Вы находитесь на странице: 1из 143

Downloaded from UvA-DARE, the institutional repository of the University of Amsterdam (UvA)

http://dare.uva.nl/document/229304

File ID 229304
Filename Thesis

SOURCE (OR PART OF THE FOLLOWING SOURCE):


Type Dissertation
Title Cervical radiculopathy: diagnostic aspects and non-surgical treatment
Author B. Kuijper
Faculty Faculty of Medicine
Year 2011
Pages 141
ISBN 978-94-6169-100-2

FULL BIBLIOGRAPHIC DETAILS:


http://dare.uva.nl/record/391610

Copyright

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or
copyright holder(s), other than for strictly personal, individual use.

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)


radiculopathy
non-surgical treatment
diagnostic aspects and
Cervical
Barbara Kuijper

Barbara Kuijper Cervical radiculopathy diagnostic aspects and


non-surgical treatment
Cervical radiculopathy;
diagnostic aspects and
non-surgical treatment
Cervical radiculopathy;
diagnostic aspects and
Barbara Kuijper
non-surgical treatment
Barbara Kuijper
ISBN: 978-94-6169-100-2

Cover: Gijs Kuijper


Layout and printing: Optima Grafische Communicatie, Rotterdam, The Netherlands

The salary of the trial nurse was paid by Dr. Eduard Hoelen Stichting.

Publication of this thesis was made possible by financial support of Dr. Eduard Hoelen
Stichting, Centrum Orthopedie Rotterdam, Sanofi Aventis BV, Boehringer Ingelheim BV.
Cervical radiculopathy;
diagnostic aspects and non-surgical treatment

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor


aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties ingestelde
commissie, in het openbaar te verdedigen in de Agnietenkapel
op donderdag 22 september 2011, te 14.00 uur

door

Barbara Kuijper
geboren te Alkmaar
Promotiecommissie

Promotores:
Prof. dr. M. de Visser
Prof. dr. F. Nollet

Copromotores:
Dr. J.Th.J. Tans
Dr. A. Beelen

Overige leden:
Prof. dr. J. Stam
Prof. dr. M. Wieringa-de Waard
Dr. J.H.T.M. Koelman
Prof. dr. C.B.L.M. Majoie
Prof. dr. R.J. de Haan
Prof. dr. W. C. Peul

Faculteit der Geneeskunde


Contents

Chapter 1 General introduction 7

Chapter 2 Degenerative cervical radiculopathy: diagnosis and 19


conservative treatment. A review

Chapter 3 Interobserver agreement on MRI evaluation of patients 35


with cervical radiculopathy

Chapter 4 Root compression on MRI compared with clinical 49


findings in patients with recent onset cervical
radiculopathy

Chapter 5 Results of needle electromyography in 176 patients 61


with cervical radiculopathy

Chapter 6 Cervical collar or physiotherapy versus wait and see 71


policy for recent onset cervical radiculopathy

Chapter 7 General discussion 91


Summary 107
Samenvatting 111

Appendix 1 Recent ontstane cervicale radiculopathie: minder pijn 115


met halskraag of fysiotherapie

Appendix 2 Physiotherapy protocol 129

List of publications 135


Dankwoord 137
Over de auteur 141
6
Chapter 1

General introduction

7
8
General introduction

Cervical radiculopathy

Degenerative cervical radiculopathy is a common cause of pain in the arm radiating


from the neck with an annual incidence rate of 83.2 per 100.000. 1 Symptoms and signs
include sensory disturbances in arm and hand with dermatomal and motor weakness
with myotomal distribution. Neck pain itself is usually present but is often minor
compared to arm pain. The pain in the arm is caused by cervical root compression due
to spondylosis or herniated discs, and is often severe, leading to function loss in the
first weeks to months and causing a considerable impact on overall health status. 23
The prognosis of degenerative cervical radiculopathy is often good with spontaneous
relief of symptoms and signs over time in the majority of patients. Consequently surgical
intervention is only needed in a minority of patients with ongoing pain. 45 However, the
severe pain in the first weeks requires pain relieving treatment modalities. In this thesis
the results of our studies on diagnostic value of neurological examination, magnetic
resonance imaging (MRI) and needle electromyography (EMG) as well as that of the
effect of non-surgical pain relieving treatment in recent onset cervical radiculopathy are
described.

Pathogenetic aspects

The complex of symptoms that we now associate with cervical radiculopathy was first
described in the thirties of the previous century. In the earliest reports the cause of root
compression was thought to be neoplasm, called enchondromas. Histopathological
examination of resected enchondromas showed fibrous connective tissue characteristic
of annulus fibrosus, which was thought to be neoplasm of the disc. 67 Mixter and Barrs
landmark report, written in 1934, changed the view on root compression and disc
disease. 8 They reported a series of 19 surgically treated patients with root compression,
four of which at the cervical level. They described histopathological findings and
concluded that ruptured discs were a more common cause than cartilaginous neoplasm.

A case of cervical radiculopathy was also described by Nachlas in 1934; pseudoangina


pectoris caused by hypertrophic changes in the cervical spine 9, later called hypertrophic
arthritis as a cause of root compression. In 1936, Turner and Oppenheimer reported 50
cases, with what they called discogenetic disease, which was the cause of thinning of
the intervertebral disc and narrowing of the intervertebral foramina. A detailed report by

9
Chapter 1

Spurling and Segerberg on cervical disc disease in 1953 mentioned two distinct causes:
osteophytic spur formation and rupture of the disc. 10 This still represents the current
opinion on causes of cervical root compression; herniated discs, foraminal stenosis or a
combination of both.

In lumbar radiculopathy herniated discs have been shown to be the most common
cause of root compression, in cervical radiculopathy this information is not available. It is
assumed that foraminal root compression by spondylosis plays a more important role in
the cervical spine than in the lumbar region. 1511-13

Diagnostic aspects

In the early reports dating from the thirties and forties of the previous century the
diagnosis of cervical root compression was based on clinical signs and symptoms. 81415
Additional diagnostic methods such as electromyography, X- rays of the cervical spine
and protein concentration in the cerebrospinal fluid were available but their usefulness
was limited. Electromyography has been used in cervical radiculopathy since the forties
of the previous century; the earliest report as far as we know is from Brazier in 1946. 16 In
her report she describes the technique of electromyography as a delicate tool capable of
showing abnormal spontaneous muscle activity in patients with root compression. 16 EMG
is still used as a diagnostic tool in this condition but well designed studies assessing its
diagnostic value are lacking. X-rays may show narrowing of the foraminal space, but this
is often an asymptomatic feature. 10 An elevated protein level in the cerebrospinal fluid
has been reported to support the diagnosis; however this finding is neither specific nor
sensitive. 1417 Intrathecal injection with radio opaque contrast agents such as poppy seed
oil, lipiodol, opamidol or iohexol were used to verify root compression on X-rays. 141518
Although potentially helpful this invasive technique was unreliable with low sensitivity
while associated with complications. 1015

The development of Computed Tomography (CT) allowed for showing the herniated disc,
spondylotic changes and root compression. CT myelography became the gold standard
for the demonstration of root compression in the eighties of the previous century. 19
Currently MRI has replaced CT-scan as the primary diagnostic tool due to its superior
imaging quality and better patient safety. 19 MRI allows better detection of the presence

10
General introduction

and cause of root compression but its true diagnostic accuracy for cervical radiculopathy
has yet to be determined.

MRI gives us detailed information on anatomy, for instance it shows us the root itself
and degenerative changes compressing the root. However, from studies in lumbar
radiculopathy we learned that asymptomatic abnormalities also occurred (false positive),
and vice versa, patients with a clinical diagnosis of lumbar radiculopathy without root
compression on the MRI were also found (false negative). 20-22 Despite the frequent use
of MRI, this crucial information, necessary for good interpretation of MRI results, is not
available for the cervical spine.

Therapeutic aspects

Several early reports describe surgical techniques performed in those days. The article
by Semmes and Murphy from 1943 provides interesting details of 3 operated cases:
under local anaesthesia the cervical laminas were exposed and the involved nerve
root was identified by the patient by gentle pressure applied on the ligamentum flavum
reproducing the exact pain. This test was done before any bone was removed. Then a
subtotal hemilaminectomy was performed and the loose fragments and the bulging parts
of the disc were removed. 15 Also conservative, non-surgical therapies were addressed.
For instance, Spurling stated in 1953 that the only indication for surgery is severe
neurological deficit. Nonetheless in his series of 110 patients with cervical radiculopathy
30% was treated surgically. 10 The conservative therapies in the forties and fifties of
the previous century were traction of the cervical spine, bed rest or massage. 101415 In
1966, a large trial was performed on non-surgical therapy of cervical radiculopathy. In
this trial different therapies were compared; cervical collar, physiotherapy with traction,
physiotherapy with exercises and placebo therapy. The study was not conclusive. 23 Later
trials using cervical collar or physiotherapy also did not sufficiently clarify the issue of
pain relieving modalities in recent onset cervical radiculopathy, as only patients with long
lasting pain were included. 2425

11
Chapter 1

Aims and outline of this thesis

Despite its frequent occurrence, little is known on diagnosis and nonsurgical treatment
of cervical radiculopathy. Contemporary literature often focuses on surgery 26-30,
although this is only performed in a minority of patients. Diagnosis and treatment of
cervical radiculopathy is done by general practitioners, neurologists, neurosurgeons
and physiotherapists. Clinical trials evaluating the use of diagnostic tools, long term
outcomes, as well as efficacy of conservative treatments, are scarce. Consequently there
is no consensus on diagnostic criteria or treatment. We have designed a multicenter
randomised controlled trial evaluating the two most commonly used non-surgical
treatments; physiotherapy and cervical collar. Within the framework of this trial we
explored the major diagnostic challenges of cervical radiculopathy.

To learn more on the existing knowledge of degenerative cervical radiculopathy we


first performed a review of the literature, the results of which are presented in chapter
2. We have reviewed studies on epidemiology, pathophysiology, clinical diagnosis,
electrophysiological examinations, imaging studies and the various types of non-surgical
treatment of cervical radiculopathy. After reviewing the existing literature we established
criteria for the clinical diagnosis of cervical radiculopathy to be used in our studies on
MRI, EMG and non-surgical treatment.

Chapters 3 and 4 deal with magnetic resonance imaging (MRI), currently the most often
used imaging technique in patients with symptoms of cervical radiculopathy. MRI is
used to demonstrate the presence and cause of root compression in patients with a
clinical diagnosis of cervical radiculopathy. Although MRI provides many details, it is
questionable if these correlate well with the clinical symptoms and signs. In our MRI
study we first determined the interobserver variability of MRI evaluation of herniated
discs, spondylotic neuroforaminal stenosis and root compression in patients with cervical
radiculopathy. Also the effect of disclosure of clinical information on interobserver
variability was examined (chapter 3). After that we studied the correlation between clinical
symptoms and signs and the abnormalities as reported by the blinded neuroradiologists
(chapter 4), data that are important for a correct interpretation of MRI results.

Chapter 5 describes the results of our needle electromyography (EMG) study. Although
many studies were performed on EMG in cervical radiculopathy since the forties, there
is still no consensus on its usefulness in daily practice. Highly varying sensitivities were

12
General introduction

reported, but blinded studies in a large study population are not available. 3132 Therefore
our electrodiagnostic consultants, who were blinded for clinical and MRI data, examined
all patients using a limited standardized electromyography protocol.

As shown in chapter 2, evidence for the effectiveness of any non-surgical therapy,


including cervical collar or physiotherapy, is lacking. Treatment in acute or subacute
cervical radiculopathy has not been studied at all. In the past non-surgical treatment of
the even more frequently occurring lumbosacral radiculopathy has been studied, showing
that standardised physiotherapy or bed rest did not produce better results than watchful
waiting and continuing daily activities as much as possible. 3334 With the results of the
trials on lumbosacral radiculopathy in mind it seemed useful to investigate whether there
is evidence for efficacy of conservative treatment of early onset degenerative cervical
radiculopathy. We conducted a randomised trial to compare the effect of cervical collar and
physiotherapy with a wait and see policy. The results of this trial are shown in chapter 6.
Appendix 1 of this thesis contains a Dutch version of the trial, published in the
Nederlands Tijdschrift voor Geneeskunde. Appendix 2 contains our physiotherapy
protocol.

The overall results of this thesis are discussed in chapter 7; which includes a general
discussion and a summary of the results.

13
Chapter 1

1. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

2. Fanuele JC, Birkmeyer NJ, Abdu WA, Tosteson TD, Weinstein JN. The impact of spinal problems on

the health status of patients: have we underestimated the effect? Spine 2000;25(12):1509-14.

3. Daffner SD, Hilibrand AS, Hanscom BS, Brislin BT, Vaccaro AR, Albert TJ. Impact of neck and arm

pain on overall health status. Spine 2003;28(17):2030-5.

4. Binder AI. Cervical spondylosis and neck pain. Bmj 2007;334(7592):527-31.

5. Carette S, Fehlings MG. Clinical practice. Cervical radiculopathy. N Engl J Med 2005;353(4):392-9.

6. Stookey B. Compression of spinal cord due to ventral extradural cervical chondromas: diagnosis and

surgical treatment. Arch Neurol Psychiatry 1927;20.

7. Elsberg CA. Tumors of the spinal cord and the symptoms of irritation and compression of the spinal cord

and nerve roots: pathology, symptomatology, diagnosis and treatment. New York: Paul B Hoeber, 1925.

8. Mixter WJ, Barr JS. Rupture of the intervertebral disc with involvement of the spinal canal. N Engl J

Med 1934;211.

9. Nachlas IW. Pseudoangina pactoris originating in the cervical spine. J Am Med Assoc 1934;103.

10. Spurling RG, Segerberg LH. Lateral intervertebral disk lesions in the lower cervical region. J Am Med

Assoc 1953;151(5):354-9.

11. Sengupta DK, Kirollos R, Findlay GF, Smith ET, Pearson JC, Pigott T. The value of MR imaging in

differentiating between hard and soft cervical disc disease: a comparison with intraoperative findings.

Eur Spine J 1999;8(3):199-204.

12. Shedid D, Benzel EC. Cervical spondylosis anatomy: pathophysiology and biomechanics.

Neurosurgery 2007;60(1 Supp1 1):S7-13.

13. Salemi G, Savattieri G, Meneghini F, Di Benedetto ME, Ragonese P, Morgante L, et al. Prevalence of

cervical spondylotic radiculopathy: a door-to-door survey in a Sicilian municipality. Acta Neurol Scand

1996;93:184-188.

14. Michelsen JJ, Mixter WJ. Pain and disability of shoulder and arm due to herniation of the nucleus

pulposus of cervical intervertebral disks. N Engl J Med 1944;231(8).

15. Semmes RE, Murphey F. Syndrome of unilateral rupture of sixth cervical intervertebral disk, with

compression of seventh cervical nerve root: report of four cases with symptoms simulating coronary

disease. J Am Med Assoc 1943;111.

16. Brazier MA, Watkins AL, Michelsen JJ. Electromyography in differential diagnosis of ruptured cervical

disk. Arch Neurol Psychiatry 1946;56(6):651-8.

17. Yoss RE, Corbin KB, Maccarty CS, Love JG. Significance of symptoms and signs in localization of

involved root in cervical disk protrusion. Neurology 1957;7(10):673-83.

18. Davies AM, Evans N, Chandy J. Outpatient lumbar radiculography: comparison of iopamidol and

iohexol and a literature review. Br J Radiol 1989;62(740):716-23.

14
General introduction

19. Kaiser JA, Holland BA. Imaging of the cervical spine. Spine 1998;23(24):2701-12.

20. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal magnetic-resonance scans of

the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am

1990;72(3):403-8.

21. van Rijn JC, Klemetso N, Reitsma JB, Majoie CB, Hulsmans FJ, Peul WC, et al. Observer

variation in MRI evaluation of patients suspected of lumbar disk herniation. AJR Am J Roentgenol

2005;184(1):299-303.

22. van Rijn JC, Klemetso N, Reitsma JB, Majoie CB, Hulsmans FJ, Peul WC, et al. Symptomatic and

asymptomatic abnormalities in patients with lumbosacral radicular syndrome: Clinical examination

compared with MRI. Clin Neurol Neurosurg 2006;108(6):553-7.

23. Pain in the neck and arm: a multicentre trial of the effects of physiotherapy, arranged by the British

Association of Physical Medicine. Br Med J 1966;1(5482):253-8.

24. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery,

physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997;22(7):751-8.

25. Persson LC, Moritz U, Brandt L, Carlsson CA. Cervical radiculopathy: pain, muscle weakness and

sensory loss in patients with cervical radiculopathy treated with surgery, physiotherapy or cervical

collar. A prospective, controlled study. Eur Spine J 1997;6(4):256-66.

26. Matz PG, Holly LT, Groff MW, Vresilovic EJ, Anderson PA, Heary RF, et al. Indications for anterior

cervical decompression for the treatment of cervical degenerative radiculopathy. J Neurosurg Spine

2009;11(2):174-82.

27. van Limbeek J, Jacobs WC, Anderson PG, Pavlov PW. A systematic literature review to identify the

best method for a single level anterior cervical interbody fusion. Eur Spine J 2000;9(2):129-36.

28. Jacobs WC, Anderson PG, Limbeek J, Willems PC, Pavlov P. Single or double-level anterior

interbody fusion techniques for cervical degenerative disc disease. Cochrane Database Syst Rev

2004(4):CD004958.

29. Bartels RH, Donk R, Verbeek AL. No justification for cervical disk prostheses in clinical practice: a

meta-analysis of randomized controlled trials. Neurosurgery;66(6):1153-60; discussion 1160.

30. Nikolaidis I, Fouyas IP, Sandercock PA, Statham PF. Surgery for cervical radiculopathy or myelopathy.

Cochrane Database Syst Rev 2010(1):CD001466.

31. American Association of Electrodiagnostic Medicine. The electrodiagnostic evaluation of patients with

suspected cervical radiculopathy: literature review on the usefulness of needle electromyography.

Muscle Nerve 1999;Supplement 8:S213-S221.

32. Haig AJ, Yamakawa K, Kendall R, Miner J, Parres CM, Harris M. Assessment of the validity of masking

in electrodiagnostic research. Am J Phys Med Rehabil 2006;85(6):475-81.

15
Chapter 1

33. Hofstee DJ, Gijtenbeek JM, Hoogland PH, van Houwelingen HC, Kloet A, Lotters F, Tans JThJ.

Westeinde sciatica trial: randomized controlled study of bed rest and physiotherapy for acute sciatica.

J Neurosurg 2002;96(1 Suppl):45-9.

34. Vroomen PC, de Krom MC, Wilmink JT, Kester AD, Knottnerus JA. Lack of effectiveness of bed rest for

sciatica. N Engl J Med 1999;340(6):418-23.

16
17
18
Chapter 2

Degenerative cervical
radiculopathy: diagnosis
and conservative
treatment. A review.
Barbara Kuijper, Jos Th J Tans, Robert Jan Schimsheimer, Bas F W van der Kallen, Anita
Beelen, Frans Nollet, Marianne de Visser

European Journal of Neurology 2009

19
Chapter 2

Abstract

Objectives
To provide a state-of-the-art assessment of diagnosis and non-surgical treatment of
degenerative cervical radiculopathy

Methods
A literature search for studies on epidemiology, diagnosis including electrophysiological
examination and imaging studies, and different types of conservative treatment was
undertaken.

Results
The most common causes of cervical root compression are spondylarthrosis and disc
herniation. Diagnosis is made mainly on clinical grounds, although there are no well-
defined criteria. Provocative tests like the foraminal compression test are widely used but
not properly evaluated. The clinical diagnosis of degenerative cervical radiculopathy can
be confirmed by magnetic resonance imaging. The role of electromyography is mainly
to rule out other conditions. Cervical radiculopathy is initially treated conservatively,
although no treatment modality has been evaluated in a randomised controlled trial.

20
Degenerative cervical radiculopathy: a review

Introduction

Cervical radiculopathy because of root compression by spondylosis or disc herniation is


a common cause of pain in neck and arm, with a considerable impact on overall health
status. 12 Yet, many controversies exist regarding diagnostic procedures and mode of
treatment. This paper reviews current knowledge of pathophysiology, diagnosis and
non-surgical treatment.

Methods

For this literature review, articles were obtained through a comprehensive Medline search
with the terms: cervical or neck AND radiculopathy, root compression, spondylosis,
disc without limitation of year of publication; the earliest dated from 1953. Articles were
also identified from reference-lists of the articles found in the Medline search. Original
articles on epidemiology, pathophysiology, diagnosis and non-surgical therapy were
selected; review articles were only included in this review if they offered new insights or
opinions. Papers not written in English were exluded. Finally 77 articles were reviewed in
this paper.

Epidemiology

Epidemiological data on cervical radiculopathy are sparse. Radhakrishnan et al


performed a large retrospective population-based study and found an annual age-
adjusted incidence rate of 83.2 per 100 000 in total (107.3 for males and 63.5 for
females). 3 Male predominance has been confirmed by others. 3-5 There was a clear
peak incidence in the fourth and fifth decade. 346 In a door-to-door survey in Sicily a
prevalence of 3.5 per 1000, with the same age- distribution was found. 7 Spondylarthrosis
as a cause of cervical radiculopathy tends to occur more in higher age, whereas disc
herniation is more common in younger patients. 347 Not much is known about risk
factors. There have been several studies suggesting that smoking is a risk factor (relative
risk 2.9). 89 Smoking is thought to have an adverse effect on vertebral blood supply
leading to decreased nutrition of the disc. 8

21
Chapter 2

Pathophysiology

The most common cause of root compression at the cervical level is narrowing of the
foraminal space secondary to spondylarthrosis. The intervertebral foramina are shaped
like a funnel: the entrance zone is the most narrow part. As root sleeves are conical, with
their origin in the central dural sac being the largest part, the nerve root is vulnerable
for compression mainly in the entrance zone of the intervertebral foramen. 1011 Laterally
herniated discs are the second cause of root compression in the cervical spine. 3
Differentiation between these two is clinically cumbersome, although neurological deficits,
especially weakness, seem to occur more often in patients with soft disc prolaps. 12-14

The most common level of root compression is C7, followed by C6. Compression of roots
C5 and C8 are less frequent. One possible explanation is that intervertebral foramina are
largest in the upper cervical region and progressively decrease in size in the middle and
lower cervical areas, with an exception of the C7-Th1 foramen, which is larger than the
other cervical intervertebral foramina. The middle and lower cervical regions are most
susceptible for mobility and stress. 341015 Simultaneous disc herniation at more than one
level is uncommon, whereas spondylarthrosis often occurs at multiple levels. Patients
who had undergone cervical fusion show accelerated degeneration at adjacent levels,
which is probably partly because of the underlying tendency towards spondylarthrosis
and partly because of the cervical fusion. 16

There is increasing evidence that inflammation in itself and/or in association with root
compression is the main cause of the symptoms and signs. 17-19 This is corroborated by
the presence of interleukins and prostaglandin in herniated discs and the spontaneous
recovery within weeks or months in the majority of patients. 2021

Clinical diagnosis

Clinical features
Cervical radiculopathy is largely a clinical diagnosis, albeit that there is a lack of
well-defined clinical criteria. 13182223 The classic clinical picture includes neck pain
with irradiation to the arm and fingers corresponding to the dermatomes involved,
paraesthesias in arm and hand in conjunction with diminished muscle tendon reflexes,
sensory disturbances and/or motor weakness with dermatomal/myotomal distribution.

22
Degenerative cervical radiculopathy: a review

Two major studies including 100 and 561 patients, respectively, report that paraesthesias
were present in 91% and 89.7%, respectively, sensory loss was found in 24% and 33.0%,
respectively, symptoms of weakness in 34% and 15.3%, signs of weakness in 70% and
64.2%, respectively, and hypoactive tendon reflexes in 72% and 84.1%, respectively. 34

Patients often report that neck movements and manoeuvres causing increased intraspinal
pressure like coughing and sneezing worsen the pain or tingling in the arm. Neck pain
itself is not necessarily present or minor compared with the arm pain. Atypical clinical
symptoms as chest, breast and jaw pain are described; in a retrospective study of 241
patients with a C7 radiculopathy, 16% had atypical symptoms as subscapular pain
and chest pain. Also pain along the medial border of the scapula is quite common.
523-25
The localizing value of pain alone is generally less than that of the distribution
of paraesthesias. A lesion of one cervical root will usually not lead to sensory loss
because of overlap. 526 Overlap in myotomal distributions and inter-individual variations
hamper determination of the level of root compression by motor weakness alone. Also
published myotomal charts are not consistent. 26-28 Conditions that may mimic cervical
radiculopathy are listed in table 1.

Provocative tests
Provocative tests elicit or worsen symptoms in the affected arm and are indicative of
cervical radiculopathy. A recent systematic review showed that only six studies evaluating
the diagnostic accuracy of provocative tests could be included. A meta-analysis was not
performed due to several methodological flaws. 29 Although no firm conclusions could be
drawn, the review suggested that, when consistent with the history and other physical
findings, the foraminal compression test, described by Spurling in 1953, and Valsalvas
manoeuvre might be indicative of a cervical radiculopathy, while a negative upper limb
tension test (ULTT) might be used to rule it out. 152930 In cases of extradural cervical
root compression, shoulder abduction often causes relief of radicular symptoms. This
shoulder abduction relief sign is also called the shoulder abduction test. 31-33

Imaging

The clinical diagnosis cervical radiculopathy can be confirmed by demonstration of


root compression using MRI which is the method of choice to detect disc protrusions,
whereas thin slices spiral CT is possibly superior in showing foraminal stenosis by bony

23
Chapter 2

Table 1. Causes of arm pain mimicking degenerative cervical radiculopathy

With paraesthesias and neck pain

Cervical root compression

Tumors

Cysts

Trauma

Cervical root inflammation

Lyme borreliosis

Herpes zoster

With paraesthesias without neck pain

Brachial plexus

Parsonage-Turner syndrome or neuralgic amyotrophy

Tumors

Radiation-induced brachial plexopathy

Thoracic outlet syndrome

Nerve

Carpal tunnel syndrome

Ulnar nerve entrapment at the elbow

With neck pain, and sometimes referred pain

Facet-joint pain, ligamentous pain

Without paraesthesias and without neck pain

Shoulder

Rotator cuff injury

Arthritis

Bursitis

Elbow

Epicondylitis

Arthritis

Heart

Myocardial ischaemic pain

compression. 34 No data are available on the diagnostic accuracy of both methods. MR-
myelography and CT-myelography are reported to achieve higher sensitivity rates. 34-37
The imaging data should always be interpreted in the clinical context as particularly MRI

24
Degenerative cervical radiculopathy: a review

often yields false positive results showing abnormalities in asymptomatic patients. 638-40
This also holds true for plain X-ray studies that exhibit degenerative changes increasing
with age unrelated to clinical signs and symptoms 41 and may therefore be considered
obsolete in the regular diagnostic work-up of a patient with cervical radiculopathy.

