This article was downloaded by: [Flinders University of South Australia]

On: 04 October 2014, At: 01:34

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Sports Sciences

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rjsp20

Lumbo-pelvic loading during fast bowling in adolescent

cricketers: The influence of bowling speed and
technique
a b a a
Helen Crewe , Amity Campbell , Bruce Elliott & Jacqueline Alderson
a
The University of Western Australia, School of Sport Science , Exercise & Health, Perth ,
Australia
b
Curtin University, School of Physiotheraphy and Curtin Health Innovation Research
Institute , Perth , Australia
Published online: 30 Jan 2013.

To cite this article: Helen Crewe , Amity Campbell , Bruce Elliott & Jacqueline Alderson (2013) Lumbo-pelvic loading during
fast bowling in adolescent cricketers: The influence of bowling speed and technique, Journal of Sports Sciences, 31:10,
1082-1090, DOI: 10.1080/02640414.2012.762601

To link to this article: http://dx.doi.org/10.1080/02640414.2012.762601

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Sports Sciences, 2013
Vol. 31, No. 10, 1082–1090, http://dx.doi.org/10.1080/02640414.2012.762601

Lumbo-pelvic loading during fast bowling in adolescent cricketers:

The influence of bowling speed and technique

HELEN CREWE1, AMITY CAMPBELL2, BRUCE ELLIOTT1,

& JACQUELINE ALDERSON1
1
The University of Western Australia, School of Sport Science, Exercise & Health, Perth, Australia, and 2Curtin University,
School of Physiotheraphy and Curtin Health Innovation Research Institute, Perth, Australia
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

(Accepted 21 December 2012)

Abstract
This study aimed to assess changes in bowling technique and lumbar load over the course of a bowling spell in adolescent
fast bowlers, and to investigate the relationship between lumbar loads during fast bowling and kinematic factors which have
previously been associated with low back injury. Three-dimensional motion analysis was carried out on forty participants
who performed an 8-over bowling spell. There were no significant changes in bowling technique or lumbar loads over the
course of the spell. Bowling with a more extended front knee, faster ball release speed and increased shoulder counter-
rotation were related to increased lumbo-pelvic loading – in particular peak transverse plane rotation moments and anterior-
posterior shear forces. These lumbar loads may be a factor in low back injury aetiology and future studies should investigate
the relationship between lumbar loading, injury incidence and other risk factors.

Keywords: biomechanics, cricket, low back

Introduction removed from the lumbar spine itself. Indeed, it is

The fast bowling action involves complex and vigor-
ous three-dimensional (3D) motion of the trunk and
upper limb as the bowler attempts to produce max-
imum ball velocity, while enduring vertical ground
reaction forces of approximately 6 body weight (BW)
(Bartlett, Stockill, Elliott, & Burnett, 1996). Large
impact forces are transferred along the kinetic chain
through the ankle, knee and hip joints to the lumbar
spine (Elliott, 2000), an area particularly prone to
injury in adolescent fast bowlers (Stretch, 1995).
Overuse injuries, such as lumbar stress fractures
(spondylolysis), result from the repetitive application
of force (Whiting & Zernicke, 2008), and it is there-
fore critical that lumbar loads are quantified during
bowling in order to better understand load-related
injury mechanisms. Only one study to date has esti-
mated lumbar loads during the bowling action,
reporting that loads were highest during the phase
of the delivery stride occurring between front foot
contact (FFC) and ball release (Ferdinands,
Kersting, & Marshall, 2009).
While low back injuries in fast bowlers have received
considerable attention in the literature, research has
focused on kinematic risk factors that are somewhat
known that fast bowlers tend to rotate the trunk
from a relatively front-on position at back foot contact,
to a more side-on position prior to front foot contact,
and this ‘shoulder counter-rotation’ has been asso-
ciated with spondylolysis incidence (Foster, John,
Elliott, Ackland, & Fitch, 1989; Portus, Mason,
Elliott, Pfitzner, & Done, 2004) and intervertebral
disc degeneration (Burnett et al., 1996; Elliott,
Davis, Khangure, Hardcastle, & Foster, 1993) in ado-
lescent fast bowlers. However the mechanism through
which this aspect of the bowling technique increases
low back injury risk has not been determined. A more
extended front knee during the front foot contact
phase has also been associated with spondylolysis
development (Foster et al., 1989; Portus et al.,
2004). It is hypothesised that flexion of the knee fol-
lowing front foot contact contributes to the attenua-
tion of large ground reaction forces, thereby reducing
the magnitude of the forces transmitted to the lumbar
spine (Elliott, 2000). However, the relationship
between front knee kinematics and lumbar load has
not been investigated. The fast bowler’s primary aim is
to release the ball at high speed. It has been demon-
strated that bowlers who achieve higher ball release
speeds experience higher peak ground reaction forces

Correspondence: Jacqueline Alderson, The University of Western Australia, School of Sports Science, Exercise and Health, Perth, Australia.
E-mail:Jacqueline.Alderson@uwa.edu.au

