Manual Therapy (2003) 8(3), 176179

r 2003 Elsevier Science Ltd. All rights reserved.

1356-689X/03/$ - see front matter
doi:10.1016/S1356-689X(03)00009-2

Technical & measurement report

Measuring range of active cervical rotation in a position of full head exion using
the 3D Fastrak measurement system: an intra-tester reliability study
M. Amiri, G. Jull, J. Bullock-Saxton
Department of Physiotherapy, The University of Queensland, Australia

SUMMARY. Most external assessments of cervical range of motion assess the upper and lower cervical regions
simultaneously. This study investigated the within and between days reliability of the clinical method used to bias
this movement to the upper cervical region, namely measuring rotation of the head and neck in a position of full
cervical exion. Measurements were made using the Fastrak measurement system and were conducted by one
operator. Results indicated high levels of within and between days repeatability (range of ICC2,1 values: 0.850.95).
The ranges of axial rotation to right and left, measured with the neck positioned in full exion, were approximately
56% and 50%, respectively of total cervical rotation, which relates well to the proportional division of rotation in
the upper and lower cervical regions. These results suggest that this method of measuring rotation would be
appropriate for use in subject studies where movement dysfunction is present in the upper cervical region, such as
those with cervicogenic headache.
r 2003 Elsevier Science Ltd. All rights reserved.

sensitivity of motion loss for cervicogenic headache.

Most external measures of cervical motion consider
movements of both the upper and lower cervical
regions simultaneously. However as cervicogenic
headache has a primary involvement in the upper
cervical segments, measurement of motion biased to
the upper cervical region could be a more relevant
measure. Dhimitri et al. (1998) investigated the
reliability of a quantitative method of measuring
upper cervical exion and extension in vivo, but to
date no research has reported the reliability of
measuring range of active rotation in upper cervical
region. The C1-2 segment is commonly involved in
cervicogenic headache (Bogduk 2001) and the segment accounts for 4060% of total range of neck
rotation (Panjabi et al. 1988). It seems reasonable
that measurement of rotation purportedly biased to
this upper cervical region could be a relevant clinical
measure in the differential diagnosis of cervicogenic
headache.
Clinically, rotation in the upper cervical region is
differentiated from that in the lower cervical region
by pre-tensioning the structures of mid-lower cervical
spine in a full exion position and then rotating the
head on the pre-exed cervical spine (Dvorak et al.
1984). The objective of this study was to investigate
the reliability of this clinical test by assessing

INTRODUCTION
Cervicogenic headache arises from musculoskeletal
dysfunction in the cervical region, predominantly
from the upper three cervical segments (Bogduk
1997). There is some symptomatic overlap between
cervicogenic and other chronic headache forms such
as tension-type headache and migraine (Sjaastad &
Bovim 1991). This realizes the importance of possessing reliable measures to identify the physical signs of
cervical musculoskeletal dysfunction to assist differential diagnosis. Reduced range of cervical motion is
one physical diagnostic criterion for cervicogenic
headache (Sjaastad et al. 1998).
Zwart (1997) measured cervical range of motion in
a comparative study of subjects with cervicogenic
headache, migraine and tension headache and
asymptomatic controls and conrmed the diagnostic
Received: 18 July 2002
Revised: 7 October 2002
Accepted: 15 January 2003
Mohsen Amiri, BPhty, MPhty, PhD student, Gwendolen Jull
MPhty, PhD, FACP, Associate Professor, Joanne Bullock-Saxton
BPhty, PhD, Senior Lecturer, Department of Physiotherapy, The
University of Queensland, Brisbane, Queensland 4072, Australia.
Corresponding author: Tel.: +61 7 33654691; Fax: +61 7
33652775; E-mail: m.amiri@shrs.uq.edu.au
176

Measuring range of active cervical rotation 177

within and between day repeatability using a threedimensional external measurement system.

METHODS
Subjects
A sample of convenience of 15 healthy subjects was
studied. Subjects were volunteers from the students
and staff of The University of Queensland and
comprised six males and nine females, ranging in
age from 20 to 50 years (27.6077.85 years). Subjects
were not considered if they had a history of headache
or neck surgery or had received treatment for neck or
shoulders conditions within the past 3 months. They
were requested to avoid strenuous exercise involving
the neck and shoulder regions 3 h prior to measurements and also to attend data collection sessions of
approximately 20 min duration on two different days.
Ethical clearance for the study was gained from the
Medical Ethics Committee of The University of
Queensland and subjects gave informed consent.
Measurements
Cervical movement was measured with a 3-Space
Fastrak (Polhemus, Navigation Science Division,
Kaiser Aerospace Vermont). The Fastrak is a
non-invasive electromagnetic device, which tracks
the positions of sensors relative to a source in three
dimensions. The system has been shown to be
accurate to within 70.21 (Pearcy & Hindle 1991).
In this study, one sensor was xed to an adjustable
semi-rigid plastic headband, which was placed
around the subjects forehead so that the sensor was
aligned with the bridge of the nose. A second sensor
was placed over the C7 spinous process using doublesided tape to prevent movement of the sensor in
relation to the skin (Fig. 1). The leads were also
secured with tape to prevent traction on the sensor.
The electromagnetic source was placed in a box
attached to the back of a wooden chair. The Fastrak
was connected to a compatible PC, which continually
recorded the position of the sensors relative to the
source during each test sequence. A software
program was written to format and process the data
for 3D analysis of the neck movement. Data was
converted into les and graphs so that the process
could be visualized in real time. Data consisted of a
3  3 matrix of direction cosines, for the orientation
of the forehead sensor relative to the sensor at C7.
This was then analysed to give a 3D measurement of
the position of the head relative to the C7.
Baseline measures of full cervical exion and
rotation were taken from a neutral starting position
(zeroed on the Fastrak). For the measurement of
head and neck rotation in the pre-exed cervical
Manual Therapy (2003) 8(3), 176179