Electrophysiologic studies

The diagnostic value of conventional needle myography (EMG) is still a matter of debate.
Reported sensitivities from unblinded studies vary from 30 to 95%. EMG results were
compared with data on root compression verified during surgery, or obtained by MRI or
clinical examination. 254243 A standard screening program of 6 to 8 limb and paraspinal
muscles representing the different myotomes was recommended. 4445 As most C6
muscles are also innervated by C5 or C7 identifying a C6 radiculopathy by EMG alone
will be difficult. 2646 Paraspinal muscle examination seems to be more sensitive but
technically more challenging with often false-positive results in elderly patients. 26444748
Needle electromyography can also be used to determine the stage and severity of the
root compression. Abnormal spontaneous activity may appear in severe cases three
weeks after onset of symptoms as a sign of axon loss. In cases of chronic radiculopathy
neurogenic recruitment patterns can be seen as a result of collateral sprouting. 26434950
Flexor carpi radialis H-reflexes were delayed or absent in 21 of 25 patients with C6 or C7
compression and normal with C5 and C8 radiculopathies. 51 F-waves did not contribute
much to diagnosing cervical radiculopathy. 4649 Probably the most important role of
neurophysiologic examination is to rule out other conditions like median or ulnar nerve
entrapment with nerve conduction studies. 2646485152

Non-surgical treatment

The natural course of spondylotic and discogenic cervical radiculopathy is generally


favourable. Disc herniations often resolve spontaneously. 11155354 Surgery is only indicated
for patients with intractable or long lasting pain. Usually there is a wait and see policy
in which the patient is treated with analgesics. Although one might expect a positive
effect of anti-inflammatory drugs like non-steroidal anti-inflammatory drugs (NSAIDs)
because of the supposedly inflammatory component in radiculopathy no evidence for
that was found. It is equally unknown if oral corticosteroids have a beneficial effect on

25
Chapter 2

pain. 5556 Well-designed trials on conservative treatment of cervical radiculopathy are


not available. Persson et al have conducted the only randomised clinical trial (RCT)
comparing three modes of treatment. 5758 Patients with long-lasting (more than 3
months) cervical radiculopathy were randomised for surgery, physiotherapy or cervical
collar. Surgery was superior for pain relief at 4 months follow-up (p<0.05). At 16 months
follow-up there was no difference between the three groups regarding pain, muscular
strength or sensory loss. In 1966 the British Association of Physical Medicine conducted
a randomised clinical trial including 493 patients with cervical root symptoms, treated
with traction, placebo traction, collar, placebo tablets or placebo heat treatment. 41 No
significant difference in pain and in ability to work was found between the five treatment
groups. Seventy-five percent of patients reported pain relief at four weeks follow-up in
all treatment arms. Although the methodology of this trial does not fulfil current criteria of
RCTs, it demonstrates the favourable spontaneous course of cervical radiculopathy.

Exercise therapy
A recent systematic review on exercise for patients with neck pain (with or without
radicular arm pain) concluded that specific exercises may be effective for the treatment
of mechanical neck disorders. 59 Four studies were included encompassing patients with
neck pain and at least some radicular findings. One of these is the study of Persson et
al, described above. In the other three studies, it remained unclear how many patients
had cervical radiculopathy and what the effect of exercise was in this subgroup. 5760-62
Yet, exercise therapy is probably one of the most widely used interventions for cervical
radiculopathy.

Manipulation
A systematic review studying the effect of manipulation and mobilisation techniques in
patients with mechanical neck disorders 63 showed that manual therapy probably results
in more short-term pain-relief than does exercise therapy or usual medical care in atypical
neck pain without radiculopathy. However, there was insufficient evidence for the effect
of manipulative techniques in the subgroup with cervical radiculopathy. 60-65 Moreover,
cervical spine manipulation carries a risk of complications like vertebral dissection and
spinal cord compression due to massive disc herniation. 66-68 Therefore, this intervention
should be discouraged in cervical radiculopathy, especially if imaging of the spine has not
yet been performed.

26
Degenerative cervical radiculopathy: a review

Cervical collar
The effectiveness of treatment with a cervical collar has not been settled. Many authors
state that the use of a collar reduces pain by minimizing motion and nerve root irritation.
12224169-71
In the previously mentioned study of Persson et al. no benefit was found of
a cervical hard collar, surgery or physiotherapy in patients with cervical radiculopathy
lasting more than 3 months. 5758 A collar is usually prescribed in the acute or subacute
stage. Longer use might cause adverse effects due to immobilisation such as atrophy
of the paraspinal muscles. 1822 No clear evidence about the different types of cervical
collars used as an intervention in cervical radiculopathy is available. Hard collars are
uncomfortable, soft collars do not give enough support. Semi-hard collars might be a
good compromise.

Nerve root blocks


Selective diagnostic nerve root blocks have been reported as a diagnostic tool in
multilevel radiculopathy as well as a therapeutic tool. 72 A cohort study by Bush and
Hiller showed favourable results of foraminal steroid injections. 5373 Sixty eight patients
with cervical radiculopathy were treated with serial periradicular/epidural corticosteroid
injections. All patients made a satisfactory recovery without the need for surgical
intervention. A control group was not available, hampering a robust conclusion about its
effectiveness. Recently, the first randomised study on transforaminal steroid injections
has been published, which failed to show any positive effect. 72 Forty patients with
cervical radiculopathy were randomised for injections with steroids/local anaesthetics or
saline/local anaesthetics. At follow-up, no difference was found in subjective symptom
reduction. Foraminal injections can cause minor complications like injection-induced
pain. 7475 Serious vascular complications causing anterior spinal artery syndrome and
cerebellar strokes have been described as a rare complication. 74-77 As long as there are
no well controlled, positive studies, this treatment with potential serious complications
74-77
should not be advocated.

Conclusion

It is truly amazing that there is lack of evidence with regard to diagnostic procedures and
treatment interventions in such a frequent condition as degenerative cervical radiculopathy.
The diagnosis, mainly made on clinical grounds, is not considered as difficult, but
establishing the exact level of root compression is problematic due to the poor localising

27
Chapter 2

value of the clinical symptoms and signs. The foraminal compression test and Valsalvas
manoeuvre are the best known tests on physical examination. The clinical diagnosis
cervical radiculopathy can be confirmed by demonstration of root compression using MRI.
Electromyography can be useful in identifying severe cases, but its main value it to rule
out other conditions. Cervical radiculopathy is initially treated conservatively, although no
treatment modality has been evaluated in a randomised controlled trial and therefore the
superiority of conservative interventions to the spontaneous recovery is unknown.

28
Degenerative cervical radiculopathy: a review

1. Fanuele JC, Birkmeyer NJ, Abdu WA, Tosteson TD, Weinstein JN. The impact of spinal problems on

the health status of patients: have we underestimated the effect? Spine 2000;25(12):1509-14.

2. Daffner SD, Hilibrand AS, Hanscom BS, Brislin BT, Vaccaro AR, Albert TJ. Impact of neck and arm

pain on overall health status. Spine 2003;28(17):2030-5.

3. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

4. Dubuisson A, Lenelle J, Stevenaert A. Soft cervical disc herniation: a retrospective study of 100 cases.

Acta Neurochir (Wien) 1993;125(1-4):115-9.

5. Yoss RE, Corbin KB, Maccarty CS, Love JG. Significance of symptoms and signs in localization of

involved root in cervical disk protrusion. Neurology 1957;7(10):673-83.

6. Jensen MV, Tuchsen F, Orhede E. Prolapsed cervical intervertebral disc in male professional drivers in

Denmark, 1981-1990. A longitudinal study of hospitalizations. Spine 1996;21(20):2352-5.

7. Salemi G, Savattieri G, Meneghini F, Di Benedetto ME, Ragonese P, Morgante L, et al. Prevalence of

cervical spondylotic radiculopathy: a door-to-door survey in a Sicilian municipality. Acta Neurol Scand

1996;93:184-188.

8. An HS, Silveri CP, Simpson JM, File P, Simmons C, Simeone FA, et al. Comparison of smoking habits

between patients with surgically confirmed herniated lumbar and cervical disc disease and controls. J

Spinal Disord 1994;7(5):369-73.

9. Kostova V, Koleva M. Back disorders (low back pain, cervicobrachial and lumbosacral radicular

syndromes) and some related risk factors. J Neurol Sci 2001;192(1-2):17-25.

10. Tanaka N, Fujimoto Y, An HS, Ikuta Y, Yasuda M. The anatomic relation among the nerve roots,

intervertebral foramina, and intervertebral discs of the cervical spine. Spine 2000;25(3):286-91.

11. Adams C. Cervical spondylotic radiculopathy and myelopathy. Amsterdam/ New York / Oxford:

Elsevier Publishing Company, 1976.

12. Dreyer SJ, Boden SD. Nonoperative treatment of neck and arm pain. Spine 1998;23(24):2746-54.

13. Shah KC, Rajshekhar V. Reliability of diagnosis of soft cervical disc prolapse using Spurlings test. Br J

Neurosurg 2004;18(5):480-3.

14. Shedid D, Benzel EC. Cervical spondylosis anatomy: pathophysiology and biomechanics.

Neurosurgery 2007;60(1 Supp1 1):S7-13.

15. Spurling RG, Segerberg LH. Lateral intervertebral disk lesions in the lower cervical region. J Am Med

Assoc 1953;151(5):354-9.

16. Seo M, Coi D. Adjacent segment disease after fusion for cervical spondylosis; myth or reality?. Br J

Neurosurg 2008;22 (2):195-199.

17. Autio RA, Karppinen J, Niinimaki J, Ojala R, Veeger N, Korhonen T, et al. The effect of infliximab, a

monoclonal antibody against TNF-alpha, on disc herniation resorption: a randomized controlled study.

Spine 2006;31(23):2641-5.

29
Chapter 2

18. Carette S, Fehlings MG. Clinical practice. Cervical radiculopathy. N Engl J Med 2005;353(4):392-9.

19. Cyteval C, Thomas E, Decoux E, Sarrabere MP, Cottin A, Blotman F, et al. Cervical radiculopathy: open

study on percutaneous periradicular foraminal steroid infiltration performed under CT control in 30

patients. AJNR Am J Neuroradiol 2004;25(3):441-5.

20. Mochida K, Komori H, Okawa A, Muneta T, Haro H, Shinomiya K. Regression of cervical disc

herniation observed on magnetic resonance images. Spine 1998;23(9):990-5; discussion 996-7.

21. Kang JD, Georgescu HI, McIntyre-Larkin L, Stefanovic-Racic M, Evans CH. Herniated cervical

intervertebral discs spontaneously produce matrix metalloproteinases, nitric oxide, interleukin-6, and

prostaglandin E2. Spine 1995;20(22):2373-8.

22. Levine MJ, Albert TJ, Smith MD. Cervical Radiculopathy: Diagnosis and Nonoperative Management. J

Am Acad Orthop Surg 1996;4(6):305-316.

23. Narayan P, Haid RW. Treatment of degenerative cervical disc disease. Neurol Clin 2001;19(1):217-29.

24. Ozgur BM, Marshall LF. Atypical presentation of C-7 radiculopathy. J Neurosurg 2003;99(2

Suppl):169-71.

25. Slipman CW, Plastaras C, Patel R, Isaac Z, Chow D, Garvan C, et al. Provocative cervical discography

symptom mapping. Spine J 2005;5(4):381-8.

26. Fisher MA. Electrophysiology of radiculopathies. Clin Neurophysiol 2002;113(3):317-35.

27. Chang H, Park JB, Hwang JY, Song KJ. Clinical analysis of cervical radiculopathy causing deltoid

paralysis. Eur Spine J 2003;12(5):517-21.

28. Levin KH, Maggiano HJ, Wilbourn AJ. Cervical radiculopathies: comparison of surgical and EMG

localization of single-root lesions. Neurology 1996;46(4):1022-5.

29. Rubinstein SM, Pool JJ, van Tulder MW, Riphagen, II, de Vet HC. A systematic review of the

diagnostic accuracy of provocative tests of the neck for diagnosing cervical radiculopathy. Eur Spine J

2007;16(3):307-19.

30. Viikari-Juntura E, Porras M, Laasonen EM. Validity of clinical tests in the diagnosis of root compression

in cervical disc disease. Spine 1989;14(3):253-7.

31. Davidson RI, Dunn EJ, Metzmaker JN. The shoulder abduction test in the diagnosis of radicular pain in

cervical extradural compressive monoradiculopathies. Spine 1981;6(5):441-6.

32. Farmer JC, Wisneski RJ. Cervical spine nerve root compression. An analysis of neuroforaminal

pressures with varying head and arm positions. Spine 1994;19(16):1850-5.

33. Beatty RM, Fowler FD, Hanson EJ, Jr. The abducted arm as a sign of ruptured cervical disc.

Neurosurgery 1987;21(5):731-2.

34. Bartlett RJ, Hill CR, Gardiner E. A comparison of T2 and gadolinium enhanced MRI with CT

myelography in cervical radiculopathy. Br J Radiol 1998;71(841):11-9.

35. Birchall D, Connelly D, Walker L, Hall K. Evaluation of magnetic resonance myelography in the

investigation of cervical spondylotic radiculopathy. Br J Radiol 2003;76(908):525-31.

30
Degenerative cervical radiculopathy: a review

36. Kaiser JA, Holland BA. Imaging of the cervical spine. Spine 1998;23(24):2701-12.

37. Fortin J, Riethmiller DW, Vilensky JA. No clear winner in differing imaging modalities for cervical

radiculopathy. Pain Physician 2002;5(3):285-7.

38. van Rijn JC, Klemetso N, Reitsma JB, Bossuyt PM, Hulsmans FJ, Peul WC, et al. Observer variation

in the evaluation of lumbar herniated discs and root compression: spiral CT compared with MRI. Br J

Radiol 2006;79(941):372-7.

39. van Rijn JC, Klemetso N, Reitsma JB, Majoie CB, Hulsmans FJ, Peul WC, et al. Observer

variation in MRI evaluation of patients suspected of lumbar disk herniation. AJR Am J Roentgenol

2005;184(1):299-303.

40. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal magnetic-resonance scans of

the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am

1990;72(3):403-8.

41. Pain in the neck and arm: a multicentre trial of the effects of physiotherapy, arranged by the British

Association of Physical Medicine. Br Med J 1966;1(5482):253-8.

42. Nardin RA, Patel MR, Gudas TF, Rutkove SB, Raynor EM. Electromyography and magnetic resonance

imaging in the evaluation of radiculopathy. Muscle Nerve 1999;22(2):151-5.

43. van den Bent MJ, Oosting J, Laman DM, van Duijn H. EMG before and after cervical anterior

discectomy. Acta Neurol Scand 1995;92(4):332-6.

44. Dillingham TR, Lauder TD, Andary M, Kumar S, Pezzin LE, Stephens RT, et al. Identification of cervical

radiculopathies: optimizing the electromyographic screen. Am J Phys Med Rehabil 2001;80(2):84-91.

45. Yaar I. The logical choice of muscles for the needle-electromyography evaluation of cervical

radiculopathy. Arch Phys Med Rehabil 2005;86(3):521-6.

46. Levin KH. Electrodiagnostic approach to the patient with suspected radiculopathy. Neurol Clin

2002;20(2):397-421, vi.

47. Gilad R, Dabby R, Boaz M, Sadeh M. Cervical paraspinal electromyography: normal values in 100

control subjects. J Clin Neurophysiol 2006;23(6):573-6.

48. Czyrny JJ, Lawrence J. The importance of paraspinal muscle EMG in cervical and lumbosacral

radiculopathy: review of 100 cases. Electromyogr Clin Neurophysiol 1996;36(8):503-8.

49. Tsao. The electrodiagnosis of cervical and lumbosacral radiculopathy. Neurol Clin 2007;25:473-494.

50. Alrawi MF, Khalil NM, Mitchell P, Hughes SP. The value of neurophysiological and imaging studies

in predicing outcome in the surgical treatment of cervical radiculopathy. European Spine Journal

2006;16(4):495-500.

51. Schimsheimer RJ, de Visser BW, Kemp B. The flexor carpi radialis H-reflex in lesions of the sixth and

seventh cervical nerve roots. J Neurol Neurosurg Psychiatry 1985;48(5):445-9.

52. Lomen-Hoerth C, Aminoff MJ. Clinical neurophysiologic studies: which test is useful and when?

Neurol Clin 1999;17(1):65-74.

31
Chapter 2

53. Bush K, Chaudhuri R, Hillier S, Penny J. The pathomorphologic changes that accompany the

resolution of cervical radiculopathy. A prospective study with repeat magnetic resonance imaging.

Spine 1997;22(2):183-6; discussion 187.

54. Vinas FC, Wilner H, Rengachary S. The spontaneous resorption of herniated cervical discs. J Clin

Neurosci 2001;8(6):542-6.

55. Haimovic IC, Beresford HR. Dexamethasone is not superior to placebo for treating lumbosacral

radicular pain. Neurology 1986;36(12):1593-4.

56. Ellenberg MR, Honet JC, Treanor WJ. Cervical radiculopathy. Arch Phys Med Rehabil 1994;75(3):342-52.

57. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery,

physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997;22(7):751-8.

58. Persson LC, Moritz U, Brandt L, Carlsson CA. Cervical radiculopathy: pain, muscle weakness and

sensory loss in patients with cervical radiculopathy treated with surgery, physiotherapy or cervical

collar. A prospective, controlled study. Eur Spine J 1997;6(4):256-66.

59. Kay TM, Gross A, Goldsmith C, Santaguida PL, Hoving J, Bronfort G. Exercises for mechanical neck

disorders. Cochrane Database Syst Rev 2005(3):CD004250.

60. Hoving JL, Koes BW, de Vet HC, van der Windt DA, Assendelft WJ, van Mameren H, et al. Manual

therapy, physical therapy, or continued care by a general practitioner for patients with neck pain. A

randomized, controlled trial. Ann Intern Med 2002;136(10):713-22.

61. Kogstad OA, Karterud S, Gudmundsen J. [Cervicobrachialgia. A controlled trial with conventional

therapy and manipulation]. Tidsskr Nor Laegeforen 1978;98(16):845-8.

62. Brodin H. Cervical pain and mobilization. Int J Rehabil Res 1984;7(2):190-1.

63. Gross AR, Hoving JL, Haines TA, Goldsmith CH, Kay T, Aker P, et al. Manipulation and mobilisation for

mechanical neck disorders. Cochrane Database Syst Rev 2004(1):CD004249.

64. Schliesser JS, Kruse R, Fallon LF. Cervical radiculopathy treated with chiropractic flexion distraction

manipulation: A retrospective study in a private practice setting. J Manipulative Physiol Ther

2003;26(9):E19.

65. Hurwitz EL, Morgenstern H, Harber P, Kominski GF, Belin TR, Yu F, et al. A randomized trial of medical

care with and without physical therapy and chiropractic care with and without physical modalities for

patients with low back pain: 6-month follow-up outcomes from the UCLA low back pain study. Spine

2002;27(20):2193-204.

66. Tseng SH, Lin SM, Chen Y, Wang CH. Ruptured cervical disc after spinal manipulation therapy: report

of two cases. Spine 2002;27(3):E80-2.

67. Hurwitz EL, Aker PD, Adams AH, Meeker WC, Shekelle PG. Manipulation and mobilization of the

cervical spine. A systematic review of the literature. Spine 1996;21(15):1746-59; discussion 1759-60.

68. Di Fabio RP. Manipulation of the cervical spine: risks and benefits. Phys Ther 1999;79(1):50-65.

32
Degenerative cervical radiculopathy: a review

69. Dillin W, Booth R, Cuckler J, Balderston R, Simeone F, Rothman R. Cervical radiculopathy. A review.

Spine 1986;11(10):988-91.

70. van Gijn J, Reiners K, Toyka KV, Braakman R. Management of cervical radiculopathy. Eur Neurol

1995;35(6):309-20.

71. Mazanec D, Reddy A. Medical management of cervical spondylosis. Neurosurgery 2007;60(1 Supp1

1):S43-50.

72. Anderberg L, Annertz M, Rydholm U, Brandt L, Saveland H. Selective diagnostic nerve root

block for the evaluation of radicular pain in the multilevel degenerated cervical spine. Eur Spine J

2006;15(6):794-801.

73. Bush K, Hillier S. Outcome of cervical radiculopathy treated with periradicular/epidural corticosteroid

injections: a prospective study with independent clinical review. Eur Spine J 1996;5(5):319-25.

74. Scanlon GC, Moeller-Bertram T, Romanowsky SM, Wallace MS. Cervical transforaminal epidural

steroid injections: more dangerous than we think? Spine 2007;32(11):1249-56.

75. Huston CW, Slipman CW, Garvin C. Complications and side effects of cervical and lumbosacral

selective nerve root injections. Arch Phys Med Rehabil 2005;86(2):277-83.

76. Huntoon MA. Anatomy of the cervical intervertebral foramina: vulnerable arteries and ischemic

neurologic injuries after transforaminal epidural injections. Pain 2005;117(1-2):104-11.

77. Rathmell JP, Aprill C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology

2004;100(6):1595-600.

33
34
Chapter 3

Interobserver agreement
on MRI evaluation of
patients with cervical
radiculopathy
Barbara Kuijper, Anita Beelen, Bas F W van der Kallen, Frans Nollet, Geert J Lycklama a
Nijeholt, Marianne de Visser, Jos Th J Tans

Clinical Radiology 2011

35
Chapter 3

Abstract

Objectives
To evaluate the interobserver agreement on magnetic resonance imaging (MRI) evaluation
of herniated discs, spondylotic neuroforaminal stenosis and root compression in patients
with recent onset cervical radiculopathy. In addition, to assess the added value of
disclosure of clinical information to interobserver agreement.

Participants
82 patients with less than one month of symptoms and signs of cervical radiculopathy.

Methods
MRIs were evaluated independently by two neuroradiologists who were unaware of
clinical findings. MRI analysis was repeated after disclosure of clinical information.
Interobserver agreement was calculated using kappa statistics.

Results
The kappa score for evaluation of herniated discs and of spondylotic foramen stenosis
was 0.59 and 0.63, respectively. A kappa score of 0.67 was found for the presence of
root compression. After disclosure of clinical information kappa scores increased slightly;
for detection of herniated discs from 0.59 to 0.62, for spondylotic foramen stenosis from
0.63 to 0.66 and for root compression from 0.67 to 0.76.

Conclusion
Interobserver reliability of MRI evaluation in patients with cervical radiculopathy was
substantial for root compression, with or without clinical information. Agreement on the
cause of the compression, i.e., herniated disc or spondylotic foraminal stenosis, was
lower.

36
Interobserver agreement on MRI evaluation

Introduction

In patients with cervical radiculopathy, magnetic resonance imaging (MRI) is the


imaging technique of choice for the detection of root compression by disc herniation
and osteophytes. 1-4 However, no data are available about the interobserver variability
of the interpretation of cervical spine MRI in patients with clinical signs of root
compression. Further, it is unknown whether the availability of clinical information
diminishes interobserver variation. In daily practice clinical information is usually
considered instrumental in assessing radiological examinations. However, there are
studies showing no significant increase in interobserver agreement after disclosure of
clinical information.56 In the present study we evaluated interobserver reliability of MRI
evaluation of herniated discs, spondylotic neuroforaminal stenosis and root compression
in patients with cervical radiculopathy. In addition, we assessed the effect of disclosure of
clinical information on interobserver variability.

Materials and methods

Subjects
This prospective study comprised a cohort of 82 patients with a clinical diagnosis of
cervical radiculopathy with an onset of less than 1 month in whom the efficacy of either a
cervical collar or physiotherapy was compared with a wait and see policy. The diagnosis
cervical radiculopathy was made by a neurologist according to the following inclusion
criteria: radiation of arm pain distal to the elbow, and at least one of the following: (1)
worsening of the arm pain by neck movements, (2) sensory symptoms in one or more
adjacent dermatomes, (3) diminished deep tendon reflexes in the affected arm, or (4)
muscle weakness in one or more adjacent myotomes. The medical ethics committees of
the participating hospitals approved the protocol. Written informed consent was obtained
from all patients.

MRI protocol
MRI was performed at 1.5 Tesla, and included sagittal proton density (PD) and T2
weighted turbo spin-echo imaging (TSE; TR/TE: 2900/23/182; echo train length (ETL):
5) and T1 weighted TSE (TR/TE 664/13; ETL: 3), using 3 mm thick slices and pixels
of 1 mm2. Axial imaging consisted of a T2 weighted flow-compensated gradient echo
sequence (TR/TE 1140/27) and T1 TSE (TR/TE 538/13, ETL: 3) using 3 mm thick slices

37
Chapter 3

and 1 mm2 pixels. Axial slices were placed perpendicular to the vertebral bodies and
were angulated in the same way. Axial imaging covered intervertebral spaces C4 to Th1.

MRI evaluation protocol


MRI examinations were evaluated independently by two neuroradiologists who were
unaware of clinical findings. The presence or absence of a herniated disc or bony
foraminal stenosis by spondylarthrosis was evaluated at each cervical level on the
right and left side. Differentiation between bony foraminal stenosis and herniated disc
was made based on appearance on axial T2* weighted flow-compensated gradient
Figuren
echo boekje
boekje
sequence images. On these images, disc material shows relatively high signal
intensity compared to adjacent bony structures (figure 1). Also, on sagittal PD weighted
Figuren boekje
Chapter
images, 33
herniated disc often has a slightly different signal intensity compared to bone. 47
Chapter 3
Examples of disc herniations and foraminal stenosis are shown in figures 1 and 2.
Figure 1 Figure 1

A B
C
C

A B
B D

Figure 1 Figure 1.
Sagittal T1 (A) and proton-density (B) weighted, and corresponding axial T2 weighted
flow-compensated
Sagittal T1 (A) and proton-density gradient
(B) weighted, echo sequence
and corresponding images
axial T2 (C) and
weighted T1 (D) weighted
flow-compensated MR
gradient
echo sequence images (C)images
and T1showing a lateral
(D) weighted disc herniation
MR images showing aatlateral
level disc
C5-C6 on the at
herniation left, best
level appreciated
C5-C6 on the
on axial T2 weighted image (C, arrow), due to relatively high signal of herniated disc.

