Clinical Radiology (2008) 63, 681e687

Fatty inltrate in the cervical extensor muscles is not a feature of chronic, insidious-onset neck pain
J. Elliotta,b,c,*, M. Sterlinga, J.T. Noteboomb, R. Darnella, G. Gallowayc, G. Julla
a

Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia, bDepartment of Physical Therapy, Rueckert-Hartman School for Health Professions, Regis University, Denver, Colorado, USA, and cCentre for Magnetic Resonance, The University of Queensland, Brisbane, Australia

Received 24 July 2007; received in revised form 23 October 2007; accepted 7 November 2007

AIM: To investigate the presence of fatty inltrate in the cervical extensor musculature in patients with insidiousonset neck pain to better understand the possible pathophysiology underlying such changes in chronic whiplash-associated disorders (WAD). MATERIALS AND METHODS: A sample of convenience of 23 women with persistent insidious-onset neck pain (mean age 29.2 6.9 years) was recruited for the study. Magnetic resonance imaging (MRI) was used to quantify fatty inltration in the cervical extensor musculature. Quantitative Sensory Testing (QST; pressure and thermal pain thresholds) was performed as sensory features are present in chronic whiplash. Self-reported pain and disability, as well as psychological distress, were measured using the Neck Disability Index (NDI) and the General Health Questionnaire-28 (GHQ-28), respectively. RESULTS: Measures were compared with those of a previous dataset of chronic whiplash patients (n 79, mean age 29.7 7.8 years). Using a classication tree, insidious-onset neck pain was clearly identied from whiplash (p < 0.001), based on the presence of MRI fatty inltrate in the cervical extensor musculature (0/102 individuals) and altered temperature thresholds (cold; 3/102 individuals). CONCLUSION: Fatty inltrates in the cervical extensor musculature and widespread hyperalgesia were not features of the insidious-onset neck pain group in this study; whereas these features have been identied in patients with chronic WAD. This novel nding may enable a better understanding of the underlying pathophysiological processes in patients with chronic whiplash. 2007 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Introduction
Chronic whiplash-associated disorder (WAD) is a complex syndrome involving changes in the sensory, motor, and psychological systems.1e3 Recently, alterations (fatty inltration) were demonstrated in the cervical extensor musculature of patients with chronic WAD using magnetic resonance imaging
* Guarantor and correspondent: J.M. Elliott, Regis University, 3333 Regis Blvd, G-4, Rueckert-Hartman School for Health Professions, Department of Physical Therapy, Denver, Colorado 80221-1099, USA. Tel.: 1 303 964 5484; fax: 1 303 964 5474. E-mail address: jelltt@regis.edu (J. Elliott).

(MRI).4 The fatty inltration was widespread and present bilaterally in all cervical extensor muscles at all segments, albeit there was signicantly more fatty inltrate in the sub-occipital and multidus muscles. The cause of this fatty inltrate could not be determined, but it was not related to age, body mass index, self-reports of pain and disability, or symptom duration. It was suggested that the fatty inltrate might be a product of either general disuse or a minor nerve injury, irritated, and subsequently, demyelinated nerve tissue resulting from an acute inammatory process or a combination of these factors.5 General disuse would seem the most reasonable explanation given the widespread

0009-9260/$ - see front matter 2007 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.crad.2007.11.011

682

J. Elliott et al.

fatty inltrate in the extensor musculature of these WAD patients. If general disuse is the cause, then it might be expected that patients with chronic neck pain of an insidious-onset would also, to some extent, demonstrate these ndings. To test this hypothesis, we undertook this preliminary study of patients with chronic insidious-onset neck pain and measured the extent of fatty inltration in the cervical extensor muscles. In addition, quantitative sensory measures of thermal and pressure pain thresholds, as well as a measure of psychological stress, were obtained to better understand the similarities and differences in presentation between these patients and those with chronic WAD. This enabled us to explore any associations between these features and fatty inltrate in muscles. Alterations in the sensory measures can indicate the presence of abnormal central pain processing,2,3 and the presence of fatty inltrate in muscle may be associated with minor and/or major nerve injury.6

