Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64

http://www.biomedcentral.com/1471-2474/11/64

RESEARCH ARTICLE Open Access

The effect of a sports chiropractic manual

therapy intervention on the prevention of back
pain, hamstring and lower limb injuries in semi-
elite Australian Rules footballers: a randomized
controlled trial
Wayne Hoskins*, Henry Pollard

Abstract
Background: Hamstring injuries are the most common injury in Australian Rules football. It was the aims to
investigate whether a sports chiropractic manual therapy intervention protocol provided in addition to the current
best practice management could prevent the occurrence of and weeks missed due to hamstring and other lower-
limb injuries at the semi-elite level of Australian football.
Methods: Sixty male subjects were assessed for eligibility with 59 meeting entry requirements and randomly
allocated to an intervention (n = 29) or control group (n = 30), being matched for age and hamstring injury
history. Twenty-eight intervention and 29 control group participants completed the trial. Both groups received the
current best practice medical and sports science management, which acted as the control. Additionally, the
intervention group received a sports chiropractic intervention. Treatment for the intervention group was
individually determined and could involve manipulation/mobilization and/or soft tissue therapies to the spine and
extremity. Minimum scheduling was: 1 treatment per week for 6 weeks, 1 treatment per fortnight for 3 months, 1
treatment per month for the remainder of the season (3 months). The main outcome measure was an injury
surveillance with a missed match injury definition.
Results: After 24 matches there was no statistical significant difference between the groups for the incidence of
hamstring injury (OR:0.116, 95% CI:0.013-1.019, p = 0.051) and primary non-contact knee injury (OR:0.116, 95%
CI:0.013-1.019, p = 0.051). The difference for primary lower-limb muscle strains was significant (OR:0.097, 95%
CI:0.011-0.839, p = 0.025). There was no significant difference for weeks missed due to hamstring injury (4 v14,
c2:1.12, p = 0.29) and lower-limb muscle strains (4 v 21, c2:2.66, p = 0.10). A significant difference in weeks missed
due to non-contact knee injury was noted (1 v 24, c2:6.70, p = 0.01).
Conclusions: This study demonstrated a trend towards lower limb injury prevention with a significant reduction in
primary lower limb muscle strains and weeks missed due to non-contact knee injuries through the addition of a
sports chiropractic intervention to the current best practice management.
Trial registration: The study was registered with the Australian and New Zealand Clinical Trials Registry
(ACTRN12608000533392).

* Correspondence: waynehoskins@iinet.net.au
Macquarie Injury Management Group, Department of Chiropractic, Faculty of
Science, Macquarie University, NSW 2109, Australia

© 2010 Hoskins and Pollard; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 2 of 11
http://www.biomedcentral.com/1471-2474/11/64

Background is lacking [7]. However, a growing body of literature, lar-

Australian Rules football is a unique body contact sport. gely of an indirect nature, suggests that several non-local
It is played on a natural grass, oval shaped field, with hamstring factors may have an association with injury
the size varying between 135 - 185 meters in length and [12-16], whilst a Cochrane systematic review of the lit-
110 - 155 meters in width. Teams consist of 18 players erature has stated that consideration should be given to
per side plus four on an unlimited interchange bench. the lumbar spine, sacroiliac and pelvic alignment and
Each game is played over four 20 minute quarters plus postural control mechanisms when managing hamstring
stoppage time. Physical requirements of players include: injuries [17]. Despite the knowledge that non-local fac-
repeated rapid acceleration and deceleration efforts tors may exist, the literature appears almost devoid of
often involving change of direction, agility, jumping, research investigating their possible identification and
bending to pick up the oval shaped ball, tackling and documenting the effects, if any, of addressing non-local
other collisions [1]. There is a continuous nature of play factors in hamstring injury management [12,16,18].
requiring high aerobic capacity, although the speed of A recent review of the literature stated that newer
the game has increased and now involves a greater approaches that incorporate manipulation in multi-
number of shorter high intensity play periods and longer modal management approaches for hamstring injury
stop periods [2]. The most important means of ball pro- prevention should be further investigated [9]. Thus it
gression is by punt kicking. Australian Rules football has was the objective of this RCT to investigate whether a
the highest rates of non-contact soft tissue injuries sports chiropractic intervention consisting of pragmati-
when compared with other body contact football codes cally and individually determined high-velocity low-
such as rugby league and rugby union [3], with the inci- amplitude (HVLA) manipulation, mobilization and/or
dence of lower limb muscle strains at the elite national supporting soft tissue therapies to the spine, pelvis and
competition, the Australian Football League (AFL), extremity could reduce local and non-local hamstring
being 35% per season [4]. injury risk factors to prevent the occurrence of ham-
Hamstring injuries are the most prevalent injury in string and other non-contact lower limb injuries and
Australian Rules football at the AFL [4] and feature pro- decrease low back pain (LBP) and alter health outcomes
minently at other levels of play [5]. Per season in the in semi-elite Australian Rules footballers.
AFL hamstring injuries afflict 16% of players, cause 3.4
missed matches per injury, account for the most time Methods
missed due to injury and have the highest rates of injury Protocol
recurrence, with one in three injuries recurring on Four of the thirteen clubs competing in the semi-elite
return to play [4]. On return to play, player performance state based Victorian Football League (VFL) were
is significantly lower [6]. Hamstring injuries are also the approached and agreed to provide players for this study
most common muscle injury in running based sports during the 2005 season. However, a change in club staff
[7]. Knowledge surrounding optimal preventative mea- resulted in two clubs withdrawing support prior to sub-
sures is therefore critical. ject recruitment. VFL players train and play in the same
The prevention of hamstring injuries has long been competition as elite AFL players not selected for first
recognized as a priority effort. By contrast, Bahr and grade competition and receive financial remuneration
Holme [8] have opined that well designed prospective without being full time in their playing and training
hamstring injury prevention studies are lacking. Recent commitments. Players were eligible to participate if they
literature reviews have been universal in their depiction were listed players on their respective VFL squad and
of the lack of evidence for the prevention of hamstring excluded on the basis of: “red flag” conditions including:
injuries and the requisite for evidence based approaches fractures, infections, inflammatory diseases, tumours and
to be determined through randomized controlled trials other causes of destructive lesions of the spine; “yellow
(RCTs) [7,9,10]. Prevention of injury becomes more cru- flag” conditions including: insurance claims, litigation;
cial as the most established predictors for hamstring history of malignant disease; clinical signs suggesting
injury in Australian Rules football are immutable in nat- inguinal or femoral hernia; vascular disease; history of
ure, namely a current or recent history of a hamstring motor vehicle accident, or other serious fall or accident
injury and age [11]. in the last three months; neurological signs and symp-
Conventional injury prevention has focused on local toms (muscle wasting, nerve root signs, bowel, bladder
hamstring factors. Orchard [11] has said that sports or sexual dysfunction); organic kidney, urinary tract or
medicine dogma advises that poor flexibility, fatigue, reproductive disease; previous recent spinal surgery (less
lack of warm up and weakness are risk factors for injury. than 2 years); club doctor or medical staff excludes the
The evidence to support this tenet for hamstring injury players participation; severe history of chronic hamstring
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 3 of 11
http://www.biomedcentral.com/1471-2474/11/64

