflect
[53] reported significantly lower PT global stiffness and heterogeneity in tendon structure and loading patterns (i.e.
modulus in volleyball athletes with tendinopathy compared tension, compression or both [68]) in patellar tendinopathy,
to asymptomatic controls. In contrast, Couppé et al. [52] wherein areas within the same tendon could display very
reported a 56% higher tendon modulus in elite badminton different mechanical and/or material properties. Similarly,
players with patellar tendinopathy compared to players with- other factors such as measurement error, due to variations
out patellar tendinopathy (1.87 ± 0.55 versus 1.2 ± 0.19 GPa, in the location and orientation of the 2D image plane and
p < 0.05). The disparity in the PT global modulus results tendon load, as well as the use of the asymptomatic limb as
is likely due to contrasting tendon cross-sectional area a control group, could also contribute to the differences in
results, which are used to convert tendon force to tendon outcomes for local PT modulus studies. Not surprisingly, we
 S. J. Obst et al.

also could not identify any clear relationship between local [71], it is important that future studies report both asympto-
and global modulus measures of the PT. For these reasons, it matic matched control and asymptomatic contralateral limb
may be necessary for future studies that use SWE to include data when examining the effect of tendinopathy on tendon
multiple measurement regions acquired in multiple image mechanical behaviour in vivo.
planes, and to report the distribution (frequency and spatial
[69]) of modulus values within these measurement regions, 4.4 Effect of Tendinopathy on the Achilles
in addition to the mean. It will also be important for stud- and Patellar Tendons
ies to compare local and global measures of in vivo tendon
mechanical behaviour [70], with a view of cross validating An important finding of this review was that we did not iden-
these measurement techniques, and better understanding the tify a similar effect of tendinopathy on the mechanical and/
effect of tendon injury on tendon function across multiple or material properties of the Achilles and patellar tendons,
measurement scales. despite similarities in the change in tendon cross-sectional
dimensions. The lack of a clear and consistent effect of ten-
4.3 Tendon Properties of the Contralateral Control dinopathy on the PT was unexpected and could reflect meth-
Limb odological differences between the studies or differences in
the structure, function and/or pathology between the two
Although not an a priori aim of the study, we noted that the lower limb tendons. As expected, the Achilles tendinopa-
use of the asymptomatic contralateral limb as the control thy groups were older (mean age range 27–58 years) com-
consistently led to smaller ES than when an asymptomatic pared to the patellar tendinopathy groups (mean age range
matched control group was used, particularly for the AT 22–35 years). It is therefore possible that normal age-related
[43, 44, 46, 48, 52] (Fig. 5). While not conclusive, there changes in tendon properties were more pronounced in the
does appear to be some evidence that people with unilateral AT compared to the PT groups [30, 74]. Similarly, although
Achilles tendinopathy have a lower global stiffness [44, 48] the mechanical and material properties of the AT and PT
and modulus [48] compared to a matched control group. are different in males and females [75–77], the majority of
These observations are consistent with studies that have studies included an age- and/or sex-matched contralateral
identified bilateral deficits in neuromuscular control [71, control group or used the contralateral limb. Furthermore,
72] and tendon morphology [4, 73] in people with unilat- there were no clear differences in the duration and/or sever-
eral tendinopathy. The exception was the study by Couppé ity of symptoms reported between the AT and PT studies,
et al. [52], who reported a 44% higher global tendon modu- with VISA-A and VISA-P scores ranging from 46.3 [49] to
lus in the asymptomatic contralateral limb in people with 79.4 [44, 48] and 53.6 [58] to 76 [53], respectively. While
unilateral patellar tendinopathy compared to a matched con- these differences in participant characteristics may limit the
trol group. While some experimental controls are afforded generalisability of our findings, they do not explain the dis-
by using the contralateral limb as a control (e.g. genetics, parity between the effects of tendinopathy on the AT com-
habitual loading, etc.), the likelihood of bilateral tendon defi- pared or the PT.
