Article Text

This article has a correction. Please see:

Download PDFPDF

MRI study of the size, symmetry and function of the trunk muscles among elite cricketers with and without low back pain
  1. J Hides1,2,
  2. W Stanton1,2,
  3. M Freke5,
  4. S Wilson3,
  5. S McMahon4,
  6. C Richardson1
  1. 1
    Division of Physiotherapy, School of Health and Rehabilitation Sciences, University of Queensland, Brisbane, Australia
  2. 2
    UQ/Mater Back Stability Clinic, Mater Health Services, South Brisbane, Queensland, Australia
  3. 3
    School of IT and Electrical Engineering, University of Queensland, Brisbane, Australia
  4. 4
    Commonwealth Bank Centre of Excellence, Brisbane, Queensland, Australia
  5. 5
    Physiotherapy Department, Second Health Support Battalion, Gallipoli Barracks, Enoggera, Brisbane, Queensland, Australia
  1. Dr J Hides, Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia, 4072; j.hides{at}shrs.uq.edu.au

Abstract

Objectives: To determine if asymmetry of trunk muscles and deficits of motor control exist among elite cricketers with and without low back pain (LBP).

Design: Single-blinded observational quasi-experimental design study

Setting: Assessments were conducted in a hospital setting.

Participants: Among a total eligible sample of 26 male elite cricketers (mean age 21.2 (SD 2.0) years), selected to attend a national training camp, 21 participated in the study.

Risk factors: The independent variables were ‘group’ (LBP or asymptomatic) and ‘cricket position’ (fast bowler versus the rest of the squad).

Main outcome measurements: The dependent variables were the cross-sectional areas (CSA) of the quadratus lumborum (QL), lumbar erector spinae plus multifidus (LES + M) and psoas muscles, the thickness of the internal oblique (IO) and transversus abdominis (TrA) muscles, and the amount of lateral slide of the anterior abdominal fascia.

Results: The QL and LES + M muscles were larger ipsilateral to the dominant arm. In the subgroup of fast bowlers with LBP, the asymmetry in the QL muscle was the greatest. The IO muscle was larger on the side contralateral to the dominant arm. No difference between sides was found for the psoas and TrA muscles. Cricketers with LBP showed a reduced ability to draw in the abdominal wall and contract the TrA muscle independently of the other abdominal muscles.

Conclusions: This study provides new insights into trunk muscle size and function in elite cricketers, and evidence of impaired motor control in elite cricketers with LBP. Rehabilitation using a motor control approach has been shown to be effective for subjects with LBP, and this may also benefit elite cricketers.

Statistics from Altmetric.com

Cricket is a sport associated with low injury rates in comparison with contact sports. Recent reports13 showed that injury rates for elite spin bowlers, wicket keepers and batsmen were among the lowest for any sport in Australia. However, fast bowlers who bowled over sustained periods were shown to incur injuries at a rate similar to Australia’s football codes, with a higher risk of injury than for any other type of player in elite cricket.13 Injuries to the lower back are common among cricket fast bowlers.4 Key factors identified as possible contributors to low back pain (LBP) include bowling technique,5 increased bowling workload,6 and other risk factors such as playing surface, footwear, age and physical preparation.7 Although the biomechanical analysis of technique2 and issues related to bowler workloads have received much focus,1 6 8 the importance of player preparation, and in particular lumbopelvic stability, has received little attention.

Many sports are asymmetrical in nature, and asymmetry has been thought possibly to be related to injuries. Owing to the proposed undesirable consequences of asymmetry, several coaching and training sources encourage players in sports involving kicking such as Australian Rules9 10 and soccer1113 to practice using both legs during training. The rationale for this practice is to minimise potential asymmetrical forces acting on joints, reduce muscle imbalances and decrease the workload of the dominant leg, which otherwise may eventually lead to overuse injuries.14 One trunk muscle that has been studied in cricketers is the quadratus lumborum (QL) muscle. Engstrom et al15 16 and Walker et al17 used MRI to reveal hypertrophy of the QL muscle volume ipsilateral to the bowling arm in fast bowlers. They hypothesised that increased muscle development created an increased mechanical load on the neural arch resulting in contralateral bone stress injuries. A mathematical model was developed to test this theory by Visser et al,18 which was used to estimate forces and moments delivered by the QL muscle on the L3 and L4 vertebrae during the bowling action. In contrast to earlier studies, which had predicted a relationship between hypertrophy of the QL muscle and development of injuries,1517 the model predicted that asymmetry of the QL muscle may help to reduce bone stresses.18 Although it has not been shown that development of muscle asymmetry relates to injury, it is nevertheless important to determine if such asymmetries actually do exist in relation to specific sports.

