Headache and the lower cervical spine: long-term, postoperative follow-up after decompressive neck surgery

Torbjørn A. Fredriksen* Andreas Stolt-Nielsen*,*** Karl Ove Skaanes* Ottar Sjaastad** Departments of Neurosurgery* and Neurology**, University of Trondheim Hospitals, ***Ski Hospital, Ski, Norway Reprint requests to: Dr T.A. Fredriksen, Department of Neurosurgery, St Olavs Hospital, N-7006 Trondheim, Norway E-mail: [email protected]

Accepted for publication: February 7, 2003

Summary A retrospective search for headache sufferers was conducted among patients operated on for cervicobrachialgia, and the operative results were evaluated. We also tried to classify the preoperative headache according to current headache classification systems. A total of 187 patients were operated on with SmithRobinson’s method, or by “simple” foraminectomy/ facetectomy. Headache was present in 23, and 17 of these could be followed up for an average post-operative observation period of 8.5 years. The operation seemed to result in headache improvement in 15 patients, i.e., excellent in 7; good in 2; moderate in 6; in other words: “excellent”/”good” in 53%. Improvement of neck pain was found in 15 patients (excellent in 5; good in 4; moderate in 6). The headache characteristics were generally compatible with the criteria for cervicogenic headache (CEH) (1990 version). CEH can probably stem from a pathology in the lower cervical spine. This should probably be taken into account when evaluating the individual patient with symptoms reminiscent of CEH. This headache may benefit from operations directed towards the cervical spine. KEY WORDS: Cervicobrachialgia, cervicogenic headache, facetec tomy, Smith-Robinson procedure.

Introduction Cervicogenic headache (CEH) is a headache caused by abnormalities located in the neck. Criteria for CEH have been set forth by The Cervicogenic Headache International Study Group (CHISG) (1). CEH as such has been accepted by the International Association for the Functional Neurology 2003; 18(1): 17-28

Study of Pain (IASP) and included in their Classification of Chronic Pain (2). The IHS (3) still uses the term cervical headache, and its criteria differ essentially from the IASP criteria. The possibility that headache can have its origin in the neck has largely been negated until recent decades. Now, however, this concept steadily seems to be gaining support. The different therapeutic approaches reflect the lack of precise insight into the pathogenesis. Therapy has been directed towards occipital nerves (4), zygapophyseal joints (5), discs (6), nerve roots (7), towards “venous lakes” around the roots (8), as well as towards the vertebral artery with its nerve supply (9). It has been a commonly held view that headache – should it stem from the neck at all – would stem from the upper cervical segments. However, available evidence suggests that not only the upper, but also the middle/lower part of the cervical spine down to C6-C7 (9), may be involved in headache pictures that – clinically – can hardly be distinguished from CEH. We have previously reported on patients in whom decompression of the spinal root at lower cervical level has alleviated headache, categorized as CEH (10,11). The present communication addresses this topic. The principal aim was to investigate whether, in a fairly large series, there may be further evidence to substantiate the idea that surgery directed towards the lower cervical roots and discs can alleviate associated headache.

Materials and methods The study is in principle retrospective. However, some of the patients (no.=3) have been followed in a prospective way for many years because of headache. One of these patients (no. 17 in the present communication) has been described in detail previously (10). The records of all patients undergoing neck surgery for cervicobrachialgia during the period 1990-1993 at the Department of Neurosurgery, Trondheim University Hospital, were scrutinized: the emphasis was on head-, neck-, and arm-pain, clinical findings, and supplementary diagnostic work-up. The records contained information concerning range of motion in the neck, estimated in a rough, clinical way. However, they generally contained little information concerning systematic testing programs, such as for the eliciting of pain similar to the spontaneous pain by the application of external pressure upon given anatomical structures. Nor had systematic anaesthetic blockades been included, as these were not a diagnostic requirement according to the original CEH criteria (12), in force when these patients were first observed. Patients who, according to the hospital records, complained of prominent headache were selected. In some

17

T.A. Fredriksen et al.