© 2013 Taylor & Francis


and also develop these forces more rapidly (Portus et
al., 2004), and this has led previous authors to suggest
that increased ball release speed may also be a con-
tributory factor in low back injury development in
some bowlers (Portus et al., 2004).
It is possible that lumbar loading changes over the
course of a long bowling spell in adolescent fast bow-
lers due to fatigue-related changes in technique.
Bowling technique has been found to remain consis-
tent during an extended spell in two studies where the
mean age of the bowlers was 18 years (Burnett,
Elliott, & Marshall, 1995) and 23 years (Portus,
Sinclair, Burke, Moore, & Farhart, 2000), however
fatigue-related technique changes have not been
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014
investigated in younger cricketers. Given the high
incidence of low back injuries (including lumbar
stress fractures) in fast bowlers under the age of 18
years, it is important to understand the potential
effects of fatigue on lumbar loading in this age
group. This knowledge may be used, firstly, to further
inform workload guidelines in junior fast bowlers and,
secondly, to determine whether a limited number of
deliveries can be used as a representation of an indi-
vidual’s technique in future investigations.
The aim of this study was to (a) assess changes in
bowling technique and lumbar load over the course
of a bowling spell in adolescent fast bowlers and (b)
investigate the relationship between lumbar loads
during fast bowling and kinematic factors that have
previously been associated with low back injury: (i)
shoulder counter-rotation, (ii) front knee kinematics,
and (iii) ball release speed, in order to further the
understanding of lumbar spinal injury mechanisms
in fast bowlers.
Methods
Participants
Forty male fast bowlers (mean age 16.2 years, height
182 cm, mass 70.8 kg) from district and/or state
junior cricket squads volunteered to participate in
the study. Ethical approval was obtained from the
University of Western Australia’s Human Research
Ethics Committee and all participants (and their
guardians, where required) provided informed, writ-
ten consent to participate in this research.
Bowling testing procedure
Data collection was performed at the biomechanics
laboratory at the School of Sport Science, Exercise
and Health at the University of Western Australia. A
12-camera VICON MX motion analysis system
(Vicon, Oxford Metrics, Oxford, UK) operating at
250 Hz and a 1.2 m × 1.2 m force plate (Advanced
Mechanical Technology Inc., Watertown, MA)
Lumbo-pelvic loading during fast bowling 1083
sampling at 2000 Hz were used to collect kinematic
and ground reaction force data. A cricket crease was
marked on the force plate to assist with the successful
collection of front foot ground reaction force data.
Retro-reflective markers were affixed to the parti-
cipants’ skin according to a customised marker set
and model for the lower limbs and pelvis (Besier,
Sturnieks, Alderson, & Lloyd, 2003). This com-
prised of single markers placed on the head of the
first and fifth metatarsals, calcaneus, anterior super-
ior iliac spines (ASIS) and posterior superior iliac
spines (PSIS). A series of marker clusters, consisting
of three markers attached to a semi-rigid plastic
baseplate, were attached bilaterally to the thigh and
lower legs. Subject specific static calibration trials
were performed using a foot-calibration rig to mea-
sure foot abduction/adduction and inversion/ever-
sion angles (Besier et al., 2003). Additionally, static
trials were conducted with markers placed on the
medial and lateral malleoli and medial and lateral
femoral condyles, and dynamic functional trials per-
formed to determine relevant knee and hip joint axes
of rotation and associated joint centres (Besier et al.,
2003). Additionally, markers were placed on the L1
and L5 spinous processes and approximately 5 cm
on either side of the spine at the level of L4 (left
lower lumbar: LLL, right lower lumbar: RLL) to
define the lumbar segment (Crewe, Campbell,
Elliot, & Alderson, in press).
Participants attended the laboratory in pairs and
after carrying out a self-directed warm-up, were
required to bowl eight overs (each over consisting
of six deliveries), aiming to bowl a good line and
length at match pace. A delivery was judged to be a
good line and length if the ball hit a target positioned
immediately behind the batsman’s stumps. The tar-
get measured 120 cm (height) × 60 cm (width) and
was placed 50 cm above the ground and in line with
the middle stump (extending approximately 45 cm
outside the off stump) (Portus et al., 2000). While
one participant bowled, the other completed a set of
shuttle runs to simulate fielding activity during a
match. These shuttle runs consisted of a walking,
jogging, running and sprinting intervals-based move-
ment patterns reported in previous research
(Petersen, Pyne, Dawson, Portus, & Kellett, 2010).
Only trials in which the bowler’s front foot landed
within the boundaries of the force plate were con-
sidered for analysis. The two successful trials with
the highest ball release speeds from overs 2, 4, 6 and
8 were then selected to analyse between-over differ-
ences in lumbar load.
Data processing
The 3D data were processed using Vicon Nexus
motion analysis software (Vicon, Oxford Metrics,
 1084 H. Crewe et al.

