Sports Med

DOI 10.1007/s40279-016-0539-4

SYSTEMATIC REVIEW

Forearm Muscle Activity in Lateral Epicondylalgia: A Systematic

Review with Quantitative Analysis
Luke J. Heales1,2 Michael J. G. Bergin1
Bill Vicenzino1 Paul W. Hodges1

Springer International Publishing Switzerland 2016

Abstract
Background Lateral epicondylalgia (LE) refers to pain at
the lateral elbow and is associated with sensory and motor
impairments that may impact on neuromuscular control
and coordination.
Objective This review aimed to systematically identify
and analyse the literature related to the comparison of
neuromuscular control of forearm muscles between individuals with and without LE.
Methods A comprehensive search of electronic databases
and reference lists using keywords relating to neuromuscular control and LE was undertaken. Studies that investigated electromyography (EMG) measures of forearm
muscles in individuals with symptoms of LE were included
if the study involved comparison with pain-free controls.
The Epidemiological Appraisal Instrument was used to
assess study quality. Data extracted from each study were
used to calculate the standardised mean difference and
95 % confidence intervals to investigate differences
between groups.

Results The search revealed a total of 1920 studies, of

which seven were included from 44 that underwent
detailed review. Due to differences in outcome measures
and tasks assessed, meta-analysis was not possible. The
seven included studies used 60 different EMG outcomes, of
which 16 (27 %) revealed significant differences between
groups. Two were properties of motor unit potentials during wrist extension. Four were measures of increased time
between recruitment of wrist extensor muscles and onset of
grip force. Seven were measures of amplitude of EMG
during tennis strokes. Three were measures of motor cortex
organisation.
Conclusion Features of neuromuscular control differ
between individuals with LE and pain-free controls. This
implies potential central nervous system involvement and
indicates that rehabilitation may be enhanced by consideration of neuromuscular control in addition to other
treatments.

Key Points

Electronic supplementary material The online version of this

article (doi:10.1007/s40279-016-0539-4) contains supplementary
material, which is available to authorized users.
& Paul W. Hodges
p.hodges@uq.edu.au
1

NHMRC Centre of Clinical Research Excellence in Spinal

Pain, Injury and Health, School of Health and Rehabilitation
Science, The University of Queensland, Brisbane, QLD 4072,
Australia
School of Human, Health and Social Sciences, Division of
Physiotherapy, Central Queensland University,
Rockhampton, QLD 4701, Australia

Some features of neuromuscular control appear

different between individuals with and without
lateral epicondylalgia and these could imply central
nervous system involvement in the condition.
Systematic review of the literature revealed evidence
of differences in 16 of 60 unique electromyography
measures in seven studies.
Widely disparate measurement methods precluded
meta-analysis of data, but reveal differences across a
range of tasks.

123

L. J. Heales et al.

1 Introduction
Lateral epicondylalgia (LE), or tennis elbow, refers to pain
at the lateral epicondyle and is characterised by pain and
decreased strength during gripping [1, 2]. This condition
impacts on participation in recreational and occupational
activities [3, 4]. Although commonly unilateral in presentation, LE is associated with bilateral changes of some
elements of the motor system. Examples include gripping
with the wrist in a less extended position than pain-free
controls [5], deficits in upper limb strength [6], altered fine
motor control (measured by the Purdue Pegboard Test and
Complete Manual Dexterity Test) [7], slower reaction time
[5, 8], and slower speed of movement [5, 8]. This diverse
array of differences in the motor system between individuals with and without LE is likely to be related to modified
neuromuscular control of forearm muscles.
Neuromuscular control is often quantified using recordings of myoelectric activity [9, 10] as a measure of the net
output of the processes in the central nervous system (CNS).
Investigation of neuromuscular control associated with LE is
likely to aid development of an understanding of potential
impairments in the motor system and inform the development of targeted treatments. There is evidence that neuromuscular control in individuals with LE differs from that of
pain-free controls, but methods and findings are diverse. The
primary aim of this study was to systematically review the
literature regarding changes in myoelectric activity of the
forearm muscles in LE. Human studies of electromyography
(EMG) recordings of forearm muscles in individuals with LE
(bilateral and unilateral) that included a comparison with
pain-free controls were included in this review.

2 Methods
2.1 Search Strategy
Electronic databases (Medlinevia Ovid, PubMed and
Scopus) were searched to identify all English-language
studies for all years up to July 2015. Keyword, title and
abstract information were used. Search terms were the
condition of interest (e.g. tennis elbow, epicondyl*) and
common terms used to describe neuromuscular control (e.
g. electromy*, EMG, motor control) (wildcard * represents any letters). The reference lists of all included studies
were systematically hand-searched to identify articles that
may have been missed by the initial screening. This process included articles not on electronic databases, articles
from networks or conferences, and theses and books.

123

2.2 Study Selection

Upon retrieval from the above search strategy, all titles and
abstracts were screened to identify studies that included
EMG outcomes in individuals with LE and comparison
with pain-free controls. It was decided a priori that any
measures of forearm muscle EMG in individuals with LE
should be included provided there was comparison with a
pain-free control group. The full text was obtained for all
eligible studies. A detailed evaluation using pre-determined
criteria based on study design, diagnosis of LE and reports
of EMG outcome measures was undertaken. Reviews, case
studies, and letters to the editor were excluded. Clarifications on matters of eligibility were made by discussion with
another investigator.
2.3 Quality Assessment
Two assessors (LH, MB) used the modified Epidemiological Appraisal Instrument (EAI) to independently score the
quality of the included studies [11]. The EAI is valid and
reliable for appraisal of observational studies [11]. The
design of the included studies precluded relevance of items
related to randomisation, follow-up and environmental
variables and these items were excluded from the assessment and the EAI was condensed to 33 items of a possible
43 items. Prior to the quality evaluation, detailed criteria to
determine each response were modified from the original
tool to match the purpose of this review and agreed upon
by all investigators. A third party (BV) was consulted for
disagreements between the two assessors. Each item was
independently scored using the standardised scale: yes
(score = 1), partial (score = 0.5), no (score = 0), unable
to determine (score = 0), or not applicable (item was
removed from scoring) and the average used as the indicator of the overall quality assessment score for each study
(range 01).
2.4 Data Extraction
LH undertook data extraction, and any queries were discussed and resolved by all investigators. Data pertaining to
the sample population and study methodology as well as
the descriptive data of the reported EMG outcome measures were extracted. All EMG measures that compared
participants with LE (symptomatic or asymptomatic limb)
and pain-free individuals were included. All EMG measures were considered relevant (e.g. root mean squared
[RMS] amplitude, median frequency). Authors were contacted if additional information was required.