problems; serious injury or surgery preventing play for non-cavitational and high velocity in intent. Mobiliza-
the remainder of the season. Before the start of the tion occurs when a joint is passively moved within its
study the subjects, coaches and medical personal were normal range of motion (usually a slow oscillatory
informed about the purpose and design of the study. movement). Treatment scheduling was also pragmati-
Club staff gave permission to participate in the study. cally determined. The minimum scheduling adhered to
was: 1 treatment per week for 6 weeks (phase 1)
Assignment followed by 1 treatment per fortnight for 3 months and
Players completed a self-reported questionnaire at their 1 treatment per month for the remainder of the season
training location prior to randomization. The question- (3 months) (phase 2).
naire consisted of the validated and reliable McGill Pain
Questionnaire (short form) (MPQ-SF) for LBP, the 39 Outcome Measures
item Health Status Questionnaire (SF-39) as well as self The study was divided into two phases. Phase 1
reported questions on knee and hamstring injury history (6 weeks) involved the late pre-season period where pre-
(incidence during the previous month, 6 months, year, 2 season matches and the intervention commenced but no
years, greater than 2 years or not at all). At each of the injury surveillance was conducted. Phase 2 (24 weeks)
two clubs after completion of the baseline questionnaire, occurred where regular season (home and away) and
players were randomly allocated into one of two groups finals matches were conducted weekly and an injury sur-
such that allocation was concealed. Eligible players were veillance was conducted. The injury surveillance com-
stratified by age and hamstring injury history and allo- menced after a period of more intense treatment
cated using a computer generated randomization list for scheduling such that the treatment effects, if any, would
each club within these strata, as these are the most be observed in a changed injury pattern. At the mid
recognized predictors for injury [11]. Randomization point of the season (12 home and away season matches,
was completed within each club to prevent an element 18 weeks of intervention) players completed the MPQ-
of randomness in a clubs injury profile each season SF and the SF-39 as secondary outcome measures at
impacting on the results of the study. After all subjects their training location.
had been allocated the two groups at each club were The injury definition and injury surveillance con-
then randomly allocated to either the intervention or ducted was a reproduction of the AFL’s injury surveil-
control with a coin toss. lance and used as a primary outcome measure for the
prevention of hamstring injuries, lower limb muscle
Intervention strains and non-contact knee injuries [4]. The definition
All of the players in both the intervention and control of an injury was: “any physical or medical condition that
group continued to receive what is considered the cur- prevents a player from participating in a regular season
rent best practice medical, paramedical and sports (home and away) or finals match”. The missed match
science management including medication, manipulative injury definition is currently considered the most reli-
physiotherapy, massage, strength and conditioning and able injury surveillance method in team sports [20]. The
rehabilitation as directed by club staff, which acted as number of games missed due to injury was also deter-
the control. All treatment from club staff was indepen- mined. Injury diagnoses were determined by club medi-
dently administered without restriction or interference cal staff who were blinded to group allocation using
from the study authors. All staff were employed by the either clinical features of injury, advanced imaging or
club and had no limitation in the number or type of both at their discretion with blinded club recorders
treatment they could render. In addition to this, the completing the injury surveillance. Clinical parameters
intervention group received a sports chiropractic of injury were also recorded including mechanism of
approach administered by a single practitioner. Treat- injury (contact or non-contact). In this way separation
ment was pragmatically and individually determined by of injuries could be made retrospectively and allocated
the therapist and could involve HVLA manipulation into groups for statistical analysis. To attain this, the
(either manual or mechanically assisted techniques), player was interviewed at the first available opportunity
mobilization and/or supporting soft tissue therapies: var- following injury. The club medical and coaching staff
ious stretching and soft tissue massage techniques to the independently determined selection in matches. There
spine, pelvis and extremity. According to Mierau et al. was no interference from the study authors.
[19] manual manipulation involves a brief, shallow, In addition a secondary injury surveillance for adverse
sudden carefully administered thrust (high velocity in outcomes resulting from the intervention was estab-
nature). Mechanically assisted manipulation is lished for the duration of the study with an injury defi-
performed through the assistance of devices (for exam- nition of: “any undue pain, discomfort or disability
ple drop pieces) or impulse type instruments, being arising during, immediately after or subsequent to
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 4 of 11
http://www.biomedcentral.com/1471-2474/11/64