content can result in altered signals on T1 or T2weighted sequences. T1-weighted images are very sensitive to the presence of fatty deposition in muscle7 and as a result, were chosen to quantify fatty inltration in the cervical extensor musculature. The measure of relative fat within the muscle was created as described previously8 by developing a pixel intensity prole using MRIcro software (www.mricro.com). The measure was the ratio between the pixel intensities of each muscle to that of a standardized region of intermuscular fat, thus allowing comparisons between individuals.4,9 For the purpose of this study, a fat score was determined for each cervical muscle by calculating the mean value across segments and sides (C3eC7). Rectus capitis posterior minor was measured at the C0eC1 level and rectus capitis posterior major was measured at the C2 level. The fat score for the sub-occipital musculature was also determined by calculating the mean value for the two muscles on each side, to provide a single fat score for these muscles.

Materials and methods

A sample of convenience encompassing 23 women (mean age 29.2 6.9 years, range of 18e45 years, mean duration of pain 33.7 20.6 months) suffering from persistent, insidious-onset neck pain was used in the study age to parallel the whiplash group. Volunteers with persistent, insidious-onset neck pain were recruited through advertising in local physiotherapy practices and within the local university fraternity. Participants were included if they suffered from insidious-onset neck pain that had persisted for 3 months or longer. Volunteers were excluded if the onset of neck pain was related to a motor vehicle crash or any other incident of trauma or if they had been diagnosed with any central or peripheral nervous system disorder; were either pregnant or thought to be pregnant; reported being claustrophobic; or did not meet institutional criteria to undergo MRI. The parallel data of the 79 WAD subjects (mean age 29.7 7.8 years, mean duration of pain 20.3 9.6 months) from a previous study were used for comparative purposes.4 This project was granted approval by Institutional Medical Research Ethical Committees. All participants provided written informed consent before inclusion in the study.

Thermal pain thresholds

Hot and cold thermal pain thresholds were measured over the cervical spine using Quantitative Sensory Testing (Medoc, Israel e www.medocweb.com). A Peltier thermode was applied directly to the glabrous skin over the posterior cervical spine and held in place manually. The thermode was preset at a baseline temperature of 32  C with the rate of temperature change being 1  C/s. To measure cold pain thresholds (CPT) and heat pain thresholds (HPT), participants were asked to push a patient-controlled button when the cold or warm sensation rst became painful.10 Triplicate recordings were obtained for each site and the mean value was used for analyses.

Pressure pain thresholds (PPT)

PPT were measured using a pressure algometer (Somedic AB, Farsta, Sweden e www.somedic.com) at two cervical spine sites bilaterally (over the spinous processes of C2 and C5) and at a remote site bilaterally (the muscle belly of the tibialis anterior). The measures for the C2 and C5 sites were combined and the mean value was used as a measure for PPT_local thresholds. The measures over the left and right tibialis anterior muscles were combined and the mean value was used as a measure for PPT_remote thresholds. These sites have been previously used in investigations of both idiopathic 11 and whiplash-induced neck pain.12 Participants were requested to push a button when the

MRI analysis
The principle sources of a MRI signal are fat and softaqueous tissue (e.g. skeletal muscle) and any abnormalities causing a change in fat or water

Fatty inltrate in the cervical extensor muscles

683

sensation changed from one of pressure to one of pressure and pain.13 Triplicate recordings were obtained at each site and the mean value for the right and left sides was used for analyses.

variables measured initially, followed by a second analysis in which fatty inltration scores were removed. The R implementation of Thernau and Atkinson18 was used in analysis.

Self-reported pain and disability

Results
Self-reported pain and disability was assessed using the Neck Disability Index (NDI), which has been shown to be reliable and valid.14,15 A higher NDI score (out of 100) indicates greater pain and disability. Table 1 presents the demographic characteristics and the results of the questionnaire on general distress (GHQ-28) for the two groups. There were no signicant group differences for age or BMI (p 0.77 and 0.17, respectively). Participants with chronic, insidious-onset neck pain had significantly lower NDI scores (p < 0.001), but neck pain of signicantly longer duration (p < 0.001) than those with chronic WAD. The groups also differed with respect to GHQ-28 scores with the insidiousonset neck pain participants reporting signicantly lower distress than the group with chronic WAD (p < 0.001).