cits in unilateral symptomatic tendinopathy must be taken The lack of clear evidence for mechanical changes in
into account. In line with other research recommendations patellar tendinopathy could therefore reflect the unique

Fig. 5  Forest plot (standardised

effect size and 95% CI) illustrat-
ing the differences in mechani-
cal properties of the Achilles
and patellar tendon (PT)
between asymptomatic con-
tralateral limb in participants
with unilateral tendinopathy and
an asymptomatic age- matched
control group. Unless otherwise
stated all effect sizes reflect data
for the Achilles tendon
Mechanical and material properties of lower limb tendinopathy

structure and function of the PT and/or the location and compared to the AT, limiting the strength of the conclusions
extent of the tendon pathology. Unlike the mid-portion of for the effect of tendinopathy on the mechanical properties
the AT that experiences predominately tensile loads, the PT of the PT. Also, we limited our data extraction to mechani-
experiences a mix of both tensile and compressive loads, cal and/or material properties measured in vivo, and can
the latter of which are most pronounced at the tendon enthe- therefore only speculate as to the influence that the loca-
ses [78] and when the knee is loaded into flexion [27, 79]. tion and nature of the tendon pathology could have on the
Because of the heterogeneous pattern of tendon loading, mechanical function of the AT or PT. Future studies should
the extent, location and nature of the pathology may differ include measures that quantify the extent and nature of the
between the two lower limb tendons [2, 78] and could influ- pathology in 3D using sophisticated image-processing meth-
ence the effect of tendinopathy on the mechanical behaviour ods, such as UTC [67], quantitative 3D ultrasound [69] and/
of each tendon. However, the similarity in tendon strain and or ultrashort echo-time magnetic resonance imaging [83],
stiffness findings between studies that included participants in both asymptomatic and symptomatic populations. Inte-
with Achilles insertional tendinopathy [46, 51] and those gration of multiple imaging techniques would enable esti-
that included mid-portion tendinopathy suggests other fac- mation of tendon mechanical, material and morphological
tors might be important. Because no studies included histo- properties across measurement scales and contribute to the
pathological data, or measures of tendon structural integrity future development of valid and patient-specific multi-scale
(e.g. using ultrasound tissue characterisation (UTC) [68]), computational models of human tendon behaviour [82, 84].
the relationship between tendon pathology and tendon We acknowledge that with the exception of one study [6], all
mechanical behaviour cannot be determined. mechanical measurements for the AT were made at the level
Although each lower limb tendon presents with similar of the tendon-aponeurosis, and not the free tendon, which
clinical signs and symptoms of tendinopathy, the resultant represents the site of pathology in Achilles tendinopathy. For
changes in tendon mechanical behaviour may not be equiv- this reason, the results may not be representative of changes
alent between the two tendons. For this reason, although in the mechanical behaviour of the Achilles free tendon, and
exercise loading alters the mechanical properties of both future studies should examine free tendon, independent of
tendons in a similar way in healthy tendons [29], this may the Achilles aponeurosis. Furthermore, we acknowledge that
be not be true for pathological tendons, and so exercise strat- comparing measurements of tendon strain between individu-
egies shown to be effective for Achilles tendinopathy may als assumes that tendon length measurements are made at
have inconsistent and potentially negative effects in patellar the same internal tendon force, and after the same rate of
tendinopathy [80]. Exercise interventions for tendinopathy force development, which may not be valid even when con-
may therefore need to be tailored to the symptomatic and/or trolled for based on external joint torque. These and other
mechanical state of the tendon [81], the latter of which can limitations of in vivo measurement of tendon mechanical
only be achieved using high-fidelity, real-time and patient- and material properties [85] must also be considered when
specific measurement of tendon mechanics in vivo [82]. interpreting the results of this review.

