BEST

Baillière’s Clinical Rheumatology B A I L L I È R E ’ S

Vol. 13, No. 2, pp 261–285, 1999

5 PRACTICE
& RESEARCH

The neck

Nikolai Bogduk MD, PhD, DSc, Dip.Anat., FAFRM

Director
Newcastle Bone and Joint Institute, University of Newcastle, Royal Newcastle Hospital, Newcastle, NSW 2300,
Australia

Tumours, infections, aneurysms and metabolic and inflammatory diseases are rare causes of
neck pain. Most cases involve neck pain of unknown origin or a whiplash-associated disorder.
Neck pain is common in the general community and more common in certain occupations. The
natural history is relatively benign, but some 10% of patients will suffer chronic, severe
symptoms. Psychosocial factors have been refuted as risk factors; the cardinal risk factors relate
to occupation. In whiplash, the severity of initial symptoms is the cardinal determinant of
chronicity. History is the major factor when considering diagnosis, physical examination adding
little to the diagnosis. Imaging is not indicated in the vast majority of cases. The available
evidence does not support most of the physical, medical and surgical therapies currently
practised. Confident reassurance is paramount and justified for acute cases. Proven options for
chronic neck pain are few.

Key words: neck pain; epidemiology; pathology; diagnosis; treatment.

Neck pain is the poor cousin of low back pain. Although neck pain is almost as
common as back pain and presents the same difficulties in diagnosis and management,
it has received far less attention by way of research. That is not to say, however, that
neck pain has not been studied: it has, but the evidence is sobering. Negative findings
concerning the traditional management of neck pain match those which have been
published for low back pain, but neck pain lags behind in terms of positive data in
support of revised, modern approaches to management.
Furthermore, neck pain is clearly divided into two categories: that arising spon-
taneously or as a result of obvious trauma, and that due to whiplash. Because of its
unique aetiology and its association with medicolegal issues, neck pain resulting from
whiplash has its own literature that is almost quarantined from the other literature on
neck pain.

DEFINITION

Neck pain is not synonymous with brachialgia or radicular pain. The latter are felt in
the upper limb and are associated with objective signs of neurological impairment.
Although they may be caused by lesions in the neck, they do not constitute neck pain.
Neck pain is pain perceived dorsally in the cervical region of the spinal column, i.e.
between the superior nuchal line and an imaginary transverse line through the spinous
process of T1.1 Nevertheless, neck pain may be referred to the head, the upper limb,
1521–6942/99/020261 + 25 $12.00/00 © 1999, Baillière Tindall
262 N. Bogduk

the dorsal scapular region or the anterior chest wall. Referred pain is perceived deeply
and is aching or pressure-like in quality. It is thus clinically distinguishable from
radicular pain, which is sharp, shooting or lancinating in quality and travels along the
upper limb in a narrow band.1

Practice point
● neck pain is not synonymous with brachialgia, radicular pain or other neuro-
logical conditions affecting the upper limb

Cervical referred pain follows a quasi-segmental distribution.2 Pain from the upper
cervical segments is referred towards the head and pain from the lower segments to
the upper limb girdle. Maps of this distribution have been defined in normal volunteers
for pain stemming from the spinal muscles, the zygapophyseal joints and the inter-
vertebral discs (Figure 1). The maps for each structure are quite similar because the
distribution of pain reflects not the structure that is the source of pain but its
segmental innervation. Thus, all structures innervated by C5–6 will refer pain to
essentially the same region over the shoulder girdle. For diagnostic purposes, the

Figure 1. Maps of the segmental pattern of referral of pain from the cervical zygapophyseal joints. Similar
maps pertain to the intervertebral discs of the same segmental numbers.
 The neck 263

referred pain maps are a reasonable guide to the segmental location of the source of
pain, but they do not indicate the structure that is the actual source.

CAUSES

The taxonomy of the International Association for the Study of Pain (IASP) lists some
60 recognized causes of neck pain.1 These, and others, can be grouped as shown in
Table 1. Tumours, infections and aneurysms constitute the ‘red flag’ conditions of the
neck as they threaten serious neurological sequelae.

Table 1. The causes of neck pain grouped according to whether they are common and
serious
Non-threatening Serious
Common Cervical spinal pain Fractures
of unknown origin
Acceleration–deceleration injuries of the neck
Zygapophyseal joint pain
Discogenic pain
Uncommon Hyoid syndrome Tumours
Rheumatoid arthritis Spinal infections
Ankylosing spondylitis Dissecting aneurysms
Reactive arthritis Spinal haematomas
Psoriatic arthritis Metabolic disorders
Crystal arthropathies

Neck pain can occur in patients with known rheumatoid arthritis, but it is unlikely
to be the sole presenting feature. Fewer than 2% of patients with rheumatoid arthritis
have neck pain as their only feature. Rheumatoid arthritis becomes potentially serious
if it affects the C1–2 joints, but even then the prognosis is favourable.
Some 10% of patients with ankylosing spondylitis may present with neck pain3, but the
rarity of ankylosing spondylitis renders it an uncommon cause of neck pain. Similarly,
although Reiter’s syndrome, psoriatic arthritis and gout can involve the neck, they are
rare causes of neck pain in patients without other manifestations of these conditions.
Missing from Table 1 are cervical spondylosis and cervical osteoarthrosis. Although
hallowed by tradition, these entities defy legitimate diagnosis. Clinically, they are
indistinguishable from any other cause of neck pain. The only available diagnostic
criteria are the radiological features of these conditions, but these features are only
age changes. They correlate poorly with neck pain.4 Indeed, cervical osteoarthrosis is
more common in patients with no neck pain.5

Practice point
● spondylosis and osteoarthritis are not legitimate diagnoses of neck pain

For this reason, the IASP recommends the rubric ‘cervical spinal pain of unknown
origin’ as an honest diagnosis for patients with neck pain whose cause is not apparent.1
264 N. Bogduk

Zygapophyseal joint pain and discogenic pain are specific subsets of what otherwise
might be known as ‘mechanical’ neck pain, but their diagnosis requires invasive
procedures such as zygapophyseal joint blocks6 and disc stimulation.7,8
Although favoured by many, there is no evidence that trigger points are a cause of
neck pain. Even in the hands of experts, the diagnosis is unreliable9, and the absence
of a criterion standard means that its validity cannot be tested. Furthermore, trigger
points in the neck do not satisfy the prescribed criteria for a trigger point. They are
characterized solely by tenderness and the reproduction of pain, in which regard they
cannot be distinguished from tenderness of the underlying zygapophyseal joints.10
Although considered common, and feared as a cause of neck pain (for medicolegal
reasons), fractures of the neck are actually uncommon. In accident and emergency
settings, only about 3% of patients suspected of having a fracture prove upon cervical
radiography to have fractures.11
Acceleration–deceleration injury, or whiplash, is perhaps the most common
traumatic basis for neck pain. Biomechanical studies and post mortem studies have
shown that a variety of lesions can affect the neck during whiplash.12 Recent Japanese
studies in normal volunteers13 have demonstrated that whiplash does not involve
flexion and extension as traditionally appreciated.12 Instead, upon impact, the trunk is
thrust upwards into the neck, and the cervical spine undergoes a sigmoid-shape
deformation. During this deformation, the lower cervical vertebrae undergo an
abnormal rotation. They rotate about an abnormally high axis of rotation such that
there is no translation of the vertebra, but the vertebra spins backwards. Anteriorly,
the vertebral bodies exhibit an abnormal separation, while posteriorly, the inferior
articular process of the moving vertebra chisels into its supporting superior articular
process (Figure 2). This motion predicts that the anterior anulus fibrosus can be torn
and that, in the zygapophyseal joints, contusion of the intra-articular meniscoids or
impaction fractures of the articular process can occur. These are the very lesions most
commonly encountered in post mortem studies of the victims of motor vehicle
accidents.
Post mortem studies in Sweden14 and Australia15 have found that, in addition to
lethal injuries of the head or craniocervical junction, victims of motor vehicle accidents
exhibit tears of the anterior anulus fibrous and contusions, haemorrhages or small
fractures of the zygapophyseal joints. Virtually none of these lesions is evident on plain
films of the neck, even upon a retrospective review of post mortem X-rays.