chiropractic therapy that resulted in missed participation trol group (n = 30) with no baseline differences for age
in a match or training session, required additional medi- (mean/SD/range intervention 20.2/1.8/18-27, control
cal consultation or treatment or was acknowledged by a 20.2/1.8/18-25), self-reported hamstring and knee injury
player as not reasonably being associated or expected history, MPQ-SF and SF-39 (all p > 0.05).
with the normal course of treatment”. If an injury
occurred, further details on the type of injury, timing of Injury Surveillance
symptom onset, duration of symptoms and severity were Table 1 presents the results for the difference in injury
to be determined. incidence between the groups at the completion of the
season (24 matches, 30 weeks of intervention). There
Statistical Analysis was no statistical difference in the prevention of
All data collected were manually entered using Micro- hamstring injuries (p = 0.051) or weeks missed due to
soft Excel and analyzed using SPSS for Windows (ver- hamstring injury (c2 1.12, p = 0.29). For primary ham-
sion 12.0) or for weeks missed due to injury, SAS string injuries, the incidence was 3.6% for the interven-
version 9.1.3 and PROC GENMOD. Pearson’s “exact” tion group and 17.2% for the control group, with the
Chi-squared test based on Monte Carlo simulation was recurrence rate being 40.0%. The intervention group
used to assess the efficacy of the intervention with missed 4 matches due to hamstring injury and the con-
respect to the number of injuries. Odds ratios and 95% trol 14 matches. The intervention group was at a statis-
confidence intervals were also included. As such this tically significant reduced risk of suffering a primary
calculation is just an approximation and is included as it lower limb muscle strain injury (p = 0.025), equating to
is believed that confidence intervals should always be 3.6% of the intervention group and 27.6% of the control
stated [21]. Negative binomial models were used to cal- group. The intervention group missed 4 matches with a
culate significance for weeks missed due to injury. Two lower limb muscle strain and the control group 21
independent sample t-tests were used to compare group matches (c2 2.66, p = 0.10). The difference in primary
age, hamstring and knee injury history, MPQ-SF and non-contact knee injury incidence was not statistically
SF-39 at baseline, or if distributions were mixed, Fisher’s significant (p = 0.051), with the incidence being 3.6% for
exact test was used. Repeated measures and regression the intervention group and 24.1% for the control group.
models were used to determine change for the MPQ-SF The intervention group missed 1 match with a primary
and SF-39. If data were extremely skewed in distribu- non-contact knee injury and the control group 24
tion, transformation of scores was required. Between matches, the difference being statistically significant
group differences were obtained from two independent (c2 6.70, p = 0.01). No players reported an adverse reac-
sample t-tests. For global statistical tests, a p value < tion to the intervention.
0.05 was considered significant.
Low Back Pain
Statistical power calculation Table 2 presents the results for the change in baseline
Based on historical AFL data [4] the assumed hamstring MPQ-SF at the mid point of the season. A positive and
incidence level for the null hypothesis is 15%. For a 5% statistical significant change for the intervention group
significance level and 80% power, a total sample size of was achieved for overall (p = 0.006) and current LBP (p
117 is required to detect a 50% reduction in the inci- = 0.026). No significant change was noted for the other
dence of hamstring injuries. components of the MPQ-SF (p > 0.05).

Ethical considerations Health Status

All players gave their written informed consent to parti- Table 3 presents the results for the change in baseline
cipate and ethical approval was obtained from the Mac- SF-39 at the mid point of the season. A positive statisti-
quarie University Human Ethics Committee (Ethics cal change for the intervention group was achieved for
Approval Number: HE27AUG2004-RO3066). role limitations due to physical health (p = 0.004), bodily
pain (p = 0.034), general health (p = 0.027), and physical
Results summary score (p = 0.013). No other statistically signifi-
Participants cant change was noted for other health status compo-
Sixty male Australian Rules football players were nents (p > 0.05).
recruited as subjects. Figure 1 shows a flow chart
describing progress of subjects through the trial for the Intervention
primary and secondary outcome measures. Players were Table 4 provides a description of the treatment rendered
randomly allocated to the intervention (n = 29) or con- to the intervention group for the course of the study.
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 5 of 11
http://www.biomedcentral.com/1471-2474/11/64

Figure 1 CONSORT flow chart indicating progress of subjects through the trial.
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 6 of 11
http://www.biomedcentral.com/1471-2474/11/64

Table 1 Difference between the intervention and control group for injury incidence at the completion of the season
(24 matches, 30 weeks of intervention)
Injury Intervention incidence Control incidence P value Odds ratio 95% CI
(n = 28) (n = 29)
Hamstring injury 1 7 0.051 0.116 0.013-1.019
1° Hamstring 1 5 0.191 0.178 0.019-1.631
2° Hamstring 0 2 - - -
1° Lower limb muscle strain 1 8 0.025* 0.097 0.011-0.839
1° Non-contact knee 1 7 0.051 0.116 0.013-1.019
* Bold type face indicates significant difference

Table 2 Lower Back Pain (as measured by the MPQ-SF): estimated marginal means for baseline and eighteen weeks by
group and estimated change within and between groups
Variable Baseline 18 weeks Δ within groups Δ between groups
intervention control intervention control intervention control mean p-value
Current mean 33.26 32.71 21.44 34.36 -11.81 1.64 -13.46 .026
95% CI 24.00, 42.52 23.62, 41.81 12.77, 30.12 25.84, 42.88 -22.90, -0.73 -8.8, 12.14 -28.35, 1.44
Overall mean 26.67 22.86 17.04 27.86 -9.63 5.00 -14.63 .006
95% CI 20.54, 32.80 16.84, 28.88 10.67, 23.41 21.60, 34.11 -16.35, -2.91 -3.07, 13.07 -24.93, -4.33 (.034)1
Sensory mean 12.57 15.26 13.02 18.20 0.45 2.94 -2.49 .461
95% CI 8.56, 16.59 11.32, 19.20 8.02, 18.03 13.29, 23.11 -3.67, 4.57 -2.54, 8.43 -9.24, 4.25
Affective mean 4.95 4.76 8.33 10.72 3.38 5.96 -2.58 .411
95% CI 1.78, 8.12 1.65, 7.87 3.96, 12.70 6.43, 15.01 -0.73, 7.48 1.07, 10.85 -8.84, 3.6
Total mean 10.53 12.56 11.81 16.19 1.28 3.63 -2.35 .436
95% CI 7.0, 14.06 9.10, 16.02 7.24, 16.38 11.70, 20.67 -2.57, 5.12 -1.14, 8.40 -8.36, 3.66
* Bold type face indicates significant difference between groups at 18 weeks.
1 Regression analysis showed a significant difference between groups: p-value calculated using regression analysis.

