Pain, 51 (1992) 43-48 @ 1992 Elsevier Science
43 Publishers
B.V. All rights
reserved
0304-3959/92/$05.00
PAIN 0211X
Cervicogenic headache, migraine without aura and tension-type headache. Diagnostic blockade of greater occipital and supra-orbital nerves Gunnar Bovim and Trond Sand Department
ofNeurology,
(Received
Trondheim University Hospitals, Regionsykehuset,
19 November
1991, revision
received
20 March
7006 Trondheim (Norway)
1992, accepted
1 May 1992)
The diagnostic value of greater occipital and supra-orbital nerve blockades in patients with Summary cervicogenic headache, migraine without aura, and tension-type headache was investigated. The pain reduction after greater occipital nerve blockade was significantly more marked in the cervicogenic headache group than in the other categories. Moreover, pain reduction in the forehead was generally only found in the cervicogenic headache patients (77%). Pain reduction (in %) was significantly more marked following the greater occipital than the supra-orbital nerve blockade. The volume effect per se was evaluated by saline injection. This procedure did not result in distinct pain reduction. The effect obtained in cervicogenic headache is, accordingly, probably due to the local anaesthesia. The present results support the postulate that different pathogenetic factors probably are responsible for cervicogenie headache, tension-type headache, and migraine without aura. Key words:
Cervicogenic Supra-orbital
headache; Tension-type headache; nerve; Nerve block (anaesthesia)
Introduction
Clinical differential diagnosis between cervicogenic headache, migraine without aura (‘common migraine’) and tension-type headache may be difficult on the basis of history and examination alone. The recommended treatment differs widely in the 3 types of headache. For this reason, it is important to improve the clinical distinction between them. The methods for, and importance of, diagnostic anaesthetic blockades have been outlined by different authors (Hunter and Mayfield 1949; Sjaastad et al. 1983; Jansen et al. 1989). According to the present pathogenetic model for cervicogenic headache, pain in the forehead may be caused by painful stimuli being transferred from the neck/posterior part of the head (mainly by the Cl-C3 nerves). In the brain stem, nerve fibres have a very
Correspondence to: Gunnar Bovim, sykehuset, 7006 Trondheim, Norway.
Dept.
of Neurology,
Region-
Migraine
without
aura;
Greater
occipital
nerve;
intimate relationship to the caudal part of the spinal trigeminal nucleus. Cortical interpretation of the impulses stemming from the posterior parts may, therefore, be a pain in the area of the ophthalmic division of the trigeminal nerve (in the following referred to as the ‘Kerr principle’) (Kerr 1961). On the basis of this model, it would be of interest to study the response to greater occipital nerve (GON) and supra-orbital nerve (SN) blockades in cervicogenic headache. The GON blockade has been included among the diagnostic criteria for cervicogenic headache (Sjaastad et al. 1990) as an ‘important’ criterion. If GON injections abate pain in the forehead in addition to the neck, i.e., outside the innervation area of the blocked nerve, this would be an indication that the Kerr principle is operative in this type of pain (Sjaastad 1990). Hence, the responses to GON and SN blockades in cervicogenic headache, migraine without aura (common migraine, in the following partly termed migraine) and tension-type headache have been investigated in the present study. Studies of this nature have, to the best of our knowledge, not been performed before.
44
TABLE
1
CHARACTERISTICS
OF THE
PATIENTS Cervicogenic
Migraine
Tension-type
headache
without aura
headache
No. of patients
24
14
I-1
Females/males
21/3
II/3
Age range (median)
IN-64
Unilaterality
without sideshift
24 *
Unilaterality
with sideshift
Bilateral
20-57
Symptoms of neck involvement
* *
0
II
0
0
3
I4
24
0
0
3
I2
21
2
3
I4
Attacks without pain in between almost continuous pain
7/7 2l-h9(4I)
(39)
0
0
pain
Fluctuating,
(435)
Pulsating pain
3 II 7 ***
Nausea, as a nearly constant accompaniment II
Phono- and/or
II
photophobia
* Five patients reported * * (Sjaastad
that headache,
et al. 1990) These
when maximal,
symptoms include
ipsilateral
upper posterior
(presence
of at least I of these factors).
* * * With maximal headache
Materials
13
I3
of severe headache Nausea, never or only sporadically
I2 I2
also could be felt on the contralateral
(1) either
neck region, (2) ipsilateral,
attack
provocation
non-radicular
neck/shoulder/arm
with
cervicogenic
graine, and 14 with tension-type (Table
headache,
I4 with
mi-
headache were included in the study
1). In some of the patients
only one of the blockades
was
carried out.
the
In the Classification stemming
and Diagnostic
Criteria
Headache
for Headache
Society (IHS),
from the neck was only summarily
genie headache
is not sufficiently
cation Committee
described:
validated”
(Headache
1988). The diagnosis of cervicogenic
the present study was, therefore,
Major
towards
into
IHS criteria sion-type
(Headache
headache
Classification
group,
11) of course without
consideration.
mainly
The
headache Committee
patients
with
diagnoses
DIAGNOSTIC
1988). In the tenthe chronic
form
were represented. avoid
contamination
of the
for more than
material
by patients
I of the headache
fulfilling
categories,
established
Disor-
(Table
I).
In this connection,
headache
‘positive’
“Cervico-
Unilaterality
Classifi-
headache
in
according to the recent
criteria,
without
was present definition. headache,
we paid particular
as far as cervicogenic sideshift
(i.e..
in all the cervicogenic All
the
whereas
headache
attention
only
CRITERIA
FOR
CERVICOGENIC
HEADACHE
(for the complete
cases in accordance
headache
in the migraine
group
II
patients patients
had
see Sjaastad et al. 1990)
of neck involvement
Pain precipitated lpsilateral
by mechanical
pressure to the ipsilateral
neck/shoulder/arm
Reduced
upper posterior
neck region or by awkward
head positioning
pain
range of motion in the cervical spine
Pain characteristics Non-clustering Moderate,
pain episodes of varying duration
usually non-throbbing
important
Anaesthetic
(or fluctuating,
continuous
pain)
pain, starting in the neck, spreading forward
criteria
blockades of the GON
and/or
the C2 nerve on the symptomatic
side abolish the pain transiently
sex
Head and/or
neck trauma
by history
Note: The effect of anaesthetic
blockades was, of course, not taken into consideration
when including patients
with
the
hilateral
had unilateral
headache
Symptoms/signs
(1)
pain always on the same side)
tension-type
criteria,
to the
is concerned.
symptoms and signs
Female
of
were based on the
II
Unilateral
Other
the
patients who could be given rather distinct diagnoses were included
ders, set forth by the International
MAIN
pressure
et al. 1990) (Table
effect
migraine without aura and tension-type
To
Diagnosis
(Sjaastad
blockade
diagnostic criteria
TABLE
or by mechanical
pain, and (3) reduced range of motion in the neck
diagnostic criteria taking
patients
side, but to a lesser degree.
by neck movement
only.
and methods
Twenty-four
3
in the present series.