4.5 Limitations

There are a number of limitations that must be considered 5 Conclusions

when interpreting the findings of this review. The overall
methodological quality of the included articles was accepta- There is consistent evidence that the in vivo mechanical
ble; however, there were inconsistencies in the use of tendon and material properties of the Achilles tendon-aponeuro-
preconditioning and incomplete reporting of methodologies, sis are adversely affected by tendinopathy. In contrast to
participant characteristics and clinical implications. In addi- Achilles tendinopathy, there is insufficient evidence in the
tion, due to the design of the included studies (i.e. case–con- literature to date to support a consistent effect of tendinop-
trol), it is not possible to interpret whether the identified athy on the mechanical or material properties of the PT.
changes in the mechanical properties of the AT and PT are Despite a similar clinical presentation, the extent, location
associated with the cause or effect of tendinopathy. Sample and nature of the tendon pathology in patellar tendinopa-
sizes of included studies were also generally small (n < 30), thy may differ from that of the AT and be insufficient to
with the exception of some of the elastography studies [49, alter the tensile behaviour of the whole PT in vivo.
56, 58]; however, the generalisability of these measurement
techniques to other measures of tendon properties cannot be Compliance with Ethical Standards 
determined. There were also fewer studies (PT = 7; AT = 13)
of the PT with fewer participants (PT = 130; AT = 381) Funding  No sources of funding were used to assist in the preparation
of this article.
 S. J. Obst et al.

Conflicts of interest Steven Obst, Luke Heales, Benjamin Schrader, 20. Obst SJ, Newsham-West R, Barrett RS. Three-dimensional mor-
Scott Davis, Keely Dodd, Cory Holzberger, Louis Beavis and Rod phology and strain of the human Achilles free tendon imme-
Barrett declare that they have no conflicts of interest relevant to the diately following eccentric heel drop exercise. J Exp Biol.
content of this review. 2015;218(Pt 24):3894–900.
21. Nuri L, Obst SJ, Newsham-West R, Barrett RS. Regional three-
dimensional deformation of human Achilles tendon during con-
ditioning. Scand J Med Sci Sports. 2017;27:1263–72.
References 22. Nuri L, Obst SJ, Newsham-West R, Barrett RS. Recovery of
human Achilles tendon three-dimensional deformation follow-
1. Khan KM, Cook JL, Kannus P, Maffulli N, Bonar SF. Time to ing conditioning. J Sci Med Sport. 2018;21:473–8.
abandon the “tendinitis” myth. Br Med J. 2002;324(7338):626–7. 23. Lichtwark GA, Cresswell AG, Newsham-West RJ. Effects of
2. Rees JD, Maffulli N, Cook J. Management of tendinopathy. Am J running on human Achilles tendon length-tension proper-
Sports Med. 2009;37(9):1855–67. ties in the free and gastrocnemius components. J Exp Biol.
3. Fredberg U, Stengaard-Pedersen K. Chronic tendinopathy tissue 2013;216(Pt 23):4388–94.
pathology, pain mechanisms, and etiology with a special focus on 24. Onambele GL, Narici MV, Maganaris CN. Calf muscle-tendon
inflammation. Scand J Med Sci Sports. 2008;18(1):3–15. properties and postural balance in old age. J Appl Physiol.
4. Grigg NL, Wearing SC, Smeathers JE. Achilles tendinopathy has 2006;100(6):2048–56.
an aberrant strain response to eccentric exercise. Med Sci Sports 25. Kujala UM, Sarna S, Kaprio J. Cumulative incidence of Achilles
Exerc. 2012;44(1):12–7. tendon rupture and tendinopathy in male former elite athletes.
5. Nuri L, Obst S, Newsham-West R, Barrett R. Three-dimensional Clin J Sport Med. 2005;15(3):133–5.
deformation of the Achilles tendon during load in people with 26. Malliaras P, Cook J, Purdam C, Rio E. Patellar tendinopa-
unilateral mid-portion Achilles tendinopathy. J Sci Med Sport. thy: clinical diagnosis, load management, and advice for
2017;20:e71. challenging case presentations. J Orthop Sports Phys Ther.