LE (11)

Controls (11)

LE (16)

Controls (14)

LE (14)

Bilateral LE
(15)
Controls (16)

Unilateral LE
(11)

Controls (13)

LE (11)

Controls (37)

38.3
(29.4)

NR

63 (69)

104 (26
521)

ND = 4

D=7

NR

ND = 4

D = 10

ND = 2

D = 11

Unilateral

NR

ND = 0
NR

D = 16

Affected
arm

Relaxation

Single-handed backhand tennis

stroke

Corticomotor excitability

Root mean squared amplitude

Corticomotor excitability

Temporal parameters

Maximum isometric grip as rapidly

as possible

Relaxation

Needle- and surface-detected motor

unit potentials

Median frequency
Mean activation times

Root mean squared amplitude

Outcome measures

Resisted wrist extension at 510 %

MVC

Single-handed backhand tennis

volley

Sustained isometric grip at 50 %

MVC

Task

ECRB, EDC

ECRB, ECRL, EDC,

FCR, PT

ECR

ECR

ECRB

ECRB, FCU, TB

ECR, EDC, FCU, FDS

Muscle recorded

Surface

Intramuscular

Surface

Surface

Surface and
intramuscular

Surface

Surface

Electrodes

D dominant, DOS duration of symptoms, ECR extensor carpi radialis, ECRB extensor carpi radialis brevis, ECRL extensor carpi radialis longus, EDC extensor digitorum communis, FCR flexor
carpi radialis, FCU flexor carpi ulnaris, FDS flexor digitorum superficialis, LE lateral epicondylalgia, MVC maximum voluntary contraction, ND non-dominant, NR not reported, PT pronator
teres, TB triceps brachii

Schabrun et al. [18]

Kelley et al. [19]

Dessureault et al.
[16]

Chourasia et al.
[20]

Calder et al. [17]

169 (130)

NR

LE (10)

LE (16)
Controls (6)

Bauer and Murray

[15]

NR

Controls (16)

Alizadehkhaiyat
et al. [6]

DOS
(weeks)

Participants
(number)

Study

Table 1 Characteristics of included studies

Forearm Muscle Activity in Lateral Epicondylalgia

123

L. J. Heales et al.

Fig. 1 Flow chart for inclusion into the review. EMG electromyographic

2.5 Statistical Methods

The reliability of the quality assessment was evaluated
using SPSS V21 software (SPSS Inc, Chicago, IL, USA).
Kappa statistics were used to report the inter-rater reliability between the two assessors. Inter-rater reliability was
considered as poor (\0.00), slight (0.000.2), fair (0.21
0.4), moderate (0.410.6), substantial (0.610.8), or almost
perfect (0.811.0) [12]. Meta-analysis was not possible due
to heterogeneity between studies with respect to the tasks
and outcome measures included in the studies (Table 1). To

123

enable consistent comparison between findings of individual studies and outcome measures, the standardised
mean difference (SMD) and 95 % confidence interval (CI)
were calculated and interpreted as small (0.2), medium
(0.5) and large (0.8) effect size [13]. If the 95 % CI did not
include zero, the outcome was considered statistically
significant. A positive SMD reflects greater values in LE
group than controls and vice versa. Because of the small
sample sizes of the included studies, we also calculated
90 % CIs. If the 90 % CI did not contain zero the outcome
was considered to be nearly significant.

Forearm Muscle Activity in Lateral Epicondylalgia

3 Results

3.3 Quality Assessment

3.1 Database Search

Agreement between the two reviewers was almost perfect

( = 0.90, p \ 0.001) [12] with 222 agreements out of 238
decisions. The quality assessment scores are presented in
Table S1 (see electronic supplementary material). The
quality assessment revealed only two studies [17, 20] (28 %)
that clearly described the study design with details assessed
using the EAI tool. No longitudinal or prospective studies
were identified. Only three studies [16, 17, 20] (43 %) clearly
described their study sample, including how and where they
recruited participants. One study [20] (14 %) reported an a
priori sample size calculation, and only one study [20] (14 %)
used diagnostic imaging as part of the eligibility criteria.
Three studies [15, 17, 19] (43 %) did not report duration of
symptoms and three studies [6, 15, 19] (43 %) did not report
data from validated pain and disability questionnaires (e.g.
Patient-Rated Tennis Elbow Evaluation Form). No studies
blinded the assessors to the condition of the participant. The
reviewers were unable to detect the timeframe over which
the participants were recruited for any of the seven studies.
Two studies [16, 18] (28 %) matched for age and sex to
facilitate comparability between groups (LE and controls).
Three studies [6, 15, 20] (43 %) displayed the age and sex of
the participants but did not use statistical tests to compare
groups. One study [17] (14 %) showed that the age of their
control group was significantly younger than their LE group.
One study [19] (14 %) used control participants from a
previous paper and did not mention their age, sex or arm
dominance. The reviewers were unable to establish the
validity or reliability of LE diagnosis in any study. Despite
the importance of using validated measures, no studies
reported the validity or reliability of their main outcome
measures. Only one study [20] (14 %) adjusted for individual
covariates (e.g. age, sex), and no studies reported the results
relative to severity of the condition.

The comprehensive search strategy yielded a total of 1920

publications from electronic databases and hand-searched
reference lists (Fig. 1). All titles and abstracts were
screened and 44 potentially relevant studies were identified. Of these, eight satisfied the inclusion criteria and
reported EMG outcomes in participants with LE and
included comparison with a pain-free control group. One
study was excluded as it reported EMG of the hamstring
muscles during the nociceptive flexion reflex in participants with LE and no recordings from the forearm muscles
[14]. Exclusion of this study left seven studies that satisfied
the criteria for inclusion in this review. Two authors were
contacted to provide additional information as data were
only presented graphically. One study [15] was unable to
provide data due to technological advances since publication (i.e. the inability to retrieve data from floppy disc
storage media). The other study [16] provided the mean
and standard deviation for the outcomes of interest and
permission to publish these data.
3.2 Study Characteristics
Study characteristics are shown in Table 1. The location of
the study, year of publication, sample size and methodology
varied widely. Three studies (43 %) were conducted in the
United States, two in Canada (28 %), and one each in the
United Kingdom (14 %) and Australia (14 %). Two studies
(28 %) were published prior to 2000; three (43 %) in the
2000s; and two (28 %) since 2010. Sample sizes varied from
16 [15] to 48 [17] participants. Overall, the number of LE
participants (total n = 104) was greater than the number of
controls (total n = 99). The mean duration of symptoms for
individuals with LE ranged from 38 [18] to 169 [17] weeks.
All studies used a case-control study design. The muscles
assessed using EMG differed between studies. All studies
described data as representative of both extensor carpi
radialis (ECR) brevis and longus combined, or selective for
extensor carpi radialis brevis (ECRB) alone. Three studies
[6, 18, 19] recorded EMG from extensor digitorum communis (EDC), two [6, 15] from flexor carpi ulnaris (FCU)
and one each from flexor carpi radialis (FCR) [19], flexor
digitorum superficialis (FDS) [6], pronator teres (PT) [19]
and triceps brachii (TB) [15]. Only one study described
EMG recording from extensor carpi radialis longus (ECRL)
separately from ECRB [19]. Five studies used surface
electrodes alone [6, 15, 16, 18, 20], one study used intramuscular electrodes alone [19], and one study used both
surface and intramuscular electrodes [17] (Table 1).