45 pain (invariably with sideshift, i.e., the pain was alternating from left to right or vice versa between or within the single headache attack) and 3 patients had bilateral pain (2 of these with side predominance) (Table I). (2) Signs of neck involvement are as stipulated in Table 1. One or more of these phenomena were present in all the cervicogenie headache cases, but not in any of the other patients. There were clear differences between the 3 groups as for nausea, photoand phonophobia, and pulsating pain quality (Table Il.
Blockades To block the GON, the needle was placed approximately 2 cm lateral and 2 cm inferior to the external occipital protuberance. The needle was first forwarded onto the periosteum of the occipital bone and then withdrawn approximately 0.5 cm before injection. Paraesthesias were frequently felt in the GON innervation area when the needle was held in this position, but their presence was not obligatory to achieve a blockade. One-half to 1.5 ml of lidocaine (20 mg/ml with 12.5 pg/ml adrenaline) was injected and the ensuing sensory deficit (pinprick) carefully assessed. By using only a small volume, a selective blockade was obtained. The topographic variability in the peripheral course of the GON (Bovim et al. 1991) made this procedure insufficient in 4 patients (2 with cervicogenic headache and 2 with tension-type headache). In these patients, a new injection was carried out at a later stage, 2 cm further down in the neck, and was then successful. To evaluate the volume effect per se, 2-ml injections of isotonic saline were made with the same location as the GON blockades in 16 of the cetvicogenic headache patients. These injections were given prior to the anaesthetic blockade and always at the spot where CON was successfully blocked. The patients were not informed as to the nature of this injection prior to the test. The SN was easier to localize than the CON, since it rather regularly seemed to enter the forehead via the supra-orbital incisure. In 2 patients, however, the nerve seemed to divide in the orbit before entering the forehead, and additional injections (more laterally) were necessary to achieve an appropriate sensory loss. One-half to 1 ml of lidocaine (20 mg/ml with 12.5 pg/ml adrenaline) was used during each injection. Accidental anaesthesia of the smaller, more medially located supratrochlear nerve cannot be ruled out with the present technique. The patients reported the effects in 2 different ways: (1) by a visual analogue scale (VAS) as to the pain intensity (prior to and every 5th minute after injection, separate recordings on the right and left sides), and (2) by pointing out the pain localization on a pre-made scheme. The sensory loss was carefully evaluated and drawn on a scheme following anaesthetic injections. The patients were asked to estimate the headache intensity in relation to headache maximum on the VAS scale. All injections were made when the pain was at least 50% of the pain maximum, evaluated by VAS. Patients with obviously ‘unstable’ headache at the time of investigation (progressing or in remission) were not included. Generally, the patients were without drugs at the time of investigation.
Statistical analysis The pain versus time (after injection) profiles were analysed with repeated measures analysis of variance (ANOVA). The effects of CON and SN blockades in the various headache groups were separately compared (cervicogenic headache vs. migraine, cervicogenic headache vs. tension-type headache, and migraine vs. tension-type headache). The difference between the effect of GON blockade and SN blockade was analysed within each headache group. In these analyses, the ‘main-effect F value’ reflects the extent of parallel displacement between pain profiles, while the ‘interaction F value’ reflects the degree of non-parallelism in the profiles. In addition, secondary (descriptive) Student’s t tests (2-sided) were applied at
Greater occipital nerve
100 [
:ot
10 i
1
‘3:
L__.
0
Supraorbital nerve
_I
5
~-
~112
IO 15 20 25 30 Minutes
U -t
L
L_.-.~
L-
0
5
10 15 20 25 30 Minutes
Fig. I. Pain relief following injections. Pain intensity registration every 5th min following greater occipital (A) and supra-orbital nerve (B) blockades. No major changes were reported between 30 and 70 min. The bars indicate S.E.M.s. (Al Secondary Student’s f tests tcervicogenic headache vs. tension-type headache) were significant after CON blockade from the 5th minute onwards (P < 0.05) and highly significant from the 20th minute onwards (P < 0.001). (B) Student’s t tests (cervicogenic headache vs. tension-type headache) were insignificant after SN blockade at each measurement.
each time point. Differences in (individual) maximal pain reduction were analysed with the Mann-Whitney test. Two-by-two tables were analysed with a chi-square test including Yates’ correction. P values less than 0.05 were regarded as significant. Intra-individual correlations between SN and GON blockades have been calculated with Pearson’s r (correlation coefficient).
Results
Side effects were not reported after the injections, and no patient experienced a change of more than 10% in pain intensity on the contralateral side. The pain intensity prior to injection was similar in the 3 headache groups (Fig. 1). Saline injections
Saline injections in the GON area in patients with cervicogenic headache (n = 16) gave no relief of pain in 12 patients whereas 4 patients reported minor relief (less than 20% improvement on VAS) of short duration (less than 20 min). The pain persisted as opposed to the reduction observed after anaesthetic GON blockade (ANOVA: main-effect F = 28.7, P < 0.0005) (Fig. 1A). Localization of the headache was not altered by this procedure. Blockade of the GON
Patients with cervicogenic headache responded better than patients with tension-type headache and migraine (Fig. 1A). The highly significant difference which was observed between the 3 headache groups (Table III) was due to the difference between the cervicogenic headache and tension-type headache groups and to the difference between the cervicogenic headache and mi-
4h
TABLE THE
improvement after GON blockade was reported by 7 migraine and 5 tension-type headache patients, as opposed to only 2 cervicogenic headache patients. Generally, the main pain reduction took place within 5 min after injection (Fig. 1A). Only minor changes in pain intensity were reported between 20 and 30 min. Increased headache was only reported by 2 migraine patients, probably due to spontaneous progression of the attack. Discontinuation or improvement of forehead pain following GON blockade (i.e., pain outside the innervation area of the nerve) was reported by 17 of 22 cervicogenic headache patients (77%) and by only 2 other patients (Table IV).
111 RESULTS
VARIANCE BEEN
FROM
TABULATED
Patients
with
tension-type
Headache
TWO
(ANOVAs)
DIFFERENT
WITH
AS F STATISTIC
cervicogenic headache
ANALYSES
REPEATED
headache,
MEASURES
OF HAVE
VALUES migraine
without
aura.
and
have been compared.