6. Nuri L, Obst SJ, Newsham-West R, Barrett RS. The tendinopathic 2015;45(11):887–98.
Achilles tendon does not remain iso-volumetric upon repeated 27. DeFrate LE, Nha KW, Papannagari R, Moses JM, Gill TJ, Li G.
loading: insights from 3D ultrasound. J Exp Biol. 2017;220(Pt The biomechanical function of the patellar tendon during in vivo
17):3053–61. weight-bearing flexion. J Biomech. 2007;40(8):1716–22.
7. Wang JH. Mechanobiology of tendon. J Biomech. 28. Wiesinger H-P, Rieder F, Kösters A, Müller E, Seynnes OR. Are
2006;39(9):1563–82. sport-specific profiles of tendon stiffness and cross-sectional
8. Jozsa L, Reffy A, Kannus P, Demel S, Elek E. Pathologi- area determined by structural or functional integrity? PLoS One.
cal alterations in human tendons. Arch Orthop Trauma Surg. 2016;11(6):e0158441.
1990;110(1):15–21. 29. Wiesinger H-P, Kösters A, Müller E, Seynnes OR. Effects of
9. Riley G. The pathogenesis of tendinopathy. A molecular perspec- increased loading on in  vivo tendon properties: a systematic
tive. Rheumatology. 2003;43(2):131–42. review. Med Sci Sports Exerc. 2015;47(9):1885–95.
10. Ireland D, Harrall R, Curry V, Holloway G, Hackney R, Hazleman 30. Narici MV, Maffulli N, Maganaris CN. Ageing of human muscles
B, et al. Multiple changes in gene expression in chronic human and tendons. Disabil Rehabil. 2008;30(20–22):1548–54.
Achilles tendinopathy. Matrix Biol. 2001;20(3):159–69. 31. Maganaris CN, Narici MV, Maffulli N. Biomechanics of the
11. Kannus P, Jozsa L. Histopathological changes preceding spontane- Achilles tendon. Disabil Rehabil. 2008;30(20–22):1542–7.
ous rupture of a tendon. A controlled study of 891 patients. J Bone 32. Lichtwark G, Barclay C. The influence of tendon compliance on
Joint Surg Am. 1991;73(10):1507–25. muscle power output and efficiency during cyclic contractions. J
12. Maffulli N, Khan KM, Puddu G. Overuse tendon condi- Exp Biol. 2010;213(5):707–14.
tions: time to change a confusing terminology. Arthroscopy. 33. Sawicki GS, Sheppard P, Roberts TJ. Power amplification in
1998;14(8):840–3. an isolated muscle–tendon unit is load dependent. J Exp Biol.
13. Archambault JM, Wiley JP, Bray RC, Verhoef M, Wiseman DA, 2015;218(22):3700–9.
Elliott PD. Can sonography predict the outcome in patients with 34. Galantis A, Woledge RC. The theoretical limits to the power out-
achillodynia? J Clin Ultrasound. 1998;26(7):335–9. put of a muscle–tendon complex with inertial and gravitational
14. Arya S, Kulig K. Tendinopathy alters mechanical and mate- loads. Proc R Soc Lond B Biol Sci. 2003;270(1523):1493–8.
rial properties of the Achilles tendon. J Appl Physiol. 35. Wren TA, Lindsey DP, Beaupre GS, Carter DR. Effects of creep
2010;108(3):670–5. and cyclic loading on the mechanical properties and failure of
15. Jozsa L, Kvist M, Balint B, Reffy A, Jarvinen M, Lehto M, et al. human Achilles tendons. Ann Biomed Eng. 2003;31(6):710–7.
The role of recreational sport activity in Achilles tendon rupture: 36. Farris DJ, Trewartha G, McGuigan MP. Could intra-tendinous
a clinical, pathoanatomical, and sociological study of 292 cases. hyperthermia during running explain chronic injury of the human
Am J Sports Med. 1989;17(3):338–43. Achilles tendon? J Biomech. 2011;44(5):822–6.
16. Lichtwark GA, Wilson AM. In vivo mechanical properties of the 37. Fouré A, Nordez A, Cornu C. Plyometric training effects on Achil-
human Achilles tendon during one-legged hopping. J Exp Biol. les tendon stiffness and dissipative properties. J Appl Physiol.