3.4 Neuromuscular Control

No outcomes were investigated by more than one study
(Table S2, electronic supplementary material). Quantitative
analysis of the seven included studies revealed a total of 60
neuromuscular control outcomes, of which 16 (28 %) were
significant using the SMD and 95 % CI. The following sections present findings according to the four identified tasks/
measures: (1) resisted wrist extension, (2) gripping, (3) tennis
strokes and (4) transcranial magnetic stimulation (TMS).
3.4.1 Resisted Wrist Extension
One study investigated morphology of motor unit action
potentials of ECRB during resisted wrist extension (Table 2)
[17]. Although this study revealed a greater duration and

123

L. J. Heales et al.
Table 2 Outcome measures and calculated SMD and 95 % CI for resisted wrist extension [17]
Outcome

Muscle

LE group

Control group

Mean

SD
426.6

SMD

Mean

SD

11

419

300.7

95 % CI

NMUP amplitude (mV)

ECRB

519

37

0.08

0.56 to 0.76

NMUP area-to-amplitude ratio

ECRB

1.62

0.59

11

1.27

0.43

37

0.73

0.04 to 1.42a

NMUP duration (ms)

ECRB

9.7

3.16

11

7.64

2.85

37

0.69

0.01 to 1.38a

NMUP firing rate (Hz)

ECRB

13.86

2.71

11

14.98

2.97

37

0.38

1.06 to 0.30

NMUP number of phases

ECRB

2.63

0.81

11

2.55

0.71

37

0.11

0.57 to 0.78

SMUP amplitude (mV)

ECRB

111.14

109.53

11

113.73

90.73

37

0.03

0.70 to 0.65

SMUP area (mm)

SMUP duration (ms)

ECRB
ECRB

502.7
19.21

518.7
5.93

11
11

593.9
19.76

507.4
5.52

37
37

0.18
0.10

0.85 to 0.50
0.77 to 0.58

CI confidence interval, ECRB extensor carpi radialis brevis, LE lateral epicondylalgia, N number of participants, NMUP needle-detected motor
unit potential, SD standard deviation, SMD standardised mean difference, SMUP surface-detected motor unit potential
a

Significant difference between LE and controls

Table 3 Outcome measures and calculated SMD and 95 % CI for grip

Study

Alizadehkhaiyat et al. [6]

Chourasia et al. [20]

Outcome

Muscle

LE group

Control group

Mean

SD

Mean

SD

SMD

95 % CI

RMS amplitude
RMS amplitude

ECR
EDC

0
11

28
36

16
16

13
13

20
32

16
16

0.52
0.06

1.23 to 0.19
0.75 to 0.64

RMS amplitude

FCU

37

32

16

32

24

16

0.17

0.52 to 0.87

RMS amplitude

FDS

32

24

16

27

16

16

0.24

0.46 to 0.93

Median frequency

ECR

16

16

16

16

0.00

0.69 to 0.69

Median frequency

EDC

15

16

15

16

0.00

0.69 to 0.69

Median frequency

FCU

14

16

15

16

0.15

0.54 to 0.85

Median frequency

FDS

12

16

15

12

16

0.33

0.52 to 0.19

Delaybilateral dominant

ECR

0.061

0.02

15

0.039

0.008

11

1.32

0.45 to 2.19a

Delaybilateral non-dominant

ECR

0.065

0.033

15

0.039

0.014

11

0.94

0.11 to 1.77a

Delayunilateral symptomatic

ECR

0.061

0.029

11

0.039

0.008

11

0.99

0.10 to 1.98a

Delayunilateral asymptomatic

ECR

0.064

0.024

11

0.039

0.014

11

1.22

0.30 to 2.15a

CI confidence interval, ECR extensor carpi radialis (brevis and longus combined), EDC extensor digitorum communis, FCU flexor carpi ulnaris,
FDS flexor digitorum superficialis, LE lateral epicondylalgia, N number of participants, RMS root mean squared, SD standard deviation, SMD
standardised mean difference
a

Significant difference between LE and controls

greater area-to-amplitude ratio of motor unit action potentials

in the LE group than in pain-free controls [17], methodological issues limit the ability to draw robust conclusions. First,
properties of motor unit action potential shape are affected by
proximity of the electrode to the muscle fibres and this was
impossible to control. Second, the LE participants were significantly older (mean age 46.6 10.7 years) than the healthy
controls (27.1 5 years). This is problematic for comparison
as motor unit morphology changes with age, usually around
the fifth decade of life [21]. Third, the force of contraction was
not standardised between groups/participants (range from 5 to
10 % maximum voluntary contraction (MVC). As motor unit
recruitment depends on force [22], participants who performed contractions at higher levels of force may have systematically recruited larger motor units. No other significant

123

differences were observed for motor unit action potentials

recorded with needle (e.g. amplitude, firing rate, number of
phases) or surface electrodes (e.g. amplitude, area and duration) (Table 2) [17].
3.4.2 Gripping
Two studies investigated aspects of neuromuscular control
during gripping (Table 3) [6, 20]. One investigated the change
in RMS amplitude over time, normalised to the start value,
and median frequency of ECR, EDC, FDS and FCU during a
submaximal grip at 50 % [range from 5 to 10 % maximum
voluntary contraction (MVC)] until exhaustion [6]. RMS
amplitude was calculated at 5-second intervals and assessed by
linear regression. Using the criterion of SMD and 95 % CI,