Discussion There are limitations in the presented study. Because

This RCT demonstrated that a sports chiropractic man- the required subject numbers as determined by the
ual therapy intervention provided at the semi-elite level power analysis was not achieved, care is needed in the
of Australian Rules football in addition to the current interpretation of the results. The late withdrawal of two
best practice multi-disciplinary medical, paramedical clubs reduced the subject numbers recruited and meant
and sports science management resulted in the preven- that the required target of subject numbers would not
tion of primary lower limb muscle strain injuries, be reached. Due to the late withdrawal it was decided to
although no statistical significance was noted for ham- continue with the study. However, the number of sub-
string injury and primary non-contact knee injury. The jects determined by the power analysis is based on an
addition of the intervention was associated with a arbitrary determined effect size, and the numbers
reduced number of matches missed due primary non- required would have been different if another effect size
contact knee injury, although no statistical significance had been chosen. Moreover, the results that are pre-
was noted for hamstring injury and primary lower limb sented report statistical significance and it is difficult to
muscle strains. In addition, reduction in LBP was determine what difference in the raw figures would be
observed along with improvements in some aspects of clinically significant. As the level of significance for pre-
the physical components of health status as measured vention of hamstring injuries and primary non-contact
by the SF-39. Treatment was predominantly directed at knee injuries was p = 0.051, given that the study was
non-local to hamstring areas, which supports the view short of the number of subjects required by the power
that several non-local factors may potentially contribute analysis, there is a strong likelihood of a type 2 error,
to hamstring and lower limb injury occurrence [7,9], especially considering how close each of these results
which may be addressed through multimodal and multi- were to p < 0.05. With regards to the fact that lower
disciplinary management [16]. These findings are impor- limb muscle strain injury incidence was significantly
tant due to their potential for injury reduction, lower while the missed weeks was not, this implies that
performance benefit and cost saving practices for a rela- many minor grade strain injuries may have been pre-
tively low cost intervention. vented, but the one injury causing 4 missed matches
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 7 of 11
http://www.biomedcentral.com/1471-2474/11/64

Table 3 Health status (as measure by the SF-39): estimated marginal means for baseline and eighteen weeks by group
and estimated change within and between groups
Variable Baseline 18 weeks Δ within groups Δ between groups
intervention control intervention control intervention control mean p-value
Physical functioning % at 100 65.50 60.00 55.60 33.30 -2.412 -4.262 1.852 .569
95%CI 48.20, 82.80 42.47, 77.53 36.86, 74.34 15.52, 51.08 -5.72, 0.90 -11.59, 3.07 -6.08, 9.79
Role limitation-physical % at 100 72.40 80.00 85.20 40.70 9.262 -18.522 27.782 .004
95%CI 56.13, 88.67 65.69, 94.31 71.81, 98.59 22.17, 59.23 -2.43, 20.94 -31.84, -5.20 10.48, 45.08
Bodily pain Mean 69.67 72.56 74.22 65.26 4.56 -7.30 11.85 .034
95%CI 63.68, 75.66 66.57, 78.55 67.43, 81.01 58.47, 72.05 -4.09, 13.21 -14.36, -0.24 0.95, 22.75
General health Mean 81.30 79.00 83.96 74.52 2.67 -4.48 7.15 .027
95%CI 76.51, 86.08 74.22, 83.78 78.51, 89.42 69.06, 79.97 -2.53, 7.86 -10.27, 1.31 -0.45, 14.74
Vitality Mean 60.37 61.11 67.04 59.44 6.67 -1.67 8.33 .050
95%CI 53.90, 66.84 54.64, 67.59 61.67, 72.40 54.08, 64.81 -0.57, 13.90 -9.15, 5.81 -1.83, 18.49
Social functioning % at 100 51.70 53.30 58.10 44.40 0.462 -5.092 5.562 .770
95%CI 33.51, 69.89 35.45, 71.15 39.49, 76.71 25.66, 63.14 -4.76, 5.68 -11.56, 1.38 -2.56, 13.67
Role limitation-emotional % at 100 75.90 83.30 92.60 66.70 8.642 -6.172 14.812 .142
95%CI 60.33, 91.47 69.95, 96.65 82.73, 102.47 48.92, 84.48 -0.75, 18.03 -16.54, 4.20 1.16, 28.47
Mental health Mean 77.63 77.93 76.59 71.11 -1.04 -6.81 5.78 .151
95%CI 72.50, 82.76 72.80, 83.06 71.61, 81.57 66.13, 76.09 -6.60, 4.52 -12.78, -0.85 -2.18, 13.74
Physical summary score Mean 52.66 52.03 53.80 49.06 1.15 -2.97 4.12 .013
95%CI 50.63, 54.68 50.00, 54.06 51.76, 55.85 47.01, 51.10 -1.14, 3.43 -5.36, -0.58 0.89, 7.35
Mental summary score Mean 50.04 50.55 51.41 48.48 1.37 -2.07 3.45 .103
95%CI 47.21, 52.87 47.72, 53.38 49.00, 53.83 46.06, 50.89 -1.38, 4.12 -5.34, 1.19 -0.72, 7.61
Depression % at 100 58.60 73.30 85.20 59.30 9.892 -3.702 13.592 .050
95%CI 40.67, 76.53 57.47, 89.13 71.81, 98.59 40.77, 77.83 1.87, 17.90 -12.15, 4.74 2.23, 24.96
* Bold type face indicates significant difference between groups at 18 weeks.