Oxford, UK). Data were filtered using a fourth-order Statistical analysis

low-pass Butterworth filter operating at a cut-off
Statistical analysis was performed using SPSS 19.0
frequency of 15 Hz for the marker trajectories and
(SPSS Inc., Chicago, IL.).
50 Hz for the ground reaction force (GRF) data.
Linear mixed models were used to determine dif-
Filter cut-off frequencies were determined using
ferences in lumbar load, bowling technique (includ-
residual analysis (Winter, 1990).
ing approach velocity, shoulder alignment at back
All lower limb anatomical and joint coordinate
foot contact, shoulder counter-rotation, front knee
systems were calculated in accordance with the stan-
flexion angles), and ball release speed during an 8-
dards outlined by the International Society of
over bowling spell. Statistical significance was set at
Biomechanics (Wu & Cavanagh, 1995) and have
P < 0.05. To ensure that the results were not influ-
been fully described (Besier et al., 2003). The lum-
enced by within-over variability, the interaction
bar segment was defined using the L5 marker to
effect between the within-over trial and the over
represent the origin of the lumbar coordinate system.
number was examined. No significant effect was
The y axis was defined using a vector from the L5 to
found, demonstrating that within-over differences
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

L1 marker, the x axis was calculated from the cross

between deliveries were insignificant, and subse-
product of the y axis, and a defining line between the
quently this was not included in the model.
LLL and RLL markers. Finally, the z axis was cal-
After confirming that all variables were normally
culated as the cross product of the y and x axes. The
distributed, Pearson’s correlation coefficients were
lumbo-pelvic joint location was defined as a virtual
used to investigate the relationship between each of
point 5% along the length of a line from the L5
the lumbar load variables and: (i) shoulder counter-
marker to the bisector of the two ASIS markers
rotation, (ii) the minimum front knee flexion angle
(Seay, Selbie, & Hamill, 2008) and served as the
occurring between front foot contact and ball
point of application for the inverse dynamics analysis
release, and (iii) ball release speed. All lumbar load
of the lumbo-pelvic segment. The lumbo-pelvic
variables with moderate (> 0.3) or large (> 0.5)
angles were decomposed using a ZXY (flexion-
statistically significant correlations (Hopkins, 2002)
extension, abduction-adduction, internal-external
with any of the three kinematic variables of interest,
rotation) Euler angle sequence, in accordance with
were then further analysed in order to confirm the
standard biomechanical methods (Wu & Cavanagh,
potential effect of these variables on lumbar load.
1995).
For this purpose, the bowlers were sub-classified
Scaled inertial parameters for the lower limb (de
into two distinct groups according to discrete tech-
Leva, 1996), pelvis and lumbar segments (Pearsall,
nique parameters. Participants whose shoulder
Reid, & Livingston, 1996) were incorporated in the
counter-rotation, maximum front knee flexion
inverse dynamics model for the calculation of
angle, and ball release speeds were determined to
lumbo-pelvic kinetics. Lumbo-pelvic angles and
lie within the 40th to 60th percentile for each of
moments were defined as positive in flexion, and
these variables (i.e. the middle eight bowlers) were
toward the non-dominant side for lateral flexion
excluded from the subsequent analysis. This allowed
and rotation. Forces were defined as positive in the
the kinematic variables for the two extremes of the
anterior direction, vertically upwards and towards
cohort to be compared, with reference to peak lum-
the dominant side medio-laterally.
bar load. Independent sample t-tests were performed
In order to assess continuous force and moment
and, following a partial Bonferroni correction to
data each trial was temporally normalised from front
address multiple comparisons; statistical significance
foot contact (0%) to ball release (100%), as this is
was set at P < 0.01. Three sets of comparisons were
the period during which peak ground reaction forces
performed between the following groups:
and lumbar loads are known to occur (Ferdinands
et al., 2009). Magnitude and rate of application are
● Shoulder counter-rotation: SCRlow and SCRhigh
two key factors that characterise load with regard to
● Knee angle: Kneeflexed and Kneestraight
injury mechanisms (Whiting & Zernicke, 2008).
● Speed: Speedlow and Speedhigh
Therefore, the variables chosen to represent ‘lumbar
load’ were peak lumbo-pelvic force (anterior-poster-
ior, medio-lateral, vertical), peak lumbo-pelvic
Results
moment (flexion-extension, lateral flexion, rotation)
and the time to peak of each of these forces and Lumbo-pelvic kinetics between front foot contact
moments, occurring between front foot contact and and ball release (Figure 1) were investigated in two
ball release. Peak forces and moments were normal- parts, the first dealing with loading throughout an 8-
ised by the individual’s mass and mass × height over spell, and secondly with respect to bowling
respectively. technique and ball release speed.
 Lumbo-pelvic loading during fast bowling 1085
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

Figure 1. Lumbo-pelvic forces and moments occurring between front foot contact (FFC) and ball release (BR) (n = 40). Solid line: mean;
dotted line: s. Forces are positive in the anterior direction, vertically upwards and towards the dominant side medio-laterally. Moments are
positive in flexion, and towards the non-dominant side for lateral flexion and rotation.