Forearm Muscle Activity in Lateral Epicondylalgia

there were no significant between-group differences in

amplitude of muscle activity over time (Table 3). This differs
from the outcome of the statistical analysis reported in the
study by Alizadehkhaiyat et al. [6], but we argue that reference
to SMD and 95 % CI is more robust. A major consideration of
this study is the interpretation that there is less RMS amplitude
in the ECR muscle during gripping in the LE group than in the
pain-free controls [6]. EMG amplitude was normalised to that
at the start of the sustained contraction and assessed by linear
regression. This method does not allow comparison of muscle
amplitude between groups as this initial value used for normalisation may differ for each group and provides a reference
that is difficult to interpret. If the participants in the LE group
commenced the contraction with greater activity (start value),
they might have less ability to increase the RMS amplitude
over time, thus affecting the slope of the EMG change and the
statistical comparison of activity over time between groups.
Additionally, this study did not record whether the included
participants did or did not experience pain during the gripping
task. Although inclusion of a task at 50 % MVC may have
resulted in pain provocation for some participants, the lack of
recorded information and our inability to find significance
between groups render this difficult to interpret.
Another study investigated the delay in time from the
onset of ECR EMG to the onset of grip force development,
measured using a specialised multi-axis dynamometer,
with the task performed in response to a visual stimulus
[20]. Participants with unilateral and bilateral LE demonstrated a longer delay between these two events, bilaterally,
than pain-free controls (i.e. both symptomatic and asymptomatic arm were affected) (Table 3) [20]. Although the
authors used the term electromechanical delay (EMD) to
describe this finding, this is not entirely accurate as EMD is
defined as the time interval between the initiation of EMG
from a muscle and the first detectable force output that the
muscle produces [23]. In this study, the onset of force was
identified from grip force, which is initiated by the finger
flexor muscles [FDS and flexor digitorum profundus
(FDP)], and does not reflect the onset of wrist extension
force. The wrist extensor muscles [ECRB, ECRL, EDC,
and extensor carpi ulnaris (ECU)] will co-activate with the
finger flexors to stabilise the wrist [24], but this measure
cannot be used as a surrogate of wrist extensor force.
Rather than simple interpretation as a delay in transduction
from EMG to force, increase in the latency between ECRB
EMG and grip force could reflect altered coordination
between finger flexor and wrist extensor muscles between
participant groups.
3.4.3 Tennis Strokes
Two studies investigated neuromuscular control during
single-handed backhand tennis strokes [15, 19]. Analysis

using our a priori criterion could only be conducted on

one study [19], as data from the other study were not
available due to technological advances since publication
(i.e. the inability to retrieve data from floppy disc storage
media) [15]. Between-group differences were observed
for muscle amplitude (expressed as a percentage of MVC)
during a single-handed backhand tennis stroke, which was
divided into six phases: (1) preparation, (2) early and (3)
late acceleration, (4) ball impact, and (5) early and (6) late
follow through [19]. During the preparation phase, the LE
group demonstrated greater ECRL and FCR EMG than
pain-free controls (Table 4). During ball impact, the LE
group demonstrated a greater ECRB, ECRL and PT EMG
than pain-free controls. During early follow through, the
LE group demonstrated a greater ECRB and PT EMG
than pain-free controls [19]. There were no betweengroup differences for the early acceleration or the late
follow-through phases using the criteria of SMD and 95 %
CI. Using a more lenient 90 % CI, several measures could
be considered nearly significant: ECRB EMG was
greater in LE during the early acceleration phase (SMD
0.90, 90 %CI 1.67 to 0.13) and FCR EMG was greater
during the late follow-through phase (SMD 0.92, 90 % CI
0.151.69) [19].
The study, which presented data that could not be used to
calculate SMD, investigated temporal parameters for ECRB,
FCU and TB during a single-handed backhand tennis volley
at three ball speeds (low 11.94 m/s, medium 17.13 m/s, and
high 22.95 m/s) and three racket head locations (central, long
axis, and torsional) [15]. LE participants had greater EMG
duration and greater integrated ECRB EMG for each ball
speed and each racket location (p [ 0.05) than pain-free
controls. When EMG times were pooled across ball speeds
and racket locations and presented as values pre- and postball impact, the LE group demonstrated an earlier ECRB
EMG onset during the pre-impact period and a longer
duration of EMG (later EMG offset) into the post-impact
period than pain-free controls [15]. No differences were
observed for FCU and TB. No studies recorded pain levels
during tennis strokes, making it difficult to interpret whether
pain influenced the forearm muscle activity or timing in
individuals with LE. Although there was no mention of pain
in either study, participants with LE may have developed
specific patterns of motor activity with repeated exposure to
previously painful activities.
3.4.4 Transcranial Magnetic Stimulation
Two studies investigated corticomotor measures using
TMS in individuals with LE and pain-free controls [16,
18]. One study revealed that, compared with pain-free
controls, participants with LE had a smaller separation
between the cortical representations of ECRB and EDC

123

L. J. Heales et al.
Table 4 Outcome measures and calculated SMD and 95 % CI for tennis strokes [19]
Outcome

Muscle

LE group
Mean

Control group
SD

Mean

SD

SMD

95 % CI

RMS amplitudepreparation

ECRB

28

23

18

11

14

0.59

0.31 to 1.48

RMS amplitudepreparation

ECRL

28

19

13

10

14

1.04

0.11 to 1.98a

RMS amplitudepreparation
RMS amplitudepreparation

EDC
FCR

13
27

9
26

8
8

11
9

7
5

14
14

0.25
1.09

0.62 to 1.12
0.15 to 2.03a

RMS amplitudepreparation

PT

14

12

13

12

14

0.08

0.79 to 0.95

RMS amplitudeearly acceleration

ECRB

28

12

62

44

14

0.90

1.82 to 0.01

RMS amplitudeearly acceleration

ECRL

36

40

35

23

14

0.03

0.84 to 0.90

RMS amplitudeearly acceleration

EDC

44

33

57

39

14

0.34

1.21 to 0.54

RMS amplitudeearly acceleration

FCR

19

18

14

17

14

0.28

0.60 to 1.15

RMS amplitudeearly acceleration

PT

11

17

16

17

14

0.28

1.16 to 0.59

RMS amplitudelate acceleration

ECRB

78

31

83

37

14

0.14

1.01 to 0.73

RMS amplitudelate acceleration

ECRL

81

47

72

38

14

0.21

0.66 to 1.08

RMS amplitudelate acceleration

EDC

81

31

77

19

14

0.16

0.71 to 1.03

RMS amplitudelate acceleration

FCR

53

47

38

27

14

0.41

0.47 to 1.29

RMS amplitudelate acceleration

PT

17

15

29

22

14

0.58

1.47 to 0.31

RMS amplitudeball impact

ECRB

94

42

40

31

14

2.02

0.93 to 3.11a

RMS amplitudeball impact

ECRL

89

29

43

43

14

1.14

0.20 to 2.09a

RMS amplitudeball impact

EDC

42

25

72

55

14

0.62

1.51 to 0.27

RMS amplitudeball impact

RMS amplitudeball impact

FCR
PT

70
60

52
26

8
8

56
26

34
25

14
14

0.33
1.29

0.55 to 1.20
0.32 to 2.26a

RMS amplitudeearly FT

ECRB

67

27

43

19

14

1.04

0.11 to 1.98a
0.17 to 1.63

RMS amplitudeearly FT

ECRL

62

25

42

27

14

0.73

RMS amplitudeearly FT

EDC

45

18

50

27

14

0.20

1.07 to 0.67

RMS amplitudeearly FT

FCR

53

28

41

24

14

0.45

0.43 to 1.33

RMS amplitudeearly FT

PT

61

32

32

20

14

1.12

0.18 to 2.06a

RMS amplitudelate FT

ECRB

28

24

18

13

14

0.55

0.34 to 1.43
0.29 to 1.50

RMS amplitudelate FT

ECRL

23

14

15

12

14

0.60

RMS amplitudelate FT

EDC

16

12

21

12

14

0.40

1.28 to 0.48

RMS amplitudelate FT

FCR

23

19

11

14

0.92

0.00 to 1.84

RMS amplitudelate FT

PT

24

26

13

14

0.62

0.27 to 1.51

CI confidence interval, ECRB extensor carpi radialis brevis, ECRL extensor carpi radialis longus, EDC extensor digitorum communis, FCR flexor
carpi radialis, FT follow through, LE lateral epicondylalgia, N number of participants, PT pronator teres, RMS root mean squared, SD standard
deviation, SMD standardised mean difference
a