Fig. 1The position of the subjects head and neck and the
location of Fastrak sensors in measurement active rotation range in
a position of full cervical exion.

position, the subjects fully exed their cervical spine

and in that position, rotated the head to the left and
then right sides. In this test, the exion range was
recorded on the X-axis of the Fastrak system as a
negative value and the head and neck rotation to the
left and right sides, as negative and positive values
respectively on the Z-axis (Fig. 2).
Procedure
Subjects received full information of the measurement protocol and tasks required, prior to commencement of the study. They were then seated on a
wooden chair with a backrest and the sensors were
attached. The subjects performed each movement
three times, which served as a familiarization and
warm-up session.
For the measurement session, subjects were instructed to sit tall with their thoracic spine resting
against the backrest of chair and to look straight
ahead. This position was marked as the zero reference
starting point on the Fastrak and prior to the
commencement of each test movement, the subjects
head was relocated to this reference position using the
real time viewing facility of the software program.
Measurements were taken of cervical exion and
rotation to the left and to the right. For the
measurement of rotation in a position of cervical
pre-exion, the subjects were requested to rst fully
ex their neck. A researcher gently maintained the
neck in this exed position and the subject was asked
to then turn their head to the left or right (Fig. 1).
Three repetitions of the movement sequence were
performed within each trial and three trials were
performed on each day with a 3-min rest between
r 2003 Elsevier Science Ltd. All rights reserved.

178 Manual Therapy

RESULTS
The mean (7SD) for the total range of cervical
exion was 66.2177.61 and for rotation to the right
and left sides were 78.6179.51 and 81.11710.31,
respectively. The means (7SD) of the ranges of
rotation measured with the neck in a pre-exed
position as well as the degree of exion in which
the movements were performed, on each day (inter
trial) and between days (test-retest) are presented in
Table 1. This table also includes the intraclass
correlation coefcient (ICC) and SEM. These results
indicate that the measures have good to excellent
repeatability between and within days, respectively
(range of ICC2,1 values 0.850.95). The SEMs were
similar in both rotation directions, ranging from
1.6 to 1.91 for within day trials and 1.4 to 2.01 for
between days trials.

Fig. 2A sample of a Fastrak trace of measuring active cervical

rotation range to the right and left sides (Z-axis) in a position of
full cervical exion (X-axis).

each trial. The order of performance of the movements was randomized between subjects on each day.
Subjects were retested on a second occasion 3 days
later.

DISCUSSION
The results of this study suggest that the clinical
method purported to bias rotation to the upper
cervical region (head and neck rotation in a position
of full neck exion) as described by Dvorak et al.
(1984) can be measured repeatedly both within and
between days, as measured by one operator in this
study. The ICCs for the measures derived by the
Fastrak measurement system revealed good to
excellent repeatability (Table 1) and ICC values are
comparable to those of other studies (Dhimitri et al.
1998, Jordan et al. 2000). The SEM indicated that
there was minimal error on repeated measurement
within and between days.
The exion position can be maintained quite well
during the test and subjects performed the head
rotation within approximately 4 and 11 (left rotation)
of full exion (Table 2). That exion was not
maintained as fully with head rotation to the right,
might account for the slightly greater proportion of
total cervical rotation in this direction found in this
study. This could reect that 2 or 31 of rotation in
the lower cervical segments contributed to the range
of head rotation to the right. Nevertheless, this was a

Data analysis
The mean of three repetitions in each trial and also
the mean of three trials on both days were calculated
in preparation for statistical analysis. The data were
inspected to ensure that statistical assumptions were
met. Calculation of skewness for all measures were
within those values accepted for reliability tests (o3).
Intra-class correlation coefcients (ICC2,1) were
calculated from a repeated measures analysis of
variance (ANOVA) to examine the within and
between day reliability of the measures of rotation
in the pre-exed cervical position (Hass 1991,
Domholdt 1993). The standard errors of measurement (SEM) were computed as a measure of absolute
reliability, by which the variability of the scores
obtained from measurement to measurement was
examined (Domholdt 1993). The SEM is calculated
from the square root of the error variance from the
ANOVA and expresses the magnitude of the measurement error in the same units (Bland et al. 1996a, b)
All analyses were computed with the SPSS program.