Table 4 Description of the treatment rendered to the intervention group

Intervention group (n = 29)
Number of treatments 487 (mean per player 17)
Amount of manipulation and/or mobilization to joint 2000 (47% total treatment, mean 4 per treatment)
regions
Location of manipulation and/or mobilization Thoracic spine 21%, knee 18%, hip 18%, lumbar spine 15%, sacroiliac joint 12%
Manipulation and mobilization breakdown HVLA manipulation only 56%, HVLA manipulation and mobilization 36%, Mobilization
only 8%
Amount of soft tissue techniques to soft tissue regions 2258 (53% total treatment, mean 4 per treatment)
Location of soft tissue techniques Gluteal region 22%, lumbar spine 12%, hip flexors 10%, knee 9%, posterior thigh 6%
* Soft tissue structures are defined as surrounding the involved joint (muscle, tendon, ligament, fascia etc.)

skewed the results and meant the comparison would not no change in hamstring injury rates using this same
be statistically significant. This is important in a small approach have been documented in the AFL’s long run-
sample study such as the prevention of one serious ning injury surveillance [4]. Corroborating this view-
injury (or not) can significantly alter the weeks lost pro- point, the hamstring injury incidence reported for the
file of a particular treatment approach. Only studies control group (17%) was very similar to that reported in
with much larger sample sizes can really effectively con- AFL players (16%) using the same methodology.
firm this important research observation. Difficulty arises in attempting to perform research on
Furthermore, a question could be raised regarding high-level professional or semi-professional athletes due
control group selection. It was felt that using the club to clubs not being overly enthusiastic for researchers to
based best practice medical, paramedical and sports perform interventions on their contracted and paid
science management as the control was valid because players, particularly if the intervention to be performed
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 8 of 11
http://www.biomedcentral.com/1471-2474/11/64

is purely for control purposes. To counter this dilemma [23]. It should be noted that none of the hamstring inju-
a pragmatic approach to research design was taken ries in the study had positive lumbar signs present at the
which created a further limitation in that subjects were time of diagnosis, but the lack of MRI diagnosis remains
not blinded to group allocation, meaning it cannot be as a limitation of the study.
ruled out that the intervention effect was due purely to The intervention applied in this study was based lar-
placebo or Hawthorne effects, particularly as there was gely on indirect evidence and speculative reasoning that
no blinding of the therapist. However, more modern local and non-local factors could potentially contribute
research design often requires new interventions to be to hamstring injury, which have been suggested to act
compared with the existing best practice approach [22], as a guide to a complete prevention program [9]. Similar
which was done in this study. hypotheses could be made regarding other lower limb
The injury surveillance for adverse reactions to treat- muscle strain injuries and non-contact knee injuries. As
ment may be limited due to the subjectivity of aspects a uni-modal approach was not adopted to address a sin-
of the injury definition. Players may not have self- gle risk factor, it is unclear as to what the specific
reported injury. If injury was delayed or transient, mechanism of improvement was or what component of
players may not have attributed injury to the interven- the protocol resulted in injury prevention. The multi-
tion, instead attributing it to training/competition activ- modal intervention was decided upon on the basis that
ities or other medical, paramedical or sports science it more accurately represents sports chiropractic clinical
management. Such a problem exists in any multi-modal practice [26-28], and because sports injuries, including
management scenario. Conversely, given the reliability hamstring injuries, result from a complex interaction of
of the missed match injury definition [20], it is highly multiple risk factors and events, of which only a fraction
unlikely that a more severe injury resulting in loss of have been identified [8]. For the reversible risk factors
competition match play was missed. that exist for hamstring injury, no definitive evidence
The diagnosis of hamstring strains is usually made on exists to support them [7]. It has been suggested that
clinical grounds [23]. Hamstring strains are commonly waiting for a substantial body of evidence to exist to
diagnosed through history (acute onset, non-contact support a risk factor in its role in injury before conduct-
mechanism) and examination (local tenderness, reprodu- ing a RCT may be considered unethical [8].
cible pain on straight leg raise testing and/or resisted Whilst a multi-modal approach was adopted, we spec-
knee flexion) [24]. In professional sport, MRI assessment ulate that the most significant difference between the
is often used to support the clinical diagnosis and provide control and the intervention groups was the inclusion of
further assessment of the extent and severity of the a significant amount of HVLA manipulation, as soft
injury. However, costs and availability preclude the use of tissue therapies were habitually administered to the ath-
this modality for routine assessment outside of profes- letes in this cohort. Although data were not recorded in
sional sport. Additionally, both clinical examination and this study, manipulation if used by manipulative phy-
MRI findings are strongly correlated with the time siotherapists (as in the control group) has a tendency to
required to return to competition, suggesting MRI is not be slow velocity or mobilization in nature and if HVLA
required for estimating the duration of rehabilitation of techniques are rendered they are characteristically done
an acute minor or moderate hamstring injury [25]. MRI so sparingly [29]. In the paper by Flynn et al. [29] they
imaging to confirm diagnosis of hamstring strains was state that in the previously reported low back pain lit-
not routinely performed in this study. There are limita- erature high velocity spinal manipulation utilization
tions in relying on both clinical methods of diagnosis and rates for low back pain to be between 2.8% and 8.9%,
MRI as hamstring injuries can appear clinically but not with rates in a heavily evidence based education system
on MRI and they also may appear on MRI but not clini- to be 36.2%. Alternatively, in the cited studies low velo-
cally [25]. As MRI was not routinely used, there is a pos- city mobilization is used between 27.2% and 72.0% of
sibility that some of the hamstring injuries in this study the time. Despite these figures being the most up to
may have been MRI negative which are often considered date yet published, these figures represent United States,
“back related”. There is some controversy regarding Ireland and United Kingdom physiotherapists and the
“back related” hamstring injuries as to whether a muscle figures may not be representative of current practice in
strain is the cause, particularly for minor strains where those geographical locations or in Australian phy-
causes for the pain may include referred pain from neu- siotherapists in particular. In contrast, the sports chiro-
romeningeal or myofascial structures such as the lumbar practic intervention provided to the intervention group
spine and sciatic nerve or from nearby muscles such as had a greater emphasis on performing HVLA manipula-
the gluteal and pirifomis [23]. However, “back related” tive techniques to both spinal and extremity joints, with
hamstring injury is an undefined term generally signify- 92% of total joint based treatment involving some form
ing both local hamstring signs and positive lumbar signs of HVLA manipulation technique. Future research
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 9 of 11
http://www.biomedcentral.com/1471-2474/11/64