Significant difference between LE and controls

and fewer peaks of excitability in the motor cortex maps

for ECRB and EDC (Table 5) [18]. The other study found
no significant differences for any measures of corticomotor excitability (e.g. motor-evoked potential [MEP],
resting motor threshold and silent period) [16].
When the data were analysed using a 90 % CI,
several additional measures approached significance.
Compared with pain-free controls, individuals with LE
showed a greater motor cortex map volume for ECRB
(SMD 0.89, 90 % CI 0.141.63) and a greater peak
MEP amplitude for ECRB (SMD 0.88, 90 % CI 0.14
1.62) [18].

123

4 Discussion
The primary aim of this review was to systematically analyse
evidence for differences in the activity of forearm muscles
between individuals with and without LE. Quantitative analysis of data from multiple studies showed 60 outcome measures, of which 16 (27 %) of these differ significantly between
individuals with LE and pain-free controls. Of these 16 outcomes, two were properties of motor unit potentials during
resisted wrist extension; four were measures of increased time
between recruitment of wrist extensor muscles and onset of
grip force; seven were measures of amplitude of EMG during

Forearm Muscle Activity in Lateral Epicondylalgia

Table 5 Outcome measures and calculated SMD and 95 % CI for corticomotor measures
Study

Dessureault et al.
[16]
Schabrun et al. [18]

Outcome

Muscle

LE group

Control group

Mean

SD

Mean

SD

SMD

95 % CI

MEPs

ECR

101.28

79.53

15

115.6

103.67

16

0.15

0.86 to 0.56

Resting motor threshold

ECR

39.4

8.76

15

40.5

7.65

16

0.13

0.84 to 0.57

Silent period

ECR

75

29.84

15

72.56

19.15

16

0.10

0.61 to 0.80

Distance between COG for

ECRB and EDC

ECRB/EDC

0.05

11

0.20

0.10

11

1.10

Cortical volume

ECRB

Cortical volume
Peak MEP amplitude

0.11

2.00 to 0.19a

15.5

13.1

11

6.4

4.9

11

0.89

0.00 to 1.77

EDC

9.3

5.5

11

8.2

5.6

11

0.19

0.65 to 1.03

ECRB

1.3

0.79

11

0.71

0.46

11

0.88

0.01 to 1.76

Peak MEP amplitude

EDC

0.82

0.35

11

0.64

0.37

11

0.48

0.37 to 1.33

Number of cortical peaks

ECRB

1.4

0.8

11

3.4

1.5

11

1.60

2.43 to 0.77a

Number of cortical peaks

EDC

1.8

1.0

11

3.2

1.5

11

1.06

1.82 to 0.30a

CI confidence interval, COG centre of gravity, ECR extensor carpi radialis (brevis and longus combined), ECRB extensor carpi radialis brevis,
EDC extensor digitorum communis, LE lateral epicondylalgia, MEP motor-evoked potential, N number of participants, SD standard deviation,
SMD standardised mean difference
a

Significant difference between LE and controls

single-handed backhand tennis strokes; and three were measures of motor cortex organisation.
4.1 Interpretation of Differences in Forearm Muscle
Activity in Lateral Epicondylalgia
4.1.1 Gripping
A surprising observation of this review is the paucity of
studies that have investigated forearm muscle activity during
gripping, despite the fact that pain during gripping is a primary clinical complaint in individuals with LE. Several
impairments in kinematics and force have been identified in
individuals with LE during gripping (e.g. altered wrist
position during gripping [5], decreased grip strength [6, 20]),
but only two studies have investigated forearm muscle
activity [6, 20]. No studies have investigated for possible
differences in EMG amplitude during gripping. Although
Alizadehkhaiyat et al. [6] reported a smaller increase in RMS
amplitude over time in ECR (normalised to the amplitude at
the start of the sustained contraction) in individuals with LE
than in controls, this did not reach our a priori defined
threshold for significance based on SMD and 95 % CI.
Further, as mentioned earlier, normalisation to the amplitude
at the start of the contraction, and the statistical analysis
using linear regression to determine between-group differences, render the data difficult to interpret.
Differences in temporal parameters have been identified
with gripping, but there is some confusion regarding the
interpretation of the data. The latency between the onset of
ECR EMG and the onset of grip force during rapid gripping

is longer, bilaterally, in individuals with unilateral and

bilateral LE than in pain-free controls [20]. Although this
was interpreted as increased EMD, the onset of force was
not that generated by the homonymous muscle. Rather than
reflecting the change in transduction of EMG to force generated by the same muscle (which is the traditional definition
of EMD), increased delay between the onset of ECR EMG
and force generated by the finger flexor muscles might
instead indicate a change in inter-muscle coordination,
mediated by either; (1) earlier activation of the ECR muscle,
or (2) delayed activation of the finger flexor muscles.
Although reaction time of finger flexor muscles was not
directly assessed during that study, the rate of force development was reduced in individuals with LE. This may be
argued to indicate a potential slower activation of finger
flexor muscles. This reduced rate of force development (and
possibly slower activation of finger flexor muscles) might be
explained by differences in the task between groups. The LE
group were instructed to grip as rapidly and as hard as
possible, but stop prior to the onset of pain (i.e. pain-free
grip), whereas pain-free participants were instructed to grip
to maximum. The objective to stop prior to the onset of pain
might prevent the individuals in the LE group from gripping
as rapidly and as hard as possible. This may be compounded
by fear of pain, which has been shown to limit maximum
contractions [25]. Although a reduced rate of force development was reported (potentially supporting this interpretation) [20], this is only observed in the symptomatic arm of
individuals with LE and cannot explain the bilateral changes
in activation. Another possible explanation is the observation that individuals with LE actively grip with less wrist