D
left, best appreciated on axial T2 weighted image (C, arrow), due to relatively high signal of herniated disc.
D
The probability of root compression, either caused by herniated disc or spondylotic
neuroforamen stenosis, was scored on a five-point scale: definitely no root
Figure
Figure 1.
1.
Sagittal
Sagittal T1
T1 (A)
compression,
(A) and
possiblyproton-density
and proton-density (B) (B) weighted,
weighted,and
no root compression, indeterminate,corresponding
and correspondingaxial
possibly root
axialT2
T2weighted
weighted
flow-compensated
flow-compensated gradient echo sequence images (C) and T1 (D) weightedMR
compression and gradient
definitely echo
root sequence
compression. images
Presence of (C) and
other T1 (D) weighted
abnormalities like MR
images
images
spinal showing
showing
canal aawith
stenosis lateral
lateral disc
disc herniation
herniation
or without at
atlevel C5-C6
leveltumors
cord compression, on
C5-C6etc. the
onwas left,
left,best
therecorded, appreciated
bestbut
appreciated
on
on axial
wasaxial T2
T2 weighted
weighted
not included
image
image (C,
in the analysis. (C, arrow),
arrow), due
dueto
torelatively
relativelyhigh
highsignal
signalofofherniated
herniateddisc.
disc.

38
Interobserver agreement on MRI evaluation

Figure 2

Sagittal T1 MR image (A) showing degenerative changes at several levels. Axial T2 (B) and T1 (C) weighted images at
level C5-C6 showing bony foraminal stenosis (arrow) with low signal intensity of the bony changes on T2 (B).

Disclosure of clinical information


Directly after the blinded evaluation, the complete MRI analysis was repeated with disclosure
of clinical information including side of arm pain and suspected level of root compression.
This clinically affected level was determined by a neurologist, based on the pattern of
irradiating pain, sensory changes, diminished deep tendon reflexes or muscle weakness.

Statistical analysis
Data on root compression were dichotomized as either root compression (comprising
the scores possibly root compression and definitely root compression) or no root
compression (definitely no root compression, possibly no root compression and
indeterminate). Interobserver agreement was calculated using kappa statistics. We
calculated overall kappas, concerning all levels and both sides. Four levels were
investigated separately on both sides (C4-5 to C7-Th1). Kappa values were calculated
for the presence of herniated disc, foraminal stenosis and root compression. The kappa
values were interpreted as follows: values between 0 and 0.2 represent poor agreement,
0.21 and 0.40 fair agreement, 0.41 and 0.60 moderate agreement, and 0.61 and 0.80
substantial agreement. A value above 0.80 is considered excellent agreement. 8-10

39
Chapter 3

Results

In 4 of the 82 patients reliable MRI-data could not be obtained due to artifacts by pain (n=2)
and claustrophobia (n=2). For the remaining 78 MRI examinations interobserver agreement
and kappa scores could be calculated at a total of 624 segments (4 levels on right and left
side). The mean age of the patients was 48.8 (SD 9.9) years. 47.4% was male (n=37).

Agreement without clinical information

All segments
With regard to detection of herniated discs at all segments the neuroradiologists
agreed in 80.8% and interobserver agreement was moderate (kappa 0.59, CI 0.40-0.77)
(table 1). For evaluation of spondylotic foramen stenosis, the neuroradiologists agreed
in 82.1% of the cases with a kappa of 0.63 (CI 0.46-0.81) (table 2). An agreement of
91.0% and a kappa of 0.67 (CI 0.43-0.90) were found for presence or absence of root
compression (table 3).

Per segment
Percentages of agreement were high at level C5 with comparatively low kappa
values. For presence of herniated discs and spondylotic foramen stenosis agreement
percentages and kappa scores were higher at the C7 than at the C6 segments, for

Table 1 Interobserver agreement on the presence of herniated disc in 78 patients with cervical
radiculopathy.

Presence of herniated discs, n (%) agreement % kappa 95%

radiologist #1 radiologist #2 confidence interval

Overall 46 (59.0) 53 (67.9) 80.8 0.59 0.40-0.77

C5 right 1 (1.28) 2 (2.6) 98.7 0.66 0.00-1.32

C5 left 0 0 100.0 *

C6 right 14 (17.9) 14 (17.9) 87.2 0.56 0.31-0.82

C6 left 14 (17.9) 13 (16.7) 91.0 0.69 0.47-0.91

C7 right 13 (16.7) 13 (16.7) 94.9 0.82 0.64-0.99

C7 left 22 (28.2) 26 (33.3) 87.2 0.70 0.53-0.87

C8 right 0 0 100 *

C8 left 0 0 100 *

40
Interobserver agreement on MRI evaluation

Table 2 Interobserver agreement on the presence of spondylotic foraminal stenosis in 78 patients


with cervical radiculopathy.

presence of spondylotic stenosis, n (%) agreement kappa 95%

radiologist #1 radiologist #2 % confidence interval

Overall 43 (55.1) 49 (62.8) 82.1 0.63 0.46-0.81

C5 right 6 (7.7) 8 (10.3) 92.3 0.53 0.17-0.89

C5 left 5 (6.4) 8 (10.3) 91.0 0.42 0.00-0.83

C6 right 29 (37.2) 33 (42.3) 84.6 0.68 0.51-0.85

C6 left 23 (29.5) 30 (38.5) 78.2 0.52 0.32-0.72

C7 right 23 (29.5) 26 (33.3) 91.0 0.79 0.65-0.94

C7 left 24 (30.8) 26 (33.3) 89.7 0.77 0.61-0.92

C8 right 0 0 100 *

C8 left 0 0 100 *

Table 3 Interobserver agreement on the presence of root compression in 78 patients with cervical
radiculopathy.

presence of root compression, n (%) agreement % kappa 95%

radiologist #1 radiologist #2 confidence interval

Overall 67 (85.9) 64 (80.8) 91.0 0.67 0.43-0.90

C5 right 1 ( 1.3) 5 (6.4) 94.9 0,32 -0.33-0.97

C5 left 1 (1.3) 2 (2.6) 96.2 -0.17 -1.00-1.00

C6 right 27 (34.6) 28 (35.9) 85.9 0.69 0.52-0.86

C6 left 25 (32.1) 25 (32.1) 87.2 0.71 0.54-0.88

C7 right 24 (30.8) 20 (25.6) 82.1 0.56 0.35-0.77

C7 left 35 (44.9) 32 (41.0) 85.9 0.71 0.56-0.87

C8 right 0 0 100 *

C8 left 0 0 100 *

* kappa statistics could not be calculated

41
Chapter 3

root compression these were almost equal (tables 1, 2, 3). Kappa scores could not be
calculated for the C8 segments due to the absence of abnormalities.

Agreement after disclosure of clinical information


After disclosure of the side and clinical expected level of root compression, the
neuroradiologists re-evaluated all MRI images. One neuroradiologist changed his
assessment once: indeterminate root compression was altered in possibly root
compression. The other neuroradiologist changed four assessments of three separate
MRI examinations. The degree of root compression was altered in two patients: in one of
these two cases a herniated disc that was seen before disclosure was now not observed,
in another case foraminal stenosis on top of an already detected herniated disc was
noted.

After disclosure of clinical information the interobserver agreement and kappa scores
did increase, but not significantly; the kappa of overall herniated discs changed from
0.59 (CI 0.40-0.77) to 0.62 (CI 0.43-0.80) and that of spondylotic foramen stenosis from
0.63 (CI 0.46- 0.81) to 0.66 (CI 0.49-0.83). For blinded MRI evaluation the kappa of root
compression was 0.67 (CI 0.43-0.90) and after disclosure 0.76 (CI 0.57-0.96).

Discussion

Our study showed an overall kappa of 0.67 for the detection of root compression on
MRI in a series of patients with well defined subacute onset cervical radiculopathy. This
can be interpreted as substantial agreement. In 9% of the 624 examined segments
no agreement was found on the clinically most important question, whether root
compression was present or not. These results allow neuroradiologists to report on this
condition with confidence. Although cervical spine MRI is routinely used as a diagnostic
tool in patients with cervical radiculopathy, in particular in those who fail conservative
treatment and are candidates for surgery, the interobserver variability of neuroradiological
interpretation has, surprisingly, never been studied to our knowledge.

Several studies assessed interobserver agreement in evaluation of cervical MRI


examinations, but patients with clinically well-defined cervical radiculopathy were not
included in any of those studies.. The kappa value in our study is slightly lower than that
found in a study where a grading system for cervical disc degeneration was evaluated in

42
Interobserver agreement on MRI evaluation

patients with unspecified cervicobrachial pain. This grading system, based on nucleus
signal intensity and structural homogenity yielded a kappa of 0.73 to 0.83. 11 Others have
used grading systems for cervical disc MRI properties as well, 1213 but none focused on
the presence of root compression making a direct comparison of the results of these
studies with ours less relevant. Apart from the fact that no grading systems for cervical root
compression are available we prefer to stay close to the clinical practice in which grading
systems are not used on a routine basis.

In earlier studies on cervical spine MRI a wide range of kappa scores was reported. In
a study assessing 10 patients with cervical myelopathy high agreement, but low kappa
scores ranging from -0.12 to 0.51 were found, mainly due to methodological choices. 14
Another study investigated interobserver agreement on cervical spine MRIs in patients
complaining of dizziness, without neck or arm pain. This study showed a moderate
agreement on spinal cord compression (kappa 0.53), but did not include herniated discs
or root compression. 15 Therefore, these studies can not be compared with our cohort
encompassing patients with cervical radiculopathy.

In a study on interobserver agreement on the evaluation of disc herniation and root


compression in patients with lumbar radiculopathy, kappa scores of 0.63 and 0.77,
respectively, were found. 5 The methodology of this study is comparable with that of ours
and the kappa values are within the same range.

We found a strikingly low prevalence of root compression at level C5 and C8 compared


to C6 and C7 which has also been shown in other studies. Reported percentages of
compression of root C5 range from 2-7% and of root C8 from 6-12%, compared to
18-19% at C6 and 46-69% at C7. 16-18 The low prevalence of abnormalities at levels C4-5
and C7-Th1 in our study can probably be explained by a selection bias since we only
included patients with pain radiating below the elbow whereas in C5 radiculopathy pain is
mostly perceived in the upper arm. We cannot explain why our series did not encompass
patients with C8 radiculopathy. At levels C5 and C8 agreement was high, but due to low
prevalence of abnormalities kappa scores were low 101920 : a kappa score at C8 could
even not be calculated due to the complete agreement on the absence of herniated
discs, spondylotic foramen stenosis and root compression.

A difference was found between percentages of agreement for the assessment of disc
herniation and spondylotic foramen stenosis (80.8% and 82.1%, respectively) and that of

43
Chapter 3

root compression (91.0%). Also kappa scores were higher for the latter. In other words,
the presence of root compression itself is more reliably detected than the cause of root
compression.

After disclosure of the level and side of the radiculopathy by the neurologist the
neuroradiologists reported root compression more frequently, but interobserver
agreement did not increase significantly. The same results were observed in a study
assessing kappa scores in lumbar herniated discs, which did not change significantly
after unblinding. More bulging discs without root compression were seen but
interobserver agreement remained unaltered. 5 A limitation of the present study is that the
evaluation with clinical information occurred directly after the blinded evaluation possibly
leading to underestimation of the differences between the two evaluations. In addition,
all our patients had a cervical radiculopathy on clinical grounds and, therefore, the initial
radiological evaluation was only partly blind. Therefore we still are of the opinion that
clinical information is a useful adjunct to the radiological evaluation.

Conclusion

There is substantial interobserver agreement between neuroradiologists for the detection


of root compression on MRI in a series of patients with well-defined recent onset cervical
radiculopathy, with or without clinical information. Agreement on the cause of root
compression i.e., disc herniation and spondylotic foraminal stenosis is less robust.

44
Interobserver agreement on MRI evaluation

1. Bartlett RJ, Hill CR, Gardiner E. A comparison of T2 and gadolinium enhanced MRI with CT

myelography in cervical radiculopathy. Br J Radiol 1998;71(841):11-9.

2. Birchall D, Connelly D, Walker L, Hall K. Evaluation of magnetic resonance myelography in the

investigation of cervical spondylotic radiculopathy. Br J Radiol 2003;76(908):525-31.

3. Fortin J, Riethmiller DW, Vilensky JA. No clear winner in differing imaging modalities for cervical

radiculopathy. Pain Physician 2002;5(3):285-7.

4. Kaiser JA, Holland BA. Imaging of the cervical spine. Spine 1998;23(24):2701-12.

5. van Rijn JC, Klemetso N, Reitsma JB, Majoie CB, Hulsmans FJ, Peul WC, et al. Observer

variation in MRI evaluation of patients suspected of lumbar disk herniation. AJR Am J Roentgenol

2005;184(1):299-303.

6. Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliability and outcome after

thrombolytic treatment--systematic review. Radiology 2005;235(2):444-53.

7. Khanna AJ, Carbone JJ, Kebaish KM, Cohen DB, Riley LH, 3rd, Wasserman BA, et al. Magnetic

resonance imaging of the cervical spine. Current techniques and spectrum of disease. J Bone Joint

Surg Am 2002;84-A Suppl 2:70-80.

8. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics

1977;33(1):159-74.

9. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority

agreement among multiple observers. Biometrics 1977;33(2):363-74.

10. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med

2005;37(5):360-3.

11. Miyazaki M, Hong SW, Yoon SH, Morishita Y, Wang JC. Reliability of a magnetic resonance

imaging-based grading system for cervical intervertebral disc degeneration. J Spinal Disord Tech

2008;21(4):288-92.

12. Christe A, Laubli R, Guzman R, Berlemann U, Moore RJ, Schroth G, et al. Degeneration of the

cervical disc: histology compared with radiography and magnetic resonance imaging. Neuroradiology

2005;47(10):721-9.

13. Kolstad F, Myhr G, Kvistad KA, Nygaard OP, Leivseth G. Degeneration and height of cervical discs

classified from MRI compared with precise height measurements from radiographs. Eur J Radiol

2005;55(3):415-20.

14. Cook C, Braga-Baiak A, Pietrobon R, Shah A, Neto AC, de Barros N. Observer agreement of spine

stenosis on magnetic resonance imaging analysis of patients with cervical spine myelopathy. J

Manipulative Physiol Ther 2008;31(4):271-6.

15. Colledge N, Sellar R, Wardlaw J, Lewis S, Mead G, Wilson J. Interobserver agreement in magnetic

resonance brain and neck imaging. J Neuroimaging 2006;16(1):47-51.

45
Chapter 3

16. Yoss RE, Corbin KB, Maccarty CS, Love JG. Significance of symptoms and signs in localization of

involved root in cervical disk protrusion. Neurology 1957;7(10):673-83.

17. Friedenberg ZB, Edeiken J, Spencer HN, Tolentino SC. Degenerative changes in the cervical spine. J

Bone Joint Surg Am 1959;41-A(1):61-70 passim.

18. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-

35.

19. Cicchetti DV, Feinstein AR. High agreement but low kappa: II. Resolving the paradoxes. J Clin

Epidemiol 1990;43(6):551-8.

20. Feinstein AR, Cicchetti DV. High agreement but low kappa: I. The problems of two paradoxes. J Clin

Epidemiol 1990;43(6):543-9.

46
47
48
Chapter 4

Root compression on
MRI compared with
clinical findings in
patients with recent onset
cervical radiculopathy
Barbara Kuijper, Jos Th J Tans, Bas F W van der Kallen, Frans Nollet, Geert J Lycklama a
Nijeholt, Marianne de Visser

Journal of Neurology, Neurosurgery & Psychiatry2011

49
Chapter 4

Abstract

Objectives
To evaluate the occurrence of symptomatic and asymptomatic root compression caused
by herniated discs and spondylotic foraminal stenosis by magnetic resonance imaging
(MRI) in patients with recent onset cervical radiculopathy.

Participants
78 patients with symptoms and signs of cervical radiculopathy of less than one months
duration.

Methods
We determined the clinically suspected level of root compression in each patient. Two
neuroradiologists independently evaluated MRIs, blinded for the clinical findings. For
each patient the level of root compression on MRI was compared with the clinically
affected level. We also examined the cause of compression: herniated disc, spondylotic
foraminal stenosis or both.

Results
In 73% of patients, the clinically affected root was compressed on MRI. In 45%, MRI
showed root compression without clinical substrate together with, or to a lesser extent
without, the coexistence of compression of the clinically affected root. MRIs were
assessed as normal in 13-15% of cases, and in 9-10% only asymptomatic roots were
compressed. Herniated discs without spondylosis were more often responsible for root
compressions only at the clinically affected level and spondylotic foraminal stenosis for
multiple root compression including compression of clinically unaffected roots.

Conclusion
MRI findings in patients with cervical radiculopathy should be interpreted together with
the clinical findings as false-positive and false-negative MRIs occur rather frequently.

50
Root compression on MRI compared with clinical findings

Introduction

Root compression without clinical substrate as a coincidental finding on MRI of the


cervical spine is well known. 1-4 However, the presence of a clinically evident cervical
radiculopathy without root compression on MRI also occurs. 5-7 Knowledge of the
occurrence of this phenomenon is important to make correct treatment decisions.The
most common causes of cervical root compression are narrowing of the foraminal space
secondary to spondylarthrosis and herniated discs. 8-10 Large studies investigating the
relationship between the radiological and clinical findings are lacking. In this study, we
assess the relation between the clinically affected level of root compression and the level
of root compression on magnetic resonance imaging (MRI) in a group of patients with
recent onset cervical radiculopathy.

Methods

Subjects
In our prospective cohort study we included patients with a clinical diagnosis of unilateral
cervical radiculopathy in whom efficacy of either a cervical collar or physiotherapy was
compared with a wait and see policy. The diagnosis cervical radiculopathy was made
by a neurologist according to the following inclusion criteria: radiation of arm pain distal
to the elbow, symptoms for less than one month, arm pain on a visual analogue scale of
40 mm or more, plus at least one of the following: (1) worsening of the arm pain by neck
movements, (2) sensory symptoms in one or more adjacent dermatomes, (3) diminished
deep tendon reflexes in the affected arm, or (4) muscle weakness in one or more adjacent
myotomes. The clinically affected level was determined by the neurologist via structured
history taking and physical examination data. Written informed consent was obtained
from all patients. The medical ethics committees of the participating hospitals approved
the protocol.

MRI protocol
MRI was performed at 1.5 Tesla, and included sagittal proton density (PD) and T2
weighted turbo spin-echo imaging (TSE; TR/TE: 2900/23/182; echo train length (ETL):
5) and T1 weighted TSE (TR/TE 664/13; ETL: 3), using 3 mm thick slices and pixels of 1
mm2. Axial imaging consisted of a T2 weighted MEDIC sequence (TR/TE 1140/27) and T1
TSE (TR/TE 538/13, ETL: 3) using 3 mm thick slices and 1 mm2 pixels. Axial slices were

51
Chapter 4

placed perpendicular to the vertebral bodies and were angulated in the same way. Axial
imaging covered intervertebral spaces C4 to T1. Per patient four levels were investigated
on both sides: C4-5 (root C5), C5-6 (root C6), C6-7 (root C7) and C7-Th1 (root C8).

MRI evaluation protocol


The MRI examinations were evaluated independently by two neuroradiologists who
were blinded to clinical findings. For each cervical level, the presence of herniated disc
and bony foraminal stenosis by spondylarthrosis was assessed. The probability of
root compression, caused by either herniated disc or spondylotic foraminal stenosis,
was scored on a five-point scale: definitely no root compression, possibly no
root compression, indeterminate, possibly root compression and definitely root
compression. Presence of other abnormalities like spinal canal stenosis with or without
cord compression, tumors, and other findings was recorded but not included in the
analysis.

Analysis
Data on root compression were dichotomized as either root compression (possibly
and definitely root compression) or no root compression (definitely and possibly no
root compression and indeterminate). For each patient we compared the level of
root compression with that of the clinically affected root. As it is not always feasible
to determine with certainty the level of root compression by history and physical
examination, we separately recorded the presence of root compression one level higher
or lower. We also examined the cause of compression: herniated disc, spondylotic
foraminal stenosis or both.

Results

From May 2005 to December 2006, 82 patients with a clinical diagnosis of recent onset
unilateral cervical radiculopathy underwent MRI of the cervical spine. Four scans were
of poor quality because of movement artifacts by pain (n=2) and claustrophobia (n=2).
Seventy-eight MRIs were fully examined.

Data on the clinical signs of radiculopathy in these 78 patients are listed in table 1.
Sensory abnormalities were found in 89.9% of the patients, diminished reflexes in 48.1%
and muscle weakness in 29.1%. Two patients (2.6%) only experienced pain. The mean

52
Root compression on MRI compared with clinical findings

Table 1 Clinical signs and symptoms in 78 patients with suspected cervical radiculopathy

No (%)

only pain 2 (2.6)

muscle weakness 23 (29.5)

diminished reflexes 38 (48.7)

sensory abnormalities 70 (89.7)

age was 47.4 year (range 24-70). There was no difference in age between patients with
herniated discs and spondylotic foraminal stenosis.

Single and multiple root compression assessments are listed in table 2. The clinically
affected root was judged as compressed in 73.1/69.2% (radiologist 1/2) of all cases.
When we consider compression of the roots adjacent to the clinically affected level as
symptomatic root compression, this percentage increased to 78.2/74.4%.

Table 2 Single and multiple root compression assessed on MRI in 78 patients with suspected
cervical radiculopathy

total n=78 radiologist 1 radiologist 2

Single root compression on MRI n=37 (47.4) n=37 (47.4)

Compression of clinically affected root only 28 (35.9) 27 (34.6)

Compression of the root one level higher or lower than that of the clinically 3 (3.8) 2 (2.6)

affected root

Compression of the root two levels higher or lower 1 (1.3) 1 (1.3)

Compression of contralateral root 5 (6.4) 7 (9.0)

Multiple root compression on MRI n=31 (39.7) n=29 (37.2)

Compression of clinically affected root together with compression one level 2 (2.6) 1 (1.3)

higher or lower

Compression of clinically affected root and contralateral root compression 18 (23.1) 13 (16.7)

Compression of clinical affected root together with one level higher or lower 9 (11.5) 13 (16.7)

and contralateral root compression

Compression of root one level higher and lower together with contralateral 1 (1.3) 2 (2.6)

root compression

Compression of multiple contralateral roots 1 (1.3) 0

Normal MRI n=10 (12.8) n=12 (15.4)

53
Chapter 4

Single-level, one-sided root compression was found in 47.4% of patients. Most of these
compressions were of the clinically affected root, according to table 2 in 35.9/34.6%
(radiologist 1/2). Multiple root compression was seen in 39.7/37.2% of cases (radiologist
1/2), mostly a combination of clinically affected and unaffected roots. One or more
clinically unaffected root compressions were reported in 44.9/46.2% of all patients. MRIs
were assessed as normal in 12.8/15.4% of cases and in 9.0/10.3% only unaffected roots
were compressed. We additionally recorded the clinical signs listed in table 1, for the
group of patients on whom the 2 neuroradiologists agreed that the MRI showed no root
compression at the clinical level or one level above or below. No difference was found
between the clinical data of patients with and without root compression on MRI.

Herniated discs were the single cause of root compression in 41.1/34.6% and
spondylotic foraminal stenosis in 35.9/21.7% of our patients (table 3). A combination of
herniated disc and spondylotic foraminal stenosis was reported in 10.3/28.2%. Herniated
discs were mainly seen in the group with one compressed root, whereas spondylotic
foraminal stenosis more often caused multiple root compressions, including compression
of clinically unaffected roots (table 3).

Table 3 Causes of root compression in 78 patient with suspected cervical radiculopathy who
underwent MRI

total n=78 radiologist 1 radiologist 2

Single root compression n=37 (47.4) n=37 (47.4)

-by herniated disc 25 (32.1) 24 (30.8)

-by spondylotic foraminal stenosis 10 (12.8) 4 (5.1)

-by both 2 (2.6) 9 (11.5)

Multiple root compression n=31 (39.7) n=29 (37.2)

-by herniated disc(s) 7 (9.0) 3 (3.8)

-by spondylotic foraminal stenosis 18 (23.1) 13 (16.7)

-by both 6 (7.7) 13 (16.7)

Normal MRI n=10 (12.8) n=12 (15.4)

54
Root compression on MRI compared with clinical findings

Discussion

To the best of our knowledge, this is the first study evaluating the correlation between
clinical signs and symptoms, and the occurrence of root compression on MRI in a
well-defined population of patients with recent onset unilateral cervical radiculopathy, in a
setting closely resembling clinical practice.

In almost three-quarters of the 78 patients, the clinical level corresponded with the level
of root compression on MRI. Assessment of the level of root compression by neurological
examination alone is reported to be difficult. 11-16 Inclusion of the root compressions one
level above or below the clinically suspected level, led to an increase of the percentage
of matching clinical and MRI levels by only 5% for both radiologists, indicating that the
localising value of the neurological examination is quite good.

Strikingly, we found false-positive root compressions in 45% of the cases, defined as


compressions on the contralateral, that is, asymptomatic, side or at least two levels
higher or lower than the clinically affected level. These asymptomatic root compressions
were often present together with root compression at the clinically affected level or
adjacent to this. In order to avoid unnecessary treatment of root compressions observed
on MRI without clinical substrate, physicians should only interpret MRI results after a
careful history and neurological examination.

Earlier studies reported a high percentage of cervical degenerative disease on MRI in


asymptomatic patients, with a higher prevalence in older age. Teresi et al studied 100
MRIs of the cervical spine of patients who were investigated for laryngeal disease. Twenty
percent of patients aged 45-54 years and 57% of patients older than 64 years had
cervical disc protrusion without clinical symptoms. 17 Another study on asymptomatic
subjects showed major abnormalities (herniated disc, foraminal stenosis, disc space
narrowing) in 28 percent of people over 40 years of age. 4 However, these studies only
investigated asymptomatic patients, whereas our study included patients based on the
presence of cervical radiculopathy at one level.