As cricket is an asymmetrical sport, it is likely that asymmetries will develop in trunk muscles other than the QL. An MRI study found that asymmetry of the psoas muscle in fast bowlers was significantly greater than that in control subjects.19 Although increased loading and hypertrophy of muscle may be associated with sporting activities, results must be interpreted carefully as pain and muscle inhibition can lead to atrophy of muscles. Asymmetry of the psoas muscle has also been observed in subjects presenting with unilateral LBP with decreased cross-sectional area (CSA) on the affected side.20 Thus athletes with LBP may present with competing influences of muscle hypertrophy due to increased activity levels and muscle atrophy due to pain and muscle inhibition. Psoas muscle size in fast bowlers with LBP has not been examined. Asymmetry of the combined volume of the lumbar erector spinae and multifidus muscles (LES+M) was not found in fast bowlers with asymmetry of the volume of the QL muscles,16 although it is widely established that changes in the multifidus muscle occur in subjects with acute/subacute and chronic LBP.2022 Real-time ultrasound imaging studies conducted on healthy, asymptomatic subjects have shown the external oblique (EO), internal oblique (IO) and rectus abdominis (RA) muscles to be symmetrical between sides23 and there was shown to be no relationship between hand dominance and asymmetry of the transversus abdominis muscle (TrA).24

Previous research has highlighted the important role played by the TrA muscle in support, stability and protection of the lumbar spine. It has been proposed that the TrA muscle may contribute to stability of the lumbopelvic region via its effects on intraabdominal pressure and by affecting fascial tension.2528 Clinical muscle testing of the TrA muscle has mainly involved observation of the abdominal wall during a cognitive “drawing in” of the abdominal wall.29 The action of the bilateral muscle bellies of the TrA muscle has been viewed during this manoeuvre using ultrasound and MRI.30 31 In a recent study conducted on elite asymptomatic cricketers,31 the muscle bellies of the TrA muscle were seen to thicken as well as shorten in length during this manoeuvre, to give the appearance of a deep muscle ‘corset’, in line with anatomical studies.26 32 The anterior abdominal fascia (AAF) was observed to slide in a lateral direction and the cross-sectional area of the trunk was found to decrease with an abdominal drawing-in manoeuvre.31 Differences in ability to perform the drawing-in manoeuvre have been documented in subjects with LBP using MRI.33 Subjects with LBP were shown to be less able to decrease the CSA of their trunk and there was less slide of the AAF when they drew in their abdominal walls. Randomised clinical trials which have focused on re-educating TrA muscle function through its “draw-in” action have been successful in decreasing lumbo-pelvic pain and disability.3436 Animal studies have provided evidence that the TrA muscle is only effective in providing stability to the lumbar spine when its action is symmetrical between sides, so assessment of symmetry of action during the muscle test for the TrA may also be of importance.27

Cricket is by nature an asymmetrical sport, whether participants are batting, bowling or throwing. It is therefore more than likely that trunk muscle asymmetry exists in elite cricketers. As patients with LBP have been shown to have deficits in motor control of the deep abdominal muscle, it is possible that these same deficits may occur in cricketers with LBP. Information of this kind could directly influence rehabilitation and training programmes for cricketers. This study aimed to investigate, using Magnetic Resonance Imaging (MRI), the size and symmetry of the quadratus lumborum (QL), psoas, lumbar erector spinae plus multifidus (LES+M), internal oblique (IO) and transversus abdominis (TrA) muscles and the ability to independently contract the TrA muscle in elite cricketers with and without LBP.

METHODS

The study was approved by the ethics committees of the relevant institutions that hosted the study. All participants gave informed consent and the rights of subjects were protected.