CON
Supra-orbital
blockade
blockade
x.5 *
2.2
3.4 **
5.8 **
category
(main effects) F(df=21 Headache
category
vs. time interaction F (df=
IO)
VAS values for pain (in the injected side of the head) between 5 and Xl min after injection
df: degrees of freedom.
have been analysed.
* P < 0.005, * * P < 0.0005.
Blockade
Note: The main effect F value is a statistical reflection of parallel blockade)
displacement and
IB
between
(supra-orbital
of the degree
the 3 curves in Fig. blockade).
The
1A (CON
interaction
terms
reflect the degree of non-parallelity.
graine groups (post-hoc Student’s t test: P < 0.01 from the 15th minute onward). On the other hand, a lesser difference was found between the pain profiles in migraine versus tension-type headache (Table III, Fig. 1A). Maximal pain reduction (median values) was 54.5% in the cervicogenic headache, 14% in the tension-type headache and 6% in the migraine group (Fig. 2A). During the 30 min following injection, 17 of 22 cervicogenie headache patients reported maximal pain relief of more than 40%. An improvement of this order of magnitude was not reached by any tension-type headache patient and by only 1 migraine patient. Migraine in this patient improved both in the front and back, possibly due to a spontaneous remission. No
Greater
A
occlpltal
90
100 Fig. 2. Maximal CON
E
Termon-type
-
he&)he
Mlgralne without aura (n=ll)
Tenwn-type he;daahe
Median
(A) and SN (B) blockades.
The maximal,
----
Following
relative pain reduction
the CON
blockade
(A)
CON
blockade:
cervicogenic
(B) SN blockade: cervicogenic
within 30 min following the injections are illustrated
it should be noted that no patient,
(1 exception discussed in the text) achieved 50% or more reduction
P < 0.0005.
nerve
0000000
pain reduction.
the pain reductions.
Supraorbltal Cerwcogenlc
he(;$$e
ii
SN
B
nerve
Mlgralne without aura (n=i4)
of the
The SN blockade differentiated poorer between the headache categories than did the GON blockade (non-significant main-effect in Table III). Overall pain reduction in the migraine group was, however. less than the reduction in the other groups (Figs. 18 and 2B) corresponding to the significant interaction in Table III. Pain levels were significantly lower in cetvicogenic headache than in migraine from the 20th minute onwards (post-hoc Student’s t test, P < 0.09. The mean pain scores were also slightly lower in cervicogenic headache than in tension-type headache from 10 min onward. These differences did not reach statistical significance in the post-hoc Student’s t tests, however. Maximal pain reduction (medians) was 28% in the cervicogenic headache, 30% in the tension-type headache, and 16% in the migraine groups. Two patients with cervicogenic headache reported complete pain relief within 20 min, whereas the best response in migraine was 30% reduction and in tension-type headache 50% reduction.
headache
of the pain. The Mann-Whitney vs. tension-type
headache vs. tension-type
headache:
headache:
either with tension-type
median
for each patient after headache
test was applied to compare
P < 0.0005;
P = 0.23; cervicogenic
cervicogenic
headache
headache vs. migraine:
or migraine
the medians of vs. migraine: P < 0.02.
47
TABLE
IV
LOCATION
OF
PAIN
REDUCTION
RELATED
TO
AREA
OF
ANAESTHESIA A: blockade of the greater
Cervicogenic
Migraine
Tension-
occipital nerve
headache
without
f” = 22)
aura
type headache
fn = 14)
(n = 13)
Frontal
pain reduction
Occipital
pain reduction
19 -,f
x
B: blockade of the supra-
Cervicogcnic
Migraine
Tension-
orbital nerve
headache
without
tn = 13)
aura
type headache
(n=Il)
tn = 9)
Frontal
pain reduction
Occipital
6
I2
h
ther in the migraine nor in the tension-type headache group (ANOVA, Student’s t test). In the cervicogenic headache group, however, the GON blockade was significantly more effective than the SN blockade when relative pain reduction was considered (P < 0.003, ANOVA). Pain levels differed significantly at 5 min (P < 0.002) and 15 min (P < 0.03) (2-sided Student’s I test). Intra-individual correlations between the responses at the 2 injection sites were calculated (Table V). Among the 30 patients who received both injections, there seemed to be a slight positive correlation (Pearson’s r = 0.36, P < O.OS),particularly in the tension-type headache patients. When the patients were split according to response strength (less vs. greater than 50% pain reduction after GON blockade) the correlation was only significant for the ‘low responders’ (Table VI.
0
pain reduction
Discussion The “urnher of patients who reported
pain reduction
back) of the head following the blockade pain reduction Frontal
by the supra-orbital
area: area innervated
Chi-square
The mmmf
of
is not taken into consideration.
area: area innervated
Posterior
is reported.
in the front for
by the greater
nerve.
occipital
nerve.
test with Yates’ correction.
xxx: P < o.otlft3: xx: P < 0.02: x: P < 0.04.
The SN blockade did not relieve pain outside the innervation area in migraine or tension-type headache groups. Some relief of pain in the posterior part of the head was reported by 6 of 13 cervicogenic headache patients (Table IV). Comparison between injections around the GON and SN
Pain reduction after GON and SN blockades was compared. No significant differences were found, neiTABLE
V
INTRA-INDIVIDUAL SPONSES NERVE
TO
CORRELATION
GREATER
BETWEEN
OCCIPITAL
VS.
THE
RE-
SUPRA-ORBITAL
BLOCKADES n
Pearson r
P value
II
- 0.0 1
I1.S.
II
0.11
fl.S.
8
0.66
0.08
All
30
0.36
0.048
Low responders
21 9
- 0.42
Cervicogenic headache Migraine without aura Tension-type headache
Responders
Patients who have got both greater blockades
responders
Il.S.
occipital
and supra-orbital
nerve
are included.
Low responders blockade;
< 0.000s
0.70
had less than 50%- pain reduction
responders
had at least 50%
were cervicogenic
headache
following
pain reduction. patients.