EPIDEMIOLOGY

In the general population, the lifetime prevalence of neck pain is about 70%16 and the
point prevalence about 10%17,18; in a year, some 30% of the population will experience
neck pain, and 17% will experience neck pain lasting longer than 6 months.19 The
prevalence increases with age and is somewhat higher in females than in males of the
same age.19 In manual workers and office workers, the prevalence is higher than in
the general population.20
In Western societies, the incidence of whiplash injuries is about 1 case per 1000
population per year.12 Most cases recover, usually well within a year, but at 12 months
20% of patients are still symptomatic and 4% suffer severe symptoms. By 2 years, 14%
remain symptomatic and a further 4% are severely disabled.12 Some 0.5–1% of the
population have chronic neck pain ascribable to a motor vehicle accident.12
The natural history of neck pain without neurological deficit and not caused by
 The neck 265

axis

S - shape 100 ms

Figure 2. Tracings of X-rays of the cervical spine at 90 ms and 110 ms after impact of volunteers under-
going a whiplash impact. The cervical spine develops an S-shaped distortion, during which the C5 vertebra
undergoes posterior sagittal rotation about an abnormally high axis of rotation. This results in separation of
the anterior ends of the vertebral bodies (large arrowhead) and impaction of the zygapophyseal joints (small
arrowhead). (Based on Kaneoka et al, 1999.13)

malignancy or rheumatoid arthritis is modestly benign. Within 10 years, some 43% of

patients will recover, but 25% will continue to have moderate symptoms, and 7%
become severely disabled.21,22

RISK FACTORS

Factors that have been refuted as risk factors for neck pain, because of a lack of
significant correlation or on the grounds of odds ratios barely greater than 1.0, are
degenerative disc disease17, zygapophyseal osteoarthrosis17, smoking16, socio-economic
status23 and prolonged sitting at a work station.24 Factors found to be positively
associated with neck pain are educational level, occupation16, physical stress at work16,
mental stress at work16 and working with machines.24
Psychological factors have been found explicitly not to be associated with neck
pain.25,26 Factors found to be related to neck pain upon univariate analysis disappear
upon multivariate analysis. The factors that survive relate not to personal psychological
factors but to occupational factors such as the ability to vary workload, co-operation
between employees and demands that are too great.27
266 N. Bogduk

Practice point
● occupational factors are risk factors for neck pain. The contribution of
psychological factors is controversial

Prognostic factors for neck pain of unknown origin have not been determined, but
for neck pain after whiplash, certain factors are predictive. In rank order of decreas-
ing significance, they are impaired neck movements, a previous history of headache, a
history of head trauma, age, the severity of the initial pain, the intensity of the initial
headache and abnormal scores on tests that measure nervousness, neuroticism and
focused attention.28 Although commonly regarded as a prognostic factor for chronic
neck pain after whiplash, degenerative changes are not significant when corrected for
age.12
Little more than speculation and anecdote suggest that the symptoms of whiplash
are the result of ‘litigation neurosis’. Competent follow-up studies of whiplash patients
report that the likelihood of the chronicity of symptoms following whiplash injury is
independent of litigation.12 The formal study of litigation neurosis or compensation
neurosis unearths little evidence in support of the concept but reveals a plethora of
reports demonstrating that compensation patients are no different from non-
compensation patients.12

Practice point
● the evidence refutes the notion that compensation neurosis is the basis of
chronic neck pain after whiplash

Some authors have maintained that pre-existing personality traits or psychiatric

problems create or contribute to somatic symptoms after whiplash injury.12 These
notions have, however, been refuted by a Swiss study29 in which consecutive whiplash
patients were assessed, using standard tests, for psychosocial stress, negative affectivity
and personality traits soon after suffering a whiplash injury. None of these factors was
found to be predictive of the persistence of symptoms at 6 months, 12 months and 2
years. A more recent study has consolidated these findings, finding that personality
traits and psychiatric symptoms measured a mean of 1.7 days after a whiplash injury
did not predict the presence of symptoms at 6 months.30 There is, therefore, no real
evidence that either malingering for financial gain, or pre-existing psychosocial factors,
contribute in any significant way to the natural history of whiplash injury. The unavoid-
able conclusion is that the majority of whiplash injuries result in real, organic lesions
in genuine patients.

HISTORY

As for any pain problem, a comprehensive history can be obtained systematically by

describing the pain in terms of the categories listed in Figure 3. For common neck pain
of unknown origin, the history will be unremarkable. The pain may or may not have an
attributable cause. It is felt high, low, unilaterally, bilaterally or all over the neck, and it
 The neck 267

Duration of illness
Circumstances of onset Duration
Mode of onset Time of onset
Site of pain Precipitating factors
Radiation Aggravating factors
Quality Relieving factors
Frequency Associated features

Figure 3. Categories of enquiry for taking a history of neck pain.

may spread to the head, chest, shoulders or upper limb. It is dull or aching in quality
and is present most of the time, aggravated by neck movements and usually somewhat
relieved by rest. There are usually no associated features.
Deviations from this archetypal pattern constitute the warning signs of red flag
conditions. Unrelenting and severe pain that is spontaneous in onset warrants the
consideration of a sinister cause, as does night pain, pain not relieved by rest and pain
not aggravated by movement. However, it is in the context of associated features that
the most valid alerting features occur.
Neurological symptoms or signs warrant investigation in their own right, regardless
of their inferred relationship to pain. However, in that regard, the use of medical
imaging or electrophysiological tests in the pursuit of a neurological condition should
not be confused with the investigation of neck pain. Those tests are appropriate for
neurological conditions but are not indicated when neck pain alone is the sole present-
ing feature.
Other warning signs are historical or clinical features of general medical conditions
affecting the cardiovascular and other body systems, which might offer clues to
conditions such as aneurysm, tumour and infection. In this respect, Table 2 provides a
checklist that serves as a reminder of the possibilities. If there are no positive
responses to this checklist, the likelihood of a red flag condition is vanishingly small.

PHYSICAL EXAMINATION

Formal studies have shown that, with respect to neurological examination of the neck
and upper limb, traditional techniques are quite reliable and have reasonable validity for
the detection of radiculopathy.31,32 However, with respect to neck pain with no associ-
ated neurological features, physical examination has proven neither reliable nor valid.
A physician might examine a patient with neck pain in terms of the location of the
pain, tenderness and the range of movement. Those trained in manual skills might
examine intersegmental motion. Short of a general medical examination, which should
be normal, little else is appropriate for a patient presenting with just neck pain.
If the patient presents with a distinct region of pain (see Figure 1 above), a reason-
able inference can be drawn on the segmental location of the pain, but the actual
source cannot be determined. Other features or the results of investigations are
required to determine whether the source is a muscle, a joint or an intervertebral
disc.
Studies of tenderness in the neck demonstrate that good agreement can be
achieved33–35, but the validity of this feature is unknown. It has not been correlated
with any known pathology or any proven source of pain.
268 N. Bogduk

Table 2. A comprehensive checklist of associated features of neck pain that

might indicate a red flag condition
System Past history Symptoms of Feature or condition
Nervous Yes No Yes No Weakness
Yes No Yes No Numbness
Yes No Yes No Bladder dysfunction
Yes No Yes No Impaired balance
Yes No Yes No Impaired vision
Yes No Yes No Altered speech
Yes No Yes No Disorientation
Yes No Yes No Altered consciousness
Cardiovascular Yes No Yes No Risk factors
Yes No Yes No Chest pain
Yes No Yes No Anticoagulants
Yes No Yes No Transient ischaemic
attacks
Respiratory Yes No Yes No Carcinoma
Yes No Yes No Tuberculosis
Yes No Yes No Cough
Yes No Yes No Weight loss
Alimentary Yes No Yes No Carcinoma
Yes No Yes No Weight loss
Yes No Yes No Loss of appetite
Yes No Yes No Dysphagia
Yes No Yes No Diarrhoea
Yes No Yes No Altered bowel habits
Urinary Yes No Yes No Incontinence
Yes No Yes No Obstruction
Reproductive Yes No Yes No Breast lump
Yes No Yes No Uterine dysfunction
Endocrine Yes No Yes No Thyroid cancer
Hyperparathyroidism
Reticuloendothelial Yes No Yes No Lymph nodes
Skin Yes No Yes No Rash
Musculoskeletal Yes No Yes No Other joint pain
Yes No Yes No Other muscle pain
Age Yes No Yes No Risk of Paget’s disease
Yes No Yes No Risk of myeloma

Research agenda
● there is a need to improve the reliability and to establish the validity of
various clinical signs of neck pain

Although trigger points might be considered to be a valid feature in fibromyalgia,

they are neither reliable nor valid in neck pain. The literature itself is internally incon-
sistent: trigger points in the neck are exempt from the classical diagnostic criteria.10
They constitute no more than sites of tenderness, which often overlie zygapophyseal
joints. Thus, trigger points cannot be distinguished from tender joints.10
 The neck 269

A restricted range of motion distinguishes patients with neck pain from asympto-
matic subjects but has no greater discriminating power. A decreased range of motion
is to be expected with any cause of neck pain and is not indicative of any particular
diagnosis.