123

L. J. Heales et al.

extension (i.e. closer to the neutral position), bilaterally, than

pain-free controls [5]. A less extended wrist might decrease
the tension of the finger flexor tendons, slowing the rate at
which the finger flexors can generate force output, thus
slowing the speed at which grip force can be achieved.
Earlier activation of ECR might also explain the increase in
latency between the onset of ECR EMG and the activation
of grip force in individuals with LE [20]. Previous work
during single-handed tennis strokes has revealed earlier
onset of ECRB than pain-free controls [15]. Whether such a
change reflects a positive adaptation or maladaptive
response is unclear. Earlier activation of ECR in the LE
group might be a purposeful strategy to compensate for a
compromised function of the ECRB muscle [26] and/or
tendon [2729]. Previous histological work has revealed
moth-eaten and necrotic muscle fibres in the symptomatic
ECRB muscle [26], and greater proportion of immature
tenocytes and neovascularisation of the symptomatic tendon
[29] of individuals with LE. Alternatively, changes in
forearm muscle activity might be influenced by regions
higher up the kinetic chain (e.g. shoulder, cervical/thoracic
spine). Although previous work has shown a higher presence
of cervical and thoracic pain in individuals with lateral
elbow pain compared with controls [30], further research is
required to investigate whether this has an impact on forearm muscle activity. Although identification of differences
in the asymptomatic arm might represent a widespread
involvement of the CNS, further investigation is required.
Due to the case-control study design, it is not possible to
determine if changes in the temporal activation of forearm
muscles in individuals with LE are related to the development of LE, or as a consequence of LE. Longitudinal
studies are required to investigate whether motor system
impairments are related to the cause of LE.
4.1.2 Tennis Strokes
A greater [19] and earlier [15] activity of forearm muscles
in individuals with LE than in pain-free controls during
tennis strokes might be explained by several possible
differences and could be related to either the cause or
effect of LE. First, individuals with LE might have a
modified afferent input or altered perception of the task.
Previous work has shown that LE is associated with a
reduced acuity of elbow proprioception than that observed
in pain-free controls [31]. In support, individuals with
rotator cuff tendinopathy consistently over estimated force
relative to pain-free controls (who consistently underestimated force) when asked to contract to a set force
without visual feedback [32]. It is possible that a greater
EMG activity and earlier activation of forearm muscles is
due to an overestimation or altered afferent input distorting the required motor control. Second, it is possible

123

that a greater EMG activation during tennis strokes in

individuals with LE might be explained by differences in
skill level, as previous work has suggested unskilled
tennis players often grip the tennis racquet too forcefully
[33]. As participants in the LE group in the study by
Kelley et al. [19] were level B and C recreational club
tennis players, and pain-free controls were either current
or former professional or division one collegiate tennis
players [19], the lower skill of the LE group may explain
their greater forearm muscle EMG amplitude. Third,
EMG analysis as a proportion of MVC requires consideration as individuals with chronic pain conditions often
do not perform a true maximum contraction for reasons
including pain and fear of pain [25]. Use of an effort
lower than maximum for EMG normalisation would lead
to overestimation of EMG amplitude in the experimental
task, and this may explain the augmented activation of
multiple wrist extensor muscles.
Evidence of a greater [19] and earlier [15] activity of the
wrist extensor muscles (e.g. ECRB) in the LE group might
contribute to the development of LE, as this condition is
thought to be caused by an overuse of the forearm extensor
muscles, particularly the ECRB [29, 3437]. Additionally,
a tighter tennis racquet grip can induce greater vibrations if
the ball is not hit in the middle of the racquet [33]. Video
recordings from Kelley et al. [19] showed individuals with
LE consistently missed the centre of the racquet (5 of 8
[63 %] participants). Whether the combination of mis-hitting and excessive grip force [33] contribute to the development of LE requires further investigation.
With respect to differences in forearm muscle activity
during tennis strokes as a consequence of LE, modified
coordination of forearm muscles might be mediated by a host
of potential mechanisms. Greater activity of the wrist extensor
muscles in LE during tennis strokes might represent a
guarding response with the aim to limit pain provocation.
This is similar to the protective activation observed for some
muscles in other musculoskeletal conditions [38, 39].
Although such muscle guarding might be a positive adaption
in the presence of acute pain to reduce pain and injury, it could
lead to persistent changes in biomechanics [40] and further
pain/injury if maintained. Additionally, it is possible that
differences in the amplitude and timing of muscle activity
might be related to changes within the muscle. For example,
recent studies have identified a strong relationship between
inflammatory response and muscle changes following back
injuries in animals [41, 42]. Evans et al. [43] identified
inflammatory markers (e.g. interleukin-1 [IL-1]) in untrained
individuals after exercise. In addition, Ljung et al. [26] identified a greater presence of type 2A muscle fibres (fast twitch)
in individuals with LE than in pain-free controls. These
changes might alter the recording of action potentials but this
requires further investigation.

Forearm Muscle Activity in Lateral Epicondylalgia

4.1.3 Transcranial Magnetic Stimulation

Although this review identified two studies using TMS to
investigate corticomotor measures, only one revealed evidence of cortical differences between individuals with LE
and pain-free controls using our a priori defined analysis
[16, 18]. Schabrun et al. [18] highlighted an absence of the
discrete cortical organisation of forearm muscles (ECRB
and EDC) in individuals with LE that was identified in
pain-free controls. This was reflected in fewer peaks of
excitability in the cortical map, and less distance between
the centres of gravity of ECRB and EDC in the LE group
than in the pain-free control group [18]. This finding is
similar to the observation that individuals with chronic low
back pain have less distance between the motor cortex
representation of different components of the lumbar
extensor muscles than pain-free controls [44]. Recent work
has shown changes in cortical areas, including the primary
motor cortex [45], that relate to modified motor behaviour
of individuals with chronic musculoskeletal pain. Whether
these changes are present prior to development of pain
(possibly even as a contributor), or whether these changes
occur after the development of pain, remains unknown and
longitudinal studies are required.
4.2 Interpretation of Features of Forearm Muscle
Activity That Do Not Differ in Lateral
Epicondylalgia
Identification of features that do not differ significantly
between groups is just as important as identifying those
that do. Such features give clinicians and researchers
insight into the similarities between individuals with LE
and pain-free controls. No studies identified statistical
differences between individuals with and without LE in
parameters of EMG median frequency (surrogate for
muscle fatigue) during gripping [6], EDC EMG amplitude
during tennis strokes [19], and features of corticomotor
excitability as tested by TMS [16, 18].
Absence of significant differences in median frequency during gripping between those with and without
LE is interpreted to reflect a lack of between-group
differences in muscle fatigability. EMG median frequency decreases during fatiguing tasks [46], and was
similar for LE and control groups in the study by Alizadehkhaiyat et al. [6].
EDC has been suggested to be the source of symptoms
for individuals with LE [47]. Although the EDC muscle
primarily acts as a finger extensor [48], previous work has
shown that it is activated during gripping [24]. According
to the findings of Kelley et al. [19], EDC was activated
during tennis strokes in both groups (LE and controls) and
this activity did not differ between groups. This does not