Eight
GON of 9
Entrapment of the GON has been thought to be one of the underlying causes to cervicogenic headache, and it was the basis for treating these patients with a neurolysis operation (Bovim et al. 1992b). In cases of entrapment, the effect of an injection in the GON area might be a double one. A volume effect per se (expanding effect, contributing to distension of the connective and muscular tissue surrounding the nerve) could be added to the anaesthetic effect. Presumably, the volume effect has been properly investigated by the saline injections carried out in the present study. Such injections seemed to be without effect. The observed effects in the present scrics are, accordingly, unlikely to be due to the volume per se. A placebo effect is also a factor to be taken into account in such contexts. The placebo effect should, however. probably be of the same magnitude in all 3 hcadachc categories and for all the different injections (saline injection, GON and SN blockades). An anaesthetic effect itself may increase the expectation of pain relief but, in such casts. would in all probability only concern pain in the anaesthetized area and not outside it (i.e., in the frontal part of the head following GON blockade). When evaluating pain intensity, spontaneous fluctuations must bc taken into consideration even when the headache seems to be stable at the time of injection. This is especially important for migraine, but also the other headache categories may have ‘attacks’ or at least exacerbations/ improvements. Correlations between the responses to the blockades at the 2 sites were calculated in order to expose possible unspecific ‘rcspondcrs’. No obvious correlation was found in patients who experienced a good effect (at least 50% pain relief) from the GON injection. There was, however, a rather strong positive correlation
among the ‘low responders’ (mainly migraine or tension-type headache patients). This observation seems to support our hypothesis that a good effect of GON blockade in cervicogenic headache is a specific (local) effect and not an unspecific (general) response to injection of anaesthetic agents. Reaction patterns to anaesthetic injections differed in several respects between the 3 headache categories studied. Cervicogenic headache seemed to be the most responsive headache, both regarding the number of patients responding and the degree of pain reduction in single patients. Migraine without aura patients gencrally responded least. Responses to the anaesthetic injections were not uniform within the 3 headache categories. In particular, there was a larger spreading of the extent of response in the cervicogenic headache group. Both GON and SN blockades gave complete pain freedom in some cervicogenic headache patients, while no pain relief was obtained in other cases. A probable explanation for this large variability is that the clinical picture of cervicogenic headache represents a ‘final common pathway’ with different etiologies. Similar clinical pictures may, for example, be caused by headache stemming from other cervical levels or the facet joint C2/C3 (Bogduk and Marsland 1986; Michler et al. 1991; Bovim et al. 1992a). The Cl root or nerve is difficult to study with injection techniques but may, according to the Kerr principle, also be a possible site of pathology in cervicogenic headache. Also the auriculotemporal (from the mandibular branch of the trigeminal nerve) and the lesser occipital nerves (from the ventral ramus of C2) are of interest for the pain generation in these headache cases. In our series, the best pain reduction was achieved with the GON blockade in the cetvicogenic headache group. A headache relief of more than 50% within 20 min following blockade was only reported in the cervicogenic headache group (except 1 migraine patient with a possible spontaneous recovery). Another interesting difference in the response between cervicogenic headache patients and patients in the other categories was the change in pain localization. According to the Kerr principle, forehead pain relief following GON blockade could be expected in cervicogenic headache and was indeed observed. Reduction in frontal pain after GON blockade was not observed in migraine and tension-type headache. This phenomenon suggests that impulses from the GON to the brain stem actually may provoke pain in the forehead in cervicogenic headache. A similar reaction pattern was observed following SN blockades in the cervicogenic headache group (i.e., pain reduction in the back of the head), although on a lesser scale. Hence, the difference between cervicogenic headache, on the one hand, and migraine and tension-type headache, on
the other
hand, as to pain relief outside the anaesthetized area seems clear. The results in the present series underline the usefulness of local anaesthetic blockade in the assessment of headache patients in the 3 categories studied. Moreover, the results do support the suggestion that pathogenetic mechanisms differ between cervicogenic headache, migraine without aura and tension-type headache. A blockade effect of a nerve does not in itself prove
that a pathological process is localized at, or distal to, the point of injection (Kiebler and Nathan 1960). The blockades performed in the present study do, therefore, not prove that cervicogenic headache actually is ‘cervicogenic’ (i.e., stemming from the neck). Nevertheless, the clinical criteria involve several factors that point to the cervical region as a probable source (mechanical precipitation of pain episodes by pressure in the neck region, but not in the frontal region; pain in shoulder and arm, etc.). This probability is, to some extent, supported by the present findings.
References Bogduk,
N. and Marsland.
headache, Bovim,
G..
J. Neural. Bonamico.
Nielsen.
A.
peripheral
and
A.,
On
Neurosurg.
the concept
L., Fredriksen, Sjaastad,
of third
occipital
Psychiat., 49 (1986) 775-780.
0..
T.A.,
Lindboe,
Topographic
course of the greater occipital
C.F.,
variations
nerve, Spine,
Stoltin the
16 (lY91)
475-478. Bovim, G.. Berg, R. and Dale.
L.G..
Cetvicogenic
nostic blockades of the cervical nerves C2-C5 C2/C3, Bovim,
Pain (1992a)
G..
headache.
in press.
Fredriksen.
T.A.,
Stolt-Nielsen.
A.
and Sjaastad,
Neurolysts of the greater occipital nerve in cervicogenic A follow up study, Headache, Headache
Classification
7 (198X) Hunter,
Criteria
Headache
for Headache
and Facial Pain. Cephalalgia,
Dis-
8, Suppl.
I -Y6.
C.R. and Mayfield.
the production
F.H..
of pain
Role of the upper cervical roots in
in the
head.
Am.
J. Surg.,
4X (1949)
I.
743-7s
Jansen, J., Markakis, attacks
E., Rama,
or permanent
root compression, Kerr.
of the International
and Diagnostic
Neuralgias
0..
headache.
32 f l992bJ 17S- 179.
Committee
Society, Classification orders, Cranial
Diag-
and the facet joint
F.W.L..
B. and lfildebrandt,
hemicrania
Cephalalgia,
Structural
relation
9
-
J.. Hemicranial
a sequel of upper
( 1089)
cervical
12% 130.
of the trigeminal
spinal
tract
to
upper cervical roots and the solitary nucleus in cat, Exp. Neural., 4 (19613 134-148. Kiebler.
R.F.
and Nathan
P.W.,
Relief
of pain and paresthesiae
nerve block distal to a lesion, J. Neural.
by
Neurosurg.
Psychiat., 23
O., Disorders
in the lower
(1YhOJ 91~08. Michlrr,
R.-P.,
cervical
Bovim, G. and Sjaastad.
spine: a cause of unilateral
liradache, Sjaastad,
31 (lY91)
O., The
headache. Sjaastad,
O.,
headache: Sjaastad.
0..
Cluster
Fredriksen,
of challenge and G. Nappi
Saunte,
of our ttmc: cervicogenic fEds.1, Cefalee
cervicoge-
press IY90, 1 I- 16. T.A.
diagnostic criteria,
Cervicogenic XX.
A case report.
S%~SSI.
headache
In: A. Ruju
netiche Milano:
headache’?
C., Hovdal,
and
Pfaffenrath,
Headache, H.. Brevik.