Practice point
● there is little that physical examination can achieve with respect to diagnosing
neck pain

In the hands of chiropractors, agreement on special signs of segmental dysfunction

is poor.36 Whatever these features mean, they are unreliable. In the hands of one
physiotherapist, the examination of passive intervertebral movements has been
reported as being valid for the detection of painful zygapophyseal joints, but the validity
of this study has been questioned.36

MEDICAL IMAGING

Fear of missing a tumour and fear of missing a fracture are perhaps the most common
reasons that medical practitioners have for requesting medical imaging of the neck.
Neither is a justified reason in the absence of historical or clinical features of a red flag
condition or serious neck trauma. In a patient with uncomplicated neck pain, there is
nothing that might be shown on plain films, computed tomography (CT) or magnetic
resonance imaging (MRI) that correlates with neck pain.
A British study of 1263 patients referred for cervical spine radiography found no
unexpected findings of malignancy or infection on any of the films.4 An American study
of 1146 radiographic examinations found that in no patient was a serious diagnosis,
including acute fracture, dislocation or tumour detected.37 A 5-year follow-up
concluded that ‘no medically dangerous diagnoses would have been missed if the
cervical spine series had not been done’.37 These studies indicate that the prevalence
of clinically unsuspected red flag conditions is less than 0.38%.

Practice point
● medical imaging is not justified for neck pain unless there are features in the
history of a serious condition

Imaging is justified only if there are clinical grounds for suspecting a red flag
condition (see Table 2 above). However, in that event, imaging should be ordered
judiciously. Plain films may not be the appropriate investigation. Depending on the
condition suspected, a bone scan or an MRI may be the investigation of choice.
In patients with a history of trauma and suspected neck injury, the prevalence of
fractures is less than 3%.11 As clinical signs of fracture, an immediate onset of pain,
midline cervical tenderness and an impaired range of motion are 100% sensitive but
only marginally or poorly specific (Table 3). These features, therefore, have only
270 N. Bogduk

Table 3. The validity of clinical signs as indicators of radiographically evident cervical spine
fractures in patients with suspected neck injury. (Based on McNamara, 1998.11) LR, likelihood ratio.
Fracture present
Sign Present Yes No Sensitivity Specificity LR
Immediate onset of pain Yes 7 119 1.00 0.65 2.9
No 0 225
Midline tenderness Yes 5 176 1.00 0.48 1.9
No 0 164
Impaired range of motion Yes 2 137 1.00 0.60 2.5
No 0 204
Neurological signs Yes 1 7 0.17 0.98 8.5
No 6 319
Head injury Yes 6 102 0.86 0.70 2.9
No 1 239
Loss of consciousness Yes 5 21 0.71 0.94 11.8
No 2 318
Neurological symptoms Yes 0 36 0.00 0.89 0.00
No 7 305

marginally useful likelihood ratios. Because of their higher specificity, neurological

signs and loss of consciousness have much higher likelihood ratios, but these are still
far too small to overcome the low pre-test probability of cervical fracture. Even in
patients with neurological signs, the chance of a fracture being the cause is still very
low.
Thus, there are no clinical means of avoiding the fear of missing a fracture. Despite
the low risk of fracture in patients with neck pain after trauma, the available data do
not permit the development of guidelines to be followed that would reduce unneces-
sary radiography without the risk of missing patients with fractures.

Research agenda
● there is an urgent need to develop and test clinical indicators that might
reduce the overuse of cervical radiography

There is no evidence to justify the use of CT scans as a primary investigation in

patients with neck pain who exhibit no history or clinical signs of the causative lesion.
The only demonstrated role of CT is the definition of lesions already evident on plain
X-rays.
No studies have demonstrated a utility for MRI in the investigation of uncomplicated
neck pain. Features such as herniated, bulging or narrowed intervertebral discs, and
foraminal stenosis, are common even in asymptomatic individuals38,39 and cannot be
held to be diagnostic. In patients with neck pain after whiplash, several studies have
repeatedly shown that MRI does not demonstrate any abnormalities consistent with
or diagnostic of the cause or source of pain.40–42 MRI reveals only those features
evident in asymptomatic individuals.
 The neck 271

There is emerging evidence that, perhaps, single position emission computed

tomography (SPECT) scanning might be of value in the early investigation of post-
traumatic neck pain. A retrospective review of 35 patients with neck pain following
motor vehicle accidents, sporting injuries or blunt head trauma compared what could
be demonstrated on SPECT scanning with what had been seen on previous plain
X-rays.43 In 16 patients with normal cervical X-rays, SPECT revealed three to have
occult fractures. In 10 patients with osteoarthrosis on cervical X-rays, SPECT revealed
five with a fracture or post-traumatic arthropathy. The nature of the fractures was not
fully specified, but one example of a fracture of the lateral mass of C1 was illustrated.
In nine patients with a suspected fracture on plain films, active fractures were excluded
in six and confirmed in two. The remaining patient exhibited non-specific activity. This
study implied that the fractures found were small and stable as none of the patients
exhibited neurological signs or deteriorated. At follow-up, 95% of the patients with
normal SPECT scans had recovered by 6 months, but of those with an abnormal
SPECT scan, 44% had continuing symptoms by the same time.
The authors recommended SPECT scanning between 2 and 6 weeks after the onset
of pain in patients with continuing symptoms.43 Although finding a small fracture may
not influence the patient’s natural history, it has a putative bearing on management.
Surgical management is not indicated for such small fractures, but immobilization may
be preferable to physiotherapy. Indeed, it was the investigators’ impression that those
patients with fractures fared better when immobilized than when treated with physio-
therapy.43

INVASIVE TESTS

The attraction of medical imaging is that the tests are readily available and involve no
stress to the patient. Conventional medical imaging does not, however, establish a
diagnosis for the vast majority of cases of neck pain. If a diagnosis is to be pursued,
invasive tests are required. These involve introducing a needle, under fluoroscopic
control, into structures believed to be the source of pain or onto their nerve supply,
in order either to provoke or to anaesthetize the putative source of pain. In the case
of neck pain, two procedures have been developed and assessed: disc stimulation and
zygapophyseal joint blocks.
Disc stimulation is the modern name for discography. The new term emphasizes
that it is not the radiographic appearance of the disc that is diagnostic but the patient’s
response to stimulation of that disc. Disc stimulation is designed to test whether an
intervertebral disc is the source of a patient’s pain. This is determined by distending
the disc with an injection of normal saline or contrast medium.8 In order to ensure
validity, multiple discs must be stimulated in order to provide control observations.
The primary criterion for a positive response is a reproduction of the patient’s pain
upon stimulation of a given disc providing that stimulation of the adjacent discs does
not produce pain.
Although well regarded by some practitioners, particularly in the US, disc
stimulation can be a capricious test. It is subject to false positive responses. Even in
normal volunteers, cervical discs can be painful when distended.44 However, the pain
is relatively minor. Nevertheless, because of this, an additional criterion for a positive
response to disc stimulation in patients is that the evoked pain must be rated as ‘7’ or
greater on a 10-point visual analogue scale. Despite this, disc stimulation can still
provide false positive results. Some 40% of patients who ostensibly have painful discs
272 N. Bogduk

have their pain relieved by local anaesthetic blocks of their zygapophyseal joints.7 This
arises, it would seem, because distension of a disc induces painful movement in the
zygapophyseal joints of the same segment.
Apart from these diagnostic exigencies, cervical disc stimulation has no demon-
strated therapeutic utility. It is used as an indication for cervical fusion, but no studies
have convincingly demonstrated the efficacy of cervical fusion in the treatment of neck
pain, or the predictive validity of disc stimulation. It may have a virtue, however, in
preventing unnecessary surgery. If disc stimulation reveals multiple symptomatic discs,
surgery is not indicated.