We also found a fair amount of false-negative MRIs. MRIs were assessed as normal in
12.8 /15.4% of cases, and in 9.0/10.3% only asymptomatic roots were compressed. So,
together we had 21.8/25.6% of false-negative MRIs.

55
Chapter 4

Patients were eligible for our study if they fulfilled strict inclusion criteria. We feel
confident that they had indeed cervical radiculopathy because all patients except two
had neurological deficits corresponding with radicular pain, and no other conditions
emerged during a 6-months follow-up period. We had low percentages of surgery so we
could not confirm our diagnosis surgically in most cases.

The cause of the root compression was most often herniated disc, particularly in those
patients with unilateral MRI abnormalities only at the clinically affected level. This is
noteworthy, since it is often assumed that spondylotic foraminal stenosis is the most
common cause of root compression in cervical radiculopathy. 8-10 Our study shows that
foraminal stenosis was more frequently asymptomatic. The occurrence of herniated discs
and spondylotic changes was not related to age.

As described in chapter 3, we previously found a high interobserver agreement of 91%


for MRI evaluation of root compression with a kappa score of 0.67. This agreement was
less for the cause of the compression, that is, herniated disc (81%) and spondylotic
foramen stenosis (82%). In the present study, the two neuroradiologists also disagreed
more on assessment of herniated discs and foraminal stenoses than on the presence
of root compression. In particular, the reported percentages of spondylotic foraminal
stenosis differed (table 3). It is known that on MRI spondylotic foraminal stenosis is often
more difficult to detect and CT-myelography techniques are probably more accurate. 5-718
Our neuroradiologists may have under-reported spondylotic foraminal stenosis, although
the 55 and 63% of patients in whom stenosis was found by radiologist 1 and 2 suggest
otherwise. (chapter 3)

Another limitation is that the neuroradiologists knew that all patients in the study had
a clinical diagnosis of cervical radiculopathy. They may have been more aware of the
possibility of root compression in general, resulting in higher percentages abnormal MRIs.
Because the radiologists had no information on the level and side of the radiculopathy,
the main results of our study on the relation between clinical and MRI findings seem
reliable.

56
Root compression on MRI compared with clinical findings

Conclusion and clinical implications

It is evident from this study that MRI findings in patients with cervical radiculopathy are
meaningful only in a clinical context. False-negative MRI results were encountered in
almost one-quarter, and false-positive results in half of the patients. Therefore, cervical
MRI is useful only when there is a clear picture of cervical radiculopathy.

57
Chapter 4

1. Jensen MV, Tuchsen F, Orhede E. Prolapsed cervical intervertebral disc in male professional drivers in

Denmark, 1981-1990. A longitudinal study of hospitalizations. Spine 1996;21(20):2352-5.

2. van Rijn JC, Klemetso N, Reitsma JB, Bossuyt PM, Hulsmans FJ, Peul WC, et al. Observer variation

in the evaluation of lumbar herniated discs and root compression: spiral CT compared with MRI. Br J

Radiol 2006;79(941):372-7.

3. van Rijn JC, Klemetso N, Reitsma JB, Majoie CB, Hulsmans FJ, Peul WC, et al. Observer

variation in MRI evaluation of patients suspected of lumbar disk herniation. AJR Am J Roentgenol

2005;184(1):299-303.

4. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal magnetic-resonance scans of

the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am

1990;72(3):403-8.

5. Fortin J, Riethmiller DW, Vilensky JA. No clear winner in differing imaging modalities for cervical

radiculopathy. Pain Physician 2002;5(3):285-7.

6. Bartlett RJ, Hill CR, Gardiner E. A comparison of T2 and gadolinium enhanced MRI with CT

myelography in cervical radiculopathy. Br J Radiol 1998;71(841):11-9.

7. Birchall D, Connelly D, Walker L, Hall K. Evaluation of magnetic resonance myelography in the

investigation of cervical spondylotic radiculopathy. Br J Radiol 2003;76(908):525-31.

8. Tanaka N, Fujimoto Y, An HS, Ikuta Y, Yasuda M. The anatomic relation among the nerve roots,

intervertebral foramina, and intervertebral discs of the cervical spine. Spine 2000;25(3):286-91.

9. Adams C. Cervical spondylotic radiculopathy and myelopathy. Amsterdam/ New York / Oxford:

Elsevier Publishing Company, 1976.

10. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

11. Fisher MA. Electrophysiology of radiculopathies. Clin Neurophysiol 2002;113(3):317-35.

12. Levine MJ, Albert TJ, Smith MD. Cervical Radiculopathy: Diagnosis and Nonoperative Management. J

Am Acad Orthop Surg 1996;4(6):305-316.

13. Yoss RE, Corbin KB, Maccarty CS, Love JG. Significance of symptoms and signs in localization of

involved root in cervical disk protrusion. Neurology 1957;7(10):673-83.

14. Ozgur BM, Marshall LF. Atypical presentation of C-7 radiculopathy. J Neurosurg 2003;99(2

Suppl):169-71.

15. Slipman CW, Plastaras C, Patel R, Isaac Z, Chow D, Garvan C, et al. Provocative cervical discography

symptom mapping. Spine J 2005;5(4):381-8.

16. Narayan P, Haid RW. Treatment of degenerative cervical disc disease. Neurol Clin 2001;19(1):217-29.

17. Teresi LM, Lufkin RB, Reicher MA, Moffit BJ, Vinuela FV, Wilson GM, et al. Asymptomatic degenerative

disk disease and spondylosis of the cervical spine: MR imaging. Radiology 1987;164(1):83-8.

18. Kaiser JA, Holland BA. Imaging of the cervical spine. Spine 1998;23(24):2701-12.

58
59
60
Chapter 5

Results of needle
electromyography
in 176 patients with
cervical radiculopathy
Barbara Kuijper, Robert Jan Schimsheimer, Anita Beelen, Freek Verheul, Frans Nollet,
Marianne de Visser, Jos Th J Tans

Submitted

61
Chapter 5

Abstract

Objectives
To investigate a standardised electromyography protocol in patients with recent onset
cervical radiculopathy.

Methods
For this prospective cross-sectional study we included patients with a clinical diagnosis
of recent onset (less than one month) cervical radiculopathy taken from a cohort of
patients participating in a clinical trial on non-surgical treatment. The electrodiagnostic
consultant was blinded for the clinically suspected level of root compression, symptom
duration and MRI-results. Standardised concentric needle electromyography was done in
five muscles of the affected arm.

Results
176 patients were included. Abnormal spontaneous activity was found in only 16.5% of
cases and neurogenic motor unit action potentials in 29.5%, yielding a total of 39.2% of
abnormal EMG examinations.

Conclusion
The low percentage of EMG abnormalities demonstrates that this standardised limited
EMG protocol is not useful as a routine investigation in patients with recent onset cervical
radiculopathy.

62
Needle electromyography

Introduction

Although many studies on electromyography (EMG) in patients with cervical


radiculopathy have been published 1-6 there is still no consensus on its diagnostic value.
It has been advised to perform EMG in a blinded fashion to properly evaluate its value. 67
However we are not aware of such studies in this condition. In this explorative study we
report our EMG findings in a large series of patients with a clinical diagnosis of cervical
radiculopathy. In every patient we performed a standardised EMG protocol consisting
of needle electrode examination of five arm muscles. The electrodiagnostic consultants
were blinded for clinical and magnetic resonance imaging (MRI) data.

Methods

For this prospective cross-sectional study we included patients with a clinical diagnosis
of recent onset (less than one month) cervical radiculopathy taken from a cohort of
patients participating in a clinical trial on non-surgical treatment. The medical ethics
committees of the participating hospitals approved the protocol. Written informed
consent was obtained from all patients. Patients were examined by the local investigator
(neurologist) who established a clinical diagnosis of cervical radiculopathy based on
history and neurological findings on examination. Inclusion criteria for the study were:
age 18 - 75 years, arm pain on a visual analogue pain scale 40 mm on a 0-100
scale, radiation of arm pain distal to the elbow, and at least one of the following: (1)
provocation of the arm pain by neck movements, (2) sensory changes in one or more
adjacent dermatomes, (3) diminished deep tendon reflexes in the affected arm, (4) muscle
weakness in one or more adjacent myotomes.

Needle electrode examination


The electrodiagnostic consultant was aware of the suspicion of cervical radiculopathy
but blinded for the clinically suspected level of root compression, symptom duration
and MRI-results. Patients were asked not to inform the EMG consultant about their
symptoms. 7 At the time of the EMG, symptom duration had to be at least 3 weeks;
otherwise the EMG was postponed as previously advised. 589 Standardised concentric
needle electromyography was done on five muscles of the affected arm, representing
C5, C6, C7 and C8 myotomes: m. brachioradialis (C5, C6), m. flexor carpi radialis (C6,
C7), m. triceps brachii (C7), m. extensor digitorum communis (C7, C8) and m. abductor

63
Chapter 5

digiti minimi (C8). 356810-12 An examined muscle was considered abnormal if abnormal
spontaneous activity (fibrillations, positive sharp waves) or neurogenic motor unit action
potential (MUAP) abnormalities (increased polyphasic motor units, increased motor unit
amplitudes (>7mV) or high frequency motor unit recruitment) were recorded.

Analyses
We recorded the occurrence of EMG abnormalities (abnormal spontaneous activity and
neurogenic MUAP abnormalities) for each examined muscle and dichotomized the overall
EMG examination as normal or abnormal. The EMG was considered abnormal if EMG
abnormalities occurred in one or more of the above mentioned muscles.

Results

Two hundred and five patients were included in the trial on conservative treatment. Of
these, 29 were excluded from the current EMG study because they refused to undergo
the procedure in advance (n=24) or they refused continuation of the EMG during the
procedure (n=5). So, the number of included EMGs was 176. The mean duration of arm
pain at the day the EMG was performed was 7.4 weeks (SD 2.9).

Clinical characteristics are listed in table 1. The pain in the arm (VAS score 69.6 mm)
was more severe than the pain in the neck (VAS score 57.5 mm). Almost all patients had
sensory disturbances (90.3%), whereas hyporeflexia and muscle weakness were present
in 44.9 and 35.2%.

Table 1 Clinical characteristics of 176 included patients

Mean age (SD) 47.2 (10.0)

Mean (SD) VAS on arm pain (mm)* 69.6 (20.2)

Mean (SD) VAS on neck pain (mm)* 57.5 (28.5)

Provocation of arm pain by neck movements, n (%) 121 (68.8)

Sensory disturbances, n (%) 159 (90.3)

Hyporeflexia, n (%) 79 (44.9)

Muscle weakness, n (%) 62 (35.2)

* Visual Analogue Scale; 100 mm scale (0=no pain; 100=worst pain ever).

64
Needle electromyography

Table 2 Needle electromyography abnormalities recorded from 176 patients with cervical
radiculopathy

n=176 all EMG abnormalities abnormal spontaneous neurogenic MUAP

n (%) activity abnormalities

n (%) n (%)

total 69 (39.2) 29 (16.5) 52 (29.5)

m brachioradialis 12 (6.8) 8 (4.5) 4 (2.3)

m. flexor carpi radialis 15 (8.5) 13 (7.4) 4 (2.3)

m. triceps brachii 20 (11.4) 6 (3.4) 15 (8.5)

m. extensor digitorum communis 25 (14.2) 10 (5.7) 17 (9.7)

m. abductor digiti minimi. 37 (21.0) 8 (4.5) 31 (17.7)

In 39.2% of the EMG examinations we found the predefined abnormalities (table 2).
Abnormal spontaneous activity (16.5%) was less often seen than neurogenic motor
unit action potentials (29.5%). The abductor digiti minimi muscle showed the highest
percentage of MUAP abnormalities.

Discussion

We evaluated EMG data obtained from a standardised protocol in patients with a firm
clinical diagnosis of cervical radiculopathy. Abnormal spontaneous activity, indicating
active or ongoing axonal loss, was found in only 16.5% of patients, neurogenic MUAPs
were seen in 29.5% of patients. As we only included recent onset cases, with a relatively
benign course of the pain from whom only 6.3% needed surgery (see also chapter 6),
it is not surprising that overall only 39.2% of EMGs showed abnormalities. In literature
there is little discussion on the existence of false-negative EMG findings in patients with
clinical evidence of cervical root compression. One explanation is that usually the sensory
fibres are compromised in cervical radiculopathy. 4-6 This was also the case in our cohort
in which 64.8% of the 176 patients had normal strength and 55.1% had normal reflexes,
whereas sensory disturbances were found in 90.2%.

For our study we used a needle electromyography screening limited to 5 muscles


because we chose a program with minimal burden for this large group of patients with a
good prognosis for whom the reason to perform the test was mainly scientific. Also, we

65
Chapter 5

did not include examination of the paraspinal muscles for the same reason. In addition,
there is ample evidence that both sensitivity and specificity are moderate. 561314 However
we might have found higher percentages abnormalities if we had used a less strict study
protocol in which more muscles per patient were examined. Dillingham et al studied
patients with an electrodiagnostically confirmed diagnosis of cervical radiculopathy,
in whom they tried to establish the optimal screening program, comparing screens
consisting of 5 muscles with screens of 6, 7 or 8 muscles. When using 5 muscles without
the paraspinals like in our patients they found EMG abnormalities (spontaneous muscle
activity and MUAP abnormalities) in 84 to 92% of the study population, this percentage
increased to 90 to 98% when paraspinals were included. 3 The high percentages of EMG
abnormalities in this study can not be compared with ours, as the inclusion of patients
was based on the presence of needle electromyographic abnormalities in whom only
65% of patients reported arm pain, whereas we included patients with a clinical diagnosis
of recent onset cervical radiculopathy, i.e., all had severe brachialgia.

Several other publications recommend a screening program of six to eight limb muscles
representing the different myotomes, including paraspinal muscles 5611 However none of
these authors based their recommendations on prospective clinical studies. On the other
hand, we may have overestimated our findings, as we considered EMG abnormalities in
one muscle consistent with involvement of a cervical root, whereas the presence of two
abnormal muscles in one myotome is often advised. 35

One of the strengths of our study is that it fulfils several criteria stated by the American
Association of Electrodiagnostic Medicine (AAEM) in their review on the value of EMG
in cervical radiculopathy. 1 For instance, we had a prospective study design and made a
diagnosis of cervical radiculopathy independent of the EMG results. Other strengths of
our study included the large study population, well-defined inclusion criteria and blinding
of the electrodiagnostic consultants.

A true gold standard for cervical radiculopathy is not available. Surgical assessment of
root compression could be considered the gold standard, but only a minority of patients
will undergo surgery as was also the case in our study (12 out of 205 patients). 1516
Cervical radiculopathy is primarily a clinical diagnosis. Most patients who have long
lasting arm pain will have MRI of the cervical spine as the method of choice. However, we
found false-positive MRI results in 45% and false-negative MRIs in almost one-quarter

66
Needle electromyography

of our patients. (chapter 4) Therefore we refrained from calculation of sensitivity and


specificity of EMG using MRI as gold standard.

The low percentage of EMG abnormalities demonstrates that this standardised limited
EMG protocol is not useful as a routine investigation in patients with recent onset cervical
radiculopathy and relatively mild neurological deficit.

67
Chapter 5

1. American Association of Electrodiagnostic Medicine. The electrodiagnostic evaluation of patients with

suspected cervical radiculopathy: literature review on the usefulness of needle electromyography.

Muscle Nerve 1999;Supplement 8:S213-S221.

2. Miller TA, Pardo R, Yaworski R. Clinical utility of reflex studies in assessing cervical radiculopathy.

Muscle Nerve 1999;22(8):1075-9.

3. Dillingham TR, Lauder TD, Andary M, Kumar S, Pezzin LE, Stephens RT, et al. Identification of cervical

radiculopathies: optimizing the electromyographic screen. Am J Phys Med Rehabil 2001;80(2):84-91.

4. Lauder TD. Physical examination signs, clinical symptoms, and their relationship to electrodiagnostic

findings and the presence of radiculopathy. Phys Med Rehabil Clin N Am 2002;13(3):451-67.

5. Dumitru D, Zwarts M. . Radiculopathy. In: Dumitru D, editor. Electrodiagnostic Medicine. 2nd ed.

Philadelphia: Hanley and Belfus, 2002:713-76.

6. Fisher MA. Electrophysiology of radiculopathies. Clin Neurophysiol 2002;113(3):317-35.

7. Haig AJ, Yamakawa K, Kendall R, Miner J, Parres CM, Harris M. Assessment of the validity of masking

in electrodiagnostic research. Am J Phys Med Rehabil 2006;85(6):475-81.

8. Levin KH. Electrodiagnostic approach to the patient with suspected radiculopathy. Neurol Clin

2002;20(2):397-421, vi.

9. Lomen-Hoerth C, Aminoff MJ. Clinical neurophysiologic studies: which test is useful and when?

Neurol Clin 1999;17(1):65-74.

10. Wilbourn AJ, Aminoff MJ. AAEE minimonograph #32: the electrophysiologic examination in patients

with radiculopathies. Muscle Nerve 1988;11(11):1099-114.

11. Yaar I. The logical choice of muscles for the needle-electromyography evaluation of cervical

radiculopathy. Arch Phys Med Rehabil 2005;86(3):521-6.

12. Levin KH, Maggiano HJ, Wilbourn AJ. Cervical radiculopathies: comparison of surgical and EMG

localization of single-root lesions. Neurology 1996;46(4):1022-5.

13. Date ES, Kim BJ, Yoon JS, Park BK. Cervical paraspinal spontaneous activity in asymptomatic

subjects. Muscle Nerve 2006;34(3):361-4.

14. Gilad R, Dabby R, Boaz M, Sadeh M. Cervical paraspinal electromyography: normal values in 100

control subjects. J Clin Neurophysiol 2006;23(6):573-6.

15. Carette S, Fehlings MG. Clinical practice. Cervical radiculopathy. N Engl J Med 2005;353(4):392-9.

16. Polston DW. Cervical radiculopathy. Neurol Clin 2007;25(2):373-85.

68
69
70
Chapter 6

Cervical collar or
physiotherapy versus
wait and see policy
for recent onset
cervical radiculopathy:
randomised trial
Barbara Kuijper, Jos Th J Tans, Anita Beelen, Frans Nollet, Marianne de Visser

British Medical Journal 2009

71
Chapter 6

Abstract

Objectives
To evaluate the effectiveness of treatment with collar or physiotherapy compared with a
wait and see policy in recent onset cervical radiculopathy.

Design
Randomised controlled trial.

Setting
Neurology outpatient clinics in three Dutch hospitals.

Participants
205 patients with symptoms and signs of cervical radiculopathy of less than one months
duration.

Interventions
Treatment with a semi-hard collar and taking rest for three to six weeks; 12 twice weekly
sessions of physiotherapy and home exercises for six weeks; or continuation of daily
activities as much as possible without specific treatment (control group).

Main outcome measures


Time course of changes in pain scores for arm and neck pain on a 100 mm visual
analogue scale and in the neck disability index during the first six weeks.

Results
In the wait and see group, arm pain diminished by 3 mm/ week on the visual analogue scale
(beta -3.1 mm, 95% CI -4.0 to -2.2) and by 19 mm in total over six weeks. Patients who were
treated with cervical collar or physiotherapy achieved additional pain reduction (collar: beta
-1.9 mm, 95% CI -3.3 to -0.5 mm, physiotherapy: beta -1.9, 95% CI -3.3 to -0.8), resulting in
an extra pain reduction compared with the control group of 12 mm after six weeks.
In the wait and see group, neck pain did not significantly decrease in the first six weeks
(beta -0.9 mm, 95% CI -2.0 to 0.3). Treatment with the collar resulted in a weekly
VAS-scale reduction of 2.8 mm (95% CI -4.2 to -1.3) amounting to 17 mm in six weeks,
whereas physiotherapy with a weekly reduction of 2.4 mm (95% CI -3.9 to -0.8) resulting
in a decrease of 14 mm after six weeks.

72
Cervical collar or physiotherapy versus wait and see policy

Compared with a wait and see policy, the neck disability index showed a significant
change with the use of cervical collar and rest (beta -0.9 mm, 95% CI -1.6 to -0.1) and a
non-significant effect with physiotherapy and home exercises.

Conclusion
A semi-hard cervical collar and rest for three to six weeks or physiotherapy accompanied
by home exercises for six weeks reduced neck and arm pain substantially as compared
to a wait and see policy in the early phase of cervical radiculopathy.

73
Chapter 6

Introduction

Cervical radiculopathy is a common disorder characterized by neck pain radiating to the


arm and fingers corresponding to the dermatome involved. On examination, diminished
muscle tendon reflexes, sensory disturbances and/or motor weakness with dermatomal/
myotomal distribution can be found. The diagnosis is made primarily on clinical grounds.
Magnetic resonance imaging of the cervical spine usually shows the cause of the
radiculopathy; which is usually spondylarthrosis or a herniated disc.
Generally, degenerative cervical radiculopathy with subacute onset has a favourable
prognosis, allowing a wait and see policy during the first six weeks. 1-4 However, as
pain is often excruciating during the first weeks to months, treatment to accelerate the
improvement of pain and function would be highly valuable. Unfortunately, evidence
is lacking for the effectiveness of any non-surgical therapy including a wait and see
policy, cervical collar or physiotherapy. Two randomised trials comparing different
non-invasive treatment modalities in chronic cervical radiculopathy showed no benefit
of physiotherapy or a cervical collar. 56 Treatment in acute or subacute cervical
radiculopathy has not yet been studied. Therefore, we evaluated the effectiveness
of a semi-hard cervical collar in combination with taking as much rest as possible or
physiotherapy and home exercises compared with a wait and see policy in recent onset
cervical radiculopathy. We hypothesised that a treatment policy (collar or physiotherapy)
would result in a faster decline in pain and improvement in function than would a wait and
see policy.

Methods

We did a prospective, randomised controlled trial among patients with less than one
month of symptoms and signs of cervical radiculopathy, to compare the time course
of pain reduction in patients treated with a cervical collar or physical therapy with the
natural course in a control group that did not receive any treatment other than tailored
analgesics, as for all patients in the study. Three Dutch hospitals in The Hague, Gouda
and Amersfoort participated. The medical ethics committees of the participating hospitals
approved the protocol.

74
Cervical collar or physiotherapy versus wait and see policy

Sample size
We calculated the sample size for this three arm trial on the basis of comparison
treatment (cervical collar or physiotherapy) versus a wait and see policy, with equal
allocation in the treatment arms and three repeated measurements (at entry and at three
and six weeks follow-up), with an estimated correlation coefficient of the measurements
of rho=0.7 and a difference in the mean value of the VAS arm of 10 mm, as a clinical
relevant standard deviation in each treatment group of 30 mm. 7 As arm pain is the main
complaint in cervical radiculopathy, we chose this outcome for calculating the sample
size. The total sample size needed to detect this difference on a 5% level of significance
with a power of 90% was 240 patients.

Participants
We asked general practitioners to refer patients with signs of recent onset (less than
one month) cervical radiculopathy. All patients were examined by the local investigator
(neurologist), who made a clinical diagnosis of cervical radiculopathy. Inclusion criteria for
the study were age 18 - 75 years, symptoms less than one month, arm pain on a visual
analogue pain scale (VAS) 40 mm, and radiation of arm pain distal to the elbow, plus at
least one of (1) provocation of the arm pain by neck movements, (2) sensory changes in
one or more adjacent dermatomes, (3) diminished deep tendon reflexes in the affected
arm, or (4) muscle weakness in one or more adjacent myotomes. Muscle weakness was
measured with the 0 (paralysis) to 5 point (normal strength) Medical Research Council
scale. 8 Exclusion criteria were clinical signs of spinal cord compression, previous
treatment with physiotherapy or a cervical collar and insufficient understanding of the
Dutch or English language.

Assignment
For each of the three participating hospitals, randomisation was based on a computer
generated sequence that was kept in a separate box with sealed envelopes. The boxes
had been prepared by an employee from the Department of Biostatistics who was not
otherwise involved in the study. No other stratification or blocking procedure was used.
All envelopes were sequentially numbered. After the patient had given informed consent,
the investigator opened the envelope with the next consecutive number and informed the
patient about the treatment allocated. Patients and investigators were not blinded to the
type of treatment.

75
Chapter 6

Treatment protocols
All patients received written and oral reassurance about the usually benign course of
the symptoms. We explained that a wait and see policy and treatment with collar or
physiotherapy might be equally effective interventions. Patients in all treatment groups
were allowed to use painkillers. Usually, we chose paracetamol with or without a
non-steroidal anti-inflammatory drug, but if necessary we prescribed opiates. We asked
patients to note their drug use, including over the counter analgesics, in a specially
designed diary during the first six weeks after randomisation. MRI and electromyography
(EMG) examination were done in all cases except for patients with contraindications,
such as claustrophobia. If the patient and doctor considered pain reduction to be
insufficient, the option of surgical treatment was discussed.

The cervical collar was a semi-hard collar (Cerviflex S, Bauerfeind) available in six sizes
that could be snugly fitted. Patients were advised to wear the collar for three weeks
during the day and take as much rest as possible. Over the next three weeks they were
weaned from the collar and after six weeks the patients were advised to take it off
completely. The patients were asked to record the time they wore the collar in a diary.

Physiotherapy with a focus on mobilising and stabilising the cervical spine was given
twice a week for six weeks, by certified physiotherapists who participated in the study.
The standardised sessions were hands off and consisted of graded activity exercises
to strengthen the superficial and deep neck muscles. The physiotherapist also educated
the patients to do home exercises. Patients were advised to practise every day and were
asked to record the duration of the home exercises in their diary. Appendix 2 in this thesis
gives the physiotherapy protocol and the list of home exercises.

Patients in the control group were advised to continue their daily activities as much as
possible. They noted in their diaries the number of parts of the day that they were able to
continue their normal activities. All patients were encouraged to contact the investigators
if they had questions.