V.,
Cetvicogenic
30 (1990) 725726. H. and Gronbmk,
headache: an hypothesis, Cephalalgia,
E.,
3 (1983) 24Y-
43 Publishers
B.V. All rights
reserved
0304-3959/92/$05.00
PAIN 0211X
Cervicogenic headache, migraine without aura and tension-type headache. Diagnostic blockade of greater occipital and supra-orbital nerves Gunnar Bovim and Trond Sand Department
ofNeurology,
(Received
Trondheim University Hospitals, Regionsykehuset,
19 November
1991, revision
received
20 March
7006 Trondheim (Norway)
1992, accepted
1 May 1992)
The diagnostic value of greater occipital and supra-orbital nerve blockades in patients with Summary cervicogenic headache, migraine without aura, and tension-type headache was investigated. The pain reduction after greater occipital nerve blockade was significantly more marked in the cervicogenic headache group than in the other categories. Moreover, pain reduction in the forehead was generally only found in the cervicogenic headache patients (77%). Pain reduction (in %) was significantly more marked following the greater occipital than the supra-orbital nerve blockade. The volume effect per se was evaluated by saline injection. This procedure did not result in distinct pain reduction. The effect obtained in cervicogenic headache is, accordingly, probably due to the local anaesthesia. The present results support the postulate that different pathogenetic factors probably are responsible for cervicogenie headache, tension-type headache, and migraine without aura. Key words:
Cervicogenic Supra-orbital
headache; Tension-type headache; nerve; Nerve block (anaesthesia)
Introduction
Clinical differential diagnosis between cervicogenic headache, migraine without aura (‘common migraine’) and tension-type headache may be difficult on the basis of history and examination alone. The recommended treatment differs widely in the 3 types of headache. For this reason, it is important to improve the clinical distinction between them. The methods for, and importance of, diagnostic anaesthetic blockades have been outlined by different authors (Hunter and Mayfield 1949; Sjaastad et al. 1983; Jansen et al. 1989). According to the present pathogenetic model for cervicogenic headache, pain in the forehead may be caused by painful stimuli being transferred from the neck/posterior part of the head (mainly by the Cl-C3 nerves). In the brain stem, nerve fibres have a very
Correspondence to: Gunnar Bovim, sykehuset, 7006 Trondheim, Norway.
Dept.
of Neurology,
Region-
Migraine
without
aura;
Greater
occipital
nerve;
intimate relationship to the caudal part of the spinal trigeminal nucleus. Cortical interpretation of the impulses stemming from the posterior parts may, therefore, be a pain in the area of the ophthalmic division of the trigeminal nerve (in the following referred to as the ‘Kerr principle’) (Kerr 1961). On the basis of this model, it would be of interest to study the response to greater occipital nerve (GON) and supra-orbital nerve (SN) blockades in cervicogenic headache. The GON blockade has been included among the diagnostic criteria for cervicogenic headache (Sjaastad et al. 1990) as an ‘important’ criterion. If GON injections abate pain in the forehead in addition to the neck, i.e., outside the innervation area of the blocked nerve, this would be an indication that the Kerr principle is operative in this type of pain (Sjaastad 1990). Hence, the responses to GON and SN blockades in cervicogenic headache, migraine without aura (common migraine, in the following partly termed migraine) and tension-type headache have been investigated in the present study. Studies of this nature have, to the best of our knowledge, not been performed before.
44
TABLE
1
CHARACTERISTICS
OF THE
PATIENTS Cervicogenic
Migraine
Tension-type
headache
without aura
headache
No. of patients
24
14
I-1
Females/males
21/3
II/3
Age range (median)
IN-64
Unilaterality
without sideshift
24 *
Unilaterality
with sideshift
Bilateral
20-57
Symptoms of neck involvement
* *
0
II
0
0
3
I4
24
0
0
3
I2
21
2
3
I4
Attacks without pain in between almost continuous pain
7/7 2l-h9(4I)
(39)
0
0
pain
Fluctuating,
(435)
Pulsating pain
3 II 7 ***
Nausea, as a nearly constant accompaniment II
Phono- and/or
II
photophobia
* Five patients reported * * (Sjaastad
that headache,
et al. 1990) These
when maximal,
symptoms include
ipsilateral
upper posterior
(presence
of at least I of these factors).
* * * With maximal headache
Materials
13
I3
of severe headache Nausea, never or only sporadically
I2 I2
also could be felt on the contralateral
(1) either
neck region, (2) ipsilateral,
attack
provocation
non-radicular
neck/shoulder/arm
with
cervicogenic
graine, and 14 with tension-type (Table
headache,
I4 with
mi-
headache were included in the study
1). In some of the patients
only one of the blockades
was
carried out.
the
In the Classification stemming
and Diagnostic
Criteria
Headache
for Headache
Society (IHS),
from the neck was only summarily
genie headache
is not sufficiently
cation Committee
described:
validated”
(Headache
1988). The diagnosis of cervicogenic
the present study was, therefore,
Major
towards
into
IHS criteria sion-type
(Headache
headache
Classification
group,
11) of course without
consideration.
mainly
The
headache Committee
patients
with
diagnoses
DIAGNOSTIC
1988). In the tenthe chronic
form
were represented. avoid
contamination
of the
for more than
material
by patients
I of the headache
fulfilling
categories,
established
Disor-
(Table
I).
In this connection,
headache
‘positive’
“Cervico-
Unilaterality
Classifi-
headache
in
according to the recent
criteria,
without
was present definition. headache,
we paid particular
as far as cervicogenic sideshift
(i.e..
in all the cervicogenic All
the
whereas
headache
attention
only
CRITERIA
FOR
CERVICOGENIC
HEADACHE
(for the complete
cases in accordance
headache
in the migraine
group
II
patients patients
had
see Sjaastad et al. 1990)
of neck involvement
Pain precipitated lpsilateral
by mechanical
pressure to the ipsilateral
neck/shoulder/arm
Reduced
upper posterior
neck region or by awkward
head positioning
pain
range of motion in the cervical spine
Pain characteristics Non-clustering Moderate,
pain episodes of varying duration
usually non-throbbing
important
Anaesthetic
(or fluctuating,
continuous
pain)
pain, starting in the neck, spreading forward
criteria
blockades of the GON
and/or
the C2 nerve on the symptomatic
side abolish the pain transiently
sex
Head and/or
neck trauma
by history
Note: The effect of anaesthetic
blockades was, of course, not taken into consideration
when including patients
with
the
hilateral
had unilateral
headache
Symptoms/signs
(1)
pain always on the same side)
tension-type
criteria,
to the
is concerned.
symptoms and signs
Female
of
were based on the
II
Unilateral
Other
the
patients who could be given rather distinct diagnoses were included
ders, set forth by the International
MAIN
pressure
et al. 1990) (Table
effect
migraine without aura and tension-type
To
Diagnosis
(Sjaastad
blockade
diagnostic criteria
TABLE
or by mechanical
pain, and (3) reduced range of motion in the neck
diagnostic criteria taking
patients
side, but to a lesser degree.
by neck movement
only.
and methods
Twenty-four
3
in the present series.