2005;208(24):4715–25. 2010;109(3):849–54.
17. Lichtwark GA, Wilson AM. Is Achilles tendon compliance opti- 38. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred
mised for maximum muscle efficiency during locomotion? J Bio- reporting items for systematic reviews and meta-analyses: the
mech. 2007;40(8):1768–75. PRISMA statement. Int J Surg. 2010;8(5):336–41.
18. Obst SJ, Newsham-West R, Barrett RS. Changes in Achilles ten- 39. Galna B, Peters A, Murphy AT, Morris ME. Obstacle cross-
don mechanical properties following eccentric heel drop exer- ing deficits in older adults: a systematic review. Gait Posture.
cise are specific to the free tendon. Scand J Med Sci Sports. 2009;30(3):270–5.
2016;26(4):421–31. 40. Cohen J. A coefficient of agreement for nominal scales. Educ
19. Obst SJ, Barrett RS, Newsham-West R. Immediate effect of exer- Psychol Meas. 1960;20(1):37–46.
cise on Achilles tendon properties: systematic review. Med Sci 41. Cohen J. Statistical power analysis current directions. Psychol Sci.
Sports Exerc. 2013;45(8):1534–44. 1992;1(3):98–101.
Mechanical and material properties of lower limb tendinopathy

42. Heales LJ, Bergin MJ, Vicenzino B, Hodges PW. Forearm mus- 60. Leppilahti J, Puranen J, Orava S. Incidence of Achilles tendon
cle activity in lateral epicondylalgia: a systematic review with rupture. Acta Orthop Scand. 1996;67(3):277–9.
quantitative analysis. Sports Med. 2016;46(12):1833–45. 61. Wang T, Lin Z, Day RE, Gardiner B, Landao-Bassonga E, Ruben-
43. Aubry S, Nueffer JP, Tanter M, Becce F, Vidal C, Michel F. son J, et al. Programmable mechanical stimulation influences
Viscoelasticity in Achilles tendonopathy: quantitative assess- tendon homeostasis in a bioreactor system. Biotechnol Bioeng.
ment by using real-time shear-wave elastography. Radiology. 2013;110(5):1495–507.
2015;274(3):821–9. 62. Maganaris CN, Narici MV, Almekinders LC, Maffulli N. Biome-
44. Chang YJ, Kulig K. The neuromechanical adaptations to Achil- chanics and pathophysiology of overuse tendon injuries: ideas on
les tendinosis. J Physiol. 2015;593(15):3373–87. insertional tendinopathy. Sports Med. 2004;34(14):1005–17.
45. Child S, Bryant AL, Clark RA, Crossley KM. Mechani- 63. Benjamin M, Toumi H, Ralphs J, Bydder G, Best T, Milz S.
cal properties of the Achilles tendon aponeurosis are altered Where tendons and ligaments meet bone: attachment sites
in athletes with Achilles tendinopathy. Am J Sports Med. (‘entheses’) in relation to exercise and/or mechanical load. J Anat.
2010;38(9):1885–93. 2006;208(4):471–90.
46. Chimenti RL, Flemister AS, Tome J, McMahon JM, Flannery 64. Maffulli N, Reaper J, Ewen SW, Waterston SW, Barrass V. Chon-
MA, Xue Y, et al. Altered tendon characteristics and mechani- dral metaplasia in calcific insertional tendinopathy of the Achilles
cal properties associated with insertional Achilles tendinopathy. tendon. Clin J Sport Med. 2006;16(4):329–34.
J Orthop Sports Phys Ther. 2014;44(9):680–9. 65. Bah I, Kwak ST, Chimenti RL, Richards MS, Ketz JP, Flemister
47. Intziegianni K, Cassel M, Rauf S, White S, Rector M, Kaplick H, AS, et  al. Mechanical changes in the Achilles tendon due to
et al. Influence of age and pathology on Achilles tendon properties insertional Achilles tendinopathy. J Mech Behav Biomed Mater.
during a single-leg jump. Int J Sports Med. 2016;37(12):973–8. 2016;53:320–8.