Lumbar load and bowling technique during an 8-over shoulder counter-rotation (35.1° ± 12.2), ball
bowling spell release speed (30.2 m · s−1 ± 3.4), and minimum
(6.8° ± 7.0) and maximum front knee flexion
Of the 40 participants, only 23 completed the 8-over
(42.6° ± 14.2) angles are presented in Table II.
bowling spell. Seventeen of the bowlers’ data was
There were a number of significant moderate to
considered void beyond the first four overs as a
large correlations between several lumbar load vari-
result of repeated marker loss due to excessive per-
ables and the selected kinematic variables of interest
spiration. Sample means at over 2, 4, 6 and 8 for
(Table III). The strongest correlation was between
bowling technique and lumbar load variables are
ball release speed and mean peak lumbo-pelvic ante-
presented in Table I. Statistical analysis revealed no
rior-posterior force (r = 0.529, 95% confidence
significant effect for ‘bowling over’ of any of the
interval (CI): 0.300 to 0.729, P < 0.01). Ball release
variables analysed (Table I), indicating that neither
speed was also correlated with mean peak lumbo-
bowling technique nor lumbar loading varied signif-
pelvic medio-lateral force (r = 0.391, CI: 0.161 to
icantly during the 8-over spell.
0.653, P <0.05) and mean time to peak vertical force
(r = –0.311, CI: –0.633 to 0.093, P < 0.05).
Bowling technique, ball release speed and lumbar load Increased ball release speed is therefore associated
with an increased lumbo-pelvic anterior-posterior
After confirming that there was no within- or
and medio-lateral shear force and more rapid devel-
between-over effect on any of the variables of inter-
opment of peak vertical force.
est, the mean of each bowler’s four deliveries from
The mean minimum front knee flexion angle was
overs 2 and 4 (selected as described earlier by high-
correlated with the mean peak anterior-posterior force
est ball release speed) was used to investigate the
(r = –0.346, CI: –0.620 to –0.064, P <0.05), peak
relationship between lumbar load and the three kine-
medio-lateral force (r = –0.353, CI: –0.586 to –0.072,
matic variables (shoulder counter-rotation, front
P < 0.05) and peak rotation moment (r = –0.498, CI:
knee angle and ball release speed). The mean
 1086 H. Crewe et al.

Table I. Bowling technique and lumbar load variables: mean of 23 bowlers at over 2, 4, 6, 8 (n = 23).

OVER 2 OVER 4 OVER 6 OVER 8

Mean s Mean s Mean s Mean s F P

−1
Approach velocity (m · s ) 5.8 0.6 5.8 0.7 5.8 0.6 5.9 0.7 0.237 0.870
Shoulder alignment at BFC (°) 230.9 15.5 229.4 15.6 230.7 15.4 230.4 15.8 0.083 0.969
Shoulder counter-rotation (°) 36.0 15.6 34.1 15.0 35.3 14.4 35.5 14.5 0.147 0.931
Front knee flexion angle at FFC (°) 14.2 6.3 14.8 6.5 14.0 6.9 14.0 7.5 0.129 0.943
Maximum front knee flexion angle (°) 41.8 15.2 41.6 15.1 43.5 16.3 43.2 15.8 0.178 0.911
Peak GRF (braking) (BW) 4.3 0.8 4.2 0.9 4.2 0.9 4.2 1.0 0.236 0.871
Peak GRF (vertical) (BW) 4.8 1.0 4.8 1.1 4.7 1.1 4.8 1.3 0.076 0.973
Ball release speed (m · s−1) 29.4 3.2 29.8 3.4 29.5 3.2 29.7 3.1 0.112 0.953
Force AP (N · kg−1) 26.6 13.6 26.7 14.4 27.3 13.7 27.7 14.3 0.067 0.977
Force V (N · kg−1) 19.8 9.1 21.6 9.8 21.5 10.2 21.1 10.2 0.322 0.809
Force ML (N · kg−1) 21.9 7.9 23.1 8.5 22.6 8.4 22.9 7.7 0.212 0.888
Moment FE (Nm · kg−1 · m−1) 10.5 5.4 10.1 5.2 10.4 5.0 10.3 5.6 0.047 0.986
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

Moment LF (Nm · kg−1 · m−1) 12.4 4.2 12.4 3.8 12.3 4.8 12.3 4.5 0.014 0.998
Moment RO (Nm · kg−1 · m−1) 10.7 3.6 10.4 3.5 9.4 3.5 9.8 3.5 1.285 0.281
Time to peak force AP (ms) 36.9 26.6 37.9 28.0 30.9 21.3 37.1 24.8 0.746 0.526
Time to peak force V (ms) 34.3 27.4 35.1 30.3 35.5 30.7 32.0 26.9 0.138 0.937
Time to peak force ML (ms) 29.8 16.2 31.0 15.5 29.2 18.6 29.1 15.6 0.130 0.942
Time to peak moment FE (ms) 61.0 18.8 58.9 19.2 61.6 16.4 58.9 18.6 0.271 0.846
Time to peak moment LF (ms) 63.1 20.1 64.3 20.4 64.4 19.8 65.7 19.1 0.140 0.936
Time to peak moment RO (ms) 78.3 14.7 77.0 14.9 76.8 14.9 77.7 16.9 0.095 0.963

BFC: back foot contact; FFC: front foot contact; GRF: ground reaction force; BW: body weight; AP: anterior-posterior; V: vertical; ML:
medio-lateral; FE: flexion-extension; LF: lateral flexion; RO: rotation

Table II. Descriptive biomechanical data for all 40 bowlers.