Outcome measures
Primary outcome measures included neck pain and arm pain on a 100 mm visual
analogue scale and disability assessed with the 100 point neck disability index (NDI).910
This index is a standardised and validated scale for neck pain with 10 five point questions
on functional activities, symptoms and concentration. The total score has to be multiplied

76
Cervical collar or physiotherapy versus wait and see policy

by two. Secondary outcome measures were treatment satisfaction, use of opiates


and working status. Treatment satisfaction was assessed on a five point scale: 1=very
satisfied, 2= satisfied, 3= equivocal, 4=unsatisfied, 5=very unsatisfied; we classified 1 and
2 as satisfied and scores of 4 and 5 as unsatisfied.

At entry and at three weeks, six weeks and six months after randomisation, the patients
filled out all the outcome scales in the presence of, but without interference from, the
research nurse who also acted as data manager. If patients were not able to visit the
outpatient clinic, the three week follow-up was done by a telephone interview. At entry
and after six weeks and six months, the investigators did a standardised neurological
history and examination.

Statistical analysis
For baseline values of continuous variables we used analysis of variance tests with post
hoc comparisons using Bonferronis method. We used Chi-square tests to compare
dichotomous variables at baseline. We used generalized estimating equations (GEE) to
examine the effect of cervical collar and physiotherapy on changes in pain scores and
NDI during the first six weeks, with adjustments for baseline scores. GEE analysis is a
linear regression analysis which takes into account the dependency of the observations
within one patient. 7 Because of expected non-normal distribution of pain scores and
NDI at six months follow-up, we used non-parametric tests (Kruskal-Wallis test with
Mann-Whitney tests for post hoc comparisons) to analyse differences between groups.
We analysed secondary outcomes with Chi-square statistics. We imputed missing values
on the basis of the last observation carried forward technique. We used SPSS software
version 16.0 for all statistical analysis except for GEE-analyses, for which we used Stata
9.0.

Results

Participant flow and follow-up


Between August 2003 and January 2007, 275 patients were assessed for eligibility.
Reasons for exclusion of 70 patients were: refusal to participate (n=9), previous treatment
with physiotherapy (n=12) or cervical collar (n=1), longer than one month duration of
arm pain (n=15), VAS arm pain less than 40 mm (n=8), no cervical radiculopathy (n=26),
language problems (n=6) and unknown (n=2). Some patients had more than one reason

77
Chapter 6

for exclusion. In January 2007 we had to terminate the recruitment of patients for
practical reasons, although at that time only 205 of the planned 240 patients had been
included. Of these 205 patients, 69 were allocated to cervical collar, 70 to physiotherapy
and 66 to the control group (figure 1).

Assessed for eligibility (n= 275)

Excluded (n= 70)


-Not meeting inclusion criteria (n=62)
-Refused to participate (n=8)

Enrolment (n=205)

Randomisation Cervical collar (n=69) Physical therapy (n=70) Control group (n=66)

Follow-Up at 3 Lost to follow-up*: 1 Lost to follow-up*: 2 Lost to follow-up*: 1


weeks Missing: 3 Missing: 2 Missing: 4

Follow-Up at 6 Total lost to follow-up*:1 Total lost to follow-up*:2 Total lost to follow-up*:1
weeks Missing: 0 Missing: 1 Missing: 0

Analysis 6 months Total lost to follow-up*:6 Total lost to follow-up*:2 Total lost to follow-up*:5
follow up

Analysis initial Analyzed: 68 Analyzed: 68 Analyzed: 65


course (first 6 Excluded: 1 (no follow-up) Excluded: 2 (no follow-up) Excluded: 1 (no follow-up)
weeks)

Follow-Up at 6 Analyzed: 63 Analyzed: 68 Analyzed: 61


Months Excluded: 6 (no follow-up) Excluded: 2 (no follow-up) Excluded: 5 (no follow-up)

Figure 1 Flow of participants

* patients who are lost in the study (drop outs)


patients who missed one follow-up but were still participating in the study and attended next follow-up visits

78
Cervical collar or physiotherapy versus wait and see policy

Table 1 Baseline characteristics of 205 patients with recent onset cervical radiculopathy. Values are
numbers (percentages) unless stated otherwise

Characteristics Collar Physiotherapy Controls

(n=69) (n=70) (n=66)

Mean (SD) age (years) 47.0 (9.1) 46.7 (10.9) 47.7 (10.6)

Male sex 38 (55) 34 (49) 32 (48)

Mean (SD) body mass index (kg/m ) 2


26.5 (4.8) 26.2 (4.4) 26.8 (4.8)

Mean (SD) duration of arm pain (weeks) 2.8 (1.4) 2.8 (1.4) 3.0 (1.5)

Mean (SD) duration of neck pain (weeks) 3.3 (2.3) 3.0 (2.1) 3.2 (2.0)

Mean (SD) VAS on arm pain (mm)* 68.2 (19.6) 72.1 (19.2) 70.8 (21.2)

Mean (SD) VAS on neck pain (mm)* 57.4 (27.5) 61.7 (27.6) 55.6 (31.0)

Mean (SD) neck disability index 41.0 (17.6) 45.1 (17.4) 39.8 (18.4)

Smoking 30 (43) 32 (46) 32 (48)

Pain in right arm 35 (51) 30 (43) 21 (32)

Sensory disturbances 60 (87) 63 (90) 64 (97)

Hyporeflexia biceps reflex 12 (17) 13 (19) 16 (24)

Hyporeflexia radial reflex 8 (12) 6 (9) 6 (9)

Hyporeflexia triceps reflex 16 (23) 24 (34) 16 (24)

Muscle weaknessm biceps 9 (13) 5 (7) 6 (9)

Muscle weaknessm triceps 7 (10) 12 (17) 6 (9)

Muscle weaknessm brachialis 8 (12) 8 (11) 8 (12)

Muscle weaknessm extensor digitorum communis 11 (16) 9 (13) 9 (14)

Muscle weaknessm deltoideus 1 (1) 2 (3) 3 (5)

Muscle weaknessm flexor digitorum 8 (12) 6 (9) 4 (6)

Muscle weaknessm abductor digiti minimi* 5 (7) 9 (13) 4 (6)

Sick leave (partial or complete) 29 (42) 31 (44) 26 (39)

Root compression on MRI 48/61 (79) 50/61 (82) 40/54 (74)

MRI=magnetic resonance imaging; VAS=visual analogue scale.


* 100 mm scale (0=no pain; 100=worst pain ever).
100 point scale specific for neck pain, with 10 items on functional status (higher scores represent worse functional
status).
As observed on standardised physical examination (MRC<5).
Measured on three point scale (no sick leave, partial sick leave, fulltime sick leave).
According to original MRI report by radiologist.

79
Chapter 6

The baseline data in table 1 show that right arm pain occurred more often in the collar
group. The other baseline characteristics were evenly distributed over the three groups,
indicating that randomisation had been successful. The mean VAS scores of 70 mm for
arm and 60 mm for neck pain indicate the severity of the complaints. Neurological deficit
consisted mainly of sensory disturbances; relatively few patients exhibiting paresis or
hyporeflexia. In the three groups 78.7, 82.0 and 74.1% of patients showed degenerative
root compression on MRI (table 1). Five of the 205 patients were not available for follow-
up at six weeks. We found no significant difference between the three groups in the
number of patients who had surgery (five in the collar group, three in the physiotherapy
group and four in the control group).

Primary outcomes
Table 2 shows the group mean values for visual analogue scale and neck disability index
scores at three and six weeks in the three groups. Table 3 summarises the comparison of
these mean weekly changes in outcome measures in the control group with those of the
collar and physiotherapy groups during the first six weeks. The amount of pain reduction
per week is expressed by the estimated beta coefficient.

Table 2 Primary outcomes for collar and physiotherapy versus control group at 3 and 6 weeks.
Values are mean (SD)

Outcome 3 weeks 6 weeks

Measure cervical physiotherapy control cervical physiotherapy control

collar collar

Arm pain 50.3 (27.7) 55.1 (26.4) 59.1 (26.4) 33.5 (30.4) 36.0 (30.7) 48.6 (31.8)

(VAS) *

Neck pain 38.0 (28.4) 44.5 (32.5) 55.0 (31.8) 31.0 (28.2) 36.2 (31.0) 51.1 (32.7)

(VAS) *

NDI 33.8 (18.7) 34.6 (16.1) 34.3 (18.8) 25.9 (19.1) 27.8 (17.7) 29.9 (20.0)

* Pain on a Visual Analogue Scale as indicated on a 100 mm scale (0 representing no pain and 100 representing the
worst pain ever
The Neck Disability Index is a 100- point scale specific for neck pain with 10 items on functional status (high scores
representing poor functional status)

The reduction in arm pain in the control group had a beta of 3.1. In other words, the
pain score for arm pain diminished by a little more than 3 mm/week, amounting to 19
mm after six weeks. In the collar and physiotherapy groups, the pain scores diminished

80
Cervical collar or physiotherapy versus wait and see policy

80

VAS score (mm)


Cervical collar
Physical therapy
60
Controls

40

20

0
0 3 6 26

Weeks
Figure 2 Arm pain over time

80
VAS score (mm)

Cervical collar
Physical therapy
60
Controls

40

20

0
0 3 6 26

Weeks

Figure 3 Neck pain over time

significantly more than in the control group, with an extra pain reduction of 2 mm/ week
or 12 mm in 6 weeks (figure 2).

The decline in neck pain in the wait and see group over a period of six weeks had an
estimated beta of 0.9 mm (table 3). Treatment with a collar or physiotherapy resulted in a
significant reduction of neck pain of 2.8 mm/week for the collar (amounting to 17 mm in
six weeks) and 2.4 mm/week for physiotherapy (14 mm in 6 weeks) (figure 3).

The neck disability index scores of patients wearing a collar showed a significant
difference in rate of improvement compared to the control group (table 3). The additional

81
Chapter 6

Table 3 Comparison of mean weekly changes* (95% confidence interval) in outcome measures in
control group with those of collar and physiotherapy groups during first six weeks

Overall weekly change Additional weekly change Additional weekly change with

with collar physiotherapy

Arm pain -3.1 (-4.0 to -2.2)mm -1.9 (-3.3 to -0.5) mm -1.9 (-3.3 to -0.8) mm

(VAS) p=<0.001 p=0.006 p=0.007

Neck pain -0.9 (-2.0 to 0.3) mm -2.8 (-4.2 to -1.3) mm -2.4 (-3.9 to -0.8) mm

(VAS) p=0.11 p=<0.001 p=0.002

NDI -1.4 (-1.9 to -0.9) -0.9 (-1.6 to -0.1) -0.8 (-1.8 to 0.2)

p=<0.001 p=0.024 p=0.090

* On basis of generalised estimating equations (beta coefficient)


See table 2 for explanation
See table 2 for explanation

Table 4 Primary outcomes for collar and physiotherapy versus control group at six months. Values
are median (interquartile range) unless stated otherwise

Outcome measures Collar Physiotherapy Control P-value *

(n=63) (n=68) (n=61)

Arm pain (VAS) (mm) 0 (0-30.0) 0 (0-46.3) 0 (0-50.0) P=0.928

Neck pain (VAS) (mm) 10.0 (0-40.0) 20.0 (0-43.8) 10 (0-50.0) P=0.680

NDI 8 (0-26.0) 10 (2-29.2) 8 (0-26.0) P=0.670

* Kruskal Wallis Median test


See table 2 for explanation
See table 2 for explanation

weekly change of the physiotherapy group showed a similar pattern but was not
significantly different from that of the control patients.

After six months, most patients had no or limited pain (table 4). The median pain score
for arm pain was zero. Median scores for neck pain were 10 to 20 mm. The pain scores in
the two treatment groups did not differ from those of the control patients.

Secondary outcomes
We found no significant differences between the three groups for the secondary
outcome measures (table 5). The satisfaction scores, which ranged between 54% and
59%, were not different at the three and six weeks follow-up. The use of non-steroidal
anti-inflammatory drugs and opiates was also equal in the three groups. Working status

82
Cervical collar or physiotherapy versus wait and see policy

Table 5 Secondary-outcomes; satisfaction-scores, non-steroidal anti-inflammatory drugs (NSAID)


and opiate-use, and working status

three weeks six weeks


Collar Physio- Control p-value* Collar Physio- Control p-value*

(n=65) therapy (n=66) (n=61) (n=68) therapy (=65) (n=62)

Satisfied (%)* 38 (58) 38 (58) 33 (54) p=0.573 39/67 (58) 39/66 (59) 35 (56) p=0.870

NSAID-use 11 (17) 10 (15) 18 (30) p=0.094 7 (10) 8 (12) 12 (19) p=0.312

opiate use (%) 20 (31) 24 (36) 19/59 (32) p=0.790 13 (19) 13/66 (20) 16 (25) p=0.630

sick leave (%) 25/64 (39) 31 (47) 19 (31) p=0.189 20 (29) 30 (45) 24 (38) p=0.215

Values are numbers (percentages) unless stated otherwise


*, and Pearson-Chi-Square
* Satisfaction was measured on a 5-point scale: 1=very satisfied, 2= satisfied , 3= not satisfied nor unsatisfied,
4= unsatisfied, 5=very unsatisfied; 1 and 2 were classified as satisfied, 4 and 5 as unsatisfied
Based on reported use in the patients diaries.
Percentage on partial or complete sick leave

showed a non-significant pattern that patients treated were more often on partial or
complete sick leave than were those in the collar or control group.

Adherence
We used the patients diaries to verify adherence to the allocated treatment. Dividing a day
into four parts (morning, afternoon, evening, and night), patients wore the collar during the
first three weeks for a mean of 2.4 parts of the day; six patients did not wear the collar at
all. In weeks four to six, the collar was worn for a mean of 1.2 parts of the day.

In addition to the 12 physiotherapy sessions, 52% (34/65) of the patients exercised at


home for more than 10 minutes a day during the first three weeks, 36% (24/65) exercised
up to 10 minutes and 12% (8/65) did not exercise at all. During weeks three to six,
exercise time decreased slightly: 43% (28/65) exercised more than 10 minutes, 43%
(28/65) exercised up to 10 minutes and 14% (9/65) did not exercise at all. Sixty-one
percent (34/65) of patients in the control group noted in their diaries that they could
continue their normal activities during three or four parts of the day.

83
Chapter 6

Discussion

Clinical significance of results


In this randomised study of patients with recent onset cervical radiculopathy, we found
that treatment with a semi-hard cervical collar in combination with taking as much rest as
possible for three weeks, with a maximum of six weeks, or standardised physiotherapy
and doing home exercises for six weeks resulted in a significant reduction in neck and
arm pain as compared with a wait and see policy. The differences in pain reduction
between the treatment and control groups varied from 12 to 17 mm on a 100 mm visual
analogue scale in six weeks and were highly statistically significant. Studies on visual
analogue scale consider this difference as clinically meaningful. 11-15 One such study in
patients with acute pain, mainly in emergency departments, showed that the meaning of
a difference in pain scores depends on the height of the scores. In patients with scores
between 34 and 66 mm, as in our patients at three and six weeks, a difference of 17 (SD
10) mm was found to be clinical meaningful. 15 However, the setting differed from our
study in which patients with subacute onset cervical radiculopathy were treated, and this
may limit the interpretation of these data for our study population.

Disability decreased 9 points on the neck disability index over six weeks in the
control patients, with an additional 5 points decrease in both the collar and the
physiotherapy group. The additional effect of the collar on disability was small but
statistically significant. Although the additional effect of physiotherapy on disability was
not significant, a favourable effect on disability occurred, presumably owing to pain
reduction. The less prominent effect on the neck disability index compared with the
pain scores may well be explained by the fact that the index predominantly measures
the disability caused by neck pain, whereas arm pain scores were highest initially and
showed the largest improvement. All differences between the groups on the visual
analogue scale and the neck disability index scores were no longer present at the six
months follow-up. Most patients had no or limited pain, confirming earlier reports of the
favourable natural course of the disease. 61617 As the patients had arm and neck pain for
a mean of three weeks before entering the study, and as they were treated for six weeks,
we have shown that both the cervical collar and physiotherapy are efficacious within
this time frame. Considering the degree of pain reduction obtained at six weeks, further
interventions after this period are not likely to be of benefit in most patients.

84
Cervical collar or physiotherapy versus wait and see policy

Comparison to other studies


We could find no evidence in the literature showing the efficacy of a cervical collar or
physiotherapy in patients with subacute onset degenerative cervical radiculopathy. In
1966, a randomised clinical trial including 493 patients with cervical root symptoms
was performed with five treatment arms (traction, positioning, collar, placebo tablets,
placebo heat treatment). No significant difference in pain and ability to work was found
between the five treatment groups. Seventy-five percent of patients reported pain relief
at four weeks follow-up. 6 However, comparison of the results of this study with ours is
hampered by the fact that the methodology as described in the 1966 article does not
meet the current standard of clinical trials - for example, validated outcome scales were
not available.

Persson et al did a randomised clinical trial comparing physiotherapy and a hard collar in
patients with chronic (more than three months, median 21 months) cervical radiculopathy
who were randomised to surgery, physiotherapy or cervical collar. Surgery was superior
for pain relief at four months follow-up. At 16 months follow-up, no difference existed
between the three groups in terms of pain, muscle strength or sensory loss. 518 Marked
differences exist between our study and that of Persson et al. Our target study group
encompassed only patients with less than one month of arm pain. Our study also
differed with respect to the type of collar used. A soft collar is reported to give insufficient
support. A hard collar, as used in Perssons study, can cause serious discomfort. 19
Therefore, we decided to compromise with a snugly fitting semi-hard collar. 51820 In
Perssons study, the collar had to be worn over a three month period. However, several
authors warn for counterproductive effects of prolonged immobilisation. 1921

Mechanisms of collar and physiotherapy


Little evidence exists on the mechanisms of collars and physiotherapy in giving pain
relief, and the explanations provided in the literature are largely hypothetical. The collar
probably reduces foraminal root compression and associated root inflammation by
immobilising the neck, which might explain the larger reduction of arm pain compared
to neck pain and neck disability as found in our study. 22 Physiotherapy aims at restoring
range of motion and strengthening of the neck musculature, 23 probably diminishing
secondary musculoskeletal problems although the mechanism of pain reduction is
unclear. Thirteen (6.3%) of the 205 patients, equally distributed over the three groups,
were surgically treated during the six months of follow-up. Considerably higher
percentages of surgery for cervical radiculopathy are reported in the literature. 1724 We

85
Chapter 6

discussed surgical treatment options with patients who had persisting or intractable
pain and referred them to our neurosurgical department. The low rate of surgery in our
cohort may be due to the fact that our patients were included at an early stage, whereas
previous studies in which more chronic cases were included encompassed a larger
number of patients not responding to non-surgical treatment. Furthermore, patients were
possibly less inclined to have surgery because they participated in a study aimed at
reducing signs and symptoms by non-surgical interventions.

Limitations
Our study has several limitations. Firstly, for obvious reasons, the patients could not be
blinded for obvious reasons, and neither were the examiners blinded. We tried to prevent
bias by asking the patients to fill out the visual analogue scale scores and neck disability
index in the presence of a research nurse who did not examine the patients.

Secondly, the positive results of a collar or physiotherapy might be partially explained


by non-specific effects - for instance, by the attention given by a physiotherapist or
the attention received due to wearing a visible collar. By far the most professional
attention, however, was given to patients receiving physiotherapy for the full six weeks,
yet the treatment effect was largest for the collar that was worn mainly during the first
three weeks. In addition, the treatment satisfaction scores did not differ between the
three groups. Use of non-steroidal anti-inflammatory drugs, opiate use and sick leave
percentages were similar as well.

Thirdly, the quality of data on adherence to treatment obtained with diaries may be
compromised in a condition that tends to improve quite rapidly. However, we conclude
that adherence to the two treatment strategies, wearing a collar and taking rest versus
exercise therapy, was reasonable. These approaches were sufficiently different from that of
the control patients who largely succeeded in continuing their normal activities of daily life.

A fourth limitation is that we included patients in our study on the basis of clinical criteria
and used standardised magnetic resonance imaging and electromyography examinations
after randomisation because we wanted to stay close to clinical practice. Degenerative
cervical radiculopathy is primarily a clinical diagnosis and magnetic resonance imaging
is usually performed if surgery is considered or if there is doubt about the cause of the
radiculopathy. In addition, magnetic resonance imaging findings are often false negative,

86
Cervical collar or physiotherapy versus wait and see policy

2526
as was also shown by our study in which 20 to 25% of the clinically careful selected
patients had no root compression on magnetic resonance imaging (chapter 4).

We can not entirely exclude the possibility that patients with other causes of arm pain
than cervical radiculopathy have participated in the study. The number will have been
small as we required a firm clinical diagnosis of cervical radiculopathy, followed by strict
inclusion criteria. Patients with other causes of arm pain are unlikely to change the results
of the study, as we would expect that they were equally distributed over the three arms.

A final limitation is that we did not reach the calculated sample size of 240 patients.
Because of the magnitude of the differences in pain reduction with comfortable 95%
confidence intervals, an extra 35 patients would have unlikely to have changed our results.

Conclusion and policy implication

The results of this randomised clinical trial show a clinically relevant short term pain
reduction in recent onset cervical radiculopathy on two therapeutic interventions- that
is, a semi-hard cervical collar combined with taking rest and standardised physiotherapy
accompanied by home exercises- compared to a wait and see policy.

We recommend a semi-hard cervical collar and taking rest in recent onset cervical
radiculopathy because the costs are lower than for physiotherapy, although
physiotherapy is a good alternative with an almost similar efficacy.

87
Chapter 6

1. Adams C. Cervical spondylotic radiculopathy and myelopathy. Amsterdam/ New York / Oxford:

Elsevier Publishing Company, 1976.

2. Bush K, Chaudhuri R, Hillier S, Penny J. The pathomorphologic changes that accompany the

resolution of cervical radiculopathy. A prospective study with repeat magnetic resonance imaging.

Spine 1997;22(2):183-6; discussion 187.

3. Spurling RG, Segerberg LH. Lateral intervertebral disk lesions in the lower cervical region. J Am Med

Assoc 1953;151(5):354-9.

4. Vinas FC, Wilner H, Rengachary S. The spontaneous resorption of herniated cervical discs. J Clin

Neurosci 2001;8(6):542-6.

5. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery,

physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997;22(7):751-8.

6. Pain in the neck and arm: a multicentre trial of the effects of physiotherapy, arranged by the British

Association of Physical Medicine. Br Med J 1966;1(5482):253-8.

7. Twisk J. Applied longitudinal data analysis for epidemiology. A practical guide. Cambridge Cambridge

University Press, 2003.

8. Medical Research Council. Aids to the Examination of the Peripheral Nervous System. London: Her

Majesty Stationerys Office, 1978.

9. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. J Manipulative Physiol

Ther 1991;14(7):409-15.

10. McCarthy MJ, Grevitt MP, Silcocks P, Hobbs G. The reliability of the Vernon and Mior neck disability

index, and its validity compared with the short form-36 health survey questionnaire. Eur Spine J

2007;16(12):2111-7.

11. Kelly AM. Does the clinically significant difference in visual analog scale pain scores vary with gender,

age, or cause of pain? Acad Emerg Med 1998;5(11):1086-90.

12. Kelly AM. The minimum clinically significant difference in visual analogue scale pain score does not

differ with severity of pain. Emerg Med J 2001;18(3):205-7.

13. Todd KH. Clinical versus statistical significance in the assessment of pain relief. Ann Emerg Med

1996;27(4):439-41.

14. Todd KH, Funk KG, Funk JP, Bonacci R. Clinical significance of reported changes in pain severity. Ann

Emerg Med 1996;27(4):485-9.

15. Bird SB, Dickson EW. Clinically significant changes in pain along the visual analog scale. Ann Emerg

Med 2001;38(6):639-43.

16. Binder AI. Cervical spondylosis and neck pain. Bmj 2007;334(7592):527-31.

17. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

88
Cervical collar or physiotherapy versus wait and see policy

18. Persson LC, Moritz U, Brandt L, Carlsson CA. Cervical radiculopathy: pain, muscle weakness and

sensory loss in patients with cervical radiculopathy treated with surgery, physiotherapy or cervical

collar. A prospective, controlled study. Eur Spine J 1997;6(4):256-66.

19. Mazanec D, Reddy A. Medical management of cervical spondylosis. Neurosurgery 2007;60(1 Supp1

1):S43-50.

20. Ellenberg MR, Honet JC, Treanor WJ. Cervical radiculopathy. Arch Phys Med Rehabil 1994;75(3):342-52.

21. Polston DW. Cervical radiculopathy. Neurol Clin 2007;25(2):373-85.

22. Naylor JR, Mulley GP. Surgical collars: a survey of their prescription and use. Br J Rheumatol

1991;30(4):282-4.

23. Rosomoff HL, Fishbain D, Rosomoff RS. Chronic cervical pain: radiculopathy or brachialgia.

Noninterventional treatment. Spine 1992;17(10 Suppl):S362-6.

24. Heckmann JG, Lang CJ, Zobelein I, Laumer R, Druschky A, Neundorfer B. Herniated cervical

intervertebral discs with radiculopathy: an outcome study of conservatively or surgically treated

patients. J Spinal Disord 1999;12(5):396-401.

25. Nardin RA, Patel MR, Gudas TF, Rutkove SB, Raynor EM. Electromyography and magnetic resonance

imaging in the evaluation of radiculopathy. Muscle Nerve 1999;22(2):151-5.

26. Sengupta DK, Kirollos R, Findlay GF, Smith ET, Pearson JC, Pigott T. The value of MR imaging in

differentiating between hard and soft cervical disc disease: a comparison with intraoperative findings.

Eur Spine J 1999;8(3):199-204.

89
90
Chapter 7

General discussion

91
92
General discussion

Introduction

Degenerative cervical radiculopathy is a common neurological disorder, with an incidence


rate of 83.2 per 100.000 1, manifesting with pain, paraesthesias and less often weakness
in one arm radiating from the neck. The symptoms and signs originate from nerve
root compression caused by degenerative disease of the cervical spine. In the general
population degeneration can be seen as a sign of normal aging; only in a minority it
causes radicular symptoms. 23

In our study we found that herniated disc was more often the cause of single root
compression, whereas spondylarthrosis more often caused multiple root compression
which could be (partly) asymptomatic. This is in contrast to a large study on root
compression by Radhakriskan et al, who retrospectively investigated 561 patients with
cervical radiculopathy. In this study, disc protrusion was the cause of root compression in
only 21.9 percent of cases, whereas 68.4% had radiculopathy of uncertain pathogenesis
associated with degenerative disease, which could be spondylosis, disc or both. The
remainder had other causes including spinal fractures. 1 This article has often been
cited and spondylosis has been considered as the most common cause of degenerative
cervical radiculopathy. 4-6 It is sometimes difficult to distinguish spondylosis from
herniated disc on MRI 17, as was also seen in our study.