Practice point
● zygapophyseal joint pain is the single, most common basis for chronic neck
pain after whiplash but can be diagnosed only by anaesthetizing the painful
joint.

Zygapophyseal joint blocks are another matter. In this test, the nerves that
innervate a putatively painful joint are anaesthetized.6 A complete relief of pain
constitutes presumptive evidence that the joint is the source of the patient’s pain.
However, single, diagnostic blocks carry a false positive rate of 27%.6 For this reason,
control tests are mandatory in each case. These controls can be comparative local
anaesthetic blocks using different agents on different occasions, or placebo controls.6
When used under strict, double-blind, controlled conditions, zygapophyseal joint
blocks have revealed that zygapophyseal joint pain is the single most common basis for
chronic neck pain after whiplash.45,46 On worst-case analysis, it accounts for some 50%
of cases and may explain up to 65% of cases in total; however, the diagnosis cannot be
established other than by controlled diagnostic blocks. The C5–6 and C2–3 levels are
the ones most commonly affected. If the diagnosis is established, patients can be
treated with percutaneous radiofrequency neurotomy (see below).

TREATMENT

Every treatment for neck pain works: a treatment would not be used unless some-
one, at least, thought it worked. The critical, supplementary questions, however, are:
does the treatment work better than something else, by how much, and for how
long?
A treatment that nonetheless works may not be worthwhile if it is more hazardous,
more complex or more expensive than a simpler treatment, nor may a treatment that
is only 5% better than a comparison treatment. A treatment may not be worthwhile
if its effects are extinguished upon cessation of the therapy, especially if it is complex,
labour-intensive and costly. In this context, ‘complex’ refers not to the intricacies of
the therapy but to aspects such as the equipment and facilities required and the extent
and duration of training required for the therapist.
A further issue is the influence of confounding factors. When a therapist claims that
a treatment works, the imputation may be false. Rather than the professed treatment,
it could be the conviction and enthusiasm of the therapist, or the time spent with the
patient, that exerts the therapeutic effect. Indeed, in the case of physical therapies,
 The neck 273

Feine and Lund47 found that time spent in therapy was the most consistent factor
associated with reported success.
Perspicacious consumers of health information will be aware of the various issues
concerning the efficacy of treatments and will not accept that a treatment works
simply because someone says it does. Instead, they will ask for the evidence that
supports this claim.

Evidence
In promoting guidelines for therapy, the National Health and Medical Research Council
of Australia48 defined four grades of evidence (Table 4). The highest ranking types of
evidence were results of systematic reviews (grade I) and the results of randomized
controlled trials (grade II). Consensus reports of learned individuals were ranked only
as grade IV.

Table 4. Rating scale for quality of evidence recommended by the Australian National Health and
Medical Research Council.48
Grade Definition
I Evidence obtained from a systematic review of all relevant randomized controlled trials
II Evidence obtained from at least one properly designed randomized controlled trial
III–1 Evidence obtained from well-designed controlled trials without randomization
III–2 Evidence obtained from well-designed cohort or case control analytical studies, preferably
from more than one centre or research group
III–3 Evidence obtained from multiple time series with or without the intervention
IV Opinions of respected authorities, based on clinical experience, descriptive studies or the
reports of expert committees

This system provides a means of determining how well sustained any claim of
efficacy might be. Claims supported by grade I and grade II evidence are more credible
than hearsay, anecdotes, case reports or even consensus reports. Grade III or IV
evidence is not a substitute for contradicting grade I and II evidence. Recognizing that
the evidence is only grade III or grade IV warns the consumer that efficacy is far from
proven.

Physical therapies
Pragmatic reviews have in the past extolled the virtues of a variety of treatments for
neck pain.49–51 These include education, rest, collars, posture control, exercises,
physical modalities, traction, manipulation, massage, analgesics, tricyclic anti-
depressants, psychological interventions, trigger point injections, occipital nerve
blocks, epidural steroid injections, neurectomy, discectomy, fusion, soft tissue tech-
nique, muscle energy technique, thrust technique, myofascial release, manipulation
under anaesthesia and craniosacral manipulation. None of the reviews, however,
provided any scientific evidence of efficacy of any of these traditional interventions.
274 N. Bogduk

Grade I evidence
The Quebec Task Force on Whiplash Associated Disorders attempted a major
synthesis of the literature on neck pain after whiplash but was unable to compose a
systematic review or meta-analysis.52 Out of 10 382 potentially relevant articles, it
found 1204 that met its criteria for preliminary screening, of which only 294 were
finally found to be suitable for further review; of these, however, only 62 were
accepted as being both relevant and scientifically meritorious. As a result of this
process, the Task Force concluded that the evidence was found to be sparse and
generally of unacceptable quality.52
For the conservative therapy of neck pain not specifically related to whiplash, there
has been only one grade I study.53 It detailed the positive and negative studies, and where
possible calculated the effect-sizes of various therapies. Some therapies were reported
as having no effect above comparison therapies; others were found to have a positive
effect. However, the review did not comment on the duration and durability of positive
effects or on their clinical significance. It simply accepted any reported positive effects,
which were usually in the form of mean or median scores on a visual analogue scale. The
review did not address other measures of outcome such as the proportion of patients
who were totally relieved of their pain. It was satisfied with statistically significant
improvement beyond control as the cardinal, if not only, measure of successful therapy.
Despite its graciousness in this regard, the review nonetheless concluded that ‘there is
early evidence to support the use of manual treatments in combination with other
treatments for short term pain relief, but in general, conservative interventions have not
been studied in enough detail to assess efficacy or effectiveness adequately’.53
There has been a grade I study of the evidence for manipulative therapy and
mobilization for neck pain.54 The study found no randomized controlled studies of
manipulation for acute neck pain and only four studies of manipulation for chronic
neck pain. For mobilization, it found three studies for acute neck pain and one for
chronic neck pain. In order to distill a statistically significant effect, the review pooled
studies on headache as well as studies on neck pain, and pooled studies that indi-
vidually failed to achieve statistical significance. The review, although positive, was
guarded in its conclusions: ‘cervical spine manipulation and mobilisation probably
provide at least short-term benefit for some patients with neck pain and headache’.54

Grade II evidence
Tables 5 and 6 summarize the results of controlled studies of physical treatments for
neck pain and for neck symptoms after whiplash. Conspicuous is the lack of any studies
of the management of chronic neck pain after whiplash and the lack of any compelling
proof of efficacy of any therapy.

Practice point
● no conservative therapy has been tested, let alone proven, for chronic neck
pain after whiplash

For acute neck pain after whiplash, certain treatments used as controls in random-
ized studies have, by implication, been demonstrated to be less effective than index
therapies. These include rest and analgesia55, combinations of transcutaneous electrical
 The neck 275

Table 5. A summary of the evidence on therapy for acute and chronic neck
pain after whiplash, without comment on the quality of the studies involved
and the magnitude of effect and its duration
Evidence
Therapy Acute neck pain Chronic neck pain
Electromagnetic therapy = Temporary benefit No data
Traction No added benefit No data
Collars = Traction and exercise No data
Rest and analgesia ? = Natural history No data
TENS or ultrasound < Rest and analgesia No data
Multimodal therapy > TENS or ultrasound No data
> Rest and analgesia No data
Mobilization > Rest and analgesia No data
Home exercise > Rest and analgesia No data
Tailored physiotherapy > Rest and analgesia No data
= Home exercise No data
TENS, transcutaneous electrical nerve stimulation.