Minimum Maximum Mean s

SCR (°) 11.9 59.2 35.1 12.2

Max knee flexion (°) 12.9 67.6 42.6 14.2
Min knee flexion (°) −10.7 17.9 6.8 7.0
Ball release speed (m · s−1) 22.5 34.5 29.9 2.8
Peak GRF vertical (BW) 2.7 7.2 4.9 1.0
Peak GRF brake (BW) 1.6 5.2 3.3 0.8

SCR: shoulder counter-rotation; GRF: ground reaction force; BW: body weight

–0.673 to –0.296, P <0.01). A more extended front
knee during the front foot contact phase is therefore
associated with increased lumbo-pelvic shear forces
and flexion and rotation moments.
Mean shoulder counter-rotation was correlated
with mean peak medio-lateral force (r = 0.326, CI:
0.024 to 0.587, P < 0.05) and time to peak vertical
force (r = –0.436, CI: –0.659 to –0.132, P < 0.01).
This suggests that shoulder counter-rotation, occur-
ring between back foot contact and front foot con-
tact, is associated with increased medio-lateral
lumbar forces and decreased time to peak vertical
force in the phase comprising front foot contact to
ball release.
Relationships between the kinematic variables were
also identified. Mean ball release speed was correlated
with shoulder counter-rotation (r = 0.541, CI: 0.267
to 0.733, P < 0.01) and minimum front knee flexion
angle (r = –0.551, CI: –0.743 to –0.339, P <0.01).
Shoulder counter-rotation and minimum front knee
flexion angle were also correlated with one another
(r = –0.436, CI: –0.665 to –0.133, P < 0.01).
Considered together, these results imply that faster
ball release speed is associated with greater shoulder
counter-rotation and a more extended front knee dur-
ing the stance phase.
Two groups of 16 bowlers were stratified with
respect to each of the three kinematic variables of
interest in the following manner:
● SCRlow (< 30.1°) and SCRhigh (> 38.7°)
● Kneeflexed (> 9.0°) and Kneestraight (< 6.6°)
−1
● Speedlow (< 29.5 m · s ) and Speedhigh
−1
(> 30.5 m · s )
Increased lumbar load was observed for each of the
variables analysed (peak anterior-posterior and
medio-lateral force, peak flexion-extension and
 Lumbo-pelvic loading during fast bowling 1087

Table III. Pearson’s correlations (r) between lumbar load variables and shoulder counter-rotation (SCR), minimum knee flexion angle
(Knee) and ball release speed (Speed).

SCR Knee Speed

r P r P r P

Peak force AP 0.174 0.283 −0.346* 0.029 0.542** 0.000

V 0.026 0.875 0.008 0.959 0.305 0.055
ML 0.326* 0.040 −0.353* 0.025 0.442* 0.004
Peak moment FE 0.262 0.102 −0.309* 0.047 0.215 0.183
LF −0.019 0.909 −0.058 0.723 0.025 0.879
RO 0.289 0.071 −0.498* 0.001 0.411* 0.008
Time to peak force AP 0.004 0.979 −0.052 0.749 −0.091 0.578
V −0.436* 0.005 0.211 0.190 −0.338* 0.033
ML −0.144 0.376 −0.030 0.853 −0.009 0.955
Time to peak moment FE −0.142 0.382 −0.119 0.466 0.018 0.910
LF 0.020 0.905 −0.151 0.352 0.012 0.941
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

RO −0.072 0.660 −0.102 0.533 −0.117 0.471

*P < 0.05, **P <0.01.

AP: anterior-posterior; V: vertical; ML: medio-lateral; FE: flexion-extension; LF: lateral flexion; RO: rotation