In the classic clinical picture patients have radicular pain and sensory disturbances in
the corresponding dermatome, weakness of the muscles of the involved myotome (for
instance m. biceps brachii for C6, m. triceps brachii for C7), and reduction or absence of
corresponding reflexes (biceps in C5 root lesions, brachioradialis in C6 lesions or triceps
reflex in C7 lesions). 489 However patients rarely present with all symptoms and signs
corresponding to the afflicted root. Unusual symptoms like radiation of pain to chest or
scapula, have been reported as signs of cervical root compression. 1011 In this thesis only
patients with typical clinical symptoms were included.

In most clinical cases, cervical radiculopathy can be easily differentiated from other
causes of arm pain, like entrapment syndromes and neuralgic amyotrophy by a careful
history and neurological examination. 49 Clinicians should be alert for signs and
symptoms indicative for myelopathy, like gait disability and bladder dysfunction, as
spinal cord compression may occur together with degenerative radiculopathy. 12 Different
diagnostic tools, such as imaging modalities and needle electromyography (EMG), are

93
Chapter 7

additionally utilized for confirmation of radiculopathy or excluding other diagnoses.


However, there is not much evidence to appreciate its added value to the clinical
observations.

Comparison of MRI with nerve conduction studies and needle electromyography yielded
that sensitivity of MRI is higher (93 vs. 42%) than electrophysiology in the preoperative
evaluation of patients with cervical radiculopathy. 24

This thesis comprises a three-armed randomised clinical trial of non-surgical treatments


(collar, physiotherapy or wait and see) in patients who were referred by their general
practitioner to the neurologist, with signs and symptoms consistent with recent onset
cervical radiculopathy. Furthermore, in this thesis, the added value of additional
diagnostic methods (MRI and EMG) to the clinical diagnosis was assessed. In this final
chapter the results obtained in the studies described in the previous chapters will be
discussed, with emphasis on their clinical implications, the applied methodology, and
implications for further research.

Diagnosis

Clinical diagnosis
For our studies on recent onset cervical radiculopathy, we first established a clinical
definition based on existing literature 613-18 in combination with our clinical experience.
Our criteria for the clinical diagnosis of cervical radiculopathy were: radiation of arm
pain distal to the elbow, plus at least one of the following: (1) provocation of the arm
pain by neck movements, (2) sensory changes in one or more adjacent dermatomes, (3)
diminished deep tendon reflexes in the affected arm, and/or (4) muscle weakness in one
or more adjacent myotomes, in the absence of clinical signs of spinal cord lesions.

We performed several provocation tests in our standardised neurological examination;


retroflexion, anteflexion, lateroflexion of the neck, Spurlings foraminal compression
test, and Lasegues arm sign (which can be seen as a variant of the shoulder abduction
sign). Of these, only Spurlings foraminal compression test and the shoulder abduction
sign have been validated. The foraminal compression has been demonstrated to have
low sensitivity and high specificity. The shoulder abduction test demonstrated low to
moderate sensitivity and moderate to high specificity. 19-21 In every patient we performed

94
General discussion

all of the above mentioned tests. The tests were considered positive when arm pain
worsened. The definition for the entry criterion was at least one positive provocation test.
Most patients had more than one positive test with retroflexion of the neck showing the
highest percentage of positive results, followed by Spurlings foraminal compression test.

In a guideline which was published in 2010 by the North American Spine Society (NASS)
a working definition for cervical radiculopathy was established (working group consensus
statement), showing great similarity with our criteria. The definition of the working group
was as follows: pain in a radicular pattern in one or both upper extremities related to
compression and/or irritation of one or more cervical nerve roots. Frequent signs and
symptoms include varying degrees of sensory, motor and reflex changes as well as
dysesthesias and paresthesias related to nerve root(s) without evidence of spinal cord
dysfunction (myelopathy). 20 This definition implies that the diagnosis of degenerative
cervical radiculopathy can be made on clinical grounds. The guideline considered
provocative tests useful in clinical evaluation of cervical radiculopathy, but did not include
them in their clinical definition. 20

Our criteria appeared to be accurate in practice, since none of the 205 patients
diagnosed by our clinical criteria for cervical radiculopathy had other causes than
herniated discs or spondylotic changes, nor did other conditions emerge during the
follow-up period of 6 months. However, we gathered no data on interobserver variability
and accuracy. For further research our criteria should be validated for accuracy and
interobserver variability.

Ancillary investigations

Magnetic Resonance Imaging


Since its introduction, MRI is often used in patients with a clinical diagnosis of cervical
radiculopathy to demonstrate the presence and cause of root compression, i.e. disc
herniation or spondylotic foraminal stenosis, and the level of compression. The previously
mentioned NASS guideline recommends MRI as the method of choice for the confirmation
of disc herniation and spondylosis in cervical spine patients who have failed conservative
therapy, and who may be candidates for interventional or surgical treatment. 20

When MRI is contra-indicated, computed tomography (CT) can be performed, and with
equivocal MRI results CT myelography can provide more information. Several studies

95
Chapter 7

have been conducted to compare MRI with other investigations, i.e., CT, CT myelography
and electromyography (EMG). MRI images were found to have an accuracy of 87% for
the diagnosis of foraminal encroachment, for CT myelography there was an accuracy
of 90%, and gadolinium-enhanced MRI had no additional benefit. 22 A study comparing
MRI to surgical findings in 13 patients with known or suspected cervical radiculopathy
showed that MRI identified 3 out of 3 herniations and 26 out of 27 other degenerative
abnormalities, not otherwise specified. Based on these results, the authors concluded
that diagnostic accuracy of MRI is sufficient for the evaluation of cervical radiculopathy. 23
Comparison of MRI with nerve conduction studies and needle electromyography yielded
that sensitivity of MRI is higher (93 vs. 42%) than electrophysiology in the preoperative
evaluation of patients with cervical radiculopathy. 24

However none of these studies took observer variability into account, and in addition
well-designed studies relating MRI abnormalities to clinical findings were not available.
To assess the value of MRI we first studied interobserver variability of MRI in a large
sample of patients with cervical radiculopathy. We found substantial interobserver
agreement (kappa 0.67) of MRI diagnosis of root compression between two experienced
neuroradiologists. Agreement on the cause of root compression, disc herniation (kappa
0.59) or spondylotic foraminal stenosis (kappa 0.63) was less robust.

In addition, we investigated the relation between the clinical and MRI level of root
compression. Using the clinically affected root as gold standard we found corresponding
root compression on MRI in 73 percent of patients. This percentage is similar to that
found in a study of Henderson et al. 25 In a retrospective series of 736 patients with
cervical radiculopathy they compared the clinically suspected level with the compressed
level found during surgery and it was shown that in 71.5% of patients the site of root
compression could already be accurately predicted by clinical signs alone. 25 It would
be worthwhile to compare the predictive value of clinical data with that of MRI in a large
surgically treated population. Such a study has not yet been performed as far as we know.

In 15% of our symptomatic patients MRI showed no root compression at all. One
possibility is that negative MRIs in patients with a clear clinical diagnosis have to be
considered as a technical failure of the MRI. Another explanation of false-negative
MRIs is that in these patients the symptoms are caused by inflammation rather than
compression by herniated discs or spondylotic foraminal stenosis. In earlier reports,
signs of inflammation were found histopathologically. 26 Inflammation therefore may

96
General discussion

be an alternative explanation for radiculopathies without confirmed root compression,


even though no root thickening is being visualised on MRI. A third explanation for our
normal MRIs is that we have misdiagnosed some patients. However, none of our patients
revealed another cause for their arm pain during the period of follow-up (6 months)
corroborating the previously established diagnosis. We consider the first explanation
the most likely one. Nonetheless, it can not be ruled out that some patients did not have
cervical radiculopathy. In our studies on MRI we could not use surgical data as the gold
standard, as suggested in the above-mentioned NASS guideline 20 ; only 13 out of our
205 patients needed surgical decompression.

In a clinical setting, performing a MRI has no clinical consequences for the majority of
patients with recent onset cervical radiculopathy since the usual course is spontaneous
recovery. We advise to perform MRI in the case of persistent pain exceeding a period of 6
to 8 weeks. This corresponds with the recommendation of the NASS guidelines. 20 Also,
MRI should be performed in cases with red flags, for instance indicating malignancy,
spondylodiscitis, or myelopathy. 42027 All MRI findings should be interpreted cautiously
and always in the clinical context.

MRI negative cases are an important subject for further research. The place for additional
CT techniques like contrast enhanced Multidetector Computed Tomography (MDCT)
should be established. It is a promising technique for better visualisation of bony
deformations. 28 With MDCT, foraminal stenosis can be detected that is missed on MRI. It
would also be worthwhile to study the usefulness of dynamic MRI in neutral, flexion and
extension position, especially but not exclusively in patients with a clinical diagnosis of
cervical radiculopathy and a negative MRI, and of dynamic upright MRI. 202930

Needle electromyography
Although MRI is the additional investigation of choice for detecting root compression,
EMG is still often used. 1631-33 The NASS guideline concluded that no recommendations
could be made for the use of EMG, as there is insufficient evidence. 20

We also found only limited usefulness of EMG for the diagnosis. In our EMG study we
examined 176 patients with a standardised needle electromyography protocol in which
the electrodiagnostic consultant was blinded for clinical and MRI data. We detected
denervation activity in only 16.5 % of patients; the total number of abnormal EMGs in our
study population was 39.2 %.

97
Chapter 7

The main drawback of our study was the standardisation and limitation of the EMG
protocol to 5 muscles. A more extensive EMG protocol, including the examination
of more arm and also paraspinal 32 muscles might have increased the percentage of
abnormal EMGs.

It can be concluded that there seems to be no place for routinely performed standardised
5- muscle needle electromyography in patients with recent onset cervical radiculopathy.
Needle EMG combined with conduction studies is useful when the diagnosis of
radiculopathy is not solid, since it can differentiate between radiculopathy and other
causes of arm pain, like compression of peripheral nerves, mainly the median and ulnar
nerve, and lesions of the brachial plexus. 493334

Further studies are needed to learn more on the value of EMG in patients with multilevel
root compression on MRI. Additionally, it is still unclear what EMG can contribute in
patients with a negative MRI; we also advise further research on this topic.

Treatment

Pharmacological
There are no studies addressing the role of pharmacological treatment in the management
of recent onset cervical radiculopathy from degenerative disorders. 2035 A Cochrane
review has shown significant efficacy of opioids in reducing the intensity of neuropathic
pain of any kind (median duration of opioid use 28 days, range 8-70 days). 3637

There is no literature available on pain reducing effect of specific neuropathic medications


like pregabalin, topiramate, tricyclic antidepressants and carbamazepin, in recent onset
radiculopathy. In chronic radiculopathy, these medications have no proven effect, except
for the combination tricyclic antidepressants /opioids, which was effective in one study. 38-40

In our study we did not investigate the effect of analgetics. All our patients were free to
use over the counter painkillers like paracetamol, non-steroidal anti-inflammatory drugs
(NSAIDs), and if necessary we prescribed opiates. Patients reported the use of painkillers
in their diaries. We compared the use of NSAIDs and opiates in the three treatment
groups and found that the prescription and the use were similar in all three arms.

98
General discussion

Collar, physiotherapy versus wait and see policy


Both negative and positive trials were published on the usefulness of physiotherapy
in lumbar and cervical radiculopathy. 41-45 However, study populations were
heterogeneous, a description of the physiotherapy protocol was lacking and in general
study methodology was substandard. We investigated the efficacy of a standardised
physiotherapy program, with an emphasis on exercises. Our protocol was based on
expert opinion of experienced physiotherapists. The complete exercise protocol has been
added to this thesis (appendix 2).

We also included collar therapy as a treatment arm in our study. Studies convincingly
showing a positive effect of the collar in cervical radiculopathy were notably absent.
However, for many physicians prescription of a collar in cervical radiculopathy was still
common practice as individual patients reported short term pain relief. 131746-48 The collar
has been a common treatment modality for patients with all kinds of neck pain and
particularly whiplash associated injuries. Several trials were conducted on the efficacy
of the collar for patients with whiplash associated injuries. 49-52 None of these studies
showed positive effects of wearing a collar in these patients. Guidelines, based on these
negative results, discouraged the use of collars. 5354 It was assumed that wearing the
collar led to a delayed recovery and wearing a collar for longer periods might lead to
atrophy-related secondary damage. 5455 Therefore, the collar has had a negative image
for decades.

Our RCT showed that treatment with a semi-hard cervical collar in combination with
taking as much rest as possible for three weeks, with a maximum of six weeks, or
standardised physiotherapy and doing home exercises for six weeks resulted in a
significant reduction in neck and arm pain as compared with a wait and see policy. We
also found a small effect of the cervical collar and rest on functionality measured with the
neck disability index (NDI), in contrast to physiotherapy and home exercises.

We have chosen to use the visual analogue scale (VAS), and the neck disability index
(NDI) as outcome scales. Both scales have often been used and tested in trials on
surgery in patients with cervical radiculopathy, and both the NDI and the VAS were
found to be reliable and valid outcome measures to assess the outcome in trials on
degenerative cervical radiculopathy. 2056-60 We used the Dutch version of the Neck
Disablity Scale (NDI-DV), which has been validated in the Netherlands for patients with
neck pain. 61-63 We considered non-disease specific quality of life scales like the SF-12

99
Chapter 7

57
, the Patient Specific Functional Scale (PSFS) 64 and the SF-36 65 less appropriate for
our population with mainly arm and neck pain, limited neurological deficit and marked
recovery within 6 weeks.

Based on the results of our trial we advise a collar for 3-6 weeks combined with rest, in
patients with recent onset cervical radiculopathy. We recommend a semi-hard collar, as
hard collars are uncomfortable, whereas soft collars give insufficient support. Alternatively
physiotherapy can be prescribed in combination with home exercises for six weeks.

Further randomised trials are needed to confirm our results. Also more research is
needed to understand the mode of action of physiotherapy and collar therapy. Timing of
the prescription of the collar, the period the collar should be worn and the effectiveness
of different types of collar, also need more scientific evaluation. Wearing a collar in the
acute phase followed by physiotherapy may very well yield better results than each of the
treatment modalities alone.

In conclusion, this thesis reveals strengths as well as weaknesses of additional diagnostic


examinations like MRI and EMG, in recent onset degenerative cervical radiculopathy.
The main diagnostic tool remains a thorough history and neurological examination.
Initial treatment of cervical radiculopathy may consist of wearing a semi-hard collar or
physiotherapy.

100
General discussion

1. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

2. Teresi LM, Lufkin RB, Reicher MA, Moffit BJ, Vinuela FV, Wilson GM, et al. Asymptomatic degenerative

disk disease and spondylosis of the cervical spine: MR imaging. Radiology 1987;164(1):83-8.

3. Irvine DH, Foster JB, Newell DJ, Klukvin BN. Prevalence of Cervical Spondylosis in a General Practice.

Lancet 1965;1(7395):1089-92.

4. Carette S, Fehlings MG. Clinical practice. Cervical radiculopathy. N Engl J Med 2005;353(4):392-9.

5. Wainner RS, Fritz JM, Irrgang JJ, Boninger ML, Delitto A, eAllison S. Reliability and diagnostic

accuracy of the clinical examination and patient self-report measures for cervical radiculopathy. Spine

2003;28(1):52-62.

6. Ellenberg MR, Honet JC, Treanor WJ. Cervical radiculopathy. Arch Phys Med Rehabil 1994;75(3):342-52.

7. Jahnke RW, Hart BL. Cervical stenosis, spondylosis, and herniated disc disease. Radiol Clin North Am

1991;29(4):777-91.

8. Adams C. Cervical spondylotic radiculopathy and myelopathy. Amsterdam/ New York / Oxford:

Elsevier Publishing Company, 1976.

9. Eubanks JD. Cervical radiculopathy: nonoperative management of neck pain and radicular symptoms.

Am Fam Physician 2010;81(1):33-40.

10. Ozgur BM, Marshall LF. Atypical presentation of C-7 radiculopathy. J Neurosurg 2003;99(2

Suppl):169-71.

11. Tanaka N, Fujimoto Y, An HS, Ikuta Y, Yasuda M. The anatomic relation among the nerve roots,

intervertebral foramina, and intervertebral discs of the cervical spine. Spine 2000;25(3):286-91.

12. Fouyas IP, Sandercock PA, Statham PF, Nikolaidis I. How beneficial is surgery for cervical

radiculopathy and myelopathy? Bmj 2010;341:c3108.

13. Dillin W, Booth R, Cuckler J, Balderston R, Simeone F, Rothman R. Cervical radiculopathy. A review.

Spine 1986;11(10):988-91.

14. Viikari-Juntura E, Porras M, Laasonen EM. Validity of clinical tests in the diagnosis of root compression

in cervical disc disease. Spine 1989;14(3):253-7.

15. Shah KC, Rajshekhar V. Reliability of diagnosis of soft cervical disc prolapse using Spurlings test. Br J

Neurosurg 2004;18(5):480-3.

16. Dumitru D, Zwarts M. . Radiculopathy. In: Dumitru D, editor. Electrodiagnostic Medicine. 2nd ed.

Philadelphia: Hanley and Belfus, 2002:713-76.

17. Levine MJ, Albert TJ, Smith MD. Cervical Radiculopathy: Diagnosis and Nonoperative Management. J

Am Acad Orthop Surg 1996;4(6):305-316.

18. Tong HC, Haig AJ, Yamakawa K. The Spurling test and cervical radiculopathy. Spine 2002;27(2):156-9.

101
Chapter 7

19. Rubinstein SM, Pool JJ, van Tulder MW, Riphagen, II, de Vet HC. A systematic review of the

diagnostic accuracy of provocative tests of the neck for diagnosing cervical radiculopathy. Eur Spine J

2007;16(3):307-19.

20. Bono CM, Ghiselli G, Gilbert TJ, Kreiner DS, Reitman C, Summers JT, et al. An evidence-based clinical

guideline for the diagnosis and treatment of cervical radiculopathy from degenerative disorders. Spine

J 2010;11(1):64-72.

21. Spurling RG, Segerberg LH. Lateral intervertebral disk lesions in the lower cervical region. J Am Med

Assoc 1953;151(5):354-9.

22. Bartlett RJ, Hill CR, Gardiner E. A comparison of T2 and gadolinium enhanced MRI with CT

myelography in cervical radiculopathy. Br J Radiol 1998;71(841):11-9.

23. Hedberg MC, Drayer BP, Flom RA, Hodak JA, Bird CR. Gradient echo (GRASS) MR imaging in cervical

radiculopathy. AJR eAm J Roentgenol 1988;150(3):683-9.

24. Ashkan K, Johnston P, Moore AJ. A comparison of magnetic resonance imaging and

neurophysiological studies in the assessment of cervical radiculopathy. Br J Neurosurg

2002;16(2):146-8.

25. Henderson CM, Hennessy RG, Shuey HM, Jr., Shackelford EG. Posterior-lateral foraminotomy as

an exclusive operative technique for cervical radiculopathy: a review of 846 consecutively operated

cases. Neurosurgery 1983;13(5):504-12.

26. Kang JD, Georgescu HI, McIntyre-Larkin L, Stefanovic-Racic M, Evans CH. Herniated cervical

intervertebral discs spontaneously produce matrix metalloproteinases, nitric oxide, interleukin-6, and

prostaglandin E2. Spine 1995;20(22):2373-8.

27. Shelerud RA, Paynter KS. Rarer causes of radiculopathy: spinal tumors, infections, and other unusual

causes. Phys Med Rehabil Clin N Am 2002;13(3):645-96.

28. Douglas-Akinwande AC, Rydberg J, Shah MV, Phillips MD, Caldemeyer KS, Lurito JT, et al. Accuracy

of contrast-enhanced MDCT and MRI for identifying the severity and cause of neural foraminal

stenosis in cervical radiculopathy: a prospective study. AJR Am J Roentgenol;194(1):55-61.

29. Miura J, Doita M, Miyata K, Marui T, Nishida K, Fujii M, et al. Dynamic evaluation of the spinal cord

in patients with cervical spondylotic myelopathy using a kinematic magnetic resonance imaging

technique. J Spinal Disord Tech 2009;22(1):8-13.

30. Harada T, Tsuji Y, Mikami Y, Hatta Y, Sakamoto A, Ikeda T, et al. The clinical usefulness of preoperative

dynamic MRI to select decompression levels for cervical spondylotic myelopathy. Magn Reson

Imaging 2010;28(6):820-5.

31. American Association of Electrodiagnostic Medicine.

The electrodiagnostic evaluation of patients with suspected cervical radiculopathy: literature review on

the usefulness of needle electromyography. Muscle Nerve 1999;Supplement 8:S213-S221.

102
General discussion

32. Dillingham TR, Lauder TD, Andary M, Kumar S, Pezzin LE, Stephens RT, et al. Identification of cervical

radiculopathies: optimizing the electromyographic screen. Am J Phys Med Rehabil 2001;80(2):84-91.

33. Fisher MA. Electrophysiology of radiculopathies. Clin Neurophysiol 2002;113(3):317-35.

34. Aminoff MJ. Electrodiagnosis in clinical neurology. 5 ed. New York: Churchill Livingstone, 2005.

35. Peloso P, Gross A, Haines T, Trinh K, Goldsmith CH, Burnie S. Medicinal and injection therapies for

mechanical neck disorders. Cochrane Database Syst Rev 2007(3):CD000319.

36. Eisenberg E, McNicol E, Carr DB. Opioids for neuropathic pain. Cochrane Database Syst Rev

2006;3:CD006146.

37. Eisenberg E, McNicol ED, Carr DB. Efficacy and safety of opioid agonists in the treatment of

neuropathic pain of nonmalignant origin: systematic review and meta-analysis of randomized

controlled trials. Jama 2005;293(24):3043-52.

38. Attal N, Cruccu G, Baron R, Haanpaa M, Hansson P, Jensen TS, et al. EFNS guidelines on the

pharmacological treatment of neuropathic pain: 2010 revision. Eur J Neurol 2010;17(9):1113-e88.

39. Khoromi S, Patsalides A, Parada S, Salehi V, Meegan JM, Max MB. Topiramate in chronic lumbar

radicular pain. J Pain 2005;6(12):829-36.

40. Khoromi S, Cui L, Nackers L, Max MB. Morphine, nortriptyline and their combination vs. placebo in

patients with chronic lumbar root pain. Pain 2007;130(1-2):66-75.

41. Hofstee DJ, Gijtenbeek JM, Hoogland PH, van Houwelingen HC, Kloet A, Lotters F, Tans JThJ.

Westeinde sciatica trial: randomized controlled study of bed rest and physiotherapy for acute sciatica.

J Neurosurg 2002;96(1 Suppl):45-9.

42. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery,

physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997;22(7):751-8.

43. Kay TM, Gross A, Goldsmith C, Santaguida PL, Hoving J, Bronfort G. Exercises for mechanical neck

disorders. Cochrane Database Syst Rev 2005(3):CD004250.

44. Jordan A, Bendix T, Nielsen H, Hansen FR, Host D, Winkel A. Intensive training, physiotherapy, or

manipulation for patients with chronic neck pain. A prospective, single-blinded, randomized clinical

trial. Spine (Phila Pa 1976) 1998;23(3):311-8; discussion 319.

45. Saal JS, Saal JA, Yurth EF. Nonoperative management of herniated cervical intervertebral disc with

radiculopathy. Spine 1996;21(16):1877-83.

46. Narayan P, Haid RW. Treatment of degenerative cervical disc disease. Neurol Clin 2001;19(1):217-29.

47. Dreyer SJ, Boden SD. Nonoperative treatment of neck and arm pain. Spine 1998;23(24):2746-54.

48. Mazanec D, Reddy A. Medical management of cervical spondylosis. Neurosurgery 2007;60(1 Supp1

1):S43-50.

49. Borchgrevink GE, Kaasa A, McDonagh D, Stiles TC, Haraldseth O, Lereim I. Acute treatment of

whiplash neck sprain injuries. A randomized trial of treatment during the first 14 days after a car

accident. Spine (Phila Pa 1976) 1998;23(1):25-31.

103
Chapter 7

50. Schnabel M, Ferrari R, Vassiliou T, Kaluza G. Randomised, controlled outcome study of active

mobilisation compared with collar therapy for whiplash injury. Emerg Med J 2004;21(3):306-10.

51. McKinney LA. Early mobilisation and outcome in acute sprains of the neck. Bmj 1989;299(6706):1006-8.

52. Rosenfeld M, Gunnarsson R, Borenstein P. Early intervention in whiplash-associated disorders: a

comparison of two treatment protocols. Spine (Phila Pa 1976) 2000;25(14):1782-7.

53. Teasell RW, McClure JA, Walton D, Pretty J, Salter K, Meyer M, et al. A research synthesis of

therapeutic interventions for whiplash-associated disorder (WAD): part 2 - interventions for acute

WAD. Pain Res Manag 2010;15(5):295-304.

54. Spitzer WO, Skovron ML, Salmi LR, Cassidy JD, Duranceau J, Suissa S, et al. Scientific monograph

of the Quebec Task Force on Whiplash-Associated Disorders: redefining whiplash and its

management. Spine (Phila Pa 1976) 1995;20(8 Suppl):1S-73S.

55. Muzin S, Isaac Z, Walker J, Abd OE, Baima J. When should a cervical collar be used to treat neck

pain? Curr Rev Musculoskelet Med 2008;1(2):114-9.

56. Fernandez-Fairen M, Sala P, Dufoo M, Jr., Ballester J, Murcia A, Merzthal L. Anterior cervical

fusion with tantalum implant: a prospective randomized controlled study. Spine (Phila Pa 1976)

2008;33(5):465-72.

57. Foley KT, Mroz TE, Arnold PM, Chandler HC, Jr., Dixon RA, Girasole GJ, et al. Randomized,

prospective, and controlled clinical trial of pulsed electromagnetic field stimulation for cervical fusion.

Spine J 2008;8(3):436-42.

58. Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the

neck disability index, patient-specific functional scale, and numeric pain rating scale in patients with

cervical radiculopathy. Am J Phys Med Rehabil;89(10):831-9.

59. Holly LT, Matz PG, Anderson PA, Groff MW, Heary RF, Kaiser MG, et al. Functional outcomes

assessment for cervical degenerative disease. J Neurosurg Spine 2009;11(2):238-44.

60. MacDermid JC, Walton DM, Avery S, Blanchard A, Etruw E, McAlpine C, et al. Measurement properties

of the neck disability index: a systematic review. J Orthop Sports Phys Ther 2009;39(5):400-17.

61. Vos CJ, Verhagen AP, Koes BW. Reliability and responsiveness of the Dutch version of the Neck

Disability Index in patients with acute neck pain in general practice. Eur Spine J 2006;15(11):1729-36.

62. Heymans WFGJ LSH, Elvers JWH, et al. . Neck Disability Index Dutch Version (NDI-DV): investigation

of reliability in patients with chronic whiplash. Ned T Fysiother 2002(112):94-99.

63. Jorritsma W, de Vries GE, Geertzen JH, Dijkstra PU, Reneman MF. Neck Pain and Disability

Scale and the Neck Disability Index: reproducibility of the Dutch Language Versions. Eur Spine J

2010;19(10):1695-701.

64. Cleland JA, Fritz JM, Whitman JM, Palmer JA. The reliability and construct validity of the Neck

Disability Index and patient specific functional scale in patients with cervical radiculopathy. Spine

2006;31(5):598-602.

104
General discussion

65. Kumar N, Gowda V. Cervical foraminal selective nerve root block: a two-needle technique with

results. Eur Spine J 2008;17(4):576-84.

105
106
Summary

Summary

This thesis deals with diagnostics aspects and non-surgical treatment options in recent
onset cervical radiculopathy.

In chapter 1 we gave a short historical introduction on cervical radiculopathy and


described the aims and outline of this thesis.

In chapter 2 we reviewed the existing literature on clinical diagnosis, ancillary


investigations and therapeutic aspects. It was remarkable to find such a lack of
knowledge on the value of often used diagnostic tools like magnetic resonance imaging
(MRI) and needle electromyography (EMG). Neither did we find any evidence for the
effectiveness of non- surgical pain relieving techniques.

To investigate the reliability of MRI we first assessed interobserver agreement on 78


patients with cervical radiculopathy, as described in chapter 3. The evaluation was
done by two experienced neuroradiologists blinded to clinical data. The kappa score
for interobserver variability of herniated discs was 0.59 (which can be interpreted as
moderate agreement) and of spondylotic foramen stenosis it was 0.63 (substantial
agreement). A kappa score of 0.67 (substantial agreement) was found for the presence of
root compression. After disclosure of clinical information kappa scores increased slightly;
for detection of herniated discs from 0.59 to 0.62, for spondylotic foramen stenosis
from 0.63 to 0.66 and for root compression from 0.67 to 0.76 (all to be interpreted as
substantial agreement).

In chapter 4 we report the results of our second MRI study, in which we relate the MRI
results to clinical findings. In 73 percent of patients the clinically affected root was found
to be compressed on MRI. In 45 percent of patients, MRI also showed asymptomatic
root compression in addition to compression of the clinically affected root. MRIs were
assessed as normal in 13-15 percent of cases and in 9-10 percent only asymptomatic
roots were compressed. Herniated discs without spondylosis were more often
responsible for root compressions only at the clinically affected level and spondylotic
foraminal stenosis for multiple root compression including compression of clinically
unaffected roots.

107
Chapter 7

In chapter 5 the results of our study on EMG are described. We performed a


standardised needle electromyography in 5 muscles representing 4 myotomes (C5, C6,
C7 and C8). The electrodiagnostic consultant was blinded to clinical and MRI data. We
examined the occurrence of abnormal spontaneous activity (signs of denervation) and of
neurogenic motor unit action potentials (signs of reinnervation). We found abnormalities
in 39.2% of the 176 needle EMGs. Signs of reinnervation (29.5%) were more frequently
seen than denervation (16.5%). These results led us to conclude that a standardised
needle EMG as a routine examination in patients with cervical radiculopathy is not very
useful.

In chapter 6 we describe the results of our randomised clinical trial in which we


compared treatment with a semi-hard collar and taking rest for three to six weeks;
12 twice weekly sessions of physiotherapy and home exercises for six weeks; or
continuation of daily activities as much as possible without specific treatment (control
group), in patient with recent onset cervical radiculopathy who were allowed to freely use
analgetics. In the wait and see group arm pain diminished by 3 mm/week on the visual
analogue scale (beta -3.1 mm, 95% CI -4.0 to -2.2) and by 19 mm in total over six weeks.
Patients who were treated with cervical collar or physiotherapy achieved additional
arm pain reduction (collar: beta -1.9 mm, 95% CI -3.3 to -0.5 mm, physiotherapy:
beta -1.9, 95% CI -3.3 to -0.8), resulting in an extra pain reduction of 12 mm after six
weeks compared with the control group. In the wait and see group, neck pain did not
significantly decrease in the first six weeks (beta -0.9 mm, 95% CI -2.0 to 0.3). Treatment
with the collar resulted in a weekly VAS-scale reduction of 2.8 mm (95% CI -4.2 to -1.3)
amounting to 17 mm in six weeks, whereas physiotherapy with a weekly reduction of 2.4
mm (95% CI -3.9 to -0.8) resulted in a decrease of 14 mm after six weeks.

108
Summary

Compared with a wait and see policy the neck disability index showed a significant
change with the use of cervical collar and rest (beta -0.9 mm, 95% CI -1.6 to -0.1) and a
non-significant effect with physiotherapy and home exercises was found.

In conclusion, we advise a semi-hard cervical collar and rest for three to six weeks or
physiotherapy accompanied by home exercises for six weeks, as it reduced neck and
arm pain substantially as compared to a wait and see policy in the early phase of cervical
radiculopathy.

The closing chapter 7 includes a general discussion which places the individual chapters
into the context of available evidence and provides recommendations for further
research.

109
110
Samenvatting

Samenvatting

Dit proefschrift beschrijft de resultaten van onze studies naar de diagnose en


conservatieve behandeling van een kort bestaand cervicaal radiculair syndroom.

In hoofdstuk 1 geven we een kort historisch overzicht betreffende de aandoening en


worden de doelstellingen van het proefschrift uiteengezet.

Hoofdstuk 2 geeft een overzicht van de bestaande literatuur betreffende klinische en


aanvullende diagnostiek, en de conservatieve behandeling van cervicale radiculopathie.
Er is weinig bekend over de aanvullende diagnostisch waarde van veel gebruikt
aanvullend onderzoek als magnetic resonance imaging (MRI) en electromyografie (EMG).
Ook is er geen bewijs voor veel gebruikte conservatieve behandelingen bij cervicale
radiculopathie. Om dit beter in kaart te brengen voerden wij enkele onderzoeken uit.

In hoofdstuk 3 beschrijven we onze studie naar de inter-beoordelaars betrouwbaarheid


bij de beoordeling van MRIs bij 78 patinten met cervicale radiculopathie. De
geobserveerde overeenkomst werd gemeten met een kappascore; voor de aanwezigheid
van een hernia vonden we een kappa van 0.59, wat beschouwd wordt als een redelijke
overeenkomst. Voor de aanwezigheid van een spondylotische foramenvernauwing was
de overeenkomst voldoende met een kappa van 0.63. Voor de beoordeling van de
aanwezigheid van wortelcompressie was de inter-beoordelaars betrouwbaarheid net iets
hoger met een kappa score van 0.67 (voldoende overeenkomst). Na de beoordeling van
de MRI werd het klinisch aangedane niveau vrijgegeven, waarna de neuroradioloog de
MRI opnieuw beoordeelde. Met klinische gegevens was de kappa score iets hoger: voor
de aanwezigheid van een hernia ging de score van 0.59 naar 0.62, voor spondylotische
wortelcompressie verhoogde de score van 0.63 naar 0.66 en de inter-beoordelaars
betrouwbaarheid van wortelcompressie ging van 0.67 naar 0.76 (voldoende
overeenkomst).

Hierop voerden we een tweede MRI studie uit, waarin we de MRI resultaten vergeleken
met de klinische bevindingen. In hoofdstuk 4 rapporteren we de resultaten van deze
studie; in 73% van de patinten was de klinisch aangedane wortel gecomprimeerd op de
MRI. In 45% van de patinten zagen we op de MRI asymptomatische wortelcompressie,
bij deze patinten was tevens de klinisch aangedane wortel was gecomprimeerd. In
13-15% van de patinten vonden we geen wortelcompressie op de MRI, en in 9-10%

111
Chapter 7

van de patinten waren alleen klinisch asymptomatische wortels gecomprimeerd. In de


gevallen waarbij we wortelcompressie alleen op het klinische aangedane niveau vonden,
was een hernia de meest voorkomende oorzaak. Spondylotische foramenstenose werd
vaker gezien bij multipele wortelcompressie waarbij vaker asymptomatische wortels
werden gecomprimeerd.

In hoofdstuk 5 beschrijven we de resultaten van de EMGs die bij een groot deel
van onze studiepopulatie is verricht. Bij 176 patinten werd een gestandaardiseerd
naaldonderzoek uitgevoerd, waarbij 5 spieren werden aangeprikt, corresponderend met
de 4 myotomen (C5, C6, C7 en C8). De klinisch neurofysioloog was geblindeerd voor
de klinische verschijnselen en de MRI resultaten. We bestudeerden de aanwezigheid
van abnormale spontane spiervezelactiviteit (tekenen van denervatie), en van neurogene
Motor Unit Action Potentials (tekenen van rennervatie). We vonden een afwijkend EMG
slechts in 39.2% van de 176 patinten. We zagen vaker tekenen van rennervatie (29.5%)
dan tekenen van denervatie (16.5%). Op basis van deze resultaten concluderen wij dat
een gestandaardiseerd EMG niet zinvol is bij patinten met een kort bestaand cervicaal
radiculair syndroom.

In hoofdstuk 6 beschrijven we onze gerandomiseerde klinische studie bij 205 patinten


met een kort bestaand cervicaal radiculair syndroom. We randomiseerden patinten
voor een behandeling met een halfharde halskraag gedurende 3-6 weken en zoveel
mogelijk rust, of met 12 fysiotherapiebehandelingen en oefeningen thuis gedurende 6
weken, of met zoveel mogelijk continueren van de dagelijkse bezigheden (afwachten;
controlebehandeling). Alle patinten mochten pijnstillers gebruiken.

De armpijn verminderde in de controle groep 3 VAS-mm per week (beta -3.1 mm, 95%
CI -4.0 tot -2.2) wat resulteerde in een pijnafname van 19 mm in totaal in 6 weken. Een
verdere afname werd verkregen met de halskraag en met fysiotherapie: in 6 weken een
extra reductie van 12 mm van pijn in de arm door beide interventies ten opzichte van
controlebehandeling. (kraag: beta -1.9 mm, 95% CI -3.3 tot -0.5 mm, fysiotherapie: beta
-1.9, 95% CI -3.3 tot -0.8).

Nekpijn verminderde niet significant in de controlegroep in de eerste 6 weken (beta -0.9


mm, 95% CI -2.0 tot 0.3). De halskraag gaf een afname van nekpijn met 2.8 VAS mm per
week (95% CI -4.2 tot -1.3) , na 6 weken was dit verschil 17 mm. Fysiotherapie gaf een
afname van 2.4 VAS mm per week (95% CI -3.9 tot - 0.8), totaal 14 mm in 6 weken.

112
Samenvatting

Vergeleken met controlebehandeling, gaf alleen behandeling met de halskraag een


significante verbetering op de Neck Disability Index. Concluderend adviseren we een
halfharde halskraag gedurende 3-6 weken in combinatie met rust, of fysiotherapie met
huiswerkoefeningen gedurende 6 weken, aangezien deze behandelingen relevante
extra afname van arm- en nekpijn gaven in de eerste 6 weken, vergeleken met een
afwachtende houding bij patinten met recent ontstane cervicale radiculopathie.

Hoofdstuk 7 sluit af met een algemene discussie waarin de individuele hoofdstukken


in de context van bestaande literatuur worden gezet, en waarin aanbevelingen worden
gedaan voor verder onderzoek.

113
114
Appendix 1

Recent ontstane cervicale


radiculopathie: minder
pijn met halskraag
of fysiotherapie
Barbara Kuijper, Jos Th J Tans, Anita Beelen, Frans Nollet, Marianne de Visser

Nederlands Tijdschrift voor Geneeskunde 2010

115
Appendix 1

Samenvatting

Doel
Het vergelijken van de effectiviteit van 3 niet-chirurgische behandelingsstrategien bij
patinten met recent ontstane cervicale radiculopathie

Opzet
Gerandomiseerde studie

Methode
205 poliklinische patinten met een cervicale radiculopathie die minder dan een
maand bestond, werden behandeld met een halfharde halskraag en zoveek mogelijk
rust gedurende 3-6 weken, of met 12 fysiotherapiebehandelingen en oefeningen thuis
gedurende 6 weken, of met zoveel mogelijk continueren van de dagelijkse bezigheden
(afwachten; controlebehandeling). Primaire uitkomstmaat was het beloop in tijd van
scores voor arm- en nekpijn op een 100 mm visueel-analoge pijnschaal (VAS) en van de
Neck Disability Index (NDI) gedurende 6 weken na inclusie. Verschillen in beloop tussen
de behandelgroepen werden geanalyseerd met Generalized Estimating Equations.

Resultaten
De armpijn verminderde in de controle groep 3 VAS-mm per week (19 mm in 6 weken).
Een verdere afname werd verkregen met de halskraag en met fysiotherapie: in 6 weken
een extra reductie van 12 mm van pijn in de arm door beide interventies ten opzichte
van controlebehandeling. Nekpijn verminderde niet significant in de controlegroep.
De halskraag gaf een afname van nekpijn met 2.8 VAS mm per week (17 mm in 6
weken) en fysiotherapie 2.4 VAS mm per week (14 mm in 6 weken). Vergeleken met
controlebehandeling gaf alleen behandeling met de halskraag een significante verbetering
op de NDI.

Conclusie
In de eerste 6 weken gaven zowel een halfharde halskraag als fysiotherapie relevante
extra afname van arm- en nekpijn vergeleken met een afwachtende houding bij patinten
met recent ontstane cervicale radiculopathie.

116
Halskraag of fysiotherapie versus een afwachtend beleid

Inleiding

Cervicale radiculopathie is een veelvoorkomende ziekte gekarakteriseerd door nekpijn


uitstralend in het aangedane dermatoom van arm, hand en vingers. Bij onderzoek
kunnen radiculaire prikkelingsverschijnselen en spierzwakte, verlaagde reflexen
en sensibiliteitsstoornissen volgens een myotomaal/dermatomaal patroon worden
gevonden. De diagnose wordt vooral gesteld op klinische gronden. Met magnetische
resonantie (MRI) kan de oorzaak van cervicale radiculopathie worden vastgesteld,
meestal is dat spondylarthrosis of hernia nucleus pulposi.

Cervicale radiculopathie heeft meestal een gunstige prognose, waardoor een afwachtend
beleid in de eerste 6 weken gerechtvaardigd is. 1-4 Omdat de pijn in deze periode vaak
heftig is, is er wel behoefte aan behandelingen, die afname van nekpijn en pijn in de
arm versnellen. Tot nu toe is er geen bewijs voor effectiviteit van niet-chirurgische
behandelingen. In 2 gerandomiseerde onderzoeken waarin verschillende non-invasieve
behandelingen bij chronische cervicale radiculopathie werden vergeleken bleken
fysiotherapie en het dragen van een halskraag niet effectief. 56

De behandeling van een kortbestaand cervicaal radiculair syndroom is tot op heden


niet onderzocht. Daarom hebben wij een onderzoek gedaan bij patinten met een
recent ontstane cervicale radiculopathie, waarbij we keken naar de effectiviteit van 3
behandelingsstrategien op vermindering van pijn en beperkingen. De eerste behandeling
bestond uit het dragen van een halskraag en zoveel mogelijk rust nemen, de tweede uit
gestandaardiseerde fysiotherapie en oefeningen thuis en de derde uit zoveel mogelijk
continueren van de dagelijkse bezigheden (afwachten).

Patinten en methoden

Huisartsen in Den Haag, Gouda en Amersfoort werd gevraagd om patinten met


verschijnselen van cervicale radiculopathie snel door te verwijzen naar een neuroloog.
Inclusiecriteria waren een klinische diagnose cervicale radiculopathie bij patinten met
een leeftijd van 18-75 jaar, verschijnselen korter dan 1 maand, pijn in de arm uitstralend
tot in de onderarm die op de visueel-analoge pijnschaal (VAS) als groter dan 40 mm
werd gewaardeerd en tenminste 1 van de volgende verschijnselen: (1) provocatie van
pijn in de arm door nekbewegingen, (2) gevoelsstoornissen in 1 of meer aangrenzende

117
Appendix 1

dermatomen, (3) verlaagde peesreflexen in de aangedane arm of (4) spierzwakte in 1 of


meer aangrenzende myotomen. Exclusiecriteria waren tekenen van myelumcompressie,
voorafgaande behandeling met halskraag of fysiotherapie en onvoldoende taalkennis.

Behandelingsprotocollen
Deelnemende patinten mochten pijnstillers nemen, paracetamol, NSAIDs en opiaten.
Patinten kregen een dagboek waarin zij alle medicatie die zij gebruikten noteerden. Als
de pijn naar het oordeel van patint en arts onvoldoende kon worden bestreden, werd de
mogelijkheid van chirurgische therapie besproken. Bij alle patinten werd in principe een
MRI van de nek en electromyografisch onderzoek van de arm uitgevoerd. De resultaten
daarvan werden besproken in hoofdstuk 3,4 en 5.

Bij de eerste groep werd een halfharde halskraag (een verstevigde zachte kraag, type
Cerviflex S, Bauerfeind), passend aangemeten. Patinten moesten de kraag 3 weken
overdag dragen en daarbij zoveel mogelijk rust nemen. In de daaropvolgende 3 weken
(week 4-6) bouwden zij het kraaggebruik geleidelijk af. In het dagboek noteerden de
patinten het aantal dagdelen dat zij de kraag droegen.

De tweede groep kreeg fysiotherapie, waarbij de nadruk werd gelegd op mobilisatie en


stabilisatie van de cervicale wervelkolom. De behandelingen vonden 2 keer per week
gedurende 6 weken plaats en werden uitgevoerd door aan het onderzoek deelnemende
fysiotherapeuten. De gestandaardiseerde behandelingen bestonden uit oefeningen om
oppervlakkige en diepe nekspieren te versterken. Tevens kregen patinten instructie voor
oefeningen die zij dagelijks thuis moesten doen (zie appendix 2 in dit proefschrift ). De tijd
die ze besteedden aan oefeningen noteerden de patinten in het dagboek.

De derde groep was de controlegroep. Deze patinten kregen een behandeling die
bestond uit het zoveel mogelijk continueren van de dagelijkse bezigheden. Het aantal
dagdelen dat zij daartoe in staat waren noteerden zij in het dagboek.

Uitkomstmaten en analyse
Primaire uitkomstmaten waren pijn in de arm en nekpijn uitgedrukt in millimeters op
de VAS en beperkingen gemeten met de Neck Disability Index (NDI). 7 8 Patinten
vulden de uitkomstschalen in bij inclusie en 3 weken, 6 weken en 6 maanden na
randomisatie. Het invullen gebeurde in aanwezigheid, maar zonder bemoeienis van een

118
Halskraag of fysiotherapie versus een afwachtend beleid

onderzoeksverpleegkundige. Bij het eerste bezoek, na 6 weken en na 6 maanden deden


wij een gestandaardiseerde anamnese en neurologisch onderzoek bij elke deelnemer.

Wij gebruikten een generalized estimating equations (GEE) analyse om het effect te
onderzoeken van de behandelingen met halskraag en met fysiotherapie op het beloop
van pijnscores en NDI gedurende de eerste 6 weken. Daarbij corrigeerden we we voor
de scores bij inclusie. GEE is een lineaire regressie-analyse, die rekening houdt met
de afhankelijkheid van metingen bij 1 patint. 9 Omdat de scores na 6 maanden een
niet-normale verdeling hadden, werden verschillen tussen groepen geanalyseerd met
niet-parametrische testen (Kruskal-Wallis). Wij deden alle analyses volgens het intention
to treat-principe. Voor ontbrekende gegevens werd de last observation carried forward
methode toegepast.

Resultaten

Patinten
In de periode augustus 2003- december 2006 werd van 275 patinten nagegaan of
zij voldeden aan de inclusiecriteria. Bij 70 was dit niet het geval omdat zij deelname
weigerden (n=9), reeds behandeld waren met fysiotherapie of halskraag (n=13), langer
dan een maand klachten hadden (n=15), de score voor pijn in de arm op de VAS <40
mm was (n=8), er geen cervicale radiculopathie was (n=26), er taalproblemen waren
(n=6) of om onbekende reden (n=2). Sommige patinten hadden meer dan n reden
voor exclusie. Van de 205 patinten kregen er 69 een halskraag, 70 fysiotherapie en 66
controlebehandeling (zie figuur).

Bij het eerste bezoek waren er geen significante verschillen in klinische kenmerken van
de patinten (tabel 1) uit de 3 groepen. Dat betekent dat de randomisatie geslaagd was.
De gemiddelde VAS`-scores waren 70 mm voor de pijn in de arm en 60 mm voor nekpijn.
Neurologische uitval bestond vooral uit gevoelsstoornissen, weinig patinten hadden
een parese of hyporeflexie. MRI van de 3 groepen toonde wortelcompressie door hernia
nucleus pulposie of foramenstenose bij respectievelijk 78,7 (halskraag-groep), 82,0
(fysiotherapie-groep) en 74,1% (controlegroep) van de patinten. Bij het controlebezoek
na 6 weken waren 200 van de 205 patinten beschikbaar. Er was geen significant verschil
tussen de 3 groepen voor het aantal patinten dat uiteindelijk werd geopereerd.

119
Appendix 1

Potentieel geschikt (n= 275)

Geexcludeerd (n= 70)


-Voldeden niet aan inclusiecriteria (n=62)
-Weigerden deelname (n=8)

Inclusie (n=205)

Randomisatie Halskraag (n=69) Fysiotherapie (n=70) Controle groep (n=66)

Controle bij 3 Verloren voor follow- Verloren voor follow- Verloren voor follow-
weken up *:1 up*: 2 up*: 1
Afwezig bij deze Afwezig bij deze Afwezig bij deze
controle : 3 controle : 2 controle: 4

Controle bij 6 Totaal verloren voor Totaal verloren voor Totaal verloren voor
weken follow-up*:1 follow-up*:2 follow-up*:1
Afwezig bij deze Afwezig bij deze Afwezig bij deze
controle : 0 controle : 1 controle : 0

Controle bij 6 Totaal verloren voor Totaal verloren voor Totaal verloren voor
maanden follow-up*:6 follow-up*:2 follow-up*:5

Analyse beloop Geanalyseerd: 68 Geanalyseerd: 68 Geanalyseerd: 65


eerste 6 weken Verloren voor follow-up: 1 Verloren voor follow-up: 2 Verloren voor follow-up 1

Analyse uitkomst Geanalyseerd: 63 Geanalyseerd: 68 Geanalyseerd: 61


6 maanden Verloren voor follow-up: 6 Verloren voor follow-up: 2 Verloren voor folllow-up:5

Figuur 1.

Patinten stroom
* patinten die helemaal niet meer op controle kwamen en niet meer participeerden in de studie
patinten die een controle misten maar nog participeerden in de studie en de verdere controles weer aanwezig
waren

120
Halskraag of fysiotherapie versus een afwachtend beleid

Tabel 1 Klinische kenmerken van 205 patinten bij eerste bezoek

Halskraag (n=69) Fysiotherapie (n=70) Controles (n=66)

Leeftijd (jaar) 47.0 (9.1) 46.7 (10.9) 47.7 (10.6)

Geslacht man (%) 38 (55.1) 34 (48.6) 32 (48.5)

Duur pijn in de arm, gemiddeld; SD (weken) 2.8 (1.4) 2.8 (1.4) 3.0 (1.5)

Duur nekpijn, gemiddeld; SD (weken) 3.3 (2.3) 3.0 (2.1) 3.2 (2.0)

VAS pijn in de arm, gemiddeld; SD (mm) * 68.2 (19.6) 72.1 (19.2) 70.8 (21.2)

VAS nekpijn, gemiddeld; SD (mm) * 57.4 (27.5) 61.7 (27.6) 55.6 (31.0)

Beperkingen (NDI), gemiddeld; SD 41.0 (17.6) 45.1 (17.4) 39.8 (18.4)

Gevoelsstoornissen no (%) 60 (87.0) 63 (90.0) 64 (97.0)

Hyporeflexie biceps reflex no (%) 12 (17.4) 13 (18.6) 16 (24.2)

Hyporeflexie radial reflex no (%) 8 (11.6) 6 (8.6) 6 (9.1)

Hyporeflexie triceps reflex no (%) 16 (23.2) 24 (34.3) 16 (24.2)

Parese m.biceps no (%) 9 (13.0) 5 (7.1) 6 (9.1)

Parese m. triceps no (%) 7 (10.1) 12 (17.1) 6 (9.1)

Parese m. brachialis no (%) 8 (11.6) 8 (11.4) 8 (12.1)

Parese m.extensor digitorum communis no (%) 11 (15.9) 9 (12.9) 9 (13.6)

Parese m.deltoideus no (%) 1 (1.4) 2 (2.9) 3 (4.5)

Parese- m.flexor digitorum no (%) 8 (11.6) 6 (8.6) 4 (6.1)

Parese- m.abductor digiti minimi no (%)* 5 (7.2) 9 (12.9) 4 (6.1)

MRI verricht (n) 61 61 54

Wortelcompressie op MRI (%) 48 (78.7 ) 50 (82.0) 40 (74.1)

* Pijn op Visueel Analoge Schaal uitgaande van een 100 mm schaal


De Neck Disability Index is een 100 punts-schaal voor nekpijn, met 10 items betreffende functionele beperkingen
Parese bij het gestandaardiseerd neurologisch onderzoek (MRC schaal kleiner dan 5)
De patintenpopulatie (n) is hier kleiner, aangezien niet van alle patinten een MRI gemaakt kon worden.
Volgens het originele verslag van de radioloog.