of these patients, the headache seemed to vary pari passu with the neck/arm pain. The series also includes 2 patients who, postoperatively, mentioned the disappearance of headache (a headache not mentioned in their admission notes). The available headache sufferers were interviewed by telephone between February and May 2000. In several cases, an extra telephone interview was carried out in October 2000. The headache questions posed concerned: duration, intensity, localization, headache pattern, and provocation mechanisms (prior to and after operation(s)). Cervicobrachialgia was invariably present. In all cases, there were additional, objective findings of pressure against the nerve roots/cord due to bony spurs or disc protrusions, demonstrated on MRI and/or cervical radiculography. The neck/arm symptoms and signs invariably constituted the indication for operation – not the headache. As regards the pre-intervention headache, the information from the telephone interview was compared with information from the patients’ records. When – in some cases – they did not concur, we did not systematically use one source or the other, but adopted a policy of reasonable personal judgement. We have nevertheless tried to give all the essential information in a condensed form in the tables. We then tried to synthesize information from the two sources and apply it, first and foremost according to the current CEH criteria (1), but also with differential diagnostic alternatives in mind. These patients are, in a way, their own controls: prior to surgery, their headache had gone on for years on end, in spite of all kinds of conservative treatment. Pre- and post-operative periods have, therefore, been compared. Five patients, i.e., no.s. 3, 13, 14, 16, and 17, formed a special category. They were subjected to a total of 12 operations, all directed towards the neck/occipital area, (in other words, as a group they underwent a total of 7 operations in addition to the final one, described herein). Patient 17 had been operated upon four times (see 10): C2 ganglion/root decompression and – twice – greater occipital nerve “liberation”, in addition to the final SmithRobinson (S-R) operation. The first operations did not give an optimal result; probably because the target was not well selected; for that reason, the later operations were carried out. In all of these patients, the final operation led to clear improvement (2: moderate; 1: good; 2: excellent effect on the headache). This experience, involving five of the patients, tends to show that an operation “at random” in this area is of little or no benefit. To be of avail, the target must be hit. This group of five patients is, in other words, subjected to sort of a control study, the patients serving as their own controls. The patients were operated upon with two principally different approaches (Table I). I: Anterior approach, according to Smith-Robinson (13,14): the disc was removed with an emphasis on removing the compression upon the cord and nerve roots. A fixation at the actual level was carried out by inserting a bone dowel. II: Posterior approach, removing parts of the ligamentum flavum and joint to liberate the root from the compression (foraminectomy + partial facetectomy). If necessary, an additional laminectomy was performed. A laminectomy was only carried out when there were roentgenological and/or clinical findings of myelopathy (2 subjects) (Table I). No structure anterior to the root

18

was removed during the foraminectomy/facetectomy procedure. Most procedures were carried out at the C5C7 levels (Table II). In some patients, a second operation was performed, due to insufficient results obtained with the first procedure. For statistical analysis, we employed, first and foremost, the Fisher exact test (two-tailed), but Student’s ttest was also used. TAF performed the telephone interviews and evaluated the operative results. He was not involved in the operative treatment. Table I - Interventions in the neck Procedures

No. of patients

Foraminectomy (facetectomy)

16

Foraminectomy combined with laminectomy

12

S-R*

15

Foraminectomy followed by S-R**

12

S-R followed by foraminectomy**

11

S-R later repeated at second level **

11

Total

17***

* S-R: Smith-Robinson procedure: disc evacuation and fixation. ** Two different operations. And for that reason, the total number of S-R operations differs in Tables I and II. Patients with S-R: n=9, Table I. S-R procedures: n=10 (Table II). *** A total of 21 operations, some of which comprised more than one level (cfr. Table II).

Table II - Cervical levels of operation No. of procedures Foraminectomy

S-R*

C3-C4

11**

10

C4-C5

12**

10

C5-C6

17**

16

C6-C7

16**

14

16**

10

Total

Procedure carried out at more than one level No. of patients Foraminectomy

S-R

2 levels

3

2***

3 levels

1

0**

* Smith-Robinson procedure: disc evacuation and fixation. ** This patient (no. 7) was primarily operated upon at the C4-C5 level (but at this operation laminectomy was also performed, and the C3-C4 level was explored). *** In one of the patients, the interventions were carried out during 2 different operations. Functional Neurology 2003; 18(1): 17-28

