- Email: [email protected]
Reliability and Normative Database of the Zebris Cervical Range-of-Motion System in Healthy Controls with Preliminary Validation in a Group of Patients with Neck Pain
RELIABILITY AND NORMATIVE DATABASE OF THE ZEBRIS CERVICAL RANGE-OF-MOTION SYSTEM IN HEALTHY CONTROLS WITH PRELIMINARY VALIDATION IN A GROUP OF PATIENTS WITH NECK PAIN Barbara Cagnie, PT, PhD,a Ann Cools, PT, PhD,b Veerle De Loose, PT,c Dirk Cambier, PT, PhD,b and Lieven Danneels, PT, PhDb
ABSTRACT Objective: The first aim of this study was to determine the reliability of the Zebris (Achen, Germany) ultrasound-based
testing of cervical range of motion (ROM). The second aim was to develop a normative database in a healthy sample of 96 volunteers. The third aim was to evaluate, with the Zebris system, the ROM in a sample of patients with chronic neck pain compared to healthy controls to determine if cervical ROM could discriminate between these groups and between subgroups of pain patients (with or without whiplash injury). Methods: The study participants were 96 healthy volunteers, 14 patients with idiopathic neck pain, and 16 patients with chronic whiplash. Cervical ROM was measured in the 3 planes with the Zebris CMS 70P ultrasound-based motion analysis system. The intra- and interrater reliability of the protocol was tested in 12 volunteers. Results: Full-cycle measurements showed high reliability (intraclass correlation coefficient, 0.80-0.94) with the SE of measurement ranging from 4.258 to 7.888. The distribution of ROM measures showed a great individual variation, with a significant age-related decrease in ROM in all directions. Range of motion was reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM. Conclusion: Results demonstrate a high degree of test-retest reliability in measuring cervical ROM. The use of normative data for ROM when evaluating patients with neck disorders needs to take age into account. The current study has demonstrated that patients with chronic neck pain demonstrate reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash. (J Manipulative Physiol Ther 2007;30:450-455) Key Indexing Terms: Neck Pain; Range of Motion; Articular; Reproducibility of Results; Whiplash Injuries
A
ssessments of cervical range of motion (ROM) are useful in the quantification of musculoskeletal impairments and may also be used to provide a basis for assessing the efficacy of therapeutic interventions
a Postdoctural research fellow, Department of Rehabilitation Sciences and Physiotherapy, Ghent University, Ghent, Belgium. b Professor, Department of Rehabilitation Sciences and Physiology, Ghent University, Ghent, Belgium. c Physical therapist, Department of Physical Therapy, Centre of Physical Medicine and Rehabilitation, Military Hospital of Base Queen Astrid, Brussels, Belgium. Submit requests for reprints to: Barbara Cagnie, PT, PhD, Department of Rehabilitation Sciences and Physiotherapy, Ghent University Hospital, De Pintelaan 185, 6K3, B - 9000 Ghent, Belgium (e-mail: [email protected]). Paper submitted February 7, 2007; in revised form April 17, 2007; accepted May 1, 2007. 0161-4754/$32.00 Copyright D 2007 by National University of Health Sciences. doi:10.1016/j.jmpt.2007.05.003
450
in patients with neck pain.1-3 In recent decades, many attempts have been made to obtain an objective method of assessing cervical spine mobility. To successfully distinguish between normal and abnormal motions of the spine, 2 conditions have to be fulfilled: the derivation of age- and sex-based normative data and high reproducibility of the findings.4,5 Several noninvasive methods for assessing ROM have been available. However, previous reviews on studies concerning reliability and validity of neck mobility have demonstrated that most of the instruments and procedures used are seriously flawed.6-8 A significant breakthrough in measuring cervical motion has taken place with the introduction of dedicated 3-dimensional motion analysis systems that can record, calculate, and display spatial head position.9 These systems have the advantages of monitoring relative changes in curvature of the spine during movement as well as limiting researcher bias because the results are displayed graphically and in tabulated forms on screen.3 One of these systems is the
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
Cagnie et al Neck Mobility in Neck Pain Patients
Table 1. Anthropometric data of the studied sample
Male
Female
Age group
Weight
Height
BMI
20-34 35-49 50-64 20-64 20-34 35-49 50-64 20-64
76.6 80.4 80.3 79.0 61.6 63.2 67.5 64.0
1.80 1.85 1.74 1.79 1.68 1.66 1.66 1.67
23.7 23.4 26.3 24.6 22.1 22.6 24.3 23.0
(.4) (6.9) (9.3) (8.4) (11.3) (8.8) (10.7) (10.4)
(0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.1)
(2.3) (1.9) (2.3) (2.5) (3.8) (2.0) (2.9) (3.1)
Mean values are given with SD in parentheses.
Zebris, an ultrasound (US)–based coordinate measuring system that was developed in Achen, Germany, and appears to be one of the best devices available at the moment to measure cervical ROM.3,6 However, discrepancies exist among similar studies using the Zebris system regarding normative data, which highlights the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects and patients.2,3,9,10 Secondly, high inter- and intrarater reliability estimates have been demonstrated in numerous studies, although a great variety of methodologies makes the comparison between studies almost impossible.1,2,6,11 According to Jordan,8 reliability of a tool can only be inferred for the protocol used. This emphasizes the need of determination of reliability and normative data before any differentiation can be made between healthy subjects and patients. The first aim of this study was to determine the reliability of the Zebris testing of cervical ROM. The second aim of the study was to develop a normative database in a healthy sample of 96 volunteers. The third aim of the study was to evaluate, with the Zebris system, the ROM in a sample of patients with chronic neck pain compared to healthy controls to determine if cervical ROM could discriminate between these groups and between subgroups of pain patients (with or without whiplash injury).
METHODS Sample Healthy volunteers (n = 96) and 30 patients with chronic neck pain took part in this study. To be included, healthy subjects should not have had neck pain for at least 1 year before testing. The healthy patients were divided into 3 age groups (20-34, 35-49, and 50-64 years), each consisting of 16 men and 16 women. Descriptive characteristics of the subjects are presented in Table 1. Patients with chronic neck pain were recruited through local advertisement within the university. These patients were divided into idiopathic and those with whiplash. Written informed consent was obtained from all the volunteers, and the study protocol was approved by the local ethics committee.