preclude differences in EDC for other measures, as has

been suggested by Schabrun et al. [18].
Analysis using our a priori criterion for SMD and
95 % CI did not identify any significant differences for
TMS measures of corticomotor excitability [16, 18].
This contrasts with the analysis outcome of Schabrun
et al. [18], who identified a significantly greater MEP
amplitude for ECRB and EDC in individuals with LE
than in pain-free controls, using an analysis of variance.
Our more lenient analysis method revealed this difference to be nearly significant. Schabrun et al. [18]
proposed that reduced cortical inhibition may account
for the greater MEP amplitude in individuals with LE.
This was based on evidence that intracortical inhibition
is reduced in chronic pain conditions [49]. The difference in outcome of analysis methods may be resolved by
inclusion of a larger number of participants in future
studies.
4.3 Clinical Implications
This systematic review highlights differences in forearm
muscle activity between individuals with LE and pain-free
controls that present as potentially modifiable factors that
may contribute to the efficacy of rehabilitation. Increasingly, the treatment of tendinopathies like LE is targeted
towards motor function as well as pain relief with an
acknowledgement that there might be subgroups who will
require different combinations of approaches targeted at
pain or function [50]. Advice and education (e.g. activity
modification) and exercises targeted towards correcting
altered forearm muscle activity are possible candidate
targets to aid resolution of pain and functional issues, and
potentially reduce recurrence rates. Possible targets for
treatment could include education for tennis players
regarding gripping and ball hitting (e.g. ball in the centre of
the racquet strings), rehabilitation of kinematics and muscle activation during wrist extension, and focus on independent activation of ECRB and EDC. Such modifications
to treatment require investigation in clinical trials before
integration into practice.
4.4 Quality Assessment and Considerations
Interpretation of the findings of this review requires consideration of several methodological issues. Meta-analysis
was not possible due to heterogeneity between tasks and
outcomes used in the included studies. The methodological
quality of the included studies was limited when assessed
using the EAI tool, with scores between 0.24 and 0.76
(mean 0.39) out of a possible 1. A major limitation was the
inability to determine the reliability and validity of the
outcome measures. Only one out of the seven (14 %)

123

L. J. Heales et al.

studies highlighted the lack of evidence of validity and

reliability of the neuromuscular control technique as a
limitation of the study [18]. Only one study (14 %) mentioned other limitations which may affect their generalisability [20].

5 Conclusion
This review highlights the limited and heterogeneous nature of research regarding forearm muscle activity in LE.
This review identified evidence of differences in forearm
muscle activity (recorded with EMG) between individuals
with LE and pain-free controls, but differences in the
methods make it impossible to draw robust conclusions.
Further studies are required with larger sample sizes and
consistent methodologies to allow for meta-analysis. Evidence of altered temporal parameters in the asymptomatic
limb of individuals with unilateral LE may indicate more
widespread motor changes, but further investigation is
required.
Compliance with Ethical Standards
Funding Funding for this work was provided by a Program Grant
from the National Health and Medical Research Council (NHMRC)
of Australia (ID631717). Paul Hodges is supported by a Senior
Principal Research Fellowship (APP1002190) and Luke Heales by an
Australian Postgraduate Award scholarship.
Conflicts of interest Luke Heales, Michael Bergin, Bill Vicenzino
and Paul Hodges declare that they have no conflicts of interest relevant to the content of this review.

References
1. Lim ECW. Pain free grip strength test. J Physiother. 2013;59
(1):59.
2. Coombes B, Bisset L, Vicenzino B. A new integrative model of
lateral epicondylalgia. Br J Sports Med. 2009;43:2528. doi:10.
113/bjsm.2008.052738.
3. Priest JD, Jones HH, Nagel DA. Elbow injuries in highly skilled
tennis players. J Sports Med. 1974;2(3):139.
4. Silverstein B, Viikari-Juntura E, Kalat J. Use of a prevention
index to identify industries at high risk for work-related musculoskeletal disorders of the neck, back, and upper extremity in
Washington State, 19901998. Am J Ind Med. 2002;41(3):149
69.
5. Bisset LM, Russell T, Bradley S, et al. Bilateral sensorimotor
abnormalities in unilateral lateral epicondylalgia. Arch Phys Med
Rehabil. 2006;87(4):4905. doi:10.1016/j.apmr.2005.11.029.
6. Alizadehkhaiyat O, Fisher A, Kemp G, et al. Upper limb muscle
imbalance in tennis elbow: a functional and electromyographic
assessment. J Orthop Res. 2007;25(12):16517. doi:10.1002/jor.
20458.
7. Skinner DK, Curwin SL. Assessment of fine motor control in
patients with occupation-related lateral epicondylitis. Man Ther.
2007;12(3):24955.