Table 6. A summary of the evidence on therapy for acute and chronic

neck pain not caused by whiplash, without comment on the quality of the
studies involved and the magnitude of effect and its duration.
Evidence
Therapy Acute neck pain Chronic neck pain
Exercise Effect uncertain No data
Collars ? = Natural history No data
TENS ? = Natural history No data
Neck school = No treatment No data
Spray and stretch = No benefit No data
Laser = Placebo No data
Traction = Analgesics No data
Magnets No data = Placebo
Acupuncture No data = Sham TENS
Physiotherapy No data = Acupuncture
Electromagnetic therapy = Moderate analgesia
Manipulation No data = Drugs
= Physiotherapy or GP
Mobilisation No data ≥ Salicylates
TENS, transcutaneous electrical nerve stimulation. GP, management by
general practitioner.
276 N. Bogduk

nerve stimulation (TENS), ultrasound and pulsed electromagnetic therapy56, and soft
collars.55,57,58 One study of traction found no added benefit when various forms of
traction were added to a regimen of instruction, moist heat and a programme of
exercises.59 Another found no added benefit from intermittent traction and exercise
instruction over 2 weeks’ rest in a standard or moulded collar.60
A good-quality study compared the effects of pulsed electromagnetic therapy
delivered from a device implanted in a collar with the effects of wearing the same collar
with a dummy device.57 All the patients wore their collars for 8 hours a day for 12
weeks, also being given analgesics and instructions to mobilize their necks hourly. For
the first 4 weeks of treatment, the active intervention group exhibited a greater drop
in pain score, but thereafter the control group also improved, so that by 12 weeks
there was no statistically significant difference in pain scores. No data were presented
concerning the condition of any patients after the cessation of treatment at 12 weeks.
The results of positive studies for acute neck pain after whiplash are summarized
graphically in Figure 4. In one study, patients received either rest and analgesia, active
outpatient physiotherapy (consisting of a tailored programme of hot and cold appli-
cations, short-wave diathermy, hydrotherapy, traction and mobilization) or a home
exercise programme.55 In terms of the severity of pain, both the physiotherapy group
and the home exercise group were more improved than the rest and analgesia group
but there was no statistically significant difference between the physiotherapy and
home exercise groups.

10

V 6

A 5 TENS, EMT, US

S 4 R&A

3 R&A

2 PT
Home Collar Multimodal
1 Mob

0 4 8 12 16 20 24

WEEKS

Figure 4. The efficacy of several physical therapies and control therapies in the treatment of acute neck
pain after whiplash, in terms of decreased median or mean scores on a visual analogue scale (VAS). The bold
lines refer to the index therapies, the dotted lines to the control treatment. TENS, transcutaneous electrical
stimulation.56 EMT, electromagnetic therapy.56 US, ultrasound.56 R&A, rest and analgesia.55,58 PT, tailored
physiotherapy.55 Home, advice on a home exercise programme.55 Mob, passive mobilization.58 Collar, collar
and active mobilization.58 Multimodal, multimodal therapy.56
 The neck 277

In another study, the active intervention was applications of ice for the first 24
hours, followed by mobilization, local heat, exercises and analgesia. The reference
treatment was rest, analgesia and a soft collar.58 At 4 weeks and 8 weeks, the active
intervention group showed statistically significant improvements in pain score and
range of movement.
A more recent study combined the apparent virtues of the therapies described in
the aforementioned studies. The intervention was ‘multimodal therapy’, consisting of
relaxation training, the reduction of cervical lordosis, psychological support, eye
fixation exercises and massage and mobilization.56 The reference therapy consisted of
applications of TENS, pulsed electromagnetic therapy, ultrasound and calcic ionto-
phoresis. Over 2 weeks, 1 month and 6 months, the median pain scores in the
reference group steadily declined, but improvements in the active intervention group
were statistically significantly greater. No patients had totally recovered at 2 weeks
after treatment; at 6 months, however, six out of 30 in the intervention group, but
none in the reference group, had fully recovered. This is the only study to have
reported the proportion of patients who fully recovered as opposed to just improved.
Noticeable in Figure 4 is that, in different studies, patients reported quite similar initial
visual analogue pain scores. Thus, the studies would seem to have treated similar
patients. Furthermore, different studies using the same or similar interventions reported
similar rates and degrees of improvement. Thus, there is a remarkable consistency
between studies. The singular exception is the poor recovery of patients treated with
TENS, ultrasound and pulsed electromagnetic therapy.56 However, the simplest inter-
vention (rest and analgesia), despite being used as a control therapy, is nevertheless asso-
ciated with substantial improvement and is equipotent to using a collar and instructions
to mobilize. Since none of these therapies is designed to modify disease or injury, their
efficacy most probably reflects the natural history of neck pain after whiplash.
Mobilization, tailored physiotherapy and multimodal therapy do achieve a greater
improvement than rest and analgesia, but the data are strong for only the initial 8
weeks after injury. The only study to extend treatment to 6 months indicates that the
improvement at 6 months is about the same as that attained at 8 weeks. Interestingly,
the same level of achievement is also attained at 12 weeks by using a collar.
Importantly, Figure 4 is illusory as it does not show the number of patients fully
recovered. The studies depicted reported neither the number of patients who
recovered fully nor the number who did not benefit at all. The graphs depict only an
improvement in median or mean pain score. That means that, although some patients
may have obtained a complete relief of pain, others could well have obtained no relief.
None of the studies provided data on the range or standard deviations of the pain
scores reported so the range or variance of response could not be illustrated.
Most critically, written advice on exercise was superior to rest and analgesia and as
efficacious as tailored physiotherapy.55 In this regard, a supplementary report61
provided data on a 2-year follow-up. It showed that the proportion of patients fully
recovered was statistically significantly greater among those who received home
exercise (77%) than among those who received rest (54%) or tailored physiotherapy
(56%), and there was no difference between the two latter groups.

Practice point
● home exercise is the only treatment for acute neck pain that has been shown
to have significantly better lasting benefit than other more passive therapies
278 N. Bogduk

In essence, the combined data on physical and manual therapy do show that active
intervention is superior to rest and analgesia, but the impact pertains only to an earlier
decrease in average pain score and is evident only at 4–8 weeks after onset of pain.
Thereafter, either there are no data or the suggestion is that patients not treated
actively eventually achieve the same degree of recovery. The impact of physical and
manual therapy is to achieve not resolution in a greater proportion of patients but a
more rapid resolution in the early weeks after the onset of pain. The only long-term
data indicate that a greater proportion of patients achieve complete recovery when
treated with home exercise.
In relation to this a recent study offered a dramatic result.62 Patients seen upon
presentation to an emergency clinic with a whiplash injury were all treated with non-
steroidal anti-inflammatory drugs (NSAIDs) for 5 days and were assigned to receive
either 14 days sick leave and a soft collar, or an index treatment. At 6 months follow-
up patients who received the index treatment were significantly better in terms of
neck pain, pain during activities of daily living, attention and memory. The index
treatment consisted of no more than instruction to act as usual and receive no sick
leave.
The literature on acute neck pain not caused by whiplash is no more encouraging
(see Table 6 above). No benefit over placebo has been found for spray and stretch
therapy63 or laser therapy.64 Adding traction offers no advantage to other inter-
ventions65, and the effects of traction plus analgesics, or of isometric exercises plus
analgesics, are no greater than those of analgesics alone.66 A small study in patients
with acute neck pain found that the pain resolved rapidly, within 1 week, with no
significant differences between the effects of manual therapy, TENS, or a soft collar
coupled with analgesics.67
For chronic neck pain, wearing magnetic necklaces has been shown to be no more
efficacious than a placebo.68 No difference in outcome was recorded in a comparison
of ‘neck school’ with no treatment69, and infra-red laser therapy was no more effective
than mock TENS.70
In one study, manipulation combined with azapropazone and the option of the
injection of tender points with corticosteroids with or without local anaesthetic was
compared with treatment with azapropazone alone.71 There was a significantly greater
immediate effect on pain and stiffness in the manipulation group but no significant
difference at 1 week or 3 weeks. In another study, there was no significant difference
at 3 weeks between patients receiving 1–3 manipulations and those treated with a
muscle relaxant.72
The one study of strong methodological quality that addressed manipulation for
chronic neck pain was one that studied both patients with neck pain and patients with
back pain.73 Data on the neck pain patients alone were not reported, but a subsequent
interrogation of the original data revealed that there were 64 patients with chronic
neck pain. No statistically significant differences were detected between those
receiving manual therapy and those receiving physiotherapy and general practice care.54
One study found diode laser therapy to be superior to sham laser therapy74,
although curiously, unlike any other study of pain therapy, this study found absolutely
no improvement in the control group; a placebo response was totally lacking. A poorly
reported study claimed that acupuncture afforded greater average improvement than
physiotherapy for cervical spondylosis75, but insufficient raw data were provided to
enable a statistical analysis. However, the proportions of patients achieving 100% relief
were not significantly different between the two treatment groups. Another study of
acupuncture found it to be no more effective than sham TENS.76,77
 The neck 279