Subjects

In total 26 young male elite cricket players selected to attend a national training camp were eligible for enrolment in the study. The sample mean (SD) age, height and weight was 21.2 (2.0) years, 183.2 (6.0) cm and 85.5 (6.6) kg, respectively. Subjects who reported a history of LBP, and indicated that it was sufficiently severe to interfere with current sporting or training performance, were allocated to the LBP group (10 subjects). In all cases, the pain was unilateral in distribution, and all subjects reported previous episodes of LBP.

Procedures

MRI and questionnaire assessments were completed in a hospital setting before the start of the programme. Subjects were asked if they were currently receiving any medical or physiotherapy treatment for musculoskeletal conditions. A research assistant helped subjects to complete pain drawings (location of pain) on a body chart. Pain intensity of LBP at the time of testing was established using a visual analogue scale (VAS) marked from 1 to 10. An activity level score was obtained for each participant using the Habitual Activity Questionnaire (HAQ).37 Subjects nominated their hand dominance.

MRI assessment

Subjects were first screened for contraindications to MRI by a medical practitioner, then they were positioned in the supine position lying with their hips and knees resting on a foam wedge. Although lying in the scanner, subjects were instructed about the muscle contraction required for the muscle test for the TrA muscle. Subjects were asked to draw in the abdominal wall without moving the spine. The CSA of the deep musculofascial system was measured first at rest and then in the contracted state. A Siemens Sonata MR system (1.5 Tesla) was used. A true fast imaging with steady-state precession (TrueFISP) sequence using 14×7 mm contiguous slices centred on the L3–4 discs was employed for static images. A cine sequence of one 7 mm slice was taken during a contraction at a frame rate of 500 ms for 7 seconds.

Rest images were performed with the subject holding the breath at mid-expiration. Contraction images were commenced after vocal initiation of the contraction by the operator. The cine series was commenced before the initiation of a contraction to ensure complete temporal coverage of the drawing-in manoeuvre.

Measurements

MRIs were saved onto compact disks for later analysis using a measurement software package on a laptop computer. Measurements on MRI included CSA of the QL, LES+M and psoas muscles, thickness of the TrA and IO muscles, slide of the transversus abdominis fascia (left and right), CSA of the trunk at rest and CSA of the trunk on contraction (figs 13): . Repeatability and reliability of linear and CSA measurements obtained using MRI have been previously reported for the trunk muscles.31 38

Figure 1 MRI of the trunk at the level of the L3–4 disc showing cross-sectional area measurements of the psoas, quadratus lumborum and multifidus + lumbar erector spinae muscles.
Figure 2 MRI of the trunk at the level of the L3–L4 disc (at rest) showing measurements of cross-sectional area of the trunk and thickness of the transversus abdominis and internal oblique muscles.
Figure 3 MRI of the trunk at the level of the L3–L4 disc (after drawing in the abdominal wall) showing measurements of cross-sectional area of the trunk, thickness of the transversus abdominis and internal oblique muscles, and slide of the AAF.

Final group

Of the 26 cricketers who attended the training camp, 22 were available for MRI assessment. Of the four who could not attend, two were from the LBP group. One participant was excluded due a musculoskeletal condition (osteitis pubis) that confounded allocation to either group, leaving 21 cases for analysis (8 in the LBP group and 13 in the non-LBP group).

Statistical analysis

Repeated measures analysis of covariance (ANCOVA) with a type I sums of squares model was conducted to examine the asymmetry of muscle size. A type I model was used because our previous research had shown that higher-order interactions involving weight and muscle size is problematic in a type III model. The dependent variables for the analyses were the CSA of the QL, LES+M and PS muscles, thickness of the IO and TrA muscles in a relaxed condition, CSA of the trunk at rest and on contraction, and amount of slide of the AAF. The data were analysed separately for each measure. The covariates in the analyses were age, height and weight. The repeated measure was asymmetry (ipsilateral or contralateral side of the abdomen relative to dominant hand), except in the analysis of trunk contraction, where the repeated measure was contraction (relaxed and contracted condition). The independent factors were group (LBP or asymptomatic), and cricket position (fast bowler versus the rest of the squad). This factor was included because fast bowlers have previously been shown to develop asymmetry in muscles such as the quadratus lumborum related to their side of bowling. There were four fast bowlers in the healthy group and five in the LBP group.