Idiopathic Neck Pain Group. The primary inclusion criterion for the study was nonspecific, frequent, or continuous neck pain lasting over 6 months. The exclusion criteria were an operation, pregnancy, depression or mental illness, and the presence of specific diseases known to cause chronic neck pain (spinal stenosis, instability or anomaly of the cervical spine, fibromyalgia, inflammatory joint diseases). The patients with chronic neck pain who volunteered to participate in the study were 14 women, in the age group of 20 to 34 years (mean, 28.3 F 5.4 years). Whiplash Group. Criteria for inclusion in the study as a patient with whiplash consisted of a period of at least 6 months since the accident, ongoing pain and disability in the neck region, severity of injury classified as Que´bec Task Force grade 2, and no other injury relating to a previous motor vehicle accident. A total of 16 women in the age group of 20 to 34 years (mean, 27.5 F 4.8 years) were included. Study Design The study design was subdivided into 3 parts. Intra- and Interrater Reliability Study. In the healthy group, 12 volunteers were recruited to participate in the study of intrarater reliability. A total of 12 other healthy subjects volunteered to take part in the interrater reliability design. The intra- and interrater reliability was assessed on 3 different occasions. One single operator assessed the intrarater reliability, whereas in the interrater study, 3 other investigators were involved. The time interval between the tests was 1 week. Tests were performed at the same time of day to minimize any effect of diurnal variation. Normative Database. To assess normative data and to determine if cervical ROM is sex- and age-dependent, we carried out a descriptive cross-sectional study in 96 volunteers. Patient Study. The results obtained from the patients with neck pain were compared with the findings derived from the normative database study to evaluate whether patients with idiopathic or traumatic chronic neck pain had decreased ROM.
Experimental Procedure Measurements of cervical motion were performed with the Zebris CMS 70P US-based motion analysis system. The
451
452
Cagnie et al Neck Mobility in Neck Pain Patients
Journal of Manipulative and Physiological Therapeutics July/August 2007
Table 2. Intraclass correlation coefficients, SEM, and SDD for intra- and interrater full- and half-cycle cervical ROM
Intrarater
Interrater
Flexion/extension Flexion Extension Rotation Right Left Side flexion Right Left Flexion/extension Flexion Extension Rotation Right Left Side flexion Right Left
ICC
SEM
SDD
0.87 0.77 0.81 0.80 0.70 0.62 0.84 0.73 0.85 0.94 0.84 0.92 0.92 0.50 0.58 0.92 0.90 0.79
5.47
15.16
9.9
7.88
21.83
12.6
5.91
16.37
16.3
5.45
15.11
10.6
5.78
16.03
9.8
4.25
11.77
12.9
measuring procedure is based on the determination of the spatial coordinates of miniature US transmitters whose position relative to a fixed system of 3 microphones is derived from the time delay between the US pulses, using triangulation. Specially designed headgear (adjustable to adapt to individual head sizes), on which a series of 3 miniature US transmitters were attached, was fixed to the subject’s head. A thoracic harness comprising a series of 3 transmitters served as a reference and isolated the cervical spine movements from the rest of the spine. Subjects sat on an adjustable height chair without a backrest with the person’s feet flat on the ground, the hip and knee flexed 908, and the hands resting freely on the thighs. The chair was positioned at a distance of approximately 80 cm from the microphones. The zero position was defined as the anatomical position of the head (vertically upright without rotation), subjectively determined by each subject. All movements were examined in the same order dictated by the system’s software routine. Before starting the experiment, the examiner demonstrated the motion to be measured and the subject then performed 3 consecutive repetitions in each direction as a warm-up. Starting with the sagittal plane, the patient was asked to perform movement of the head toward maximal flexion, continue to maximal motion toward extension, and return back to the reference position. This movement sequence was repeated 3 times at a comfortable, self-selected pace. After a 10- to 20-second pause, the test was continued using the same protocol for measuring cervical motion in the transverse plane, followed by the frontal plane. For each direction, the mean of 3 repetitions of full-cycle ROM and half-cycle ROM were recorded.
Statistical Analysis Analysis was performed using the SPSS version 12 statistical software (SPSS Inc, Chicago, Ill). All data were
SDD %
normally distributed (Kolmogorov-Smirnov test) and thus parametric tests were used for analyzing the results. P b .05 was considered to be statistically significant. Intra- and Interrater Reliability Study. For analysis of intra- and interrater reliability, intraclass correlation coefficients (ICC), standard error of measurement (SEM), and smallest real difference were used. Defined with respect to a 95% level of confidence, the smallest real difference is equal to 1.96M2 SEM. Because full-cycle ROM showed more reliable ICCs compared with half-cycle ranges, only full-cycle measurements were further evaluated. Normative Database. The results are expressed as means with SD. Two-way analysis of variance (ANOVA) with post hoc analysis was used to determine the significance of difference in cervical ROM for the main factors sex and age. Bivariate correlations between strength and weight, height, and body mass index were studied using Pearson correlation analysis. Patient Study. A 1-way ANOVA was performed to analyze differences in ROM between healthy subjects (age group of 20-34 years), patients with whiplash, and patients with idiopathic neck pain. Post hoc analysis was used to determine the significance of difference in cervical ROM between each group.
RESULTS Intra- and Interrater Reliability Study The single-measure ICC ranged from 0.80 to 0.94 for full-cycle measurements (Table 2). For half-cycle measurements, the ICC was between 0.50 and 0.92. Because fullcycle ROM showed more reliable ICCs compared with half-cycle ranges, only full-cycle measurements were further evaluated. The SEM ranged from 4.258 to 7.888.
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
Cagnie et al Neck Mobility in Neck Pain Patients
Table 3. Bivariate correlations between ROM and weight by Pearson correlation analysis Weight Flexion/extension Rotation Side flexion
0.404 0.184 0.239
P b.001 .074 .020
Fig. 1. Mean cervical ROM in the sagittal, transverse, and frontal planes according to age groups. The ROM in all directions decreased significantly with age ( P values ranging from .039 to b.001). Rotation yielded higher SE estimates than flexion/extension and side flexion because of higher maximum ROM values.