In a methodologically strong study78, pulsed electromagnetic therapy delivered

through a collar was found to have a significantly greater effect on pain than a collar
alone. The difference arose, however, because of a greater proportion of patients in
the active treatment group reporting ‘moderately better’ as opposed to ‘no change’ in
the control group. The number of patients ‘completely well’ or ‘much better’ was not
significantly different in the two groups. The study did not report the duration of effect
for any period after the cessation of treatment.
A comparison study found in favour of active as opposed to passive physiotherapy
but only on the grounds of marginally greater improvements in the strength of certain,
but not all, movements of the neck, and in the recurrence rates of neck pain and
headache.79
A study compared treatment with salicylate alone, salicylate plus TENS, traction,
massage and information, and salicylate, information and mobilization. It reported a
significantly greater proportion of patients in the mobilization group reporting slight
or no pain, but only at 1 week after the cessation of treatment.80 An analysis of the
data provided, however, reveals that the significant difference arises only when
treatment with mobilization is compared with treatment with massage and traction.
There was no statistically significant difference in outcome between the patients
treated with mobilization and those treated with salicylate alone and put on a waiting
list. In a later publication of the same data81, greater details of the grades of outcome
were provided. Although the author reported a statistically significant trend in favour
of the mobilization group, the published data showed no significant difference between
the groups. Such differences as tended to occur were caused by the poorer outcome
in the TENS traction massage group.
In essence, for acute neck pain, commonly used therapies are unsupported by the
literature. For some therapies, it is not that they have not been studied; they have—and
they have been refuted. Shown not to be efficacious are electromagnetic therapy,
traction, collars, TENS and ultrasound, neck school, spray and stretch, and laser therapy.
These interventions are no better than rest and analgesia or no treatment at all.
Mobilization, multimodal therapy and tailored physiotherapy are all better than rest
and analgesia but are equipotent to home exercise, and home exercise has been the
only therapy shown to achieve a significantly greater proportion of patients with
complete resolution of their neck pain in the long term.
For chronic neck pain, no studies have shown any physical or manual therapy to be
of benefit for neck pain that persists after whiplash. Nor does the literature on the
treatment of neck pain unassociated with whiplash provide data to fill this deficiency.
Physiotherapy is as good as acupuncture, and acupuncture is a good as sham TENS.
The literature on manipulation has not shown a superiority over analgesics.

Research agenda
● if existing therapies are to be vindicated, criteria need to be developed to
determine which patients are most likely to respond. Otherwise, totally novel
approaches are required

Intent on verifying a therapeutic effect of manipulation/mobilization, some authors

have recognized that perhaps the published studies were individually too small to have
enough statistical power to demonstrate a therapeutic effect.54 Accordingly, they
280 N. Bogduk

pooled data from the available studies, but the data were liberally pooled from studies
of neck pain and studies of headache. As a result, they calculated a pooled effect of
16.2, that is, that manual therapy for chronic neck pain can be expected to achieve an
extra 16.2% reduction in pain. This calculation indicates that manual therapy may well
have a therapeutic effect, but it does not substitute for a demonstration of this effect
in a randomized controlled trial with sufficient statistical power of its own. Moreover,
although a therapeutic effect has been calculated, the calculation offers no indication
of the longevity of the effect and no indication of how many patients are completely
relieved of their pain.

Medical and surgical therapy

There have been no primary studies of analgesics or NSAIDs for neck pain. Such
information as is available is secondary in nature in that drugs have often been used as
the comparison or control therapy in studies of physical and manual therapy. Salicylate
appears to be effective for neck pain, but an effect distinguishable from placebo cannot
be discerned.80

Practice point
● there are no proven medical therapies for neck pain

Research agenda
● unpublished, anecdotal reports maintain that atraumatic neck pain, particularly
in the elderly, responds well to intra-articular steroids. This requires formal
assessment

The use of intra-articular injections of corticosteroids in the treatment of chronic

cervical zygapophyseal joint pain has been shown to confer no benefit over the use of
local anaesthetic alone.82 Although some patients report lasting relief, regardless of the
agent used, they constitute a minority. Approximately 80% of patients lose their pain
relief within 20 days of treatment, only some 10% of patients sustaining relief for more
than 80 days.
Although cervical fusion has been advocated for neck pain83,84, there are no
controlled studies that indicate an effect superior to that of conservative therapy or
simply analgesia. A study of 20 patients followed for 1–8 years after anterior interbody
fusion found that two patients achieved a good result, 9 patients a fair result and 9
poor results.85

Radiofrequency neurotomy
Percutaneous radiofrequency medial branch neurotomy is the one surgical therapy for
neck pain that has been subjected to a double-blind, controlled trial, and it is the only
treatment for chronic neck pain after whiplash that has been rigorously tested and
reported.86 The procedure confers complete relief in around 70% of patients
diagnosed as suffering from cervical zygapophyseal joint pain on the basis of placebo-
 The neck 281

controlled, diagnostic blocks of the painful joint, and its efficacy is not the result of a
placebo effect. Although profound and complete, the pain relief is not permanent; it
lasts 263 days, but relief can be reinstated by repeat neurotomy. The procedure, how-
ever, is limited as it is arduous and there is a possibility of technical failure. Its utility
can be corrupted if the technique is used by untrained or inexperienced operators.

Practice point
● radiofrequency neurotomy for zygapophyseal joint pain is the only treatment
for neck pain that affords a complete relief of pain and that has survived a
placebo-controlled, randomized, double-blind trial

SUMMARY

In terms of evidence-based therapy, medical therapies offer no superiority over

physical and manual therapies. Both disciplines lack compelling evidence of their
efficacy. At best, the following guidelines could be recommended:
1. For acute neck pain:
● simple analgesics or NSAIDs might be used to provide analgesia while patients
undergo natural recovery, but in the knowledge that the efficacy of these agents
may be no greater than that of a placebo;
● there is no evidence to justify the use of major tranquillizers or tricyclic anti-
depressants.
2. In the first 8 weeks after onset:
● patients could be treated with rest and analgesia, but
● a more rapid resolution of pain might be achieved by the use of ice, and passive
mobilization; although
● a home exercise programme offers just as much chance of rapid resolution and a
greater chance of being pain free at 2 years.
● instruction to resume normal activities is as efficacious as other measures.
3. In the face of refuting data, and given the availability of a proven alternative:
● traction, electromagnetic therapy, collars, TENS, ultrasound, neck school, spray
and stretch, laser therapy and traction are not justified.
4. For chronic neck pain:
● there is to date insufficient scientific evidence on which to judge the role of
physical therapy;
● the use of analgesics might be justified on humanitarian grounds but not on the
basis of evidence of efficacy;
● If NSAIDs are used, their efficacy should be determined and patients carefully
monitored for side-effects;
● intra-articular injections of corticosteroids into the cervical zygapophyseal joints
are not indicated;
● disc excision and fusion for neck pain without neurological signs might be
entertained on the basis of uncontrolled clinical trials of this procedure, but this
form of therapy is best reserved for the context of clinical trials approved by an
ethics committee and designed to determine its efficacy and safety.
 282 N. Bogduk

● radiofrequency neurotomy can be used to provide a complete relief of pain in

patients rigorously diagnosed as suffering from cervical zygapophyseal joint pain.

Acknowledgement
The preparation of this chapter was supported by the National Musculoskeletal Medicine Intitiative, a
project funded by the Commonwealth Government of Australia in the interests of fostering evidence-based
medical practice.