48. Kulig K, Chang YJ, Winiarski S, Bashford GR. Ultrasound-based 66. Wren TA, Beaupre GS, Carter DR. Mechanobiology of tendon
tendon micromorphology predicts mechanical characteristics of adaptation to compressive loading through fibrocartilaginous
degenerated tendons. Ultrasound Med Biol. 2016;42(3):664–73. metaplasia. J Rehabil Res Dev. 2000;37(2):135.
49. Ooi CC, Schneider ME, Malliaras P, Chadwick M, Connell DA. 67. Docking SI, Cook J. Pathological tendons maintain sufficient
Diagnostic performance of axial-strain sonoelastography in con- aligned fibrillar structure on ultrasound tissue characterization
firming clinically diagnosed Achilles tendinopathy: comparison (UTC). Scand J Med Sci Sports. 2016;26(6):675–83.
with B-mode ultrasound and color Doppler imaging. Ultrasound 68. Docking S, Samiric T, Scase E, Purdam C, Cook J. Relationship
Med Biol. 2015;41(1):15–25. between compressive loading and ECM changes in tendons. Mus-
50. Wang HK, Lin KH, Su SC, Shih TT, Huang YC. Effects of ten- cle Ligaments Tendons J. 2013;3(1):7–11.
don viscoelasticity in Achilles tendinosis on explosive perfor- 69. Obst SJ, Boyd R, Read F, Barber L. Quantitative 3-D ultrasound of
mance and clinical severity in athletes. Scand J Med Sci Sports. the medial gastrocnemius muscle in children with unilateral spas-
2012;22(6):e147–55. tic Cerebral Palsy. Ultrasound Med Biol. 2017;43(12):2814–23.
51. Chimenti RL, Bucklin M, Kelly M, Ketz J, Flemister AS, Rich- 70. Martin JA, Biedrzycki AH, Lee KS, DeWall RJ, Brounts SH,
ards MS, et al. Insertional Achilles tendinopathy associated with Murphy WL, et al. In vivo measures of shear wave speed as a
altered transverse compressive and axial tensile strain during predictor of tendon elasticity and strength. Ultrasound Med Biol.
ankle dorsiflexion. J Orthop Res. 2017;35(4):910–5. 2015;41(10):2722–30.
52. Couppe C, Kongsgaard M, Aagaard P, Vinther A, Boesen M, 71. Heales L, Lim E, Hodges P, Vicenzino B. Sensory and motor
Kjaer M, et al. Differences in tendon properties in elite badminton deficits exist on the non-injured side of patients with unilateral
players with or without patellar tendinopathy. Scand J Med Sci tendon pain and disability—implications for central nervous sys-
Sports. 2013;23(2):e89–95. tem involvement: a systematic review with meta-analysis. Br J
53. Helland C, Bojsen-Moller J, Raastad T, Seynnes OR, Moltub- Sports Med. 2014;48(19):1400–6.
akk MM, Jakobsen V, et al. Mechanical properties of the patellar 72. Grigg NL, Wearing SC, O’Toole JM, Smeathers JE. Achilles ten-
tendon in elite volleyball players with and without patellar tendi- dinopathy modulates force frequency characteristics of eccentric
nopathy. Br J Sports Med. 2013;47(13):862–8. exercise. Med Sci Sports Exerc. 2013;45(3):520–6.
54. Kongsgaard M, Qvortrup K, Larsen J, Aagaard P, Doessing S, 73. Girish G, Lobo LG, Jacobson JA, Morag Y, Miller B, Jamadar
Hansen P, et al. Fibril morphology and tendon mechanical prop- DA. Ultrasound of the shoulder: asymptomatic findings in men.
erties in patellar tendinopathy: effects of heavy slow resistance Am J Roentgenol. 2011;197(4):W713–9.
training. Am J Sports Med. 2010;38(4):749–56. 74. Kubo K, Morimoto M, Komuro T, Tsunoda N, Kanehisa H,
55. Lee WC, Zhang ZJ, Masci L, Ng GYF, Fu SN. Alterations in Fukunaga T, et al. Age-related differences in the properties of
mechanical properties of the patellar tendon is associated with the plantar flexor muscles and tendons. Med Sci Sports Exerc.
pain in athletes with patellar tendinopathy. Eur J Appl Physiol. 2007;39(3):541–7.