Normative Database The multivariate analysis of variance indicated a statistically significant multivariate effect for age ( P b .001), but not sex (flexion/extension, P = .410; rotation, P = .609; side flexion, P = .280). Therefore, normative data were only distributed according to age. The distribution of ROM measures showed a great individual variation. Post hoc analysis revealed that the ROM in all directions decreased significantly with age ( P values ranging from .039 to b.001) (Fig 1). There was a significant correlation between weight and ROM showing the greater the weight, the lesser the ROM. Correlation coefficients are presented in Table 3.
Patient Study The results indicated that ROM was reduced in patients with chronic whiplash in all primary movements compared to healthy subjects (Fig 2). The results also demonstrated a significant reduction in flexion/extension and side bending in the whiplash group compared to patients with idiopathic neck pain. Only for rotation, patients with idiopathic neck pain showed reduced ROM compared to healthy subjects.
DISCUSSION A considerable number of reliability and validity studies examining cervical mobility have been identified in the literature and, from this point of view, the present study cannot claim novelty. The strength of this study is that the results of the neck patient group are preceded by a reliability study and determination of a normative database. Discrepancies exist among similar studies regarding normative data, which highlight the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects
Fig. 2. Cervical ROM in healthy subjects, patients with idiopathic neck pain, and patients with whiplash. Asterisk indicates P b .05 (1-way ANOVA with post hoc analysis). and patients. This emphasizes the need to determine reliability and normative data before any differentiation can be made between healthy subjects and patients.
Intra- and Interrater Reliability Study High ICC values were derived for all directions. The single-measure ICC ranged from 0.80 to 0.94 for full-cycle measurements and between 0.50 and 0.92 for half-cycle measurements, which are in accordance with those of Malmstrom et al,1 Mannion et al,2 Antonaci et al,6 and Hoving et al.11 Full-cycle ROM yielded higher reliability than when separated into its component parts, which has been demonstrated in previous studies.3,4,8,12 Therefore, only full-cycle ROMs were further analyzed in this study. The main reason for this trend is the inability of the subjects to assume reliably the same neutral head position relative to the thorax in measuring half-cycle ROM. However, this fact is a limitation for clinical application of ROM assessment as sensitivity decreases and difficulties in distinguishing the affected side rises. In agreement with previous studies, the reliability values were better in the sagittal plane than in the other 2 planes.3,13 However, classifying ICC values higher than 0.80 as good or acceptable reproducibility sometime conveys little meaning. The SEM allows clinicians to determine whether a clinically significant change (on group or individual level) occurred or the observed difference in scores on retesting were within the measurement error, which is essentially the objective of reproducibility analysis.5 The SEM and smallest detectable difference (SDD) values from
453
454
Cagnie et al Neck Mobility in Neck Pain Patients
the present experiments are comparable to those from some previous reports.3,14 As calculated by Strimpakos et al,3 SDD ranged between 6.4% and 55.2% in previous studies, whereas in this study, all SDD values were less than 16.4%. These results indicate acceptable clinical applicability for all movements. According to Christensen and Nilsson,15 the individual natural variation in asymptomatic subjects is quite large for active ROM. When measuring individual patients, one should allow for a natural variation of 128 to 208. We have no idea if this instrument produces reliable results in patients because we only evaluated the reliability in a healthy population. However, Dvir and Prushansky5 and Hoving et al11 found similar results in patients with chronic disorders of the cervical spine compared to healthy subjects.
Normative Database Knowledge of normative values for cervical ROM is essential for design and execution of rehabilitation programs. In the present study, a normal database of cervical spine motion ranges was carried out in 96 asymptomatic volunteers. The global motion ranges obtained were in good agreement with the ones reported in the literature. However, the mean ROM values obtained from the present experiments could be compared only with studies using similar instruments and protocols. The studies of Strimpakos et al3 and Castro et al10 yielded similar results with the current values. The present values were higher than the results presented by Mannion et al2 and Dvir et al.9 No adequate explanations could be found for these discrepancies among similar studies using similar devices because the sample characteristics were not very different. This fact highlights the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects and patients. The differences among the planes of motion are in agreement with all previous reports apart from technologies or protocols used. Neck mobility was greatest in the transverse plane followed by the sagittal plane, whereas the frontal plane movements presented the smallest ROM. Differences in Age. The age effect was statistically significant across all primary movements. The present study revealed that asymptomatic subjects exhibited a reduction in the active cervical ROM throughout the years. Literature reports do not agree on the effect of age on active cervical ROM: although most investigations found significant agerelated decrements,10,16-19 some contradictory results have been listed.20 According to Wolfenberger et al,19 this oftenfound age-related phenomenon is probably associated with the decreasing flexibility of cartilage that is attributed to the greater number of covalent cross-links in the collagen of older people and the reduced tensile strength of ligaments with age.
Journal of Manipulative and Physiological Therapeutics July/August 2007
Differences in Sex. No significant differences in sex were detected, which made that all subsequent analyses were considered for men and women collectively. The effect of sex on cervical ROM is still discussed. Some studies report no sex-related variations,2,7 whereas others indicate that women display higher values than men, which they attribute to hormonal differences.19 However, the differences in most studies are small and not generally statistically significant.7 Castro et al10 found only significant differences between women and men in older subjects between the ages of 70 and 79 years in flexion and extension and in lateral bending. It was also found that women in advanced age were significantly more mobile than the respective group of men of the same age.10 Because we did not measure this age group, we cannot compare the results. Correlation with Physical Measurements. To the best of our knowledge, only one study has examined the ROM of the cervical spine as a function of body weight and height.10 Our results are in comparison with the results of the study by Castro et al indicating that the ROM of the cervical spine decreases with increasing weight. Patient Study Few studies are available investigating the difference in cervical ROM between patients with chronic neck pain and healthy subjects of a similar age range. The results indicate that ROM is reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM. In a study of Antonaci et al,6 all neck movements, with the exception of extension, were significantly reduced in patients with whiplash compared to controls, whereas the results of Dall’Alba et al21 indicated that sagittal plane movements were proportionally the most affected in patients with persistent whiplash. According to Dvir et al and Prushansky,5 patients with chronic effects of whiplash have a homogenous reduction of 25% to 35% along each of the motion directions compared to normal controls. We found a reduction of 25.8% (flexion/extension), 28.6% (rotation), and 17.1% (side flexion) in the whiplash group, indicating a similar reduction in cervical ROM compared with results from previous studies. Both sagittal and transverse planes were the most affected. The influence of idiopathic neck pain on cervical ROM has hardly ever been investigated. Hagen et al22 and Lee et al23 investigated the association between subclinical neck pain and ROM. In both studies, rotation showed to be the most affected direction of movement in the cervical spine differing between healthy subjects and patients with neck pain. This is in agreement with the results of the idiopathic neck pain group where only rotation was found to be significant when compared to healthy subjects.