REFERENCES

1. Merskey H & Bogduk N (eds) (1994) Classification of Chronic Pain. Descriptions of Chronic Pain
Syndromes and Definition of Pain Terms, 2nd edn. Seattle: IASP Press.
2. Dwyer A, Aprill C & Bogduk N (1990) Cervical zygapophyseal joint pain patterns. I. A study in normal
volunteers. Spine 15: 453–457.
3. Hochberg M, Borenstein D & Arnett F (1978) The absence of back pain in classic ankylosing spondy-
litis. Johns Hopkins Medical Journal 143: 181–183.
* 4. Heller CA, Stanley P, Lewis-Jones B & Heller RF (1983) Value of x ray examinations of the cervical
spine. British Medical Journal 287: 1276–1278.
5. Friedenberg ZB & Miller WT (1963) Degenerative disc disease of the cervical spine. A comparative
study of asymptomatic and symptomatic patients. Journal of Bone and Joint Surgery 45A: 1171–1178.
* 6. Bogduk N & Lord SM (1998) Cervical zygapophysial joint pain. Neurosurgery Quarterly 8: 107–117.
7. Bogduk N & Aprill C (1993) On the nature of neck pain, discography and cervical zygapophysial joint
pain. Pain 54: 213–217.
8. Bogduk N, Aprill C & Derby R (1995) Discography. In White AH (ed.) Spine Care, vol. 1, Diagnosis and
Conservative Treatment, pp 219–238. St Louis: CV Mosby.
9. Wolfe F, Simons DG, Fricton J et al (1992) The fibromyalgia and myofascial pain syndromes: a
preliminary study of tender points and trigger points in persons with fibromyalgia, myofascial pain and
no disease. Journal of Rheumatology 19: 944–951.
10. Bogduk N & Simons DG (1993) Neck pain: joint pain or trigger points. In Vaeroy H & Merskey H (eds)
Progress in Fibromyalgia and Myofascial Pain, pp 267–273. Amsterdam: Elsevier.
11. McNamara RM (1998) Post-traumatic neck pain: a prospective and follow-up study. Annals of Emergency
Medicine 17: 906–911.
*12. Barnsley L, Lord S & Bogduk N (1998) The pathophysiology of whiplash. In Malanga GA (ed.) Cervical
Flexion–Extension/Whiplash Injuries. Spine: State of the Art Reviews, vol. 12, pp. 209–242. Philadelphia:
Hanley & Belfus.
13. Kaneoka K, Ono K, Inami S & Hayashi K (1999) Motion analysis of cervical vertebrae during whiplash
loading. Spine 24: 763–770.
14. Jonsson H, Cesarini K, Sahlstedt B & Rauschning W (1994) Findings and outcomes in whiplash-type
neck distortions. Spine 19: 2733–2743.
15. Taylor JR & Twomey LT (1993) Acute injuries to cervical joints: an autopsy study of neck sprain. Spine
9: 1115–1122.
*16. Makela M, Heliovaara M, Sievers K et al (1991) Prevalence, determinants and consequences of chronic
neck pain in Finland. American Journal of Epidemiology 124: 1356–1367.
17. Van der Donk J, Schouten JSAG, Passchier J et al (1991) The associations of neck pain with radiological
abnormalities of the cervical spine and personality traits in a general population. Journal of Rheumatology
18: 1884–1889.
18. Lawrence JS (1969) Disc degeneration: its frequency and relationship to symptoms. Annals of the
Rheumatic Diseases 28: 121–138.
19. Bovim G, Schrader H & Sand T (1994) Neck pain in the general population. Spine 19: 1307–1309.
20. Vasseljen O & Westgaard RH (1995) A case-control study of trapezius muscle activity in office and man-
ual workers with shoulder and neck pain and symptom-free controls. International Archives of
Occupational and Environmental Health 67: 11–18.
21. Gore DR, Sepic SB, Gardner GM & Murray MP (1987) Neck pain: a long-term follow-up of 205 patients.
Spine 12: 1–5.
22. Less F & Turner JWA (1963) Natural history and prognosis of cervical spondylosis. British Medical Journal
2: 1607–1610.
 The neck 283

23. Andersen JH & Gaardboe O (1993) Prevalence of persistent neck and upper limb pain in a historical
cohort of sewing machine operators. American Journal of Industrial Medicine 24: 677–687.
24. Kamwendo K, Linton SJ & Moritz U (1991) Neck and shoulder disorders in medical secretaries. I. Pain
prevalence and risk factors. Scandinavian Journal of Rehabilitation Medicine 23: 127–133.
25. Vasseljen O, Westgaard RH & Larsen S (1995) A case-control study of psychological and psychosocial
risk factors for shoulder and neck pain at the workplace. International Archives of Occupational and
Environmental Health 66: 375–382.
*26. Pietri-Taleb F, Riihimaki H, Viikari-Juntura E & Lindstrom K (1994) Longitudinal study on the role of
personality characteristics and psychological distress in neck trouble among working men. Pain 58:
261–267.
27. Linton SJ & Kamwendo K (1989) Risk factors in the psychosocial work environment for neck and
shoulder pain in secretaries. Journal of Occupational Medicine 31: 609–613.
28. Radanov BP & Sturzenegger M (1996) Predicting recovery from common whiplash. European Neurology
36: 48–51.
29. Radanov BP, Begre S, Sturzenegger M & Augustiny KF (1996) Course of psychological variables in
whiplash injury—a 2-year follow-up with age, gender and education pair-matched patients. Pain 64:
429–434.
30. Borchgrevink GE, Stiles TC, Borchgrevink PC et al (1997) Personality profile among symptomatic and
recovered patients with neck sprain injury, measured by MCMI-I acutely and 6 months after car
accidents. Journal of Psychosomatic Research 42: 357–367.
31. Viikari-Juntura E (1987) Interexaminer reliability of observations in physical examinations of the neck.
Physical Therapy 67: 1526–1532.
32. Viikari-Juntura E, Porras M & Laasonen EM (1989) Validity of clinical tests in the diagnosis of root
compression in cervical disc disease. Spine 14: 253–257.
33. Levoska S, Keinanen-Kiukaanniemi S & Bloigu R (1991) Repeatability of measurement of tenderness
in the neck–shoulder region by a dolorimeter and manual palpation. Clinical Journal of Pain 9: 229–
235.
34. Andersen JH & Gaardboe O (1993) Musculoskeletal disorders of the neck and upper limb among swing
machine operators: a clinical investigation. American Journal of Industrial Medicine 24: 689–700.
35. Hubka MJ & Phelan SP (1994) Interexaminer reliability of palpation for cervical spine tenderness. Journal
of Manipulative and Physiological Therapeutics 17: 591–595.
*36. Gross AR, Aker PD & Quartly C (1996) Manual therapy in the treatment of neck pain. Rheumatic
Diseases Clinics of North America 22: 579–598.
37. Johnson MJ & Lucas GL (1997) Value of cervical spine radiographs as a screening tool. Clinical
Orthopaedics and Related Research 340: 102–108.
38. Boden SD, McCowin PR, Davis DG et al (1990) Abnormal magnetic-resonance scans of the cervical
spine in asymptomatic subjects: a prospective investigation. Journal of Bone and Joint Surgery 72A:
1178–1184.
39. Teresi LM, Lufkin RB, Reicher MA et al (1987) Asymptomatic degenerative disk disease and spondylosis
of the cervical spine: MR imaging. Radiology 164: 83–88.
40. Ellertsson AB, Sigurjonsson K & Thorsteinsson T (1978) Clinical and radiographic study of 100 cases
of whiplash injury. Acta Neurologica Scandinavica 5 (supplement 67): 269.
41. Borchgrevink GE, Smevik O, Nordby A et al (1995) MR imaging and radiography of patients with
cervical hyperextension–flexion injuries after car accidents. Acta Radiologica 36: 425–428.
42. Ronnen HR, de Korte PJ, Brink PRG et al (1996) Acute whiplash injury: is there a role for MR imaging?
A prospective study of 100 patients. Radiology 201: 93–96.
43. Seitz JP, Unguez CE, Corbus HF & Wooten WW (1995) SPECT of the cervical spine in the evaluation
of neck pain after trauma. Clinical Nuclear Medicine 20: 667–673.
44. Schellhas KP, Smith MD, Gundry CR & Pollei SR (1996) Cervical discogenic pain: prospective
correlation of magnetic resonance imaging and discography in asymptomatic subjects and pain sufferers.
Spine 21: 300–312.
45. Barnsley L, Lord SM, Wallis BJ & Bogduk N (1995) The prevalence of chronic cervical zygapophysial
joint pain after whiplash. Spine 20: 20–26.
46. Lord SM, Barnsley L, Wallis BJ & Bogduk N (1996) Chronic cervical zygapophysial joint pain after
whiplash: a placebo-controlled prevalence study. Spine 21: 1737–1745.
*47. Feine JS & Lund JP (1997) An assessment of the efficacy of physical therapy and physical modalities for
the control of chronic musculoskeletal pain. Pain 71: 5–23.
48. Quality of Care and Health Outcomes Committee (1995) Guidelines for the Development and
Implementation of Clinical Practice Guidelines. Canberra: National Health and Medical Research Council.
49. Bisbee LA & Hartsell HD (1993) Physiotherapy management of whiplash injuries. Spine: State of the Art
Reviews 7: 501–516.
 284 N. Bogduk