123

8. Pienimaki T, Kauranen K, Vanharanta H. Bilaterally decreased

motor performance of arms in patients with chronic tennis elbow.
Arch Phys Med Rehabil. 1997;78(10):10925.
9. Konrad P. The ABC of EMG: A practical introduction to kinesiological electromyography. USA: Noraxon USA Inc; 2005.
10. Rudroff T. Kinesiological fine wire EMG: A practical introduction
to fine wire EMG applications. USA: Noraxon USA Inc; 2008.
11. Genaidy AM, LeMasters GK, Lockey J, et al. An epidemiological
appraisal instrumenta tool for evaluation of epidemiological
studies.
Ergonomics.
2007;50(6):92060.
doi:10.1080/
00140130701237667.
12. Landis JR, Koch G. The measurement of observer agreement for
categorical data. Biometrics. 1977;33(1):15974. doi:10.2307/
2529310.
13. Cohen J. Statistical power analysis for the behavioral sciences.
2nd ed. Hillsdale: Erlbaum; 1988.
14. Lim ECW, Sterling M, Pedler A, et al. Evidence of spinal cord
hyperexcitability as measured with nociceptive flexion reflex
(NFR) threshold in chronic lateral epicondylalgia with or without
a positive neurodynamic test. J Pain. 2012;13(7):67684.
15. Bauer JA, Murray RD. Electromyographic patterns of individuals
suffering from lateral tennis elbow. J Electromyogr Kinesiol.
1999;9(4):24552.
16. Dessureault L, Tremblay F, Bilodeau M. Corticomotor
excitability and proprioceptive acuity in chronic lateral epicondylalgia [Thesis]. Ottawa: University of Ottawa; 2008.
17. Calder KM, Stashuk DW, McLean L. Motor unit potential morphology differences in individuals with non-specific arm pain and lateral
epicondylitis. J Neuroeng Rehabil. 2008;5(December):Article 34.
18. Schabrun SM, Hodges PW, Vicenzino B, et al. Novel adaptations
in motor cortical maps: The relationship to persistent elbow pain.
Med Sci Sports Exerc. 2014;47(4):68190.
19. Kelley JD, Lombardo SJ, Pink M, et al. Electromyographic and
cinematographic analysis of elbow function in tennis players with
lateral epicondylitis. Am J Sports Med. 1994;22(3):35963.
20. Chourasia AO, Buhr KA, Rabago DP, et al. Effect of lateral
epicondylosis on grip force development. J Hand Surg Am.
2012;25(1):2737. doi:10.1016/j.jht.2011.09.003.
21. Roos MR, Rice CL, Vandervoort AA. Age-related changes in
motor unit function. Muscle Nerve. 1997;20(6):67990.
22. Linnamo V, Moritani T, Nicol C, et al. Motor unit activation patterns during isometric, concentric and eccentric actions at different
force levels. J Electromyogr Kinesiol. 2003;13(1):93101.
23. Georgoulis AD, Ristanis S, Papadonikolakis A, et al. Electromechanical delay of the knee extensor muscles is not altered
after harvesting the patellar tendon as a graft for ACL reconstruction: Implications for sports performance. Knee Surg Sports
Traumatol Arthrosc. 2005;13:43743.
24. Snijders CJ, Volkers AC, Mechelse K, et al. Provocation of
epicondylalgia lateralis (tennis elbow) by power grip or pinching.
Med Sci Sports Exerc. 1987;19(5):51823.
25. Lindstroem R, Graven-Nielsen T, Falla D. Current pain and fear
of pain contribute to reduced maximum voluntary contraction of
neck muscles in patients with chronic neck pain. Arch Phys Med
Rehabil. 2012;93(11):20428.
26. Ljung BO, Lieber RL, Friden J. Wrist extensor muscle pathology
in lateral epicondylitis. J Hand Surg Am. 1999;24B(2):17783.
27. Levin D, Nazarian LN, Miller TT, et al. Lateral epicondylitis of
the elbow: US findings. Radiology. 2005;237(1):2304.
28. Heales LJ, Broadhurst N, Mellor R, et al. Diagnostic ultrasound
imaging for lateral epicondylalgia: a case control study. Med Sci
Sports
Exerc.
2014;46(11):20706.
doi:10.1249/MSS.
0000000000000345.
29. Nirschl RP. Elbow tendinosis/tennis elbow. Clin Sports Med.
1992;11(4):85170.

Forearm Muscle Activity in Lateral Epicondylalgia

30. Berglund KM, Persson BH, Denison E. Prevalence of pain and
dysfunction in the cervical and thoracic spine in persons with and
without lateral elbow pain. Man Ther. 2008;13(4):2959. doi:10.
1016/j.math.2007.01.015.
31. Juul-Kristensen B, Lund H, Hansen K, et al. Poorer elbow proprioception in patients with lateral epicondylitis than in healthy
controls: a cross-sectional study. J Shoulder Elbow Surg. 2008;17
(1 SUPPL.):S7281.
32. Maenhout AG, Palmans T, De Muynck M, et al. The impact of
rotator cuff tendinopathy on proprioception, measuring force
sensation. J Shoulder Elbow Surg. 2012;21(8):10806.
33. Hatze H. Forces and duration of impact, and grip tightness during
the tennis stroke. Med Sci Sports Exerc. 1976;8(2):8895.
34. Nirschl RP, Ashman ES. Elbow tendinopathy: tennis elbow. Clin
Sports Med. 2003;22(4):81336. doi:10.1016/S0278-5919(03)
00051-6.
35. Nirschl RP, Ashman ES. Tennis elbow tendinosis (epicondylitis).
Instr Course Lect. 2004;53:58798.
36. Budoff JE, Nirschl RP. Resection and repair of lateral tennis
elbow. Oper Tech Sports Med. 2001;9(4):2116.
37. Marx RG, Sperling JW, Cordasco FA. Overuse injuries of the
upper extremity in tennis players. Clin Sports Med. 2001;20
(3):43951.
38. Arendt-Nielsen L, Graven-Nielsen T, Svarrer H, et al. The
influence of low back pain on muscle activity and coordination
during gait: a clinical and experimental study. Pain. 1996;64
(2):23140. doi:10.1016/0304-3959(95)00115-8.
39. van der Hulst M, Vollenbroek-Hutten MM, Rietman JS, et al.
Lumbar and abdominal muscle activity during walking in subjects with chronic low back pain: Support of the guarding
hypothesis? J Electromyogr Kinesiol. 2010;20(1):318. doi:10.
1016/j.jelekin.2009.03.009.
40. Verbunt JA, Seelen HA, Vlaeyen JW, et al. Disuse and deconditioning in chronic low back pain: concepts and hypotheses on
contributing mechanisms. Eur J Pain. 2003;7(1):921. doi:10.
1016/S1090-3801(02)00071-X.

41. Hodges PW, James G, Blomster L, et al. Can proinflammatory

cytokine gene expression explain multifidus muscle fiber changes
after an intervertebral disc lesion? Spine. 2014;39(13):10107.
doi:10.1097/BRS.0000000000000318.
42. Hodges PW, James G, Blomster L, et al. Multifidus muscle
changes after back injury are characterized by structural remodeling of muscle, adipose and connective tissue, but not muscle
atrophy: molecular and morphological evidence. Spine. 2015;40
(14):105771. doi:10.1097/BRS.0000000000000972.
43. Evans WJ, Meredith CN, Cannon JG, et al. Metabolic changes
following eccentric exercise in trained and untrained men. J Appl
Physiol. 1986;61(5):18648.
44. Tsao H, Danneels LA, Hodges PW. ISSLS prize winner:
smudging the motor brain in young adults with recurrent low
back pain. Spine. 2011;36(21):17217.
45. Pelletier R, Higgins J, Bourbonnais D. Is neuroplasticity in the
central nervous system the missing link to our understanding of
chronic musculoskeletal disorders? BMC Musculoskelet Disord.
2015;16:25. doi:10.1186/s12891-015-0480-y.
46. Lowery MM, OMalley MJ. Analysis and simulation of changes
in EMG amplitude during high-level fatiguing contractions. IEEE
Trans Biomed Eng. 2003;50(9):105262.
47. Fairbank SM, Corlett RJ. The role of the extensor digitorum
communis muscle in lateral epicondylalitis. J Hand Surg Br.
2002;27(5):4059. doi:10.1054/jhsb.2002.0761.
48. Saladin KS. Human Anatomy. 3rd ed. New York: McGraw Hill;
2011.
49. Lefaucheur JP, Drouot X, Menard-Lefaucheur I, et al. Motor
cortex rTMS restores defective intracortical inhibition in chronic
neuropathic pain. Neurology. 2006;67(9):156874.
50. Coombes BK, Bisset L, Vicenzino B. Management of lateral
elbow tendinopathy: one size does not fit all. J Orthop Sports
Phys Ther. 2015;45(11):93849. doi:10.2519/jospt.2015.5841.

123