Does Prior Hamstring Strain Injury Affect Hamstring Muscle

Activation Patterns in Amateur Football Players? A Feasibility Study

Ahmed Ahmed

Queen Mary University of London, Mile End Road, Bethnal Green, London, E1 4NS, U.K.

Keywords: Hamstring, Activation, Injury, Football, EMG.

Abstract: Hamstring stains are very common injuries among athletes. Re-injury rate for hamstring strains is high and

factors effecting hamstring re-injury is a research topic of interest. The aim of this study was to determine if

there is a relationship between prior hamstring strain injury and medial:lateral (M:L) hamstring activation

ratio in amateur football players during normal gait. Six male amateur football players with a history of

unilateral hamstring strain injury volunteered for participation in this study. EMG data from the

semitendinosus and biceps femoris muscles was recorded across a full gait cycle. The results of this

preliminary, feasibility study show no significant difference between mean M:L hamstring activation ratios

in previously injured hamstrings (Mean (M)=2.54, Standard Deviation (SD)=1.56) and uninjured contralateral

hamstrings (M=3.06, SD=2.86) across the full gait cycle; t(10) = 0.73, p>0.05. Mean M:L activation ratios

during ‘Stance phase’ show no significant difference between case and control hamstrings; t(10)=0.88,

p>0.05. During ‘Swing phase’ there is no significant difference in mean M:L activation ratios between

previously injured and uninjured hamstrings; t(10)=0.61, p>0.05.

1 INTRODUCTION

Acute hamstring strains are the most common form

of injury in professional soccer players, making up

12% of all injuries (Ekstrand et al., 2009).

Biomechanical analysis of running suggests that the

injury occurs during terminal swing phase of the gait

cycle (Chumanov et al., 2007). The hamstrings must

change from contracting eccentrically to decelerate

knee extension, to contracting concentrically acting

as an active hip extensor (Thelen et al., 2005). This

rapid change from eccentric to concentric contraction

is when the muscle is most vulnerable to injury

(Verrall et al., 2001). The impact on the individual

and the resultant time-off is variable and often

prolonged due to issues surrounding rehabilitation

and recovery. Moreover, the recurrence rate of

injuries is very high, with re-injury occurring in up to

63% of hamstring strains (Brukner et al., 2013). Since

these types of injuries are so common and difficult to

prevent, hamstrings strains have proven to be a

prominent area of research.

Numerous studies have investigated the

relationship between the ratio of hamstring to

quadricep activation and how they relate to injury

occurrence. A systematic review by Opar et al.

investigated a potential relationship between prior

hamstring strain injuries and increased risk of future

anterior cruciate ligament tear (Opar et al., 2014).

One study included in this review looked at the

functional deficits caused by previous hamstring

strains. They found that ‘athletes with a history of

hamstring strain injury displayed lower

hamstrings:quadriceps strength ratio in the previously

injured limb when compared to the uninjured limb.

In 2015, Ardern et al. investigated hamstring

strength imbalances in professional soccer players in

Australia (Ardern et al., 2015). Hamstring strength

tests were performed on 42 players, 24% of which

were discovered to have hamstring strength

imbalances. One of the major findings was that the

strength imbalances were almost invariably found in

the stance leg of athletes. In this paper the researchers

were only assessing the prevalence of hamstring

strength imbalances, whereas further studies go on to

explore this topic in more depth, considering

aetiology and sequelae of strength imbalances.

Bourne et al. aimed to determine if eccentric knee

flexor strength imbalance is a risk factor for

hamstring strain injuries in rugby union (Bourne et

al., 2015). In this prospective cohort study 178 rugby

players had strength tests performed at the start of the

playing season with the primary outcome measure of

Ahmed, A.

Does Prior Hamstring Strain Injury Affect Hamstring Muscle Activation Patterns in Amateur Football Players? A Feasibility Study.