RESULTS

There was no statistically significant difference (p>0.05) in the ages, heights, weights or activity level (HAQ) scores of those who were in the LBP group versus the remainder. The mean pain VAS score of participants in the LBP group was 5.75 (SD 2.46).

Results for each muscle are shown in table 1 for the elite cricketers with and without LBP, and for the abdominal side contralateral and ipsilateral to their dominant hand. Table 2 shows the amount of trunk contraction achieved by those with or without LBP.

Table 1 Results for muscle cross-sectional areas and thickness contralateral and ipsilateral to dominant hand for elite cricketers with or without low back pain
Table 2 Results for cross-sectional area of the trunk (relaxed and contracted) for elite cricketers with or without low back pain (LBP)

For the CSA of the QL, LES+M and psoas muscles, results of ANCOVA showed that the elite cricketers had larger CSAs on the side ipsilateral to hand dominance for QL (mean difference of 0.63 cm, 7.3%; F = 16.9, p = 0.001) and LES+M (mean difference of 1.46 cm, 5.3%; F = 8.2, p = 0.013), but there was no significant effect for Psoas (F = 0.004, p = 0.95). In addition, there was a significant three-way interaction effect of group, side and role so that fast bowlers with LBP had greater asymmetry in QL muscle size (24.9%) than fast bowlers without LBP (3.0%) and the rest of the squad with LBP (7.3%) or without LBP (8.7%) (F = 14.7, p = 0.002).

In regard to the thickness of the IO and TrA muscles, the results showed that cricketers had greater thickness of the IO muscle on the side contralateral to hand dominance (mean difference of .15 cm,9.8%; F = 5.4, p = 0.036), but no significant effect for the TrA muscle (p>0.05). A result that approached statistical significance for this relatively small sample indicated a trend for cricketers with LBP to have greater thickness of the IO muscle on the side contralateral to the dominant hand (15.3%) than those without LBP (4.7%) (p = 0.09).

Results of the analysis of slide of the AAF indicated that those cricketers with LBP could not perform the muscle test for the TrA muscle (slide of the AAF 0.54 cm) as well as those without LBP (slide of the AAF 1.55 cm) (F = 34.8, p<0.001). In the case of measurement of the CSA of the trunk (table 2), results showed that cricketers with LBP could not decrease the CSA of the trunk on contraction (1.8% relaxed to contracted condition) as well as those without LBP (8.1%) (F = 23.1, p<0.001).

DISCUSSION

Although previous studies have reported asymmetry of the QL muscle in fast bowlers,1517 this study has shown (1) that asymmetry may be evident in all cricketers, and (2) that among fast bowlers this is related to the presence of LBP. Fast bowlers with LBP had the greatest asymmetry, whereas fast bowlers without LBP had no greater evidence of asymmetry of the QL muscle than did other cricketers (not fast bowlers) in the squad. Although our sample sizes are small, this result supports the findings of Engstrom et al,16 who showed a strong association between asymmetry of the QL muscle and the development of symptomatic unilateral L4 pars lesions. Previous studies of fast bowlers have proposed that a hypertrophied QL muscle could provide a destructive influence on adjacent bony structures.1517 However, a recent mathematical model has predicted that the forces exerted by an enlarged QL muscle would have a negligible effect on the bone in times of high stress.18 Our results do not indicate whether or not asymmetry of QL is problematic, but highlights the importance of collecting data from the whole squad (not just the fast bowlers) and collecting data related to the presence of pain and injury in athletes when conducting morphological investigations.