Psychological questionnaire
The General Health Questionnaire-28 (GHQ-28) is a 28-item measure of emotional distress in a medical setting. The total score can be used as a measure of psychological distress. The GHQ-28 has been used in previous research of WAD.16

Statistical analysis MRI analysis

One-way analysis of variance (ANOVA) was used to investigate any difference between the measures of total average fatty inltrate in the sub-occipital and cervical extensor muscles between the insidious-onset and WAD groups, as well as measures of self-reported pain and disability (NDI), psychological stress (GHQ-28), and of the quantitative sensory tests (thermal and pressure). Analyses of covariance (ANCOVAs) were also performed with the total fat scores (TOTAL_UPPER and TOTAL_FAT) as the dependent variable and NDI as the covariate into the nal between groups analysis. Lastly, an analysis was undertaken to determine whether insidious-onset neck pain could be discriminated from WAD on the basis of average fat (fat indices averaged by all muscles), sensory tests, and scores on the NDI and GHQ-28. In the rst instance, a forward-step regression model was used; however, logistic regression failed, due to the distinct differences in fat measures between the insidiousonset neck pain and WAD groups. Therefore, a classication tree approach17 was taken using all There was a difference in the fat indices of all muscles with WAD participants demonstrating signicantly higher amounts of total fatty inltration when compared with the insidious-onset neck pain participants (p < 0.001). Table 2 presents the total average fat index scores across the cervical extensor muscles for the participants with insidiousonset neck pain and chronic WAD. Fig. 1 illustrates the mean (SD) fat index scores for the cervical extensor musculature between the two groups. Fig. 2 illustrates the outlined region of interest for the multidus muscle at the C6 segmental level on axial MRI for a participant with insidious-onset neck pain and chronic WAD.

Thermal pain thresholds

There was a signicant difference between the cold (CPT) and hot (HPT) pain thresholds between the two groups with participants with chronic WAD demonstrating reduced CPT (p < 0.001) and HPT

Table 1 Group

Group demographics and results of the General Health Questionnaire-28 (GHQ-28) Age (years) 29.2 (6.9) 29.7 (7.8) Body mass index (kg/m2) 23.3 (4.9) 25.1 (5.7) Neck Disability Indexa 21.9 (7.5) 45.5 (15.9) Duration (months)a 33.7 (20.6) 20.3 (9.6) GHQ-28-totala 15.6 (7.5) 30.5 (12.8)

Insidious-onset neck pain (n 23) Whiplash-associated disorders (n 79) Data are mean (SD). a p < 0.001.

684

J. Elliott et al.

Table 2 Total average fat indices across cervical extensor muscle for the participants with insidious-onset neck pain and whiplash-associated disorders (all signicant at p < 0.001) Suboccipitals Insidious-onset neck pain Whiplash-associated disorders 0.24 (0.02) 0.23e0.25 0.42 (0.07) 0.4e0.43 Multidus 0.17 (0.02) 0.16e0.18 0.33 (0.05) 0.32e0.34 Semispinalis cervicis 0.16 (0.02) 0.15e0.17 0.29 (0.04) 0.28e0.3 Semispinalis capitis 0.16 (0.02) 0.15e0.17 0.28 (0.04) 0.27e0.29 Splenius capitis 0.17 (0.02) 0.16e0.18 0.28 (0.04) 0.27e0.029 Upper trapezius 0.18 (0.02) 0.17e0.19 0.22 (0.04) 0.22e0.23

Data are reported as mean (SD) and 95% condence intervals.

thresholds (p 0.001) when compared with participants with chronic, insidious-onset neck pain (Table 3).