DOI: 10.5220/0010117702170222

In Proceedings of the 8th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2020), pages 217-222

ISBN: 978-989-758-481-7

217

prospective hamstring strain injuries. The results of

their study showed that a between-limb imbalance in

eccentric knee flexor strength of ≥20% increased the

risk of a hamstring strain 3.4-fold (95% CI, 1.5 -7.6;

p = 0.003). This provides evidence to support the

rational that strength imbalances should be corrected,

particularly in those with previous hamstring strains

to prevent future injury.

A study by Sole et al. investigated EMG activity

of thigh muscles in people with a recent hamstring

injury during a weight bearing exercise (Sole et al.,

2011). They found that EMG onsets of the hamstrings

were significantly earlier in those with a previous

hamstring injury when compared with a control

group. These results indicate that hamstring strains

can effect muscle activation patterns post injury.

The insight gained from a review of the literature

provided the bases for developing a research question

and in choosing methodology. It has already been

shown that prior hamstring strain injury can result in

strength imbalances. This is true when comparing

ipsilateral, injured and contralateral, uninjured sides,

as well as when comparing quadriceps to hamstring

strength ratios on ipsilateral injured side (Opar et al.,

2014; Ardern et al., 2015). Moreover, studies have

shown that an imbalance of strength in antagonistic

muscles of the thigh increased the risk of hamstring

strain injury (Bourne et al., 2015). However, to our

knowledge changes in activation ratios of the

synergistic hamstring muscle group, post-injury has

not been investigated.

Our study will focus on amateur football players

with previous hamstring strain injuries. The aim of

our study is to assess the medial:lateral hamstring

activation ratio in amateur football players during on-

ground walking.

The primary objective is to identify if there is an

association between previous hamstring strain injury

and altered hamstring activation patterns when

comparing medial and lateral hamstrings. We

hypothesise that a prior hamstring strain injury results

in altered medial:lateral activation ratio when

compared to the contralateral uninjured side. It is

hoped that a better understanding of how a hamstring

strain injury affects muscle activation and the

different hamstring groups may lead to lower injuries

and recurrence rates.

2 METHODS

2.1 Sample Size

This study was designed as a feasibility study, with

the aim of determining whether further testing may

be applicable, if a relationship between prior

hamstring strain injury and medial:lateral (M:L)

hamstring activation ratio is observed. As such a

relatively small sample size was used; including only

6 study participants.

2.2 Inclusion & Exclusion Criteria

Criteria were chosen with the aim of limiting

confounding factors and generating a homogenous

sample group. Six, male, amateur level football

players, aged 22 with a history of a unilateral

hamstring strain injury sustained within the previous

24 months were included. Participants were excluded

if they had: a current symptomatic hamstring strain

injury, a history of bilateral hamstring strains or a

history of cardiovascular disease.

2.3 Participant Characteristics

Six amateur football players (from St Bartholomew

and The Royal London 1st XI football team)

volunteered for this study. Players trained once per

week, playing in competitive matches in an amateur

level league twice per week. All participants were 22

year old males with a mean height of 178.7cm ±

3.2cm and an average body mass of 72.35kg ±9.8kg.

All recruited individuals had a history of a unilateral

hamstring strain injury within the previous 24

months. Hamstring strains were self-reported and

defined as an acute onset posterior hamstring pain

that occurred during sprinting and resulted in time off

sport. All participants provided written consent prior

to participating in the study. Ethical approval was

given by the Queen Mary Committee of Research

Ethics. were chosen with the aim of limiting

confounding factors and generating a homogenous

sample group. Six, male, amateur level football

players, aged 22 with a history of a unilateral

hamstring strain injury sustained within the previous

24 months were included. Participants were excluded

if they had: a current symptomatic hamstring strain

injury, a history of bilateral hamstring strains or a

history of cardiovascular disease.

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

218

Table 1: Baseline Characteristics of Study Participants.