Asymmetry was also found in the IO muscle, which was larger on the side contralateral to the dominant hand. Investigations of side strain in cricket bowlers using MRI found that the injury consistently occurred on the non-bowling side and tended to affect abdominal muscles (IO, external oblique and TrA muscles) rather than the QL muscle.39 This could suggest that the asymmetrical muscular demands of the repetitive bowling action creates hypertrophy in the torque-producing muscles of trunk rotation and side flexion.40 However, the TrA muscle remained symmetrical in size. This finding is consistent with research that has shown that optimum stabilisation of the trunk requires bilateral activity of TrA.27 Despite symmetry of the thickness of the TrA muscle, a deficit in motor control was noted in the cricketers with LBP, as shown by a reduced ability to decrease the CSA of the trunk and a reduced slide of the AAF during the muscle test for the TrA muscle. In addition, there was a trend towards the cricketers with LBP having thicker IO muscles. This finding may be explained by subjects with LBP compensating for decreased lumbopelvic stability (in light of the findings for the TrA muscle). Similar strategies have been noted in non-sporting populations of subjects with LBP.41 42 Programmes of therapeutic exercise to improve the motor control of the TrA muscle have resulted in reduction in severity and recurrence of LBP symptoms in subjects with chronic LBP.34 35 36 Although bowler workloads1 6 8 and technique2 43 44 must be considered, re-education of TrA muscle function in fast bowlers may help to negate the large forces induced on the spine when bowling,2 and may act to reduce the incidence and severity of LBP in that population.

Other MRI studies measuring the psoas muscle have shown asymmetry in this muscle in fast bowlers19 and in subjects with unilateral LBP.20 In contrast, our study did not reveal significant psoas asymmetry in the cricketers examined. However, there was evidence of asymmetry of the LES+M muscles, which were larger on the dominant side. This finding may reflect muscle hypertrophy related to the action of throwing or bowling in cricket, and contrasts with the findings of Engstrom et al,16 who did not find asymmetry of these muscles between sides. However, the methods of the two studies differed, as Engstrom et al16 measured muscle volumes (compared with measurement of muscle CSAs at the mid lumbar region in the current study). The location of the slice used in the current study was also not in an optimum position to detect specific multifidus changes associated with LBP, which are known to most commonly occur at the lowest lumbar vertebral levels.45

What is already known on this topic

  • Injuries to the low back are common among cricket fast bowlers.

  • Cricket is by nature an asymmetrical sport, and in some sports, there has been a proposed link between development of muscle asymmetry and injuries. This has led to training programmes aiming to minimise the development of asymmetry.

  • Hypertrophy of the quadratus lumborum muscle has been shown ipsilateral to the bowling arm in fast bowlers.

  • Previous research has highlighted the important role of the transversus abdominis muscle in stabilisation of the lumbopelvic region; however, motor control of this muscle has not been examined in elite cricketers.

What this study adds

  • Muscle asymmetry was found in several trunk muscles in elite cricketers.

  • The quadratus lumborum muscle was asymmetrical in all cricketers, not only fast bowlers.

  • Fast bowlers with low back pain had the greatest asymmetry of the quadratus lumborum muscle, whereas fast bowlers without low back pain showed no greater asymmetry than did other cricketers.

  • The internal oblique muscle was larger on the side contralateral to the dominant hand, consistent with reports of side strain in cricket bowlers.

  • The lumbar erector spinae and multifidus muscles were larger ipsilateral to the dominant side.

  • An impairment in the motor control of the transversus abdominis muscle was found in cricketers with low back pain.

A limitation of this study was the small sample size, though it is comparable with other studies conducted on elite athletes. Furthermore, for the lumbar muscles, the multifidus muscle was measured in combination with LES at the level measured. We did not examine the muscles of the lower lumbar spine, and future studies could investigate for specific atrophy of the multifidus muscle related to the presence of LBP.

CONCLUSION

Although we are unable to comment on cause and effect in relation to trunk muscle asymmetries and LBP, the results of this study confirm that muscle asymmetries exist in elite cricketers. Future studies may be able to clarify the complex relationships between pain and pathology and the specific muscle adaptations that occur in individual sports. This may provide a clearer way forward for implementation of optimum training techniques, injury prevention and rehabilitation of sporting injuries.

Acknowledgments

We thank the members of the cricket squad, the UQ/Wesley MRI unit, K McMahon, M Bryant and T Johnsen (scientific officer).

REFERENCES

Footnotes

  • Funding: This study was funded by the Cricket Australia Sports Science Medicine Research Program.

  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.