45 pain (invariably with sideshift, i.e., the pain was alternating from left to right or vice versa between or within the single headache attack) and 3 patients had bilateral pain (2 of these with side predominance) (Table I). (2) Signs of neck involvement are as stipulated in Table 1. One or more of these phenomena were present in all the cervicogenie headache cases, but not in any of the other patients. There were clear differences between the 3 groups as for nausea, photoand phonophobia, and pulsating pain quality (Table Il.
Blockades To block the GON, the needle was placed approximately 2 cm lateral and 2 cm inferior to the external occipital protuberance. The needle was first forwarded onto the periosteum of the occipital bone and then withdrawn approximately 0.5 cm before injection. Paraesthesias were frequently felt in the GON innervation area when the needle was held in this position, but their presence was not obligatory to achieve a blockade. One-half to 1.5 ml of lidocaine (20 mg/ml with 12.5 pg/ml adrenaline) was injected and the ensuing sensory deficit (pinprick) carefully assessed. By using only a small volume, a selective blockade was obtained. The topographic variability in the peripheral course of the GON (Bovim et al. 1991) made this procedure insufficient in 4 patients (2 with cervicogenic headache and 2 with tension-type headache). In these patients, a new injection was carried out at a later stage, 2 cm further down in the neck, and was then successful. To evaluate the volume effect per se, 2-ml injections of isotonic saline were made with the same location as the GON blockades in 16 of the cetvicogenic headache patients. These injections were given prior to the anaesthetic blockade and always at the spot where CON was successfully blocked. The patients were not informed as to the nature of this injection prior to the test. The SN was easier to localize than the CON, since it rather regularly seemed to enter the forehead via the supra-orbital incisure. In 2 patients, however, the nerve seemed to divide in the orbit before entering the forehead, and additional injections (more laterally) were necessary to achieve an appropriate sensory loss. One-half to 1 ml of lidocaine (20 mg/ml with 12.5 pg/ml adrenaline) was used during each injection. Accidental anaesthesia of the smaller, more medially located supratrochlear nerve cannot be ruled out with the present technique. The patients reported the effects in 2 different ways: (1) by a visual analogue scale (VAS) as to the pain intensity (prior to and every 5th minute after injection, separate recordings on the right and left sides), and (2) by pointing out the pain localization on a pre-made scheme. The sensory loss was carefully evaluated and drawn on a scheme following anaesthetic injections. The patients were asked to estimate the headache intensity in relation to headache maximum on the VAS scale. All injections were made when the pain was at least 50% of the pain maximum, evaluated by VAS. Patients with obviously ‘unstable’ headache at the time of investigation (progressing or in remission) were not included. Generally, the patients were without drugs at the time of investigation.
Statistical analysis The pain versus time (after injection) profiles were analysed with repeated measures analysis of variance (ANOVA). The effects of CON and SN blockades in the various headache groups were separately compared (cervicogenic headache vs. migraine, cervicogenic headache vs. tension-type headache, and migraine vs. tension-type headache). The difference between the effect of GON blockade and SN blockade was analysed within each headache group. In these analyses, the ‘main-effect F value’ reflects the extent of parallel displacement between pain profiles, while the ‘interaction F value’ reflects the degree of non-parallelism in the profiles. In addition, secondary (descriptive) Student’s t tests (2-sided) were applied at
Greater occipital nerve
100 [
:ot
10 i
1
‘3:
L__.
0
Supraorbital nerve
_I
5
~-
~112
IO 15 20 25 30 Minutes
U -t
L
L_.-.~
L-
0
5
10 15 20 25 30 Minutes
Fig. I. Pain relief following injections. Pain intensity registration every 5th min following greater occipital (A) and supra-orbital nerve (B) blockades. No major changes were reported between 30 and 70 min. The bars indicate S.E.M.s. (Al Secondary Student’s f tests tcervicogenic headache vs. tension-type headache) were significant after CON blockade from the 5th minute onwards (P < 0.05) and highly significant from the 20th minute onwards (P < 0.001). (B) Student’s t tests (cervicogenic headache vs. tension-type headache) were insignificant after SN blockade at each measurement.
each time point. Differences in (individual) maximal pain reduction were analysed with the Mann-Whitney test. Two-by-two tables were analysed with a chi-square test including Yates’ correction. P values less than 0.05 were regarded as significant. Intra-individual correlations between SN and GON blockades have been calculated with Pearson’s r (correlation coefficient).
Results
Side effects were not reported after the injections, and no patient experienced a change of more than 10% in pain intensity on the contralateral side. The pain intensity prior to injection was similar in the 3 headache groups (Fig. 1). Saline injections
Saline injections in the GON area in patients with cervicogenic headache (n = 16) gave no relief of pain in 12 patients whereas 4 patients reported minor relief (less than 20% improvement on VAS) of short duration (less than 20 min). The pain persisted as opposed to the reduction observed after anaesthetic GON blockade (ANOVA: main-effect F = 28.7, P < 0.0005) (Fig. 1A). Localization of the headache was not altered by this procedure. Blockade of the GON
Patients with cervicogenic headache responded better than patients with tension-type headache and migraine (Fig. 1A). The highly significant difference which was observed between the 3 headache groups (Table III) was due to the difference between the cervicogenic headache and tension-type headache groups and to the difference between the cervicogenic headache and mi-
4h
TABLE THE
improvement after GON blockade was reported by 7 migraine and 5 tension-type headache patients, as opposed to only 2 cervicogenic headache patients. Generally, the main pain reduction took place within 5 min after injection (Fig. 1A). Only minor changes in pain intensity were reported between 20 and 30 min. Increased headache was only reported by 2 migraine patients, probably due to spontaneous progression of the attack. Discontinuation or improvement of forehead pain following GON blockade (i.e., pain outside the innervation area of the nerve) was reported by 17 of 22 cervicogenic headache patients (77%) and by only 2 other patients (Table IV).
111 RESULTS
VARIANCE BEEN
FROM
TABULATED
Patients
with
tension-type
Headache
TWO
(ANOVAs)
DIFFERENT
WITH
AS F STATISTIC
cervicogenic headache
ANALYSES
REPEATED
headache,
MEASURES
OF HAVE
VALUES migraine
without
aura.
and
have been compared.