Characteristic Mean

(±SD)

Minimum

Maximum

Age 22 22

Body Mass (kg)

72.35

(±9.8)

60.7

87.6

Height (cm)

178.73

(±3.17)

173.5

183.0

BMI

22.62

(±2.81)

18.7

26.9

2.4 Instrumentation

Surface Electromyography (EMG) data were

collected with DELSYS Trigno Lab Wireless Surface

16 EMG system (16 channel) (DELSYS INC,

Massachusetts, USA). The sampling rate was

2000Hz. Wireless EMG electrodes were applied over

the semitendinosus muscle at the mid-point of the

ischial tuberosity and the medial epicondyle of the

tibia. Electrodes for the biceps femoris muscle were

applied at the mid-point of ischial tuberosity and

lateral epicondyle of the tibia. Electrodes for the

rectus femoris muscle were applied at the mid-point

of the anterior superior iliac spine and the superior

patella. Electrodes were attached using self-adhesive

tape and placement was based on the

recommendations of SENIAM (“The Seniam

Project”, 2020). Raw EMG data was then transferred

to MATLAB (v2009a, Mathworks, Natwick, MA,

USA) where it was filtered rectified and smoothed.

The data from motion analysis was used to

determine ‘Heel Strike ’and ‘Toe Off ’stages of the

gait cycle enabling distinction of the stance and swing

phases of the cycle. This was achieved by mapping

the vertical height of the heel markers and the 5th

metatarsal markers. Kinematics were measured using

a CodaMotion system (Charnwood Dynamics Ltd.,

Leicestershire, United Kingdom). Four CX1 Series

Codamotion 3d Scanners were used to capture the

markers placed on the study subject. Clusters

consisting of four active markers were used on the

lower limbs to minimise marker movement. The

sampling rate was 200Hz. The software used for

measurement and analysis was ODIN Codamotion

3D Modelling & Control Software (v1.05).

2.5 Protocol

Upon arriving at the Human Performance Laboratory,

participants were briefed on proceedings after which

informed consent was given. For each participant

weight, height and age were recorded prior to data

collection. After these baseline measurements, EMG

electrodes were placed onto the relevant muscles. The

medial hamstring EMG data was recorded from the

semitendinosus muscle. The lateral hamstring EMG

data was recorded from the biceps femoris muscle.

The recordings from the previously injured leg will

be compared to the uninjured leg of the individual as

a control. Motion analysis markers were placed

around the subjects pelvis and feet with marker

clusters placed on the lateral aspect of the thighs and

shanks. Once all EMGs and markers were placed,

anatomical landmarks were digitised using a CLSTR-

PTR-02 pointer. The anatomical landmarks digitised

included: lateral femoral epicondyle, lateral

malleolus, medial femoral epicondyle and medial

malleolus (bilaterally).

Figures 1 and 2 below are images of a study

participant post instrumentation setup. They illustrate

the cluster markers on the lateral aspect of the thighs

and shanks bilaterally, and EMG surface electrodes

on the posterior aspect of the thigh.

Figure 1: Study Participant, Post-Instrumentation (Anterior

View).

Figure 2: Study Participant, Post-Instrumentation (Posterior

View).

Does Prior Hamstring Strain Injury Affect Hamstring Muscle Activation Patterns in Amateur Football Players? A Feasibility Study

219

With instrumentation setup complete, participants

were then allowed to familiarise themselves with the

experimental protocol. Participants walked for a

distance of 3 meters without footwear at a self

selected speed to ensure a full gait cycle was

completed. Participants were asked to place their

hands on the alternate shoulder as they walked to

avoid obstruction of marker visibility. This was

repeated 3 times before data collection trials began.

For data collection trials subjects were again asked to

walk for 3 metres at a comfortable speed. This was

repeated until two acceptable trials were obtained.

Criteria for adequate trials was marker visibility

>80% in all markers across a full gait cycle. EMG

data was only analysed from the first full gait cycle

completed during the 3 meter walk with start point

and end point of the gait cycle defined as right foot

‘heel strike ’and left foot ‘toe off ’respectively.