50. Greenman PE (1993) Manual and manipulative therapy in whiplash injuries. Spine: State of the Art Reviews
7: 517–530.
51. Teasell RW, Shapiro AP & Mailis A (1993) Medical management of whiplash injuries. Spine: State of the
Art Reviews 7: 481–499.
*52. Spitzer WO, Skovron ML, Salmi LR et al (1995) Scientific monograph of the Quebec task force on
whiplash-associated disorders: redefining ‘whiplash’ and its management. Spine 20: 1S–73S.
*53. Aker PD, Gross AR, Goldsmith CH & Peloso P (1996) Conservative management of mechanical neck
pain: systematic overview and meta-analysis. British Medical Journal 313: 1291–1296.
54. Hurwitz EL, Aker PR, Adams AH et al (1996) Manipulation and mobilization of the cervical spine: a sys-
tematic review of the literature. Spine 21: 1746–1760.
55. McKinney LA, Dornan JO & Ryan M (1989) The role of physiotherapy in the management of acute neck
sprains following road-traffic accidents. Archives of Emergency Medicine 6: 27–33.
56. Provinciali L, Baroni M, Illuminati L & Ceravolo G (1996) Multimodal treatment to prevent the late
whiplash syndrome. Scandinavian Journal of Rehabilitation Medicine 28: 105–111.
57. Foley-Nolan D, Moore K, Codd M et al (1992) Low energy high frequency pulsed electromagnetic
therapy for acute whiplash injuries. Scandinavian Journal of Rehabilitation Medicine 24: 51–59.
58. Mealy K, Brennan H & Fenelon GCC (1986) Early mobilisation of acute whiplash injuries. British Medical
Journal 292: 656–657.
59. Zylbergold RS & Piper MC (1985) Cervical spine disorders: a comparison of three types of traction.
Spine 10: 867–871.
60. Pennie BH & Agambar LJ (1990) Whiplash injuries: a trial of early management. Journal of Bone and Joint
Surgery 72B: 277–279.
61. McKinney LA (1989) Early mobilisation and outcomes in acute sprains of the neck. British Medical Journal
299: 1006–1008.
62. Borchgrevink GE, Kaasa A, McDonagh D et al (1998) Acute treatment of whiplash neck sprain injuries:
a randomized trial of treatment during the first 14 days after a car accident. Spine 23: 25–31.
63. Snow CJ, Aves Wood R, Dowhopouuk V et al (1992) Randomized controlled clinical trial of spray and
stretch for relief of back and neck myofascial pain. Physiotherapy Canada 44: S8.
64. Thorsen H, Gam AN, Svensson BH et al (1992) Low level laser therapy for myofascial pain in the neck
and shoulder girdle. A double-blind, cross-over study. Scandinavian Journal of Rheumatology 21: 139–142.
65. Goldie I & Landquist A (1970) Evaluation of the effects of different forms of physiotherapy in cervical
pain. Scandinavian Journal of Rehabilitation Medicine 2–3: 117–121.
66. Caldwell JW & Krusen EM (1962) Effectiveness of cervical traction in treatment of neck problems:
evaluation of various methods. Archives of Physical Medicine and Rehabilitation 43: 214–221.
67. Nordemar R & Throner C (1981) Treatment of acute cervical pain—a comparative group study. Pain
10: 93–101.
68. Hong CZ, Lin JC, Bender LF et al (1982) Magnetic necklace: its therapeutic effectiveness on neck and
shoulder pain. Archives of Physical Medicine and Rehabilitation 63: 426–466.
69. Kamwendo K & Linton SJ (1991) A controlled study of the effect of neck school in medical secretaries.
Scandinavian Journal of Rehabilitation Medicine 23: 143–152.
70. Lewith GT & Machin D (1981) A randomised trial to evaluate the effect of infra-red stimulation of local
trigger points, versus placebo on the pain caused by cervical osteoarthrosis. Acupuncture and Electro-
Therapeutics Research International Journal 6: 277–284.
71. Howe DH, Newcombe RG & Wade MT (1983) Manipulation of the cervical spine—pilot study. Journal
of the Royal College of General Practitioners 33: 574–579.
72. Sloop PR, Smith DS, Goldberg E & Dore C (1982) Manipulation of chronic neck pain: a double-blind
controlled study. Spine 7: 532–535.
73. Koes BW, Bouter LM, van Mameren H et al (1992) The effectiveness of manual therapy, physiotherapy,
and treatment by the general practitioner for nonspecific back and neck complaints: a randomized
clinical trial. Spine 17: 28–35.
74. Ceccherellia F, Altafini L, Lo Castro G et al (1989) Diode laser in cervical myofascial pain: a double-
blind study versus placebo. Clinical Journal of Pain 5: 301–304.
75. Loy TT (1983) Treatment of cervical spondylosis: electroacupuncture versus physiotherapy. Medical
Journal of Australia 2: 32–34.
76. Petrie JP & Langley GB (1983) Acupuncture in the treatment of chronic cervical pain: a pilot study.
Clinical and Experimental Rheumatology 1: 333–335.
77. Petrie JP & Hazleman BL (1986) A controlled study of acupuncture in neck pain. British Journal of
Rheumatology 25: 271–275.
78. Foley-Nolan D, Barry C, Coughlan RJ et al (1990) Pulsed high frequency (27 mhz) electromagnetic
therapy for persistent pain: a double blind, placebo-controlled study of 20 patients. Orthopaedics 13:
445–451.
 The neck 285

79. Levoska S & Keinanen-Kiukaanniemi S (1993) Active or passive physiotherapy for occupational cervico-
brachial disorders? A comparison of two treatment methods with a 1-year follow-up. Archives of Physical
Medicine and Rehabilitation 74: 425–430.
80. Brodin H (1984) Cervical pain and mobilization. International Journal of Rehabilitation Research 7:
190–191.
81. Brodin H (1985) Cervical pain and mobilization. Manual Medicine 2: 18–22.
82. Barnsley L, Lord SM, Wallis BJ & Bogduk N (1994) Lack of effect of intra-articular corticosteroids for
chronic cervical zygapophysial joint pain. New England Journal of Medicine 330: 1047–1050.
83. Kikuchi S, Macnab I & Moreau P (1981) Localisation of the level of symptomatic cervical disc
degeneration. Journal of Bone and Joint Surgery 63B: 272–277.
84. Whitecloud TS & Seago RA (1987) Cervical discogenic syndrome. Results of operative intervention in
patients with positive discography. Spine 12: 313–316.
85. Algers G, Pettersson K, Hildingsson C & Toolanen G (1993) Surgery for chronic symptoms after
whiplash injury. Follow-up of 20 cases. Acta Orthopaedica Scandinavica 64: 654–656.
*86. Lord SM, Barnsley L, Wallis BJ et al (1996) Percutaneous radio-frequency neurotomy for chronic
cervical zygapophyseal-joint pain. New England Journal of Medicine 335: 1721–1726.
286 N. Bogduk