would benefit from recording the nature of the control golgi tendon organs greater than that achieved by slow
interventions in order to clarify the differences between velocity mobilization [36]. Panjabi [37] has hypothesized
interventions or to specifically address the role of HVLA that injured spinal mechanoreceptors may alter afferent
based manipulative techniques. Future studies could input, effecting motor unit recruitment. Alterations in
specifically document the scope of the manual treatment the recruitment of motor units of the deep lumbopelvic
delivered by all treating practitioners in both groups, muscles may result in altered lumbopelvic stabilization
which would assist in comparing outcomes. A criticism strategies and insufficient force generated by the ham-
of manual therapy interventions is that its effects are strings and other muscles attached to the pelvis, or may
short term in nature. Because of this, it was decided result in excessive force production, causing subsequent
that an ongoing treatment approach with adequate spa- injury. Alterations in hamstring motor units may also
cing of treatments during the season would be applied. occur. Stimulation of mechanoreceptors by HVLA
This would also best manage ongoing injury and sub- manipulation may improve afferent feedback required to
clinical micro-trauma or gradual onset injury that could update and modify motor functions. This may improve
occur to players over the course of the season. The deci- neuromuscular control of the lumbopelvic region and/or
sion on the minimum scheduling of treatment decided the coordination of hamstring and pelvic muscle func-
upon for the intervention group was made such that tion, preventing injury. In support of such a view, Solo-
there would be a likely treatment effect. Treatment monow et al. [38] have demonstrated that discharge of
scheduling in this pragmatic arrangement was then spinal proprioception can produce change in multifidus
based upon current and previous player medical history, activation. Additionally, HVLA spinal manipulation has
examination findings, practicality, player preference and been shown to produce significant improvements in feed
practitioner experience. As the intervention was pro- forward activation times of deep abdominal musculature
vided by a single practitioner, this removed issues asso- [39], whilst case reports have shown it may improve the
ciated with inter-practitioner reliability. As mentioned in ability to perform transversus abdominus [40] and multi-
the results there was an average of 17 treatment consul- fidus contraction [41]. Collectively these deficiencies
tations administered per player in the intervention have been found to be associated with LBP, with trans-
group, but due to the pragmatic nature of the design, verses abdominus and multifidus being key stabilizers in
not all players received the same amount of treatment. lumbopelvic stabilization [30,42]. Studies have also indi-
Sherry and Best [18] have suggested that neuromuscu- cated that HVLA manipulation may improve muscle
lar control of the lumbopelvic region is needed to create function through either facilitation or disinhibition of
optimal function of the hamstrings. They further suggest neural pathways [35]. These effects, combined with spinal
that changes in neuromuscular control could lead to manipulation improving hamstring strength [12], and
changes in length tension relationships or force-velocity increased joint mobility through mechanical stretching
relationships of the hamstrings, predisposing injury. and neurophysiological mechanisms [35], may have lead
This hypothesis could extend to other muscle groups to improvements in hamstring and other lower limb
including quadriceps and groin muscles. Other authors muscle functioning and subsequent injury prevention
have also hypothesized that dysfunction of the axial ske- noted in this study. Due to the complex multi-factorial
leton may predispose abnormal hamstring functioning etiology underlying hamstring and lower limb muscle
that may relate to a greater incidence of injuries injury, it is probable that more than one possibly inter-
[7,9,16], which is supported by evidence documenting acting mechanism occurred to prevent injury. Addition-
lumbopelvic factors as risk factors for hamstring injury ally, the targeted inclusion of soft tissue therapies and
[12-15]. Supporting this mechanism of injury is the extremity joint mobilization and manipulation stretching
large body of literature showing that LBP is associated soft tissues and improving joint mobility may have poten-
with changes in lumbopelvic muscle activation and tially contributed to injury prevention.
recruitment [30,31], including early activation of biceps The trend towards reduction in primary non-contact
femoris and alteration in neuromuscular control strate- knee injuries and significant improvements in weeks
gies [32], all of which could contribute to injury. In ath- missed due to these injuries may appear surprising.
letes, changes in lumbopelvic stabilization exist However, recent literature has documented the more
following clinical recovery of LBP [33]. Noteworthy is precise details of the biceps femoris anatomy, which
the high prevalence, frequency and severity of LBP have not previously been appreciated [43]. The authors
occurring in the subjects recruited for this study [34]. hypothesized that there may be a synergistic effect
Although the neurophysiological mechanisms underly- between biceps femoris and popliteus, signifying bicep
ing HVLA manipulation are not fully known or under- femoris’ important role in knee joint stabilization [43].
stood [35], evidence exists showing it is capable of This may highlight the important bidirectional inter-
stimulating muscle spindles, pacinian corpuscles and play between hamstring and knee function. Thus, soft
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 10 of 11
http://www.biomedcentral.com/1471-2474/11/64