Classication tree
The classication tree analysis included the following discriminating variables: GHQ-28, NDI, duration, CPT, HPT, PPT_local, PPT_remote, and average cervical fat index score. The analysis revealed that the strongest features distinguishing the two groups were (1) the fat index scores and (2) CPT. In the rst instance, the classication tree yielded a 0% misclassication rate based on fatty inltrates; indicating that the insidious-onset neck pain group was clearly distinguished from the chronic WAD group. The cut-off value for average fat index for insidious-onset neck pain was <0.24. The fatty inltrate data were then removed, and with the remaining data, the classication tree yielded a 4% misclassication rate based on CPT. Only three participants with insidious-onset neck pain demonstrated similar CPT to those with chronic WAD. Specically, a CPT of more than 12.7  C would indicate membership of the WAD group. Only three participants with insidious-onset neck pain demonstrated a CPT of greater than 12.7  C. The lowest CPT value for WAD was 13.2  C and the highest CPT value for insidious-onset neck pain was 19.5  C. None of the other variables improved the classication.

Pressure pain thresholds

There was a signicant difference in both the PPT_local and PPT_remote pain thresholds between the two groups with participants with chronic WAD demonstrating reduced thresholds when compared with participants with chronic, insidious-onset neck pain (p < 0.001; Table 3).

Associations between NDI and fat measurements

Within groups there was no association between NDI and fat levels TOTAL_UPPER_FAT; p 0.15; TOTAL_FAT; p 0.94). Thus, NDI does not signicantly inuence the amount of fat inltrate in the cervical extensor musculature and approximately 70% of the total fat measure variance is explained by other factors (R2 0.67 and 0.69). Fig. 3a and b are scatter plots of NDI and TOTAL_UPPER_FAT and TOTAL_FAT measures for both groups, respectively.
0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0
id us tis ls s s ci iti ita ap i vi C ap ip tif C er C Tr ap cc ul ez iu s

MRI Muscle:Fat Indices

WAD Insidious

Discussion
This study indicates that there are distinct differences between the participants with insidiousonset neck pain and those with chronic WAD. The presence of widespread fatty inltrate in the sub-occipital and cervical extensor musculature was not a feature of the insidious-onset neck pain group. Notably, the scores for the fat indices in this group mirror those of asymptomatic subjects as reported in a previous study.4 In addition, the insidious-onset neck pain group differed on the sensory and psychological measures when compared with the group with WAD. The classication tree determined that the insidious-neck pain group

is

bO

lis

na l

le n Sp

iu U

is

Figure 1 Total MRI muscle:fat indices for the cervical extensor musculature between the insidious-onset neck pain and whiplash groups. Data presented are means SD (all signicant at p < 0.001).

Se

Se m

is

pp er

pi na

Su

pi

Fatty inltrate in the cervical extensor muscles

685

Figure 2

Segmental multidii musculature (outlined) at C6 on T1-weighted, axial MRI image.

could be clearly distinguished from the WAD group based on average muscular fat and cold pain thresholds. It is difcult to denitively ascribe duration of symptoms and disuse as either the primary or the only cause of fatty inltrate in the WAD group based on this study for two reasons. First, the insidious group showed low values of fatty inltrate, similar to previous data from healthy subjects4 and second, the insidious group had an average symptom duration 1 year greater than the WAD group (insidious group 2.9 years and the WAD group 1.8 years). If disuse were the primary or only cause of fatty inltrate, then it would be expected to be present in those suffering neck pain of a longer duration. Nevertheless, it is possible that the factors involved in disuse are more closely related to higher levels of pain and disability, which affect activity levels, rather than the longer duration of symptoms. It is notable that the WAD group, consistent with previous studies, had higher levels of pain and disability when compared with the insidious-onset group, 11,19 but the NDI score was shown not to signicantly inuence the amount of fat inltrate in the cervical

extensor muscles. As illustrated in Fig. 3, many of the WAD participants had NDI scores <40/100, mirroring all of the participants with insidious-onset neck pain. It is recognized that the small number of insidious-onset neck pain participants potentially decreases the validity, but this preliminary nding suggests the existence of other unknown factors that may inuence fat content in chronic WAD. Studies comparing large numbers of patients with chronic WAD and insidious-onset neck pain with a broader range of NDI scores might better inform on the potential roles of disuse, structural response to frank injury and possibly, neuropathic origins of the fatty inltrate. In this study, the activity levels were not measured, but an inverse relation between activity levels and pain would have been expected. Further research should include measures that would accurately represent activity/exercise levels specic to the head and neck before denitive conclusions into the underlying mechanisms of muscular degeneration can be drawn. The insidious-onset neck pain group did not display the sensory hypersensitivity that was apparent in