rotation moment, time to peak vertical force) in the
SCRhigh, Kneestraight and Speedhigh groups compared
with the SCRlow, Kneeflexed and Speedlow groups,
respectively (Table IV). However, these differences
were only statistically significant for the peak rotation
moment (Kneestraight 11.3 Nm · kg−1 · m−1 ± 2.9,
Kneeflexed 8.1 Nm · kg−1.m−1 ± 1.6, t = 3.879,
P < 0.01) and peak anterior-posterior force
(Kneestraight 30.2 Nm · kg−1 ± 9.0, Kneeflexed 20.2
Nm · kg−1 ± 9.9, t = 2.999, P <0.01; Speedhigh 31.5
Nm · kg−1 ± 10.1, Speedlow 20.7 Nm · kg−1 ± 8.7,
t = 3.236, P < 0.05).
Discussion
The findings of this study confirm previous research
demonstrating the substantial lumbo-pelvic loads
experienced by cricket fast bowlers (Ferdinands
et al., 2009). In comparison with a study on tread-
mill running that used a similar model (Seay et al.,
2008), the peak lumbo-pelvic vertical force during
bowling was approximately twice as large. More
notably, lumbar shear forces were approximately
40–50 times greater. Peak lumbar moments were
also substantially higher than those shown during
running, by approximately 6 times in the sagittal
plane, 25 times for transverse plane rotation and up
to 100 times for lateral flexion. These increased
lumbar loads can be attributed to greater ground
reaction forces during bowling (4.9 BW vertical
and 3.0 BW anterior-posterior in the current study,
versus 2.5–3 BW vertical and 0.5 BW anterior-pos-
terior in running (Cavanagh & Lafortune, 1980)),
and the asymmetrical nature of the bowling action.
The neural arch is the major structure of the vert-
ebra that resists anterior-posterior shear (Adams,
2002) and this is the site where spondylolysis occurs.
Also, it has been shown that the stress in the neural
arch is highest when a compression force is applied in
combination with extension, rotation or lateral flexion
(Chosa, Totoribe, & Tajima, 2004). These extreme
shear forces and moments during bowling are there-
fore implicated in the high incidence of lumbar stress
fractures in fast bowlers. Asymmetric motion of the
trunk, which is characteristic of the fast bowling action,
is known to increase medio-lateral shear force on the
vertebrae (Marras & Granata, 1997; Natarajan,
Lavender, An, & Andersson, 2008). Intervertebral
discs are especially vulnerable to such shear forces
(Marras & Granata, 1997; Natarajan et al., 2008)
and this type of lumbar loading may be a key contri-
buting factor to premature lumbar disc degeneration in
fast bowlers (Annear, Chakera, Foster, & Hardcastle,
1992; Elliott, Hardcastle, Burnett, & Foster, 1992;
Ranson, Kerslake, Burnett, Batt, & Abdi, 2005).
The first aim of this study was to assess whether
bowling speed, technique or lumbar loading change
during the course of an 8-over spell in adolescent
fast bowlers. In the present cohort, no significant
differences were found in any of these variables.
This is in agreement with previous investigations
involving older groups of bowlers (Burnett et al.,
1995; Portus et al., 2000). In these earlier studies,
the authors found no changes in bowling speed or
technique, but did note that bowlers using a front-on
action tended to increase their SCR during an
extended bowling spell. In the current study only
two participants were classified as using a front-on
bowling technique and we are therefore unable to
add further clarity to this finding. A limitation of the
current study is that the participants performed the
8-over bowling spell in laboratory conditions. While
every attempt was made to achieve ecological validity
in terms of bowling intensity and between-over
 1088 H. Crewe et al.

Table IV. Lumbar load differences for groups stratified by (a) shoulder counter-rotation, (b) minimum knee flexion angle, (c) ball release
speed.

(a) SCR

SCRlow SCRhigh

Mean s Mean s t P
−1
Peak force (Nm · kg ) AP 25.1 11.7 28.8 8.9 1.032 0.310
Peak force (Nm · kg−1) ML 20.1 5.9 25.3 8.7 1.964 0.059
Peak moment (Nm · kg−1 · m−1) FE 8.6 4.4 11.4 4.5 1.753 0.090
Peak moment (Nm · kg−1 · m−1) RO 9.3 2.6 11.2 2.9 2.077 0.046
Time to peak force (ms) V 32.0 15.9 23.9 8.3 1.804 0.081

(b) Minimum knee flexion angle

Kneeflexed Kneestraight
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014

Mean s Mean s t P

Peak force (Nm · kg−1) AP 20.2 9.9 30.2 9.0 2.999* 0.005
Peak force (Nm · kg−1) ML 19.5 6.3 24.4 8.6 1.841 0.076
Peak moment (Nm · kg−1 · m−1) FE 7.8 2.8 10.4 4.6 1.915 0.065
Peak moment (Nm · kg−1 · m−1) RO 8.1 1.6 11.3 2.9 3.879* 0.001
Time to peak force (ms) V 29.3 10.8 24.5 8.3 1.435 0.162

(c) Ball release speed

Speedlow Speedhigh

Mean s Mean s t P
−1
Peak force (Nm · kg ) AP 20.7 8.7 31.5 10.1 3.236* 0.003
Peak force (Nm · kg−1) ML 21.4 7.7 25.5 8.2 1.448 0.158
Peak moment (Nm · kg−1 · m−1) FE 9.1 4.3 11.4 4.4 1.485 0.148
Peak moment (Nm · kg−1 · m−1) RO 9.1 2.5 11.5 2.9 2.423 0.022
Time to peak force (ms) V 28.8 16.7 26.2 9.1 0.522 0.606

*P < 0.01.
AP: anterior-posterior; V: vertical; ML: medio-lateral; FE: flexion-extension; LF: lateral flexion; RO: rotation