CON
Supra-orbital
blockade
blockade
x.5 *
2.2
3.4 **
5.8 **
category
(main effects) F(df=21 Headache
category
vs. time interaction F (df=
IO)
VAS values for pain (in the injected side of the head) between 5 and Xl min after injection
df: degrees of freedom.
have been analysed.
* P < 0.005, * * P < 0.0005.
Blockade
Note: The main effect F value is a statistical reflection of parallel blockade)
displacement and
IB
between
(supra-orbital
of the degree
the 3 curves in Fig. blockade).
The
1A (CON
interaction
terms
reflect the degree of non-parallelity.
graine groups (post-hoc Student’s t test: P < 0.01 from the 15th minute onward). On the other hand, a lesser difference was found between the pain profiles in migraine versus tension-type headache (Table III, Fig. 1A). Maximal pain reduction (median values) was 54.5% in the cervicogenic headache, 14% in the tension-type headache and 6% in the migraine group (Fig. 2A). During the 30 min following injection, 17 of 22 cervicogenie headache patients reported maximal pain relief of more than 40%. An improvement of this order of magnitude was not reached by any tension-type headache patient and by only 1 migraine patient. Migraine in this patient improved both in the front and back, possibly due to a spontaneous remission. No
Greater
A
occlpltal
90
100 Fig. 2. Maximal CON
E
Termon-type
-
he&)he
Mlgralne without aura (n=ll)
Tenwn-type he;daahe
Median
(A) and SN (B) blockades.
The maximal,
----
Following
relative pain reduction
the CON
blockade
(A)
CON
blockade:
cervicogenic
(B) SN blockade: cervicogenic
within 30 min following the injections are illustrated
it should be noted that no patient,
(1 exception discussed in the text) achieved 50% or more reduction
P < 0.0005.
nerve
0000000
pain reduction.
the pain reductions.
Supraorbltal Cerwcogenlc
he(;$$e
ii
SN
B
nerve
Mlgralne without aura (n=i4)
of the
The SN blockade differentiated poorer between the headache categories than did the GON blockade (non-significant main-effect in Table III). Overall pain reduction in the migraine group was, however. less than the reduction in the other groups (Figs. 18 and 2B) corresponding to the significant interaction in Table III. Pain levels were significantly lower in cetvicogenic headache than in migraine from the 20th minute onwards (post-hoc Student’s t test, P < 0.09. The mean pain scores were also slightly lower in cervicogenic headache than in tension-type headache from 10 min onward. These differences did not reach statistical significance in the post-hoc Student’s t tests, however. Maximal pain reduction (medians) was 28% in the cervicogenic headache, 30% in the tension-type headache, and 16% in the migraine groups. Two patients with cervicogenic headache reported complete pain relief within 20 min, whereas the best response in migraine was 30% reduction and in tension-type headache 50% reduction.
headache
of the pain. The Mann-Whitney vs. tension-type
headache vs. tension-type
headache:
headache:
either with tension-type
median
for each patient after headache
test was applied to compare
P < 0.0005;
P = 0.23; cervicogenic
cervicogenic
headache
headache vs. migraine:
or migraine
the medians of vs. migraine: P < 0.02.
47
TABLE
IV
LOCATION
OF
PAIN
REDUCTION
RELATED
TO
AREA
OF
ANAESTHESIA A: blockade of the greater
Cervicogenic
Migraine
Tension-
occipital nerve
headache
without
f” = 22)
aura
type headache
fn = 14)
(n = 13)
Frontal
pain reduction
Occipital
pain reduction
19 -,f
x
B: blockade of the supra-
Cervicogcnic
Migraine
Tension-
orbital nerve
headache
without
tn = 13)
aura
type headache
(n=Il)
tn = 9)
Frontal
pain reduction
Occipital
6
I2
h
ther in the migraine nor in the tension-type headache group (ANOVA, Student’s t test). In the cervicogenic headache group, however, the GON blockade was significantly more effective than the SN blockade when relative pain reduction was considered (P < 0.003, ANOVA). Pain levels differed significantly at 5 min (P < 0.002) and 15 min (P < 0.03) (2-sided Student’s I test). Intra-individual correlations between the responses at the 2 injection sites were calculated (Table V). Among the 30 patients who received both injections, there seemed to be a slight positive correlation (Pearson’s r = 0.36, P < O.OS),particularly in the tension-type headache patients. When the patients were split according to response strength (less vs. greater than 50% pain reduction after GON blockade) the correlation was only significant for the ‘low responders’ (Table VI.
0
pain reduction
Discussion The “urnher of patients who reported
pain reduction
back) of the head following the blockade pain reduction Frontal
by the supra-orbital
area: area innervated
Chi-square
The mmmf
of
is not taken into consideration.
area: area innervated
Posterior
is reported.
in the front for
by the greater
nerve.
occipital
nerve.
test with Yates’ correction.
xxx: P < o.otlft3: xx: P < 0.02: x: P < 0.04.
The SN blockade did not relieve pain outside the innervation area in migraine or tension-type headache groups. Some relief of pain in the posterior part of the head was reported by 6 of 13 cervicogenic headache patients (Table IV). Comparison between injections around the GON and SN
Pain reduction after GON and SN blockades was compared. No significant differences were found, neiTABLE
V
INTRA-INDIVIDUAL SPONSES NERVE
TO
CORRELATION
GREATER
BETWEEN
OCCIPITAL
VS.
THE
RE-
SUPRA-ORBITAL
BLOCKADES n
Pearson r
P value
II
- 0.0 1
I1.S.
II
0.11
fl.S.
8
0.66
0.08
All
30
0.36
0.048
Low responders
21 9
- 0.42
Cervicogenic headache Migraine without aura Tension-type headache
Responders
Patients who have got both greater blockades
responders
Il.S.
occipital
and supra-orbital
nerve
are included.
Low responders blockade;
< 0.000s
0.70
had less than 50%- pain reduction
responders
had at least 50%
were cervicogenic
headache
following
pain reduction. patients.