Table 3 Thermal pain threshold (cold and heat) and pressure pain thresholds (local and remote) for the participants with insidious-onset neck pain and whiplash Cold pain thresholda ( C) Insidious-onset neck pain Whiplash-associated disorders 8.8 (3.8) 7.1e10.4 25 (3.8) 24.1e25.8 Heat pain thresholdb ( C) 42 (2.4) 40.9e43.1 39.8 (2.9) 39.2e40.5 Local pressure pain thresholda (kPa) 235.5 (60.5) 209.3e261.6 115.6 (40.9) 106.4e124.7 Remote pressure pain thresholda (kPa) 382.4 (150.2) 317.4e447.3 246.5 (77.4) 229.2e263.9

Data are mean (SD) and 95% condence intervals. a p < 0.001. b p 0.001.

686

J. Elliott et al.

Figure 3 (a) Scatter plot for NDI scores and TOTAL_UPPER_FAT and (b) TOTAL_FAT scores between participants with chronic WAD and chronic insidious-onset neck pain.

those with chronic whiplash. Although group differences were apparent for all stimuli, it is notable that cold pain thresholds most clearly distinguished the groups in the classication tree. The other sensory measures of pressure and heat pain thresholds did not emerge as signicant factors, but care should be taken with interpretation, as this could be a factor of power of the study. Nevertheless, cold allodynia or intolerance has been shown to be a consistent predictor of higher levels of pain and disability in chronic WAD.12,20 Additionally, decreased cold pain thresholds (cold hyperalgesia) are a feature of neuropathic pain.21,22 The co-occurrence of cold allodynia and muscle fatty inltrate in chronic WAD is interesting and may suggest common underlying mechanisms, possibly involving peripheral nerve injury. It is known that muscular fatty inltration, which results in increased signal on T1-weighted MRI sequences, is a hallmark of chronicity6 and may represent chronic denervation.23e25 Further research into the possibility of nerve injury may help to better understand the processes underlying chronic whiplash pain. In addition the temporal development of fatty inltrate in muscle needs to be established. It is noteworthy that participants with insidiousneck pain, despite long-standing symptoms, did not show specic signs of sensory hypersensitivity, psychological distress, or fatty changes in their cervical extensors. The difference in sensory presentation between the two groups has been previously observed11 and could indicate that augmented central pain processing does not play a large role in neck pain of insidious-onset. This

is in contrast to chronic whiplash where such processes are thought to be important.2,3,10,26 It appears that insidious-onset neck pain has a less complex presentation than whiplash and may reect a condition primarily involving peripheral nociceptive activity. In conclusion, the results of this study provide preliminary data that female patients (18e45 years) suffering from persistent insidious-onset neck pain do not show quantiable MRI changes in the fat content of the cervical extensor musculature and that their levels of fat mirror those with no history of neck pain. In addition to a lack of muscle changes, participants with insidious-onset neck pain did not demonstrate widespread sensitivity to sensory stimuli and the presence of these features is consistent with abnormal central painprocessing mechanisms; such as seen in some subjects with acute and chronic WAD.

Acknowledgements
The authors thank Kathy Francis for her contributions in completing this study.

References
1. Radanov BP, Sturzenegger M, Di Stefano G. Long-term outcome after whiplash injury. A 2-year follow-up considering features of injury mechanism and somatic, radiologic, and psychosocial ndings. Medicine (Baltimore) 1995;74:281e97.