activity, the competitive environment cannot be per-
fectly replicated in a laboratory.
Faster ball release speed and a more extended
front knee between front foot contact and ball
release were found to be associated with each
other, in agreement with the findings of several pre-
vious studies (Elliott, Foster, & Gray, 1986; Portus
et al., 2004; Loram et al., 2005). Flexion of the knee
plays a substantial role in attenuating impact forces
that are transmitted to the lumbar spine (Nigg, 1983;
Elliott, 2000) and this action is therefore desirable
with respect to reducing low-back injury risk.
However, increased knee flexion may also inhibit
energy transfer to the ball resulting in slower ball
release speeds, and may therefore be disadvanta-
geous to bowling performance (Portus et al., 2004).
Consequently, it is not surprising that in the current
study both of these factors were associated with
increased lumbar loads – in particular, the peak
lumbo-pelvic rotation moment and anterior-poster-
ior force were higher in participants who bowled
faster and with a more extended front knee.
Anterior shear forces (Adams, 2002) and compres-
sive load in combination with rotation (Chosa et al.,
2004) are known to load the pars interarticularis of
the vertebra, which may lead to the development of
spondylolysis. The specific lumbo-pelvic kinetic vari-
ables identified here in relation to bowling speed and
front knee kinematics may therefore explain why a
more flexed front knee has previously been asso-
ciated with reduced low-back injury incidence
(Portus et al., 2004) and provides some insight into
the mechanism through which bowling speed and
front knee kinematics may contribute to lumbar
injury risk.
Shoulder counter-rotation is the component of the
bowling action that has been most consistently asso-
ciated with lumbar injury in fast bowlers (Burnett
et al., 1996; Elliott et al., 1993; Foster et al., 1989;
Portus et al., 2004), yet only two of the 12 lumbar
variables selected to represent lumbar load were sig-
nificantly correlated with shoulder counter-rotation.
However, this is not an entirely unexpected finding,
as several authors have pointed out that as shoulder
counter-rotation is a global measurement of
shoulder alignment, it does not take into account
the relative alignment of the lower trunk or pelvis,
and the measure is completed prior to front foot