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
The reason for limited ROM is unknown. Structural changes in the tissues, pain inhibition, or other factors are hypothesized in the literature.5,21,24 Although it is tempting to think in terms of an inverse relationship between pain and motion, specific pain-motion paradigms have not been indicated with respect to the current groups of patients.
CONCLUSION In this study, a US-based motion analysis system was shown to be a reliable tool for continuous tracking of cervical spine primary motions. Results demonstrate a high degree of test-retest reliability in measuring cervical ROM. We established a normal database of these movements in a population of 96 asymptomatic volunteers, which may be used as a reference in clinical settings. Age should be taken into account for optimal discrimination between asymptomatic and symptomatic persons. The current study has demonstrated that patients with chronic neck pain demonstrate reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash.
Practical Applications ! The use of normative data for ROM when evaluating patients with neck disorders needs to take age into account. ! Patients with chronic neck pain in this study demonstrated reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash. ! Range of motion was reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM.
REFERENCES 1. Malmstrom E, Karlberg M, Meande A, Magnusson M. Zebris versus Myrin: a comparative study between a three-dimensional ultrasound movement analysis and an inclinometer/ compass method. Spine 2003;28:433-44. 2. Mannion A, Klein G, Dvorak J, Lanz C. Range of global motion of the cervical spine: intraindividual reliability and the influence of measurement device. Eur Spine J 2000; 9:379-85. 3. Strimpakos N, Sakellari V, Gioftsos G, et al. Cervical spine ROM measurements: optimizing the testing protocol by using a 3D ultrasound-based motion analysis system. Cephalalgia 2005;25:1133-45. 4. Dvir Z, Gal-Eshel N, Shamier B, Prushansky T, Pevzener E, Peretz C. Cervical motion in patients with chronic disorders of the cervical spine: a reproducibility study. Spine 2006; 31:394-9.
Cagnie et al Neck Mobility in Neck Pain Patients
5. Dvir Z, Prushansky T. Reproducibility and instrument validity of a new ultrasonography based system for measuring cervical spine kinematics. Clin Biomech 2000;15:658-64. 6. Antonaci F, Bulgheroni M, Ghirmai S, et al. 3D kinematic analysis and clinical evaluation of neck movements in patients with whiplash injury. Cephalalgia 2002;22:533-42. 7. Chen J, Solinger A, Poncet J, Lantz C. Meta-analysis of normative cervical motion. Spine 1999;24:1571-8. 8. Jordan K. Assessment of published reliability studies for cervical spine range of motion measurement tools. J Manipulative Physiol Ther 2000;23:180-95. 9. Dvir Z, Werner V, Peretz C. The effect of measurement protocol on active cervical motion in healthy subjects. Physiother Res Int 2002;7:136-45. 10. Castro W, Sautmann A, Schilgen M, Sautmann M. Noninvasive three-dimensional analysis of cervical spine motion in normal subjects in relation to age and sex. Spine 2000; 25:443-9. 11. Hoving J, Pool J, van Mameren H, Deville´ W, Assendelft W, de Vet H, et al. Reproducibility of cervical range of motion in patients with neck pain. BMC Musculoskelet Disord 2005;6:59. 12. Nilsson N. Measuring passive cervical motion: a study of reliability. J Manipulative Physiol Ther 1995;18:293-7. 13. Jordan K, Dziedzic K, Jones W, Ong B, Dawes P. The reliability of the three-dimensional FASTRAK measurement system in measuring cervical spine and shoulder range of motion in healthy subjects. Rheumatology 2000;39:382-8. 14. Petersen C, Johnson R, Schuit D. Reliability of cervical range of motion using the OSI CA 6000 spine motion analyser on asymptomatic and symptomatic subjects. Man Ther 2000;5:82-8. 15. Christensen H, Nilsson N. The reliability of measuring active and passive cervical range of motion. An observer-blinded and randomized repeated-measures design. J Manipulative Physiol Ther 1998;21:341-7. 16. Feipel V, Rondelet B, Le Pallec J, Rooze M. Normal global motion of the cervical spine: an electrogoniometric study. Clin Biomech 1999;14:462-70. 17. Sforza C, Grassi G, Fragnito N, Turci M, Ferrario V. Threedimensional analysis of active head and cervical spine range of motion: effect of age in healthy male subjects. Clin Biomech 2002;17:611-4. 18. Wang S, Teng C, Lin K. Measurement of cervical range of motion pattern during cyclic neck movement by an ultrasoundbased motion system. Man Ther 2005;10:68-72. 19. Wolfenberger V, Bui W, Batenchuk G. A comparison of methods of evaluating cervical range of motion. J Manipulative Physiol Ther 2002;25:154-60. 20. Mayer T, Brady S, Bovasso E, Pope P, Gatchel R. Noninvasive measurement of cervical tri-planar motion in normal subjects. Spine 1993;18:2191-5. 21. Dall’Alba P, Sterling M, Treleaven J, Edwards S, Jull G. Cervical range of motion discriminate between asymptomatic persons and those with whiplash. Spine 2001;26:2090-4. 22. Hagen K, Harms-Ringdahl K, Enger N, Hedenstad R, Morten H. Relationship between subjective neck disorders and cervical spine mobility and motion-related pain in male machine operators. Spine 1997;22:1501-7. 23. Lee H, Nicholson L, Adams R. Cervical range of motion associations with subclinical neck pain. Spine 2003;29:33-40. 24. Ylinen J, Takala E, Kautiainen H, et al. Association of neck pain, disability and neck pain during maximal effort with neck muscle strength and range of movement in women with chronic non-specific neck pain. Eur J Pain 2004;8:473-8.