tissue treatments delivered to the popliteus and knee Competing interests

The authors declare that they have no competing interests.
region and HVLA manipulation to the knee may have
assisted with knee function and therefore led to preven- Received: 27 October 2009 Accepted: 8 April 2010
tion of knee injury. Lastly, research has shown that Published: 8 April 2010
HVLA spinal manipulation can reduce knee extensor
References
inhibition associated with anterior knee pain [44], which 1. Hoskins WT, Pollard H: Injuries in Australian rules football: A review of the
could lead to improvements in knee function and injury literature. Australas Chiropr Osteopathy 2003, 11(2):49-56.
prevention. Of interest, LBP has also been associated 2. Norton KI, Craig NP, Olds TS: The evolution of Australian football. J Sci
Med Sport 1999, 2(4):389-404.
with inhibition of the knee extensor muscles [45], which
3. Seward H, Orchard J, Hazard H, Collinson D: Football injuries in Australia
could imply a link between lumbopelvic and knee func- at the élite level. Med J Aust 1993, 159(5):298-301.
tion. The potential role of the knee in hamstring injury 4. Orchard J, Seward H: Epidemiology of injuries in the Australian Football
League, seasons 1997-2000. Br J Sports Med 2002, 36(1):39-44.
has been discussed elsewhere in further detail [7].
5. McManus A, Stevenson M, Finch CF, Elliot B, Hamer P, Lower A, Bulsara M:
Due to constraints in manuscript size, we are unable to Incidence and risk factors for injury in non-elite Australian Football. J Sci
describe the entire treatment provided in this study or to Med Sport 2004, 7(3):384-391.
6. Verrall GM, Kalairajah Y, Slavotinek JP, Spriggins AJ: Assessment of player
speculate on all proposed mechanisms of improvement,
performance following return to sport after hamstring muscle strain
which will be the subject of a subsequent publication. It injury. J Sci Med Sport 2006, 9(1-2):87-90.
should be noted that treatment provided was patient spe- 7. Hoskins W, Pollard H: The management of hamstring injury–Part 1: Issues
in diagnosis. Man Ther 2005, 10(2):96-107.
cific and addressed both kinetic and kinematic chain vari-
8. Bahr R, Holme I: Risk factors for sports injuries–a methodological
ables. The treatment was representative of the ‘modern approach. Br J Sports Med 2003, 37(5):384-392.
multimodal’ (MMM) chiropractic approach that has been 9. Hoskins W, Pollard H: Hamstring injury management–Part 2: Treatment.
Man Ther 2005, 10(3):180-190.
described in the sports chiropractic literature [26,27], and 10. Petersen J, Hölmich P: Evidence based prevention of hamstring injuries
recommended in selecting a chiropractor for the man- in sport. Br J Sports Med 2005, 39(6):319-23.
agement of athletic injuries [28]. 11. Orchard JW: Intrinsic and extrinsic risk factors for muscle strains in
Australian football. Am J Sports Med 2001, 29(3):300-303.
12. Cibulka MT, Rose SJ, Delitto A, Sinacore DR: Hamstring muscle strain
Conclusions treated by mobilizing the sacroiliac joint. Phys Ther 1986, 66(8):1220-1223.
Based on the limitations of this study which includes a 13. Watson AW: Sports injuries in footballers related to defects of posture
and body mechanics. J Sports Med Phys Fitness 1995, 35(4):289-294.
low sample size, this RCT demonstrated that a sports
14. Watson AW: Sports injuries related to flexibility, posture, acceleration,
chiropractic intervention comprising significant amounts clinical defects, and previous injury, in high-level players of body
of HVLA manipulation and soft tissue therapies provided contact sports. Int J Sports Med 2001, 22(3):222-225.
15. Verrall GM, Slavotinek JP, Barnes PG, Fon GT, Spriggins AJ: Clinical risk
in addition to the current best practice medical, parame- factors for hamstring muscle strain injury: a prospective study with
dical and sports science management appears to be bene- correlation of injury by magnetic resonance imaging. Br J Sports Med
ficial for the prevention of lower limb muscle strain 2001, 35(6):435-439.
16. Hoskins WT, Pollard HP: Successful management of hamstring injuries in
injuries, weeks missed due to primary non-contact knee
Australian Rules footballers: two case reports. Chiropr Osteopat 2005,
injuries and reduction of LBP and improvement in physi- 13(1):4.
cal components of health status. In addition, although 17. Mason DL, Dickens V, Vail A: Rehabilitation for hamstring injuries.
Cochrane Database Syst Rev 2007, 24(1):CD004575.
not statistically significant, there was a trend towards pre- 18. Sherry MA, Best TM: A comparison of 2 rehabilitation programs in the
vention of hamstring and primary non-contact knee inju- treatment of acute hamstring strains. J Orthop Sports Phys Ther 2004,
ries and there were no reported adverse outcomes from 34(3):116-125.
19. Mierau D, Kirkaldy-Willis WH, Theil HW: Spinal manipulation. Managing low
the intervention. The interesting trend in results but
back pain Philadelphia: Churchill LivingstoneKirkaldy-Willis W 1999, 326-31.
non-statistically significance should be replicated using a 20. Orchard J, Hoskins W: For debate: consensus injury definitions in team
larger sample size to remove the short comings of this sports should focus on missed playing time. Clin J Sports Med 2007,
17(3):192-196.
study. Based on the findings of this study due considera-
21. McNeil D: Mantel-Haenszel. Methods. Epidemiological Research Methods
tion should be given for the inclusion of sports chiro- New York: WileyMcNeil D 1996, 97.
practic in the management options of elite athletes. 22. Abalos E, Carroli G, Mackey ME: The tools and techniques of evidence-
based medicine. Best Pract Res Clin Obstet Gynaecol 2005, 19(1):15-26.
23. Verrall GM, Slavotinek JP, Barnes PG, Fon GT, Spriggins AJ: Clinical risk
factors for hamstring muscle strain injury: a prospective study with
Acknowledgements
correlation of injury by magnetic resonance imaging. Br J Sports Med
We would like to acknowledge the statistical assistance provided by Ann
2001, 35(6):435-9.
Eyland and Don McNeil and the participation and assistance of the staff and
24. Gibbs NJ, Cross TM, Cameron M, Houang MT: The accuracy of MRI in
players at the Northern Bullants and the Coburg Tigers football clubs.
predicting recovery and recurrence of acute grade one hamstring
muscle strains within the same season in Australian Rules football
Authors’ contributions
players. J Sci Med Sport 2004, 7(2):248-58.
WH and HP conceived the idea of the study. WH performed recruitment of
25. Schneider-Kolsky ME, Hoving JL, Warren P, Connell DA: A comparison
subjects, data entry and the intervention. WH and HP contributed to writing
between clinical assessment and magnetic resonance imaging of acute
draft documents and the final manuscript. Both authors read and approved
hamstring injuries. Am J Sports Med 2006, 34(6):1008-15.
the final manuscript.
Hoskins and Pollard BMC Musculoskeletal Disorders 2010, 11:64 Page 11 of 11
http://www.biomedcentral.com/1471-2474/11/64