contact where maximum ground reaction forces and
lumbar load are experienced (Portus et al., 2004;
Ranson, Burnett, King, Patel, & O’Sullivan, 2008).
The current results suggest that shoulder counter-
rotation’s direct contribution to lumbar loading dur-
ing the front foot contact phase of bowling is perhaps
less substantial than that of front limb kinematics,
and supports the theory that large shoulder counter-
rotation may cause the bowler to adopt a position in
which the ability of the vertebrae to tolerate the
applied loads may be reduced (Burnett, Barrett,
Marshall, Elliott, & Day, 1998).
The limitations of modelling the lumbar spine as a
rigid segment, using surface markers placed on the
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014
skin, must be acknowledged. Also, the inertial seg-
ment parameters used were based on a small sample
of participants (Pearsall et al., 1996). However, these
parameters allowed us to match our kinematic model
with the inertial inputs for kinetic calculations, as
they were defined according to vertebral level.
Despite the limitations of this type of model, careful
interpretation of the results can assist researchers in
gaining further insight into the load placed on the
lumbar spine during cricket fast bowling.
Conclusion
Adolescent fast bowlers experience considerable
lumbo-pelvic loads during bowling, evidenced by par-
ticularly large shear forces and lateral flexion and
transverse plane rotational moments, when compared
with those experienced during running. Further, the
adolescent bowlers’ techniques remained consistent
and lumbar loads did not change significantly over
the course of an 8-over bowling spell. A straighter
front knee, faster ball release speed and increased
shoulder counter-rotation were related to aspects of
lumbo-pelvic loading – in particular peak transverse
plane rotation moments and anterior-posterior shear
forces. Future studies should investigate, firstly,
whether increased lumbar loads are associated with
increased lumbar injury incidence and, secondly,
whether these loads can be reduced through techni-
que modification without compromising ball release
speed, as it is an essential performance factor.
Alternatively, faster bowlers with a fully extended
front knee technique should be closely monitored
for other injury risk factors, such as bowling workload
and musculoskeletal preparation.
References
Adams, M. A. (2002). The biomechanics of back pain. Edinburgh:
Churchill Livingstone.
Annear, P., Chakera, T. M., Foster, D., & Hardcastle, P. (1992).
Pars interarticularis stress and disc degeneration in cricket's
potent strike force: The fast bowler. Australian and New
Zealand Journal of Surgery, 62, 768–773.
Lumbo-pelvic loading during fast bowling 1089
Bartlett, R. M., Stockill, N. P., Elliott, B. C., & Burnett, A. F.
(1996). The biomechanics of fast bowling in men's cricket: A
review. Journal of Sports Sciences, 14, 403–424.
Besier, T. F., Sturnieks, D. L., Alderson, J. A., & Lloyd, D. G.
(2003). Repeatability of gait data using a functional hip joint
centre and a mean helical knee axis. Journal of Biomechanics, 36,
1159–1168.
Burnett, A. F., Barrett, C. J., Marshall, R. N., Elliott, B. C., &
Day, R. E. (1998). Three-dimensional measurement of lumbar
spine kinematics for fast bowlers in cricket. Clinical
Biomechanics, 13, 574–583.
Burnett, A. F., Elliott, B. C., & Marshall, R. N. (1995). The effect
of a 12-over spell on fast bowling technique in cricket. Journal
of Sports Sciences, 13, 329–341.
Burnett, A. F., Khangure, M. S., Elliott, B. C., Foster, D. H.,
Marshall, R. N., & Hardcastle, P. H. (1996). Thoracolumbar
disc degeneration in young fast bowlers in cricket: A follow-up
study. Clinical Biomechanics, 11, 305–310.
Cavanagh, P. R., & Lafortune, M. A. (1980). Ground reaction
forces in distance running. Journal of Biomechanics, 13, 397–
406.
Chosa, E., Totoribe, K., & Tajima, N. (2004). A biomechanical
study of lumbar spondylolysis based on a three-dimensional
finite element method. Journal of Orthopaedic Research, 22,
158–163.
Crewe, H., Campbell, A., Elliot, B., & Alderson, J. (in press).
Kinetic sensitivity of a new lumbo-pelvic model to variation in
segment parameter input. Journal of Applied Biomechanics.
de Leva, P. (1996). Adjustments to Zatsiorky-Seluyanov's segment
inertia parameters. Journal of Biomechanics, 29, 1223–1230.
Elliott, B. C. (2000). Back injuries and the fast bowler in cricket.
Journal of Sports Sciences, 18, 983–991.
Elliott, B. C., Davis, J. W., Khangure, M. S., Hardcastle, P., &
Foster, D. (1993). Disc degeneration and the young fast bowler
in cricket. Clinical Biomechanics, 8, 227–234.
Elliott, B., Foster, D., & Gray, S. (1986). Biomechanical and
physical factors influencing fast bowling. Australian Journal of
Science and Medicine in Sport, 18, 16–21.
Elliott, B., Hardcastle, P., Burnett, A. F., & Foster, D. (1992).
The influence of fast bowling and physical factors in radiologic
features in high performance young fast bowlers. Sports
Medicine Training and Rehabilitation, 3, 113–130.
Ferdinands, R. E. D., Kersting, U., & Marshall, R. N. (2009).
Three-dimensional lumbar segment kinetics of fast bowling in
cricket. Journal of Biomechanics, 42, 1616–1621.
Foster, D., John, D., Elliott, B., Ackland, T., & Fitch, K. (1989).
Back injuries to fast bowlers in cricket: A prospective study.
British Journal of Sports Medicine, 23, 150–154.
Hopkins, W. G. (2002). A new view of statistics. Retrieved from
http://www.sportsci.org/resource/stats/index.html
Loram, L. C., McKinon, W., Wormgoor, S., Rogers, G. G.,
Nowak, I., & Harden, L. M. (2005). Determinants of ball
release speed in schoolboy fast-medium bowlers in cricket.
Journal of Sports Medicine and Physical Fitness, 45,
483–490.
Marras, W. S., & Granata, K. P. (1997). Spine loading during
trunk lateral bending motions. Journal of Biomechanics, 30,
697–703.
Natarajan, R. N., Lavender, S. A., An, H. A., & Andersson, G. B.
J. (2008). Biomechanical response of a lumbar intervertebral
disc to manual lifting activities: A poroelastic finite element
model. Spine, 33, 1958–1965.
Nigg, B. (1983). The load on the lower extremity in selected
sports activities. In G.Wood (ed.), Collected papers on sports
biomechanics (pp. 62–72). Nedlands, WA: University of
Western Australia.
Pearsall, D. J., Reid, J. G., & Livingston, L. A. (1996). Segmental
inertial parameters of the human trunk as determined from
 1090 H. Crewe et al.
computed tomography. Annals of Biomedical Engineering, 24,
198–210.
Petersen, C., Pyne, D., Dawson, B., Portus, M., & Kellett, A.
(2010). Movement patterns in cricket vary by both position and
game format. Journal of Sports Sciences, 28, 45–52.
Portus, M. R., Mason, B. R., Elliott, B. C., Pfitzner, M. C., &
Done, R. P. (2004). Technique factors related to ball release
speed and trunk injuries in high performance cricket fast
bowlers. Sports Biomechanics, 3, 263–284.
Portus, M. R., Sinclair, P. J., Burke, S. T., Moore, D. J. A., &
Farhart, P. J. (2000). Cricket fast bowling performance and
technique and the influence of selected physical factors during
an 8-over spell. Journal of Sports Sciences, 18, 999–1011.
Ranson, C. A., Burnett, A. F., King, M., Patel, N., & O'Sullivan,
P. B. (2008). The relationship between bowling action classifi-
cation and three-dimensional lower trunk motion in fast bow-
lers in cricket. Journal of Sports Sciences, 26, 267–276.
Downloaded by [Flinders University of South Australia] at 01:34 04 October 2014
Ranson, C. A., Kerslake, R. W., Burnett, A. F., Batt, M. E., &
Abdi, S. (2005). Magnetic resonance imaging of the lumbar
spine in asymptomatic professional fast bowlers in cricket.
Journal of Bone and Joint Surgery [Br], 87, 1111–1116.
Seay, J., Selbie, W. S., & Hamill, J. (2008). In vivo lumbo-sacral
forces and moments during constant speed running at different
stride lengths. Journal of Sports Sciences, 26, 1519–1529.
Stretch, R. A. (1995). The seasonal incidence and nature of
injuries in schoolboy cricketers. South African Medical Journal,
85, 1182–1184.
Whiting, W. C., & Zernicke, R. F. (2008). Biomechanics of muscu-
loskeletal injury. Champaign, IL: Human Kinetics.
Winter, D. A. (1990). Biomechanics and motor control of human
movement. New York, NY: John Wiley & Sons.
Wu, G., & Cavanagh, P. R. (1995). Recommendations for stan-
dardization of the reporting of kinematic data. Journal of
Biomechanics, 28, 1257–1261.