2017;117(5):1039–45. 75. Kubo K, Kanehisa H, Fukunaga T. Gender differences in the
56. Zhang ZJ. Ng GY-f, Lee WC, Fu SN. Changes in morphological viscoelastic properties of tendon structures. Eur J Appl Physiol.
and elastic properties of patellar tendon in athletes with unilateral 2003;88(6):520–6.
patellar tendinopathy and their relationships with pain and func- 76. Ochala J, Valour D, Pousson M, Lambertz D, Van Hoecke J. Gen-
tional disability. PLoS One. 2014;9(10):e108337. der differences in human muscle and joint mechanical properties
57. Coombes BK, Tucker K, Vicenzino B, Vuvan V, Mellor R, Heales during plantar flexion in old age. J Gerontol A Biol Sci Med Sci.
L, et  al. Achilles and patellar tendinopathy display opposite 2004;59(5):441–8.
changes in elastic properties: a shear wave elastography study. 77. Onambele GN, Burgess K, Pearson SJ. Gender-specific in vivo
Scand J Med Sci Sports. https​://doi.org/10.1111/sms.12986​. measurement of the structural and mechanical properties of the
58. Dirrichs T, Quack V, Gatz M, Tingart M, Kuhl CK, Schrading S. human patellar tendon. J Orthop Res. 2007;25(12):1635–42.
Shear wave elastography (SWE) for the evaluation of patients with 78. Cook JL, Purdam C. Is compressive load a factor in the develop-
tendinopathies. Acad Radiol. 2016;23(10):1204–13. ment of tendinopathy? Br J Sports Med. 2012;46(3):163.
59. Maffulli N, Testa V, Capasso G, Ewen SW, Sullo A, Benazzo F, 79. Pearson SJ, Mohammed ASA, Hussain SR. Patellar tendon
et al. Similar histopathological picture in males with Achilles and in  vivo regional strain with varying knee angle. J Biomech.
patellar tendinopathy. Med Sci Sports Exerc. 2004;36(9):1470–5. 2017;61:45–50.
 S. J. Obst et al.

80. Malliaras P, Barton CJ, Reeves ND, Langberg H. Achilles and 83. Robson M, Benjamin M, Gishen P, Bydder G. Magnetic resonance
patellar tendinopathy loading programmes: a systematic review imaging of the Achilles tendon using ultrashort TE (UTE) pulse
comparing clinical outcomes and identifying potential mecha- sequences. Clin Radiol. 2004;59(8):727–35.
nisms for effectiveness. Sports Med. 2013;43(4):267–86. 84. Smith DW, Rubenson J, Lloyd D, Zheng M, Fernandez J, Besier T,
81. Cook J, Purdam CR. Is tendon pathology a continuum? A pathol- et al. A conceptual framework for computational models of Achil-
ogy model to explain the clinical presentation of load-induced les tendon homeostasis. Wiley Interdiscip Rev Syst Biol Med.
tendinopathy. Br J Sports Med. 2009;43(6):409–16. 2013;5(5):523–38.
82. Pizzolato C, Lloyd DG, Barrett RS, Cook JL, Zheng MH, Besier 85. Seynnes OR, Bojsen-Møller J, Albracht K, Arndt A, Cronin NJ,
TF, et al. Bioinspired technologies to connect musculoskeletal Finni T, et al. Ultrasound-based testing of tendon mechanical prop-
mechanobiology to the person for training and rehabilitation. erties: a critical evaluation. J Appl Physiol. 2015;118(2):133–41.
Front Comput Neurosc. 2017. https​://doi.org/10.3389/fncom​
.2017.00096​.