Eight
GON of 9
Entrapment of the GON has been thought to be one of the underlying causes to cervicogenic headache, and it was the basis for treating these patients with a neurolysis operation (Bovim et al. 1992b). In cases of entrapment, the effect of an injection in the GON area might be a double one. A volume effect per se (expanding effect, contributing to distension of the connective and muscular tissue surrounding the nerve) could be added to the anaesthetic effect. Presumably, the volume effect has been properly investigated by the saline injections carried out in the present study. Such injections seemed to be without effect. The observed effects in the present scrics are, accordingly, unlikely to be due to the volume per se. A placebo effect is also a factor to be taken into account in such contexts. The placebo effect should, however. probably be of the same magnitude in all 3 hcadachc categories and for all the different injections (saline injection, GON and SN blockades). An anaesthetic effect itself may increase the expectation of pain relief but, in such casts. would in all probability only concern pain in the anaesthetized area and not outside it (i.e., in the frontal part of the head following GON blockade). When evaluating pain intensity, spontaneous fluctuations must bc taken into consideration even when the headache seems to be stable at the time of injection. This is especially important for migraine, but also the other headache categories may have ‘attacks’ or at least exacerbations/ improvements. Correlations between the responses to the blockades at the 2 sites were calculated in order to expose possible unspecific ‘rcspondcrs’. No obvious correlation was found in patients who experienced a good effect (at least 50% pain relief) from the GON injection. There was, however, a rather strong positive correlation
among the ‘low responders’ (mainly migraine or tension-type headache patients). This observation seems to support our hypothesis that a good effect of GON blockade in cervicogenic headache is a specific (local) effect and not an unspecific (general) response to injection of anaesthetic agents. Reaction patterns to anaesthetic injections differed in several respects between the 3 headache categories studied. Cervicogenic headache seemed to be the most responsive headache, both regarding the number of patients responding and the degree of pain reduction in single patients. Migraine without aura patients gencrally responded least. Responses to the anaesthetic injections were not uniform within the 3 headache categories. In particular, there was a larger spreading of the extent of response in the cervicogenic headache group. Both GON and SN blockades gave complete pain freedom in some cervicogenic headache patients, while no pain relief was obtained in other cases. A probable explanation for this large variability is that the clinical picture of cervicogenic headache represents a ‘final common pathway’ with different etiologies. Similar clinical pictures may, for example, be caused by headache stemming from other cervical levels or the facet joint C2/C3 (Bogduk and Marsland 1986; Michler et al. 1991; Bovim et al. 1992a). The Cl root or nerve is difficult to study with injection techniques but may, according to the Kerr principle, also be a possible site of pathology in cervicogenic headache. Also the auriculotemporal (from the mandibular branch of the trigeminal nerve) and the lesser occipital nerves (from the ventral ramus of C2) are of interest for the pain generation in these headache cases. In our series, the best pain reduction was achieved with the GON blockade in the cetvicogenic headache group. A headache relief of more than 50% within 20 min following blockade was only reported in the cervicogenic headache group (except 1 migraine patient with a possible spontaneous recovery). Another interesting difference in the response between cervicogenic headache patients and patients in the other categories was the change in pain localization. According to the Kerr principle, forehead pain relief following GON blockade could be expected in cervicogenic headache and was indeed observed. Reduction in frontal pain after GON blockade was not observed in migraine and tension-type headache. This phenomenon suggests that impulses from the GON to the brain stem actually may provoke pain in the forehead in cervicogenic headache. A similar reaction pattern was observed following SN blockades in the cervicogenic headache group (i.e., pain reduction in the back of the head), although on a lesser scale. Hence, the difference between cervicogenic headache, on the one hand, and migraine and tension-type headache, on
the other
hand, as to pain relief outside the anaesthetized area seems clear. The results in the present series underline the usefulness of local anaesthetic blockade in the assessment of headache patients in the 3 categories studied. Moreover, the results do support the suggestion that pathogenetic mechanisms differ between cervicogenic headache, migraine without aura and tension-type headache. A blockade effect of a nerve does not in itself prove
that a pathological process is localized at, or distal to, the point of injection (Kiebler and Nathan 1960). The blockades performed in the present study do, therefore, not prove that cervicogenic headache actually is ‘cervicogenic’ (i.e., stemming from the neck). Nevertheless, the clinical criteria involve several factors that point to the cervical region as a probable source (mechanical precipitation of pain episodes by pressure in the neck region, but not in the frontal region; pain in shoulder and arm, etc.). This probability is, to some extent, supported by the present findings.
References Bogduk,
N. and Marsland.
headache, Bovim,
G..
J. Neural. Bonamico.
Nielsen.
A.
peripheral
and
A.,
On
Neurosurg.
the concept
L., Fredriksen, Sjaastad,
of third
occipital
Psychiat., 49 (1986) 775-780.
0..
T.A.,
Lindboe,
Topographic
course of the greater occipital
C.F.,
variations
nerve, Spine,
Stoltin the
16 (lY91)
475-478. Bovim, G.. Berg, R. and Dale.
L.G..
Cetvicogenic
nostic blockades of the cervical nerves C2-C5 C2/C3, Bovim,
Pain (1992a)
G..
headache.
in press.
Fredriksen.
T.A.,
Stolt-Nielsen.
A.
and Sjaastad,
Neurolysts of the greater occipital nerve in cervicogenic A follow up study, Headache, Headache
Classification
7 (198X) Hunter,
Criteria
Headache
for Headache
and Facial Pain. Cephalalgia,
Dis-
8, Suppl.
I -Y6.
C.R. and Mayfield.
the production
F.H..
of pain
Role of the upper cervical roots in
in the
head.
Am.
J. Surg.,
4X (1949)
I.
743-7s
Jansen, J., Markakis, attacks
E., Rama,
or permanent
root compression, Kerr.
of the International
and Diagnostic
Neuralgias
0..
headache.
32 f l992bJ 17S- 179.
Committee
Society, Classification orders, Cranial
Diag-
and the facet joint
F.W.L..
B. and lfildebrandt,
hemicrania
Cephalalgia,
Structural
relation
9
-
J.. Hemicranial
a sequel of upper
( 1089)
cervical
12% 130.
of the trigeminal
spinal
tract
to
upper cervical roots and the solitary nucleus in cat, Exp. Neural., 4 (19613 134-148. Kiebler.
R.F.
and Nathan
P.W.,
Relief
of pain and paresthesiae
nerve block distal to a lesion, J. Neural.
by
Neurosurg.
Psychiat., 23
O., Disorders
in the lower
(1YhOJ 91~08. Michlrr,
R.-P.,
cervical
Bovim, G. and Sjaastad.
spine: a cause of unilateral
liradache, Sjaastad,
31 (lY91)
O., The
headache. Sjaastad,
O.,
headache: Sjaastad.
0..
Cluster
Fredriksen,
of challenge and G. Nappi
Saunte,
of our ttmc: cervicogenic fEds.1, Cefalee
cervicoge-
press IY90, 1 I- 16. T.A.
diagnostic criteria,
Cervicogenic XX.
A case report.
S%~SSI.
headache
In: A. Ruju
netiche Milano:
headache’?
C., Hovdal,
and
Pfaffenrath,
Headache, H.. Brevik.
V.,
Cetvicogenic
30 (1990) 725726. H. and Gronbmk,
headache: an hypothesis, Cephalalgia,
E.,
3 (1983) 24Y-