2.6 Statistical Analysis

Data was extracted using Matlab version 9.0

(Mathworks, Inc. Massachusetts, USA). Raw EMG

data was initially processed by application of a notch

filter at 50Hz to remove electrical noise. The signal

was then rectified and smoothed with a 4th order FIR

low-pass filter with a cut off frequency of 10Hz to

produce a ‘linear envelope ’with mean RMS values

calculated. This technique has been used numerous

times in previous studies (Arendt-Nielsen et al., 1991;

Neumann et al., 1996; Ng et al., 1997). EMG data

were taken from the same muscles, on the same day

with subjects acting as their own controls and

comparisons between individuals being made in-

terms of activation ratios. As such, EMG

normalisation was not thought to be necessary

(Soderberg et al., 2000).

The mean root mean square (RMS) for the

processed EMG signals were then calculated for the

stance phase, the swing phase and the complete gait

cycle. Medial:lateral hamstring activations ratio was

defined as the ratio of non-normalised, mean RMS of

EMG wave amplitude between the biceps femoris

and semitendinosus muscles, taken at various stages

of the gait cycle. The primary outcome measure was

mean medial:lateral hamstring activation ratio during

the complete gait cycle. Secondary outcome

measures were medial:lateral ratios during the

‘Stance ’and ‘Swing ’phase. Comparisons were made

between the previously injured legs and control legs

under these three parameters. Student’s independent

two-sided t-test was then carried out to compare data

from previously injured and previously uninjured

(control) legs. Statistical analysis was performed

using SPSS version 17.0 (SPSS Inc, Chicago, IL).

Statistical significance was set at p < 0.05.

3 RESULTS

The inter-subject variation in M:L activity ratios was

relatively high ranging from ratios of 0.66 during

swing phase of the gait cycle to 10.3 in the stance

phase. During the full gait cycle mean hamstring M:L

activation ratios of participants showed very little

discrepancy between previously injured and control

hamstrings, the latter having a lower ratio on average

(-0.24 ±2.79). Table 2 illustrates the mean M:L

activity ratios across the gait cycle. Figure 3

graphically represents M:L activation ratios across

the full gait cycle.

Table 2: Mean M:L Hamstring Activation Ratios (±SD)

Across One Complete Gait Cycle.

Injured Side Uninjured (Control)

Side

Stance Phase 2.52 (1.17) 22

Swing Phase 3.10 (2.23)

60.7

Full Gait Cycle 2.54 (1.56)

173.5

Students independent two-sided t test revealed that

there is no significant difference between mean M:L

hamstring activation ratios in hamstrings with prior

strain injuries and previously uninjured hamstrings

across a full gait cycle; t(10)=0.73, p>0.05. There was

no significant difference found between previously

injured and control hamstrings during the stance

phase of the gait cycle. t(10)=0.88, p>0.05.

Moreover, there was no significant difference found

between previously injured and control hamstrings

during the swing phase of the gait cycle; t(10)=0.61,

p>0.05.

Figure 3: Comparing M:L activation ratios of injured

hamstrings vs. controls across a full gait cycle.

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

220

4 DISCUSSION

4.1 Main Findings

The main finding of our study was that there is no

significant difference in hamstring medial:lateral

activation ratios in previously injured hamstrings

compared to non-injured hamstrings. These results

disprove the original research hypothesis. It is

difficult to state whether these results are in keeping

with similar studies as the number of papers

investigating individual hamstring muscle activation

patterns post-injury are very limited. However, a

study by Sole et al. compared EMG activity of gluteal

and thigh muscles of sportspeople with a recent injury

with uninjured controls during a weight-bearing task

(Sole et al, 2011). They found that the EMG onsets of

biceps femoris and medial hamstrings were

significant earlier for the injured group compared to

the control group. Although this study focussed on

EMG onset rather than mean EMG activity, their

results suggest alterations in neuromuscular control

of hamstrings post-injury. Following this, it could be

postulated that injury alters the M:L activity ratios of

hamstrings, which contradicts the findings from our

study.