Table 1. The means7standard deviations (SD) and inter-trial (within day 1 and day 2) and testretest (between days) intraclass correlation
coefcients (ICC) for active cervical rotation range in a position of full cervical exion
Range of active rotation in a position of full cervical exion (deg)
Right
Day

Mean7SD

One
Two
One vs Two
Allied exion

44.376.3
44.276.8
44.376.4

Left

SEM

ICC

Mean7SD

1.7
1.9
1.4
62.178.1

0.93
0.92
0.95

39.775.6
39.975.3
39.875.1

SEM
1.6
1.7
2.0
65.179.6

ICC
0.92
0.90
0.85

The degree of cervical exion at the point of measurement of cranio-cervical rotation is also recorded (n=15 subjects).
r 2003 Elsevier Science Ltd. All rights reserved.

Manual Therapy (2003) 8(3), 176179

Measuring range of active cervical rotation 179

Table 2. The mean7standard deviations (SD) for the measures of
total cervical exion and rotation (n=15)
Total neck movement

Mean7SD (deg)

Flexion
Rotation to the right
Rotation to the left

66.1777.63
78.5779.48
81.14710.29

the assessment of patients with dysfunction in the

upper cervical region such as those with cervicogenic
headache, for purposes of differential diagnosis and
assessment of treatment outcomes.

References
systematic difference, which occurred both within
and between days, as reected by the consistency of
the measure within and between days (Table 1) and
may reect an anatomical variation.
This study could not address the issue of validity of
the clinical test, that is, whether testing rotation in a
position of pre-exion of the cervical spine, limits the
rotation motion to the upper cervical region. In this
study, there was no attempt to externally xate C2 or
the lower cervical region, other than by positioning
and gently assisting the subject to maintain the full
exion position while they turned their head. There is
a high likelihood that some rotation occurred below
the C1-2 segment in the measurement of head
rotation. Thus the present results must be interpreted
with caution, bearing in mind the clinical assumption
that the rotation was performed predominantly in
upper cervical region. The conclusion of the current
study might be strengthened if such an assumption
could be examined in future using a form of
radiographic measurement to establish the validity
of the measurement.

CONCLUSION
This study has indicated that the measure of head
and neck rotation in a position of full cervical exion
can be performed reliably. It would be reasonable
to recommend that this clinical assessment be used in

Manual Therapy (2003) 8(3), 176179

Bland JM, Altman DG 1996a Measurement error. British Medical

Journal 313: 744753
Bland JM, Altman JM 1996b Measurement error and correlation
coefcients. British Medical Journal 313: 4142
Bogduk N 1997 Headache and the neck. In: Goadsby P, Silberstein
SD (eds) Headache, 17th edn. Butterworth-Heinemann,
Melbourne, Ch 22, pp 369381
Bogduk N 2001 Cervicogenic headache: Anatomical basis and
pathophysiological mechanisms. Journal of Current Pain
Headache Report 5: 382386
Dhimitri K, Brodeur S, Croteus M, Richard S, Seymour C 1998
Reliability of the cervical range of motion device in measuring
upper cervical motion. Journal of Manual and Manipulative
Therapy 6: 3136
Domholdt E 1993 Physical Therapy Research: Principles and
Applications, 1st edn. W.B. Sanders, Philadelphia, pp 153157
Dvorak J, Dvorak V, Schneider W 1984 Manual Medicine, 1st edn.
Springer, Berlin, Heidelberg Ch 2, pp 7079
Hass M 1991 Statistical methodology for reliability studies. Journal
of Manipulative and Physiological Therapeutics 14: 119128
Jordan K, Dziedzic K, Jones PW, Ong BN, Dawes PT 2000 The
reliability of the three-dimensional FASTRAK measurement
system in measuring cervical spine and shoulder range of
motion in healthy subjects. Journal of Rheumatology 39:
382388
Panjabi M, Dvorak J, Duranceau J, Yamamoto I, Gerber M,
Rauschning WG 1988 Three-dimensional movements of the
upper cervical spine. Spine 13: 726731
Pearcy MJ, Hindle RJ 1991 Axial rotation of lumbar intervertebral
joints in forward Flexion. Proceeding of the Institution of
Mechanical Engineers Part H- Journal of Engineering in
Medicine 205: 205209
Sjaastad O, Bovim G 1991 Cervicogenic headache. The Differentiation from common migraine: An overview. Journal of
Functional Neurology 6: 93100
Sjaastad O, Fredriksen TA, Pfaffenrath V 1998 Cervicogenic
headache: Diagnostic criteria. Journal of Headache 38: 442445
Zwart JA 1997 Neck Mobility in different headache disorders.
Journal of Headache 37: 611

r 2003 Elsevier Science Ltd. All rights reserved.