Resultaten van behandeling


In tabel 2 staan de gemiddelde VAS- en NDI-scores van de 3 behandelgroepen, na 3 en 6
weken. De resultaten van GEE-analyse zijn samengevat in tabel 3. De mate van pijnafname
per week is uitgedrukt in de cofficint bta. In de controlegroep was Bta voor afname
van pijn in de arm 3.1 VAS-punten per week. Dat betekent dat de VAS-score voor pijn
in de arm verminderde ruim 3 mm per week tot 19 mm in 6 weken. In de halskraag- en
in de fysiotherapie groep namen de VAS-scores per week een extra 2 mm per week af,

121
Appendix 1

Tabel 2 Uitkomsten pijn in de arm, nekpijn en beperkingen na 3 en 6 weken.

Uitkomstmaat 3 weken gemiddeld (SD) 6 weken gemiddeld (SD)

Halskraag Fysiotherapie Controle Halskraag Fysiotherapie Controle

behandeling behandeling

Pijn in de arm 50.3 (27.7) 55.1 (26.4) 59.1 (26.4) 33.5 (30.4) 36.0 (30.7) 48.6 (31.8)

(VAS) *

Nekpijn 38.0 (28.4) 44.5 (32.5) 55.0 (31.8) 31.0 (28.2) 36.2 (31.0) 51.1 (32.7)

(VAS) *

NDI 33.8 (18.7) 34.6 (16.1) 34.3 (18.8) 25.9 (19.1) 27.8 (17.7) 29.9 (20.0)

* Pijn op Visueel Analoge Schaal uitgaande van een 100 mm schaal


De Neck Disability Index is een 100 punts-schaal voor nekpijn, met 10 items betreffende functionele beperkingen;
hoe hoger de score hoe groter de beperking

Tabel 3 Vergelijking van gemiddelde wekelijkse verandering van uitkomstmaten met


controlebehandeling en halskraag of fysiotherapie gedurende de eerste 6 weken

Wekelijkse afname van pijn/ Extra wekelijkse afname Extra wekelijkse afname

beperkingen in controle groep door halskraag door fysiotherapie

(95% BI) (95% BI) (95% BI)

Pijn in de arm (VAS) -3.1 (-4.0 tot -2.2) mm -1.9 (-3.3 tot -0.5) mm -1.9 (-3.3 tot -0.8) mm

Nekpijn (VAS) -0.9 (-2.0 tot 0.3) mm -2.8 (-4.2 tot -1.3) mm -2.4 (-3.9 tot -0.8) mm

Beperkingen (NDI) -1.4 (-1.9 tot -0.9) -0.9 (-1.6 tot -0.1) -0.8 (-1.8 tot 0.2)

* gebaseerd op Generalized Estimating Equations. 95% BI: 95% betrouwbaarheids interval


Zie tabel 2 voor uitleg
Zie tabel 2 voor uitleg

tot 12 mm extra in 6 weken. Nekpijn verminderde niet significant bij de patinten met
controlebehandeling (bta -0.9). De behandelingen met halskraag of fysiotherapie gaven
wel een significante afname van nekpijn; 2.8 VAS-mm per week (17 mm in 6 weken) met
de halskraag en 2.4 VAS-mm per week (14 mm in 6 weken) met fysiotherapie.

De NDI scores van patinten met een halskraag verbeterden significant sneller dan van
patienten die de controlebehandeling kregen. Er was ook een extra wekelijkse verbetering
door fysiotherapie maar deze was niet significant verschillend van de controlegroep. Na 6
maanden hadden de meeste patinten weinig of geen pijn (tabel 4). De mediane VAS-
score voor pijn in de arm was nul, voor nekpijn 10-20 mm. De pijnscores verkregen met
de onderzoeksbehandelingen verschilden ook niet van die van de controlebehandeling.

122
Halskraag of fysiotherapie versus een afwachtend beleid

Tabel 4 Uitkomsten pijn in de arm, nekpijn en beperkingen na 6 maanden

Mediaan P-waarde *

(p25-p75)

Halskraag Fysiotherapie Controle

n=63 n=68 behandeling

n=61

Pijn in de arm (VAS) 0 (0-30.0) mm 0 (0-46.3) mm 0 (0-50.0) mm 0.928

Nekpijn (VAS) 10.0 (0-40.0) mm 20.0 (0-43.8) mm 10 (0-50.0) mm 0.680

Beperkingen (NDI) 8 (0-26.0) 10 (2-29.2) 8 (0-26.0) 0.670

Mediane waarden (25e en 75e Percentiel).


* Kruskal Wallis test.
Zie tabel 2 voor uitleg
Zie tabel 2 voor uitleg

Beschouwing

Patinten met een recent ontstane cervicale radiculopathie die behandeld werden met
een halfharde halskraag en rust of met fysiotherapie en oefeningen thuis hadden in de
eerste 6 weken significant minder arm- en nekpijn dan patinten in de controlegroep, die
geen behandeling ondergingen. Deze verschillen in pijnreductie tussen onderzoek- en
controlebehandeling waren niet alleen statistisch significant, maar ook klinisch relevant
zoals aangetoond in eerdere studies waarin de VAS-score werd gebruikt. 1011

Na 6 weken waren de beperkingen van patinten in de controlegroep met 9 punten


verminderd op de 100-punts NDI en met nog eens 5 punten extra door behandeling met
halskraag of fysiotherapie. De veranderingen in NDI waren kleiner dan in de pijnscores,
waarschijnlijk omdat de NDI vooral beperkingen meet als gevolg van nekpijn, terwijl pijn
in de arm het belangrijkste onderdeel is cervicale radiculopathie; de initile pijnscores
waren het hoogst voor pijn in de arm en zij toonden de grootste verbetering.

Alle verschillen tussen de 3 groepen waren verdwenen na 6 maanden, aangezien de


meeste patinten toen nauwelijks pijn meer hadden. Het eerder beschreven gunstige
natuurlijk beloop van de aandoening werd hiermee bevestigd. 61213 Aangezien de
patinten voor inclusie al gemiddeld 3 weken pijn in de arm hadden en zij 6 weken
werden behandeld, is aangetoond dat beide behandelingsprotocollen effectief zijn

123
Appendix 1

binnen deze periode. Na 6 weken was aanzienlijke pijnreductie bereikt zodat verdere
behandeling na deze tijd weinig zinvol lijkt. Slechts bij 13 van de 205 patinten, gelijk
verdeeld over de 3 groepen, was operatie nodig tijdens de onderzoeksperiode van 6
maanden, eveneens bewijs voor het goedaardig karakter van cervicale radiculopathie.

Vergelijking met eerder onderzoek


In de literatuur hebben wij geen bewijs gevonden voor werkzaamheid van halskraag
of fysiotherapie bij een kort bestaand cervicaal radiculair syndroom. In een
gerandomiseerde studie uit 1966 werden 5 behandelingen (tractie, houdingstherapie,
halskraag, placebomedicatie en warmtetherapie) vergeleken bij 493 patinten met een
chronische cervicale radiculopathie. 6 Er werd geen verschil gevonden in hoeveelheid
pijn en geschiktheid om te werken tussen de 5 groepen. Door de methodologische
tekortkomingen in die tijd, zo waren er bijvoorbeeld geen gevalideerde uitkomstmaten,
zijn de resultaten nauwelijks vergelijkbaar met de onze. In een andere studie werden
patinten met een chronische cervicale radiculopathie, langer dan 3 maanden met een
mediane duur van 21 maanden, gerandomiseerd voor operatie, fysiotherapie of harde
halskraag. Na 4 maanden gaf operatie de beste pijnbestrijding, na 16 maanden was er
geen verschil meer in pijn tussen de 3 groepen. 514

Wij hebben gekozen voor een halfharde halskraag, omdat een zachte kraag onvoldoende
steun geeft en een harde kraag zeer onbehaaglijk kan zijn. 15 We zorgden dat de kraag
goed passend was en dat deze niet langer dan 6 weken gedragen werd, gezien de
nadelige effecten van te lange immobilisatie. 1516

Beperkingen van ons onderzoek


Een beperking van het onderzoek is dat het onmogelijk was om patinten te blinderen;
ook de onderzoekers waren niet blind voor het type behandeling. Wij hebben getracht
vooringenomenheid te vermijden door patinten zelf de uitkomstschalen te laten invullen
in aanwezigheid van de onderzoeksverpleegkundige, die de patinten niet onderzocht.

Een volgend probleem is dat de positieve resultaten van halskraag en fysiotherapie


gedeeltelijk kunnen berusten op het placebo-effect. Echter, de pijnafname het grootst
bij patinten die 3 weken een halskraag droegen, terwijl patinten die 6 weken
fysiotherapeutische behandeling ondergingen verreweg de meeste aandacht kregen.
Ook de secundaire uitkomstmaten, die te lezen zijn in hoofdstuk 6 in dit proefschrift,
ondersteunen onze opvatting dat het placebo-effect beperkt was.

124
Halskraag of fysiotherapie versus een afwachtend beleid

Tenslotte kan het een beperking zijn dat wij patinten includeerden op basis van klinische
criteria. Meestal wordt namelijk pas MRI van de nek verricht als men chirurgische
interventie overweegt. Bovendien zijn de MRI bevindingen niet zelden fout negatief 1718,
zoals ook bleek uit het ontbreken van wortelcompressie in 20-25% bij onze op klinische
criteria zorgvuldig geselecteerde patinten.

Aanbevelingen voor de praktijk


Wij adviseren om patinten met een recent ontstane cervicale radiculopathie gedurende
3 weken te behandelen met een halfharde halskraag en rust. Dit bleek een effectieve
behandeling. Een goed alternatief is fysiotherapie met mobiliserende en stabiliserende
oefeningen in combinatie met oefeningen die de patint thuis uitvoert.

125
Appendix 1

1. Adams C. Cervical spondylotic radiculopathy and myelopathy. Amsterdam/ New York / Oxford:

Elsevier Publishing Company, 1976.

2. Bush K, Chaudhuri R, Hillier S, Penny J. The pathomorphologic changes that accompany the

resolution of cervical radiculopathy. A prospective study with repeat magnetic resonance imaging.

Spine 1997;22(2):183-6; discussion 187.

3. Spurling RG, Segerberg LH. Lateral intervertebral disk lesions in the lower cervical region. J Am Med

Assoc 1953;151(5):354-9.

4. Vinas FC, Wilner H, Rengachary S. The spontaneous resorption of herniated cervical discs. J Clin

Neurosci 2001;8(6):542-6.

5. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery,

physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997;22(7):751-8.

6. Pain in the neck and arm: a multicentre trial of the effects of physiotherapy, arranged by the British

Association of Physical Medicine. Br Med J 1966;1(5482):253-8.

7. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. J Manipulative Physiol

Ther 1991;14(7):409-15.

8. McCarthy MJ, Grevitt MP, Silcocks P, Hobbs G. The reliability of the Vernon and Mior neck disability

index, and its validity compared with the short form-36 health survey questionnaire. Eur Spine J

2007;16(12):2111-7.

9. Twisk J. Applied longitudinal data analysis for epidemiology. A practical guide. Cambridge Cambridge

University Press, 2003.

10. Todd KH. Clinical versus statistical significance in the assessment of pain relief. Ann Emerg Med

1996;27(4):439-41.

11. Todd KH, Funk KG, Funk JP, Bonacci R. Clinical significance of reported changes in pain severity. Ann

Emerg Med 1996;27(4):485-9.

12. Binder AI. Cervical spondylosis and neck pain. Bmj 2007;334(7592):527-31.

13. Radhakrishnan K, Litchy WJ, OFallon WM, Kurland LT. Epidemiology of cervical radiculopathy. A

population-based study from Rochester, Minnesota, 1976 through 1990. Brain 1994;117 ( Pt 2):325-35.

14. Persson LC, Moritz U, Brandt L, Carlsson CA. Cervical radiculopathy: pain, muscle weakness and

sensory loss in patients with cervical radiculopathy treated with surgery, physiotherapy or cervical

collar. A prospective, controlled study. Eur Spine J 1997;6(4):256-66.

15. Mazanec D, Reddy A. Medical management of cervical spondylosis. Neurosurgery 2007;60(1 Supp1

1):S43-50.

16. Polston DW. Cervical radiculopathy. Neurol Clin 2007;25(2):373-85.

17. Nardin RA, Patel MR, Gudas TF, Rutkove SB, Raynor EM. Electromyography and magnetic resonance

imaging in the evaluation of radiculopathy. Muscle Nerve 1999;22(2):151-5.

126
Halskraag of fysiotherapie versus een afwachtend beleid

18. Sengupta DK, Kirollos R, Findlay GF, Smith ET, Pearson JC, Pigott T. The value of MR imaging in

differentiating between hard and soft cervical disc disease: a comparison with intraoperative findings.

Eur Spine J 1999;8(3):199-204.

127
128
Appendix 2
Physiotherapy protocol

129
130
Physiotherapy protocol

Hands-off exercise therapy

Exercise 1: Chest Press, sitting position


(Technogym)
Purpose: warming-up
2 10 repetitions with 5 kg.

Exercise 2: Lateral Pull-down, sitting position


(Technogym)
Purpose: warming-up
2 10 repetitions with 20 kg

Exercise 3: Low-back flies


Purpose: warming-up
Dumbbells in every hand and make flying movements, bent forward standing position or
in roman chair
2 10 repetitions with 1 kg

Exercise 4: Neck-press
Purpose: stability
Push dumbbells from the shoulder above the head, standing position
2 10 repetitions with 1 kg

Exercise 5: Front-raises
Purpose: stability
Elevate dumbbells forwards to shoulder height, standing position
2 10 repetitions with 1 kg

Exercise 6: Upright row


Purpose: strength
Rowing up bar with weights; elbows finish above shoulder height and wrists finish at
shoulder height, standing position
2 10 repetitions with 7.5 kg

131
Appendix 1

Exercise 7: Weight rotation


Purpose: strength and stability
In standing position, keep bar with weights on top in vertical position, bottom part stays
on the ground, with stretched arms and rotate to left and right
3 repetitions: 5 to left, 5 to right with 7.5 kg

Home exercises

Once a day 210 repetitions.

Exercise 1
Purpose: mobility
Standing position. In neutral position: withdraw chin.

Exercise 2
Purpose: mobility
Lying on the back. Withdraw chin while keeping head on the ground.

Exercise 3
Purpose: mobility
Standing position. Withdraw chin and turn head to one side as far as possible. Repeat in
the opposite direction.

Exercise 4
Purpose: stability and muscle strength
Standing position. Withdraw chin, place the palm of the hand against the head (left or
right side of the forehead), and give resistance against the hand with the head (do not
allow any movements of the head).

Exercise 5
Purpose: stability and muscle strength
Standing position. Place right hand against the head behind the right ear, left hand on the
left side of the forehead. Rotate the head to the right against the resistance of the hands.
Reverse hand positions and repeat to the left. No movements of the head.

132
Physiotherapy protocol

Exercise 6
Purpose: stability and muscle strength
Standing position. Withdraw chin, place both hands on the back of the head, and push
the head against the hands. No movement of the head allowed.

Exercise 7
Purpose: stability and muscle strength
Standing position. Withdraw chin, place the right hand on the right side of the head and
move the head to the right against resistance. Repeat to the left.

Exercise 8
Purpose: stability and muscle strength
Lying on the back. Lift the head a little from the ground and move the chin just a little bit
towards the chest.

Exercise 9
Purpose: stability and muscle strength
Lying on the back, lift the head a little from the ground and turn the head to the right.
Repeat to the left.

Exercise 10
Purpose: relaxation.
Sitting on a chair. Keep both arms down. Pull back the shoulders and relax again.

133
134
List of publications

List of publications

Root compression on MRI compared with clinical findings in patients with recent onset
cervical radiculopathy.
Kuijper B, Tans JThJ, van der Kallen BFW, Nollet F, Lycklama A Nijeholt GJ, de Visser M.
Journal of Neurology, Neurosurgery & Psychiatry.
2011 May;82(5):561-3. Epub 2010 Nov 3.

Interobserver agreement on MRI evaluation of patients with cervical radiculopathy.


Kuijper B, Beelen A, van der Kallen BFW, Nollet F, Lycklama A Nijeholt GJ, de Visser M,
Tans JThJ.
Clinical Radiology.
2011 Jan;66(1):25-9.

Recent ontstane cervicale radiculopathie: minder pijn met halskraag of fysiotherapie


Kuijper B, Tans JThJ, Beelen A, Nollet F, de Visser M.
Nederlands Tijdschrift voor Geneeskunde.
2010;154:A1283

Cervical collar or physiotherapy versus wait and see policy for recent onset cervical
radiculopathy: randomised trial.
Kuijper B, Tans JThJ, Beelen A, Nollet F, de Visser M.
British Medical Journal.
2009 Oct 7; 339:b3883.

Degenerative cervical radiculopathy: diagnosis and conservative treatment. A review.


Kuijper B, Tans JThJ, Schimsheimer RJ, van der Kallen BFW, Beelen A, Nollet F, de Visser M.
European Journal of Neurology.
2009 Jan;16(1):15-20

Results of needle electromyography in 176 patients with cervical radiculopathy


Kuijper B , Schimsheimer RJ, Beelen A, Verheul F, Nollet F, de Visser M, Tans JThJ
submitted

135
List of publications

Referral pattern of neurological patients to psychiatric Consultation-Liaison Services in 33


European hospitals.
de Jonge P, Huyse FJ, Herzog T, Lobo A, Malt U, Opmeer BC, Kuijper B, Krabbendam A.
General Hospital Psychiatry
2001 May-Jun;23(3):152-7.

The impact of counselling on HIV-infected women in Zimbabwe.


Krabbendam AA, Kuijper B, Wolffers IN, Drew R.
AIDS Care.
1998 Apr;10 Suppl 1:S25-37.

136
Dankwoord

Dankwoord
Een proefschrift schrijven in de avonduren kan best, maar niet zonder de hulp,
betrokkenheid en steun van velen. Heel veel dank aan iedereen.

Dr. Jos Tans,


Vanaf de allereerste seconde is dit ons gezamenlijk project geweest. Al die tijd was
je een geweldige begeleider. Ik waardeer je concrete aanpak, efficintie en heldere
visie. De eerste jaren vanuit het Westeinde Ziekenhuis, en de laatste jaren via email
vanuit Wassenaar of Frankrijk; je bent altijd dichtbij en betrokken geweest. Bovendien
respecteerde je mijn werktempo; dit is voor mij van cruciaal belang geweest om de
eindstreep te halen.

Prof. Marianne de Visser,


Je bent veel meer voor dit proefschrift gaan betekenen dan ik op voorhand had kunnen
voorzien. Je enthousiasme, maar ook je kritische academische blik hielpen mij om mijn
eigen wetenschappelijke grenzen op te zoeken; je hebt me geprikkeld en gestimuleerd
om dit proefschrift op een hoger plan te brengen.

Prof. Frans Nollet,


Als tweede promotor was je niet minder belangrijk, jouw inzichten kwamen vanuit een
andere achtergrond, namelijk vanuit de revalidatiegeneeskunde. Ik waardeerde je frisse
ideen en stimulerende opmerkingen.

Dr. Anita Beelen,


Jij hielp me niet alleen met het uitvoeren van de analyses, maar het was ook bijzonder fijn
dat je het geduld had om mij de achtergronden van de statistische analyses uit te leggen.

De patinten die het bereid waren mee te doen met het onderzoek,
Jullie gaven de studiepopulatie een gezicht.

De verwijzende huisartsen, neurologen, arts-assistenten, Freek Verheul namens de


Goudse neurologen, Willem Oerlemans vanuit Amersfoort,
Dankzij jullie inzet kon ik 205 patinten includeren, en dat binnen mijn opleidingstijd.

De collega-assistenten neurologie uit het Westeinde,


Jullie zeurden nooit als ik voor mijn onderzoekspoli naar beneden ging. Jullie vervingen
mij in noodgevallen op deze poli en bekeken de laatste follow-up patinten na mijn
vertrek naar Rotterdam. Bovendien zorgden jullie voor een top-opleidingstijd.

137
Dankwoord

De klinisch neurofysiologen; Al de Weerd, Rob Jan Schimsheimer, Viviane van Kasteel,


De neuroradiologen Bas van der Kallen en Geert Lycklama a Nijeholt,
De deelnemende fysiotherapeuten,
Dank voor het vele extra werk wat jullie speciaal voor dit onderzoek hebben verzet.

De neurologen, neurochirurgen, klinisch neurofysiologen, kinderneurologen van het


Medisch Centrum Haaglanden,
Ik kon mijn onderzoekspoli altijd continueren tijdens de verschillende stages. Jullie
hebben me bovendien opgeleid tot een veelzijdig neuroloog.

Mijn geweldige onderzoeksverpleegkundige Ella de Wolf,


Je bent een kei in regelen, administratie en logistiek, waardoor het promotietraject
behoorlijk soepel liep.

Het secretariaat neurologie van het Maasstad Ziekenhuis,


De laatste loodjes, met het daarbij horende administratieve werk, waren een makkie
dankzij jullie hulp.

Bibliothecarissen Thomas Vissers van het MCH Westeinde en Jacco van Meel van het
Maasstad Ziekenhuis,
Jullie bijdrage aan dit proefschrift is onmiskenbaar, door het snel aanleveren van de door
mij opgevraagde artikelen, regelmatig zelfs op zondagavond.

Overige leden van de promotiecommissie,


Dank voor het plaats nemen in de commissie en het beoordelen van het manuscript.

Collegas van de maatschap neurologie van het Maasstad Ziekenhuis,


Overdag werken in een goede sfeer met fijne collegas is belangrijk om s avonds de rust
te vinden om aan een proefschrift te kunnen schrijven. Dank voor jullie geduld, begrip en
samenwerking.

Vele lieve mensen die ver van Rotterdam wonen maar altijd dichtbij voelen: mijn broers
Maarten en Gijs, vrienden, schoonzussen, zwagers,
De vriendschap, interesse en gezelligheid die ik bij jullie vind maken van mij een gelukkig
mens. Ik hoef jullie niet elke week te zien of te spreken om jullie vriendschap te voelen.
In het bijzonder wil ik nog Marille noemen, mijn schoonzus,
Altijd betrokken en liefdevol. Misschien zorgden de urenlange telefoongesprekken soms
voor enige vertraging.

138
Dankwoord

En: Irene, Didericke, Anne,


Om in Marizas termen te spreken: jullie zijn mijn hartsvriendinnen. Voor altijd.

Ouders, schoonmoeders en schoonvader,


Lieve pa en mam, dank voor jullie onvoorwaardelijke steun, liefde en interesse.
Rianne en Erika, bij jullie vind ik inspiratie: boeken, reizen, leven.
Egbert en Wil, door jullie ben ik gaan hardlopen en sta ik in topconditie tegenover de
commissie.

Mijn paranimf Gijs,


Je bent er altijd voor me, lieve grote broer. Je creatieve geest zorgde ook nu weer voor
een mooi plaatje. Fijn dat je naast me en achter me staat.

Mijn paranimf Anne,


Dank voor je jarenlange vriendschap, gezamenlijke opleiding en wetenschappelijke
carrire; voor wetenschappelijk onderzoek gingen we naar de Sloterplas, naar een
huisartsenpraktijk voor illegalen, of naar Zimbabwe in ieder geval altijd samen. Tuurlijk
moet jij als paranimf dit ook samen met mij doen.

Liefste Mariza en liefste Joram,


Jullie zijn mijn grootste trots. Ik ben blij dat het schrijven aan dit boek nooit ten koste is
gegaan van mijn tijd met jullie. Mijn liefde voor jullie is oneindig veel groter dan die voor
de wetenschap. De energie die ik kreeg van mijn uren met jullie, in speeltuin, dierentuin,
zwembad of gewoon thuis op de bank met een voorleesboek, kon ik s avonds gebruiken
om verder te schrijven. Daardoor hebben jullie een grote bijdrage geleverd aan dit boek!

Mijn rots en grote liefde Enno,


Zo, nu is die van mij ook af, de laptop mag nu echt onder de bank. Nadat jouw
proefschrift klaar was, merkte ik dat we elkaar ook wetenschappelijk aanvullen. Je hielp
me met het aanscherpen van mijn gedachten, en beantwoordde vele met het proefschrift
samenhangende vragen.
Maar mijn grote dank voor jou reikt vele malen verder dan dit. Zonder jouw liefde, geduld,
aanwezigheid naast me op de bank, was het proefschrift misschien wel afgekomen, maar
was deze promovenda niet zo gelukkig geweest als zij nu is met jou en met ons heerlijke
gezin.

Barbara

139
140
Over de auteur

Over de auteur

Barbara Kuijper werd geboren op 15 juni 1974 in Alkmaar,


als jongste in een gezin met broers Maarten en Gijs. Ze
groeide op in Heiloo en ging in Alkmaar naar naar het
Petrus Canisius College. Na het behalen van het VWO-
diploma in 1992, begon ze haar studie geneeskunde aan
de Vrije Universiteit te Amsterdam. In 1999 werd ze arts,
hierna werkte ze bijna twee jaar als agnio neurologie:
in het Lucas-Andreas Ziekenhuis en in het Academisch Medisch Centrum, beide te
Amsterdam. In 2001 kon ze beginnen aan de opleiding neurologie in het Medisch
Centrum Haaglanden (opleiders dr J.Th.J Tans en later dr Ch.Vecht). In 2003 zette ze
samen met dr. Tans een onderzoek op naar het cervicaal radiculair syndroom, waarvoor
ze elke dinsdagochtend patinten met een cervicaal radiculair syndroom zag op de
speciale onderzoekspoli. Na het afronden van haar opleiding in december 2006 verhuisde
ze naar Rotterdam-Hillegersberg, om te gaan werken in het Medisch Centrum Rijnmond-
Zuid wat nu het Maasstad Ziekenhuis heet. De eerste jaren als neuroloog werkte ze
gestaag verder aan het verwerken van de data verkregen tijdens de studieperiode, wat in
2011 resulteerde in dit proefschrift. Ze is getrouwd met Enno Wildschut, samen zijn ze de
ouders van Mariza (2004) en Joram (2007).

141