Emami et al. conducted a study investigating

EMG activity pattern of the lumbo-pelvic muscles

during prone hip extension in athletes with and

without hamstring strain injury (Emami et al., 2014).

Rather than calculating activity ratios, EMG data was

processed and normalised with muscle activity being

expressed as percentage of maximal voluntary

electrical activity (MVE). Their results showed that

there were significant differences in EMG activity of

the gluteus maximus and the medial hamstring

between the injured group and the control group.

However, no significant difference was found in

activity of the lateral hamstring between the two

groups. This would suggest that a previous hamstring

injury will alter the medial:lateral hamstring

activation ratio. This again is not in keeping with the

results of our study.

4.2 Limitations

One of the major limitations to this study was the

small sample size. Adequate sample sizes ensure high

statical power in null hypothesis testing (Liu et al.,

2013). Access to the target population of male

amateur football players with a unilateral hamstring

strain injury occurring in the past 2 years was limited

within the time frame that this study was conducted.

Future work on this topic should be carried out on a

larger scale to further assess the link between

hamstring strain and muscle activation patterns.

Another limitation to this study was the self-

reported nature of the injuries. Gabbe et al.

investigated the validity of self reported sports injury

histories. (Gabbe et al., 2003). Their results showed

that at 12 months post-injury 80% of participants

were able to accurately recall the body regions

injured, but not the exact diagnosis, with only 61% of

participants able to recall the exact body region and

diagnosis of injury sustained. Participants of this

study were included if they had sustained a hamstring

strain any time in the past 24 months. As such details

on the injury including severity, and exact location

had a decreased validity due to probable recall bias.

The inclusion criteria specified only male subjects

could be recruited. This criterion was chosen because

of the different neuromuscular control strategies and

movement patterns in women including lower

hamstring muscle activation (Malinzak et al., 2001).

As such, a limitation to this study is that the findings

are not transferable and can not be applied to females.

Furthermore, EMG data collected in our study

was not normalised. Normalisation was not necessary

for this study and non-normalised signals allowed us

to answer the research question. However this is a

weakness to our study as it limits comparability of our

results to similar studies using normalised EMG

signals.

4.3 Future Work

Future studies could investigate how strain injuries of

the different hamstring components lead to different

physiological response of the hamstrings. It has

previously been suggested that following hamstring

strain the biceps femoris compensates for the lack of

endurance capacity of the semitendinosus, increasing

the re-injury risk (Schuermans et al., 2014).

Furthermore, Opar et al. recently showed that

previously strained hamstrings show less

improvement in eccentric strength following a

training programme when compared to uninjured

hamstrings (Opar et al., 2014). However, it is not

clear whether this is the cause of or the result of the

injury. This is another aspect of this topic that has

potential to be investigated in the future. Fully

understanding the biomechanics and neuromuscular

response following hamstring strain injury is

necessary before prevention and rehabilitation

techniques can be improved. The potential benefits of

research in this area to lower rehabilitation time and

re-injury rates, resulting in less time-off sport.

Does Prior Hamstring Strain Injury Affect Hamstring Muscle Activation Patterns in Amateur Football Players? A Feasibility Study

221

4.4 Conclusion

To conclude, previous hamstring strain injuries do not

result in a significantly different medial:lateral

hamstring activation ratio when compared with

uninjured hamstrings. This is true for both ‘Stance

phase ’and “Swing phase ’of gait cycle as well as

across the entire gait cycle. However, due to various

limitations of this study, further research on this topic

is needed to comprehensively establish the

association between between previous hamstring

injury and medial:lateral hamstring activation ratios.

REFERENCES

Ekstrand, J., Hagglund, M. and Walden, M., 2009. Injury

incidence and injury patterns in professional football:

the UEFA injury study. British Journal of Sports

Medicine, 45(7), pp.553-558.