455
ABSTRACT Objective: The first aim of this study was to determine the reliability of the Zebris (Achen, Germany) ultrasound-based
testing of cervical range of motion (ROM). The second aim was to develop a normative database in a healthy sample of 96 volunteers. The third aim was to evaluate, with the Zebris system, the ROM in a sample of patients with chronic neck pain compared to healthy controls to determine if cervical ROM could discriminate between these groups and between subgroups of pain patients (with or without whiplash injury). Methods: The study participants were 96 healthy volunteers, 14 patients with idiopathic neck pain, and 16 patients with chronic whiplash. Cervical ROM was measured in the 3 planes with the Zebris CMS 70P ultrasound-based motion analysis system. The intra- and interrater reliability of the protocol was tested in 12 volunteers. Results: Full-cycle measurements showed high reliability (intraclass correlation coefficient, 0.80-0.94) with the SE of measurement ranging from 4.258 to 7.888. The distribution of ROM measures showed a great individual variation, with a significant age-related decrease in ROM in all directions. Range of motion was reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM. Conclusion: Results demonstrate a high degree of test-retest reliability in measuring cervical ROM. The use of normative data for ROM when evaluating patients with neck disorders needs to take age into account. The current study has demonstrated that patients with chronic neck pain demonstrate reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash. (J Manipulative Physiol Ther 2007;30:450-455) Key Indexing Terms: Neck Pain; Range of Motion; Articular; Reproducibility of Results; Whiplash Injuries
A
ssessments of cervical range of motion (ROM) are useful in the quantification of musculoskeletal impairments and may also be used to provide a basis for assessing the efficacy of therapeutic interventions
a Postdoctural research fellow, Department of Rehabilitation Sciences and Physiotherapy, Ghent University, Ghent, Belgium. b Professor, Department of Rehabilitation Sciences and Physiology, Ghent University, Ghent, Belgium. c Physical therapist, Department of Physical Therapy, Centre of Physical Medicine and Rehabilitation, Military Hospital of Base Queen Astrid, Brussels, Belgium. Submit requests for reprints to: Barbara Cagnie, PT, PhD, Department of Rehabilitation Sciences and Physiotherapy, Ghent University Hospital, De Pintelaan 185, 6K3, B - 9000 Ghent, Belgium (e-mail: [email protected]). Paper submitted February 7, 2007; in revised form April 17, 2007; accepted May 1, 2007. 0161-4754/$32.00 Copyright D 2007 by National University of Health Sciences. doi:10.1016/j.jmpt.2007.05.003
450
in patients with neck pain.1-3 In recent decades, many attempts have been made to obtain an objective method of assessing cervical spine mobility. To successfully distinguish between normal and abnormal motions of the spine, 2 conditions have to be fulfilled: the derivation of age- and sex-based normative data and high reproducibility of the findings.4,5 Several noninvasive methods for assessing ROM have been available. However, previous reviews on studies concerning reliability and validity of neck mobility have demonstrated that most of the instruments and procedures used are seriously flawed.6-8 A significant breakthrough in measuring cervical motion has taken place with the introduction of dedicated 3-dimensional motion analysis systems that can record, calculate, and display spatial head position.9 These systems have the advantages of monitoring relative changes in curvature of the spine during movement as well as limiting researcher bias because the results are displayed graphically and in tabulated forms on screen.3 One of these systems is the
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
Cagnie et al Neck Mobility in Neck Pain Patients
Table 1. Anthropometric data of the studied sample
Male
Female
Age group
Weight
Height
BMI
20-34 35-49 50-64 20-64 20-34 35-49 50-64 20-64
76.6 80.4 80.3 79.0 61.6 63.2 67.5 64.0
1.80 1.85 1.74 1.79 1.68 1.66 1.66 1.67
23.7 23.4 26.3 24.6 22.1 22.6 24.3 23.0
(.4) (6.9) (9.3) (8.4) (11.3) (8.8) (10.7) (10.4)
(0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.1)
(2.3) (1.9) (2.3) (2.5) (3.8) (2.0) (2.9) (3.1)
Mean values are given with SD in parentheses.
Zebris, an ultrasound (US)–based coordinate measuring system that was developed in Achen, Germany, and appears to be one of the best devices available at the moment to measure cervical ROM.3,6 However, discrepancies exist among similar studies using the Zebris system regarding normative data, which highlights the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects and patients.2,3,9,10 Secondly, high inter- and intrarater reliability estimates have been demonstrated in numerous studies, although a great variety of methodologies makes the comparison between studies almost impossible.1,2,6,11 According to Jordan,8 reliability of a tool can only be inferred for the protocol used. This emphasizes the need of determination of reliability and normative data before any differentiation can be made between healthy subjects and patients. The first aim of this study was to determine the reliability of the Zebris testing of cervical ROM. The second aim of the study was to develop a normative database in a healthy sample of 96 volunteers. The third aim of the study was to evaluate, with the Zebris system, the ROM in a sample of patients with chronic neck pain compared to healthy controls to determine if cervical ROM could discriminate between these groups and between subgroups of pain patients (with or without whiplash injury).
METHODS Sample Healthy volunteers (n = 96) and 30 patients with chronic neck pain took part in this study. To be included, healthy subjects should not have had neck pain for at least 1 year before testing. The healthy patients were divided into 3 age groups (20-34, 35-49, and 50-64 years), each consisting of 16 men and 16 women. Descriptive characteristics of the subjects are presented in Table 1. Patients with chronic neck pain were recruited through local advertisement within the university. These patients were divided into idiopathic and those with whiplash. Written informed consent was obtained from all the volunteers, and the study protocol was approved by the local ethics committee.