Fatty inltrate in the cervical extensor muscles

687

2. Curatolo M, Petersen-Felix S, Arendt-Nielsen L, et al. Central hypersensitivity in chronic pain after whiplash injury. Clin J Pain 2001;17:306e15. 3. Sterling M, Jull G, Kenardy J. Physical and psychological factors maintain long-term predictive capacity post-whiplash injury. Pain 2006;122:102e8 [epub 9 Mar 2006]. 4. Elliott J, Jull G, Noteboom JT, et al. Fatty inltration in the cervical extensor muscles in persistent whiplash-associated disorders: a magnetic resonance imaging analysis. Spine 2006;31:E847e55. 5. Nukada H, McMorran PD, Shimizu J. Acute inammatory demyelination in reperfusion nerve injury. Ann Neurol 2000; 47:71e9. 6. Lovitt S, Moore SL, Marden FA. The use of MRI in the evaluation of myopathy. Clin Neurophysiol 2006;117:486e95 [epub 27 Jan 2006]. 7. Murphy WA, Totty WG, Carrol JE. MRI of normal and pathologic skeletal muscle. AJR Am J Roentgenol 1986;146: 565e74. 8. Rorden C, Brett M. Stereotaxic display of brain lesions. Behav Neurol 2000;12:191e200. 9. Elliott JM, Galloway GG, Jull GA, et al. Magnetic resonance imaging analysis of the upper cervical spine extensor musculature in an asymptomatic cohort: an index of fat within muscle. Clin Radiol 2005;60:355e63. 10. Sterling M, Jull G, Vicenzio B, et al. Sensory hypersensitivity occurs soon after whiplash injury and is associated with poor recovery. Pain 2003;104:509e17. 11. Scott D, Jull G, Sterling M. Widespread sensory hypersensitivity is a feature of chronic whiplash-associated disorder but not chronic idiopathic neck pain. Clin J Pain 2005;21: 175e81. 12. Sterling M, Treleaven J, Jull G. Responses to a clinical test of mechanical provocation of nerve tissue in whiplash associated disorder. Man Ther 2002;7:89e94. 13. Brennum J, Kjeldsen M, Jensen K, et al. Measurements of human pressureepain thresholds on ngers and toes. Pain 1989;38:211e7.

14. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. J Manipulative Physiol Ther 1991;14:409e15. 15. Pietrobon R, Coeytaux RR, Carey TS, et al. Standard scales for measurement of functional outcome for cervical pain or dysfunction: a systematic review. Spine 2002;27:515e22. 16. Gargan M, Bannister G, Main C, et al. The behavioural response to whiplash injury. J Bone Joint Surg Br 1997;79: 523e6. 17. Breiman L, Friedman JH, Olshen RA, et al. Classication and regression trees. New York: Chapman & Hall/CRC Press; 1984. 18. Therneau TM, Atkinson EJ. An introduction to recursive partitioning using the RPART routines. Rochester MN, USA: Mayo Foundation; 1997. 19. Jull G, Kristjansson E, DallAlba P. Impairment in the cervical exors: a comparison of whiplash and insidious onset neck pain patients. Man Ther 2004;9:89e94. 20. Kasch H, Qerama E, Bach FW, et al. Reduced cold pressor pain tolerance in non-recovered whiplash patients: a 1-year prospective study. Eur J Pain 2005;9:561e9. 21. Jorum E, Warncke T, Stubhaug A. Cold allodynia and hyperalgesia in neuropathic pain: the effect of N-methyl-D-aspartate (NMDA) receptor antagonist ketamineda double-blind, cross-over comparison with alfentanil and placebo. Pain 2003;101:229e35. 22. Bennett G. Can we distinguish between inammatory and neuropathic pain? Pain Res Manag 2006;11(Suppl. A):11e5. 23. Fleckenstein JL, Watumull D, Cooner KE, et al. Denervated human skeletal muscle: MR imaging evaluation. Radiology 1993;187:213e8. 24. Andary MT, Hallgren RC, Greenman PE, et al. Neurogenic atrophy of suboccipital muscles after a cervical injury: a case study. Am J Phys Med Rehabil 1998;77:545e9. 25. Dulor JP, Cambon B, Vigneron P, et al. Expression of white adipose tissue genes in denervation induced skeletal muscle fatty degeneration. FEBS Lett 1998;439:89e92. 26. Banic B, Petersen-Felix S, Andersen O, et al. Evidence for spinal cord hypersensitivity in chronic pain after a whiplash injury and in bromyalgia. Pain 2004;107:7e15.