26. Hoskins W, McHardy A, Pollard H, Windsham R, Onley R: Chiropractic

treatment of lower extremity conditions: a literature review. J
Manipulative Physiol Ther 2006, 29(8):658-671.
27. Pollard H, Hoskins W, McHardy A, Bonello R, Garbutt P, Swain M,
Dragasevic G, Pribicevic M, Vitiello A: Australian chiropractic sports
medicine: half way there or living on a prayer? Chiropr Osteopat 2007,
15:14.
28. Hoskins W, Pollard H, Garbutt P: How to select a chiropractor for the
management of athletic conditions. Chiropr Osteopat 2009, 17:3.
29. Flynn TW, Wainner RS, Fritz JM: Spinal manipulation in physical therapist
professional degree education: A model for teaching and integration
into clinical practice. J Orthop Sports Phys Ther 2006, 36(8):577-587.
30. Wilke H, Wolfe S, Claes L, Arand M, Wiesend A: Stability increase of the
lumbar spine with different muscle groups. Spine 1995, 20(2):192-98.
31. Demoulin C, Distrée V, Tomasella M, Crielaard JM, Vanderthommen M:
Lumbar functional instability: a critical appraisal of the literature. Ann
Readapt Med Phys 2007, 50(8):677-84.
32. Hungerford D, Gilleard W, Hodges P: Evidence of altered lumbopelvic
muscle recruitment in the presence of sacroiliac joint pain. Spine 2003,
28(14):1593-1600.
33. Cholewicki J, Greene HS, Polzhofer GK, Galloway MT, Shah RA, Radebold A:
Neuromuscular function in athletes following recovery from a recent
acute low back injury. J Orthop Sports Phys Ther 2002, 32(11):568-75.
34. Hoskins W, Pollard H, Daff C, Odell A, Garbutt P, McHardy A, Hardy K,
Dragasevic G: Low back pain status in elite and semi-elite Australian
football codes: a cross-sectional survey of football (soccer), Australian-
Rules, rugby league, rugby union and non-athletic controls. BMC
Musculoskelet Disord 2009, 10(1):38.
35. Pickar JG: Neurophysiological effects of spinal manipulation. Spine J 2002,
2(5):357-371.
36. Pickar JG, Wheeler JD: Response of muscle proprioceptors to spinal
manipulative-like loads in the anesthetized cat. J Manipulative Physiol Ther
2001, 24(1):2-11.
37. Panjabi MM: A hypothesis of chronic back pain: ligament subfailure
injuries lead to muscle control dysfunction. Eur Spine J 2006,
15(5):668-676.
38. Solomonow M, Zhou B, Harris M, Lu Y, Baratta RV: The ligamento-muscular
stabilizing system of the spine. Spine 1998, 23(23):2552-2562.
39. Marshall P, Murphy B: The effect of sacroiliac joint manipulation on feed-
forward activation times of the deep abdominal musculature. J
Manipulative Physiol Ther 2006, 29(3):196-202.
40. Gill NW, Teyhen DS, Lee IE: Improved contraction of the transversus
abdominis immediately following spinal manipulation: A case study
using real-time ultrasound imaging. Man Ther 2007, 12(3):280-285.
41. Brenner AK, Gill NW, Buscema CJ, Kiesel K: Improved activation of lumbar
multifidus following spinal manipulation: a case report applying
rehabilitative ultrasound imaging. J Orthop Sports Phys Ther 2007,
37(10):613-9.
42. Hodges P, Richardson C: Inefficient muscular stabilization of the lumbar
spine associated with low back pain: a motor control evaluation of
transversus abdominus. Spine 1996, 21(22):2640-50.
43. Tubbs RS, Caycedo FJ, Oakes WJ, Salter EG: Descriptive anatomy of the
insertion of the biceps femoris muscle. Clin Anat 2006, 19(6):517-521.
44. Suter E, McMorland G, Herzog W, Bray R: Conservative lower back
treatment reduces inhibition in knee-extensor muscles: a randomized
controlled trial. J Manipulative Physiol Ther 2000, 23(2):76-80.
45. Suter E, Lindsay D: Back muscle fatigability is associated with knee
extensor inhibition in subjects with low back pain. Spine 2001, 26(16):
E361-366. Submit your next manuscript to BioMed Central
and take full advantage of:
Pre-publication history
The pre-publication history for this paper can be accessed here: http://www.
• Convenient online submission
biomedcentral.com/1471-2474/11/64/prepub
• Thorough peer review
doi:10.1186/1471-2474-11-64
• No space constraints or color figure charges
Cite this article as: Hoskins and Pollard: The effect of a sports
chiropractic manual therapy intervention on the prevention of back • Immediate publication on acceptance
pain, hamstring and lower limb injuries in semi-elite Australian Rules
• Inclusion in PubMed, CAS, Scopus and Google Scholar
footballers: a randomized controlled trial. BMC Musculoskeletal Disorders
2010 11:64. • Research which is freely available for redistribution

Submit your manuscript at

www.biomedcentral.com/submit