Idiopathic Neck Pain Group. The primary inclusion criterion for the study was nonspecific, frequent, or continuous neck pain lasting over 6 months. The exclusion criteria were an operation, pregnancy, depression or mental illness, and the presence of specific diseases known to cause chronic neck pain (spinal stenosis, instability or anomaly of the cervical spine, fibromyalgia, inflammatory joint diseases). The patients with chronic neck pain who volunteered to participate in the study were 14 women, in the age group of 20 to 34 years (mean, 28.3 F 5.4 years). Whiplash Group. Criteria for inclusion in the study as a patient with whiplash consisted of a period of at least 6 months since the accident, ongoing pain and disability in the neck region, severity of injury classified as Que´bec Task Force grade 2, and no other injury relating to a previous motor vehicle accident. A total of 16 women in the age group of 20 to 34 years (mean, 27.5 F 4.8 years) were included. Study Design The study design was subdivided into 3 parts. Intra- and Interrater Reliability Study. In the healthy group, 12 volunteers were recruited to participate in the study of intrarater reliability. A total of 12 other healthy subjects volunteered to take part in the interrater reliability design. The intra- and interrater reliability was assessed on 3 different occasions. One single operator assessed the intrarater reliability, whereas in the interrater study, 3 other investigators were involved. The time interval between the tests was 1 week. Tests were performed at the same time of day to minimize any effect of diurnal variation. Normative Database. To assess normative data and to determine if cervical ROM is sex- and age-dependent, we carried out a descriptive cross-sectional study in 96 volunteers. Patient Study. The results obtained from the patients with neck pain were compared with the findings derived from the normative database study to evaluate whether patients with idiopathic or traumatic chronic neck pain had decreased ROM.
Experimental Procedure Measurements of cervical motion were performed with the Zebris CMS 70P US-based motion analysis system. The
451
452
Cagnie et al Neck Mobility in Neck Pain Patients
Journal of Manipulative and Physiological Therapeutics July/August 2007
Table 2. Intraclass correlation coefficients, SEM, and SDD for intra- and interrater full- and half-cycle cervical ROM
Intrarater
Interrater
Flexion/extension Flexion Extension Rotation Right Left Side flexion Right Left Flexion/extension Flexion Extension Rotation Right Left Side flexion Right Left
ICC
SEM
SDD
0.87 0.77 0.81 0.80 0.70 0.62 0.84 0.73 0.85 0.94 0.84 0.92 0.92 0.50 0.58 0.92 0.90 0.79
5.47
15.16
9.9
7.88
21.83
12.6
5.91
16.37
16.3
5.45
15.11
10.6
5.78
16.03
9.8
4.25
11.77
12.9
measuring procedure is based on the determination of the spatial coordinates of miniature US transmitters whose position relative to a fixed system of 3 microphones is derived from the time delay between the US pulses, using triangulation. Specially designed headgear (adjustable to adapt to individual head sizes), on which a series of 3 miniature US transmitters were attached, was fixed to the subject’s head. A thoracic harness comprising a series of 3 transmitters served as a reference and isolated the cervical spine movements from the rest of the spine. Subjects sat on an adjustable height chair without a backrest with the person’s feet flat on the ground, the hip and knee flexed 908, and the hands resting freely on the thighs. The chair was positioned at a distance of approximately 80 cm from the microphones. The zero position was defined as the anatomical position of the head (vertically upright without rotation), subjectively determined by each subject. All movements were examined in the same order dictated by the system’s software routine. Before starting the experiment, the examiner demonstrated the motion to be measured and the subject then performed 3 consecutive repetitions in each direction as a warm-up. Starting with the sagittal plane, the patient was asked to perform movement of the head toward maximal flexion, continue to maximal motion toward extension, and return back to the reference position. This movement sequence was repeated 3 times at a comfortable, self-selected pace. After a 10- to 20-second pause, the test was continued using the same protocol for measuring cervical motion in the transverse plane, followed by the frontal plane. For each direction, the mean of 3 repetitions of full-cycle ROM and half-cycle ROM were recorded.
Statistical Analysis Analysis was performed using the SPSS version 12 statistical software (SPSS Inc, Chicago, Ill). All data were
SDD %
normally distributed (Kolmogorov-Smirnov test) and thus parametric tests were used for analyzing the results. P b .05 was considered to be statistically significant. Intra- and Interrater Reliability Study. For analysis of intra- and interrater reliability, intraclass correlation coefficients (ICC), standard error of measurement (SEM), and smallest real difference were used. Defined with respect to a 95% level of confidence, the smallest real difference is equal to 1.96M2 SEM. Because full-cycle ROM showed more reliable ICCs compared with half-cycle ranges, only full-cycle measurements were further evaluated. Normative Database. The results are expressed as means with SD. Two-way analysis of variance (ANOVA) with post hoc analysis was used to determine the significance of difference in cervical ROM for the main factors sex and age. Bivariate correlations between strength and weight, height, and body mass index were studied using Pearson correlation analysis. Patient Study. A 1-way ANOVA was performed to analyze differences in ROM between healthy subjects (age group of 20-34 years), patients with whiplash, and patients with idiopathic neck pain. Post hoc analysis was used to determine the significance of difference in cervical ROM between each group.
RESULTS Intra- and Interrater Reliability Study The single-measure ICC ranged from 0.80 to 0.94 for full-cycle measurements (Table 2). For half-cycle measurements, the ICC was between 0.50 and 0.92. Because fullcycle ROM showed more reliable ICCs compared with half-cycle ranges, only full-cycle measurements were further evaluated. The SEM ranged from 4.258 to 7.888.
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
Cagnie et al Neck Mobility in Neck Pain Patients
Table 3. Bivariate correlations between ROM and weight by Pearson correlation analysis Weight Flexion/extension Rotation Side flexion
0.404 0.184 0.239
P b.001 .074 .020
Fig. 1. Mean cervical ROM in the sagittal, transverse, and frontal planes according to age groups. The ROM in all directions decreased significantly with age ( P values ranging from .039 to b.001). Rotation yielded higher SE estimates than flexion/extension and side flexion because of higher maximum ROM values.
Normative Database The multivariate analysis of variance indicated a statistically significant multivariate effect for age ( P b .001), but not sex (flexion/extension, P = .410; rotation, P = .609; side flexion, P = .280). Therefore, normative data were only distributed according to age. The distribution of ROM measures showed a great individual variation. Post hoc analysis revealed that the ROM in all directions decreased significantly with age ( P values ranging from .039 to b.001) (Fig 1). There was a significant correlation between weight and ROM showing the greater the weight, the lesser the ROM. Correlation coefficients are presented in Table 3.