Chumanov E, Heiderscheit B, Thelen D. The effect of speed

and influence of individual muscles on hamstring

mechanics during the swing phase of sprinting. Journal

of Biomechanics 2007;40(16):3555-3562.

Thelen D, Chumanov E, Best T, et al. Simulation of biceps

femoris musculotendon mechanics during the swing

phase of sprinting. Medicine & Science in Sports &

Exercise 2005;37(11):1931-1938.

Verrall GM, Slavotinek JP, Barnes PG, et al. Clinical risk

factors for hamstring

muscle strain injury: a prospective study with correlation of

injury by magnetic

resonance imaging. British Journal of Sports Medicine

2001;35:435–40.

Brukner P, Nealon A, Morgan C, et al. Recurrent hamstring

muscle injury: applying the limited evidence in the

professional football setting with a seven-point

programme. British Journal of Sports Medicine

2013;48(11):929-938.

Opar D, Serpell B. Is there a potential relationship between

prior hamstring strain injury and increased risk for

future anterior cruciate ligament injury?. Archives of

Physical Medicine and Rehabilitation 2014;95(2):401-

405.

Ardern C, Pizzari T, Wollin M, et al. Hamstrings strength

imbalance in professional football (soccer) players in

Australia. Journal of Strength and Conditioning

Research 2015;29(4):997-1002.

Bourne M, Opar D, Williams M, et al. Eccentric knee flexor

strength and risk of hamstring injuries in rugby union:

A prospective study. The American Journal of Sports

Medicine 2015;43(11):2663-2670.

Sole G, Milosavljevic S, Nicholson H, et al. Altered muscle

activation following hamstring injuries. British Journal

of Sports Medicine 2011;46(2):118-123.

The SENIAM Project [Internet]. SENIAM. 2000 [cited 17

May 2016]. Available from: http://www.seniam.org

Arendt-Nielsen L, Sinkjaer T, Nielsen J, et al.

Electromyographic patterns and knee joint kinematics

during walking at various speeds. Journal

Electromyography and Kinesiology 1991;1:89–95.

Neumann DA. Hip abductor muscle activity in persons with

a hip prosthesis while carrying loads in one hand.

Journal of The American Physical Therapy Association

1996;76:1320–1330.

Ng JK, Richardson CA, Jull GA. Electromyographic

amplitude and frequency changes in the iliocostalis

lumborum and multifidus muscles during a trunk

holding test. Journal of The American Physical

Therapy Association 1997;77:954–961.

Soderberg GL, Knutson LM. A guide for use and

interpretation of kinesiologic electromyographic data.

Journal of The American Physical Therapy Association

2000;80(5), 485-498.

Emami M, Arab AM, Ghamkhar L. The activity pattern of

the lumbo‐ pelvic muscles during prone hip extension

in athletes with and without hamstring strain injury.

International Journal of Sports Physical Therapy

2014;9(3):312-319.

Liu X. Comparing sample size requirements for

significance tests and confidence intervals. Counseling

Outcome Research and Evaluation 2013;4(1):3-12.

Gabbe B, Finch CF, Bennell KL, et al. How valid is a self

reported 12 month sports injury history?. British

Journal of Sports Medicine 2003;37(6):545-547.

Malinzak R, Colby S, Kirkendall D, et al. A comparison of

knee joint motion patterns between men and women in

selected athletic tasks. Clinical Biomechanics

2001;16(5):438-445.

Schuermans J, Van Tiggelen D, Danneels L, et al. Biceps

femoris and semitendinosus-teammates or competitors?

New insights into hamstring injury mechanisms in male

football players: a muscle functional MRI study. British

Journal of Sports Medicine 2014;48(22):1599-1606.

Opar D, Williams M, Timmins R, et al. The effect of

previous hamstring strain injuries on the change in

eccentric hamstring strength during preseason training

in elite australian footballers. The American Journal of

Sports Medicine 2014;43(2):377-384.

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

222