Patient Study The results indicated that ROM was reduced in patients with chronic whiplash in all primary movements compared to healthy subjects (Fig 2). The results also demonstrated a significant reduction in flexion/extension and side bending in the whiplash group compared to patients with idiopathic neck pain. Only for rotation, patients with idiopathic neck pain showed reduced ROM compared to healthy subjects.
DISCUSSION A considerable number of reliability and validity studies examining cervical mobility have been identified in the literature and, from this point of view, the present study cannot claim novelty. The strength of this study is that the results of the neck patient group are preceded by a reliability study and determination of a normative database. Discrepancies exist among similar studies regarding normative data, which highlight the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects
Fig. 2. Cervical ROM in healthy subjects, patients with idiopathic neck pain, and patients with whiplash. Asterisk indicates P b .05 (1-way ANOVA with post hoc analysis). and patients. This emphasizes the need to determine reliability and normative data before any differentiation can be made between healthy subjects and patients.
Intra- and Interrater Reliability Study High ICC values were derived for all directions. The single-measure ICC ranged from 0.80 to 0.94 for full-cycle measurements and between 0.50 and 0.92 for half-cycle measurements, which are in accordance with those of Malmstrom et al,1 Mannion et al,2 Antonaci et al,6 and Hoving et al.11 Full-cycle ROM yielded higher reliability than when separated into its component parts, which has been demonstrated in previous studies.3,4,8,12 Therefore, only full-cycle ROMs were further analyzed in this study. The main reason for this trend is the inability of the subjects to assume reliably the same neutral head position relative to the thorax in measuring half-cycle ROM. However, this fact is a limitation for clinical application of ROM assessment as sensitivity decreases and difficulties in distinguishing the affected side rises. In agreement with previous studies, the reliability values were better in the sagittal plane than in the other 2 planes.3,13 However, classifying ICC values higher than 0.80 as good or acceptable reproducibility sometime conveys little meaning. The SEM allows clinicians to determine whether a clinically significant change (on group or individual level) occurred or the observed difference in scores on retesting were within the measurement error, which is essentially the objective of reproducibility analysis.5 The SEM and smallest detectable difference (SDD) values from
453
454
Cagnie et al Neck Mobility in Neck Pain Patients
the present experiments are comparable to those from some previous reports.3,14 As calculated by Strimpakos et al,3 SDD ranged between 6.4% and 55.2% in previous studies, whereas in this study, all SDD values were less than 16.4%. These results indicate acceptable clinical applicability for all movements. According to Christensen and Nilsson,15 the individual natural variation in asymptomatic subjects is quite large for active ROM. When measuring individual patients, one should allow for a natural variation of 128 to 208. We have no idea if this instrument produces reliable results in patients because we only evaluated the reliability in a healthy population. However, Dvir and Prushansky5 and Hoving et al11 found similar results in patients with chronic disorders of the cervical spine compared to healthy subjects.
Normative Database Knowledge of normative values for cervical ROM is essential for design and execution of rehabilitation programs. In the present study, a normal database of cervical spine motion ranges was carried out in 96 asymptomatic volunteers. The global motion ranges obtained were in good agreement with the ones reported in the literature. However, the mean ROM values obtained from the present experiments could be compared only with studies using similar instruments and protocols. The studies of Strimpakos et al3 and Castro et al10 yielded similar results with the current values. The present values were higher than the results presented by Mannion et al2 and Dvir et al.9 No adequate explanations could be found for these discrepancies among similar studies using similar devices because the sample characteristics were not very different. This fact highlights the concerns regarding the comparison between different investigations as well as the published normative data and cutoff points between healthy subjects and patients. The differences among the planes of motion are in agreement with all previous reports apart from technologies or protocols used. Neck mobility was greatest in the transverse plane followed by the sagittal plane, whereas the frontal plane movements presented the smallest ROM. Differences in Age. The age effect was statistically significant across all primary movements. The present study revealed that asymptomatic subjects exhibited a reduction in the active cervical ROM throughout the years. Literature reports do not agree on the effect of age on active cervical ROM: although most investigations found significant agerelated decrements,10,16-19 some contradictory results have been listed.20 According to Wolfenberger et al,19 this oftenfound age-related phenomenon is probably associated with the decreasing flexibility of cartilage that is attributed to the greater number of covalent cross-links in the collagen of older people and the reduced tensile strength of ligaments with age.
Journal of Manipulative and Physiological Therapeutics July/August 2007
Differences in Sex. No significant differences in sex were detected, which made that all subsequent analyses were considered for men and women collectively. The effect of sex on cervical ROM is still discussed. Some studies report no sex-related variations,2,7 whereas others indicate that women display higher values than men, which they attribute to hormonal differences.19 However, the differences in most studies are small and not generally statistically significant.7 Castro et al10 found only significant differences between women and men in older subjects between the ages of 70 and 79 years in flexion and extension and in lateral bending. It was also found that women in advanced age were significantly more mobile than the respective group of men of the same age.10 Because we did not measure this age group, we cannot compare the results. Correlation with Physical Measurements. To the best of our knowledge, only one study has examined the ROM of the cervical spine as a function of body weight and height.10 Our results are in comparison with the results of the study by Castro et al indicating that the ROM of the cervical spine decreases with increasing weight. Patient Study Few studies are available investigating the difference in cervical ROM between patients with chronic neck pain and healthy subjects of a similar age range. The results indicate that ROM is reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM. In a study of Antonaci et al,6 all neck movements, with the exception of extension, were significantly reduced in patients with whiplash compared to controls, whereas the results of Dall’Alba et al21 indicated that sagittal plane movements were proportionally the most affected in patients with persistent whiplash. According to Dvir et al and Prushansky,5 patients with chronic effects of whiplash have a homogenous reduction of 25% to 35% along each of the motion directions compared to normal controls. We found a reduction of 25.8% (flexion/extension), 28.6% (rotation), and 17.1% (side flexion) in the whiplash group, indicating a similar reduction in cervical ROM compared with results from previous studies. Both sagittal and transverse planes were the most affected. The influence of idiopathic neck pain on cervical ROM has hardly ever been investigated. Hagen et al22 and Lee et al23 investigated the association between subclinical neck pain and ROM. In both studies, rotation showed to be the most affected direction of movement in the cervical spine differing between healthy subjects and patients with neck pain. This is in agreement with the results of the idiopathic neck pain group where only rotation was found to be significant when compared to healthy subjects.
Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6
The reason for limited ROM is unknown. Structural changes in the tissues, pain inhibition, or other factors are hypothesized in the literature.5,21,24 Although it is tempting to think in terms of an inverse relationship between pain and motion, specific pain-motion paradigms have not been indicated with respect to the current groups of patients.
CONCLUSION In this study, a US-based motion analysis system was shown to be a reliable tool for continuous tracking of cervical spine primary motions. Results demonstrate a high degree of test-retest reliability in measuring cervical ROM. We established a normal database of these movements in a population of 96 asymptomatic volunteers, which may be used as a reference in clinical settings. Age should be taken into account for optimal discrimination between asymptomatic and symptomatic persons. The current study has demonstrated that patients with chronic neck pain demonstrate reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash.
Practical Applications ! The use of normative data for ROM when evaluating patients with neck disorders needs to take age into account. ! Patients with chronic neck pain in this study demonstrated reduced ROM, which differs between patients with idiopathic neck pain and those with chronic whiplash. ! Range of motion was reduced in patients with chronic whiplash in all primary movements, compared to healthy subjects, whereas in patients with idiopathic neck pain, only rotation showed reduced ROM.
REFERENCES 1. Malmstrom E, Karlberg M, Meande A, Magnusson M. Zebris versus Myrin: a comparative study between a three-dimensional ultrasound movement analysis and an inclinometer/ compass method. Spine 2003;28:433-44. 2. Mannion A, Klein G, Dvorak J, Lanz C. Range of global motion of the cervical spine: intraindividual reliability and the influence of measurement device. Eur Spine J 2000; 9:379-85. 3. Strimpakos N, Sakellari V, Gioftsos G, et al. Cervical spine ROM measurements: optimizing the testing protocol by using a 3D ultrasound-based motion analysis system. Cephalalgia 2005;25:1133-45. 4. Dvir Z, Gal-Eshel N, Shamier B, Prushansky T, Pevzener E, Peretz C. Cervical motion in patients with chronic disorders of the cervical spine: a reproducibility study. Spine 2006; 31:394-9.
Cagnie et al Neck Mobility in Neck Pain Patients
5. Dvir Z, Prushansky T. Reproducibility and instrument validity of a new ultrasonography based system for measuring cervical spine kinematics. Clin Biomech 2000;15:658-64. 6. Antonaci F, Bulgheroni M, Ghirmai S, et al. 3D kinematic analysis and clinical evaluation of neck movements in patients with whiplash injury. Cephalalgia 2002;22:533-42. 7. Chen J, Solinger A, Poncet J, Lantz C. Meta-analysis of normative cervical motion. Spine 1999;24:1571-8. 8. Jordan K. Assessment of published reliability studies for cervical spine range of motion measurement tools. J Manipulative Physiol Ther 2000;23:180-95. 9. Dvir Z, Werner V, Peretz C. The effect of measurement protocol on active cervical motion in healthy subjects. Physiother Res Int 2002;7:136-45. 10. Castro W, Sautmann A, Schilgen M, Sautmann M. Noninvasive three-dimensional analysis of cervical spine motion in normal subjects in relation to age and sex. Spine 2000; 25:443-9. 11. Hoving J, Pool J, van Mameren H, Deville´ W, Assendelft W, de Vet H, et al. Reproducibility of cervical range of motion in patients with neck pain. BMC Musculoskelet Disord 2005;6:59. 12. Nilsson N. Measuring passive cervical motion: a study of reliability. J Manipulative Physiol Ther 1995;18:293-7. 13. Jordan K, Dziedzic K, Jones W, Ong B, Dawes P. The reliability of the three-dimensional FASTRAK measurement system in measuring cervical spine and shoulder range of motion in healthy subjects. Rheumatology 2000;39:382-8. 14. Petersen C, Johnson R, Schuit D. Reliability of cervical range of motion using the OSI CA 6000 spine motion analyser on asymptomatic and symptomatic subjects. Man Ther 2000;5:82-8. 15. Christensen H, Nilsson N. The reliability of measuring active and passive cervical range of motion. An observer-blinded and randomized repeated-measures design. J Manipulative Physiol Ther 1998;21:341-7. 16. Feipel V, Rondelet B, Le Pallec J, Rooze M. Normal global motion of the cervical spine: an electrogoniometric study. Clin Biomech 1999;14:462-70. 17. Sforza C, Grassi G, Fragnito N, Turci M, Ferrario V. Threedimensional analysis of active head and cervical spine range of motion: effect of age in healthy male subjects. Clin Biomech 2002;17:611-4. 18. Wang S, Teng C, Lin K. Measurement of cervical range of motion pattern during cyclic neck movement by an ultrasoundbased motion system. Man Ther 2005;10:68-72. 19. Wolfenberger V, Bui W, Batenchuk G. A comparison of methods of evaluating cervical range of motion. J Manipulative Physiol Ther 2002;25:154-60. 20. Mayer T, Brady S, Bovasso E, Pope P, Gatchel R. Noninvasive measurement of cervical tri-planar motion in normal subjects. Spine 1993;18:2191-5. 21. Dall’Alba P, Sterling M, Treleaven J, Edwards S, Jull G. Cervical range of motion discriminate between asymptomatic persons and those with whiplash. Spine 2001;26:2090-4. 22. Hagen K, Harms-Ringdahl K, Enger N, Hedenstad R, Morten H. Relationship between subjective neck disorders and cervical spine mobility and motion-related pain in male machine operators. Spine 1997;22:1501-7. 23. Lee H, Nicholson L, Adams R. Cervical range of motion associations with subclinical neck pain. Spine 2003;29:33-40. 24. Ylinen J, Takala E, Kautiainen H, et al. Association of neck pain, disability and neck pain during maximal effort with neck muscle strength and range of movement in women with chronic non-specific neck pain. Eur J Pain 2004;8:473-8.
455