Lower cervical spine and headache

Results A total of 187 patients had been operated upon because of cervicobrachialgia with signs of compression of cervical structures on neuroimaging (Table III). Headache was present in 23 (12%) of these patients. Considerable underreporting, and – even more – underrecording/underestimation of symptoms could be of importance, given the fact that this symptom did not constitute an integral part of the indication for operation. One of the 23 patients with associated headache had died during the observation period, and five were lost to follow-up; the remaining 17 patients form the basis of the present study (7 females and 10 males). Clinically relevant characteristics are presented in Tables III and IV. The average preoperative headache duration was ca. 7 years (less than 3 years in 3 cases). All had neck/occipital area pain. However, in all except one, the maximum headache intensity was in the temporofrontal area. The mean age at operation was 52 years (Table III). The mean post-operative observation period was 8.5 years (range 6-10). In those undergoing 2 operations, the postoperative period was reckoned from the last operation. Unilaterality of head pain was present in 9 patients (Table IV, see over). A lesser degree of asymmetry was present in 2 cases. In one case, no. 13, headache became unilateral after the 1st operation (Table V, see p. 22). Bilateral headache (without obvious asymmetry) was present in 5 cases. The arm pain had a radiating quality in 9 patients (in the sense that it radiated to one or more fingers). In 8 cases, the arm pain was somewhat more diffuse. No patient presented a “shooting” pain in the arm. The headache was continuous with only relatively small fluctuations in 1 patient; continuous, but with marked exacerbations (attacks) in 14 patients; and characterized by typical attacks, with totally or mainly pain-free periods in-between in 2 patients. Headache was a prominent symptom in all patients (the most important in 3, as important as arm and/or neck pain in 10; arm pain and neck pain were each the most important symptom in 2 patients). Semi-specific tests, like diagnostic blockades, had only exceptionally been carried out in our cases – these blockades were not, in fact, a diagnostic requirement according to the original version of the CEH criteria (12). All the patients had limited range of motion in the neck and exhibited precipitation mechanisms, in that their headache was triggered or worsened by exercise or awkward neck movements (also during the night). Four patients had nausea/vomiting and 1 patient nausea only during the most severe headache attacks, and two of those exhibiting vomiting also had other “migrainous” features, such as photo- and phonophobia. Headache/cervicobrachialgia was in some cases antedated by neck traumata (Table IV). In six cases, there seemed to be a close temporal and a possible/probable causal relationship between the two (e.g., no. 17), but not in the other cases. The different combinations of head/arm pain are set forth in Table V. In 11 cases, there was unilaterality of headache and ipsilateral, upper extremity involvement. In another case (no. 13), there was a varying side preponderance of both headache and upper extremity inFunctional Neurology 2003; 18(1): 17-28

Table III - Patient series Total no. patients

187

Patients with headache

23

(12%)

Patients with headache, studied

17

(9%)

Headache duration (y.)

0.5->20

(mean: appr. 7)

41-66

(mean: 52)

Age at operation (y.) Females

7

Males

10

Postoperative observation period (y.)

6-10

(mean 8.5)

The target group (no.=17) represents 74% of the patients, who, in addition to the neck/arm pain suffered from headache.

volvement. A f t e r the first operation, only left-sided head- and arm-pain persisted. Table VI (p. 24-5) summarises the patients’ view, and our own grading of, the results of the operation(s). The assessments of the operative results are, in our view, on the conservative side. A summary of the operative results is presented in Table VII (see p. 26). In patients who underwent 2 operations (no.s 3,13,14, and 16), the result at follow-up has been attributed to the last operation. In these cases, the headache in the period prior to the first operation (taken to reflect the preoperative status) has been detailed. Both operative techniques improved the headache to some extent in most patients. “Excellent”/“good” results were obtained in 53% of the patients, and if we also include the “moderate” cases, this percentage rises to 88%. The results seem to be clearly better following the S-R, as opposed to the foraminectomy procedure: ratios excellent/good vs moderate/no improvement were: 6/2 and 3/6, respectively. With such a small sample size, these figures are not significantly different. If the sample size had been extended, and provided this trend had persisted, the results would have attained significance. The results with respect to neck and arm pain seemed rather similar following the two surgical techniques. Surprisingly, the results seemed somewhat better for the headache than for the neck/arm pain that constituted the indication for operation. The results of the operations are, in our view, quite satisfactory. This is also reflected in the fact that all the patients except one (patient no. 8) would have repeated the procedure(s), had they had the choice over again.

Case histories Patient 11. A female, born 1937, was examined at 54, because of left-sided headache, neck and arm pain. She had had headache for approximately 10 years. The pain was located in the occipital/neck region and spread to the eye. Vomiting accompanied two-day-long attacks. The initial diagnosis was migraine, but the attacks were uninfluenced by ergotamine and sidelocked, and she was well over 40 at onset. However,

19

T.A. Fredriksen et al.

Table IV - Clinical characteristics Patient Age Preoperative Sex in year duration of 2000 headache (years)

Postop. obs. period (years)

Laterality headache

Laterality neck/arm pain

Localization Temporal headache pattern

Operative Trauma, procedure. head/neck Year of operation. Level(s) operated

Diagnostic work-up

1.

64

4

M

8

R

R

N-O-T

Continuous fluctuating

Fo 92 C6-C7 R

0

MR: Degen. C6C7; less C5-C6. R>>Lmyelo: C5C6 as marked as C6-C7*

2.

67

20

M

9

B

B

N-O-Fr

Continuous fluctuating

S-R 91 C5-C6

0

Myelo: Degen. C5-C6, short C7 root L

3.

54

9

M

8

L

L>>>R

N-O-Fr

Continuous with exacerbations (attacks)

Fo 91 C6-C7 L S-R 92 C5-C6 C6-C7

Fall on MR: Degen. C5neck, C6, C6-C7 L concussion

4.

66

4

M

8

L

L

N-O-Fr

Attacks

Fo 92 C6-C7 L

0

MR: Degen. C6C7. Reduced foramen. C6, C7 roots L, both sides C8 root

5.

51

4

M

10

L>R

L>R

N-O

Continuous fluctuating

Fo 90 C5-C6 L

Whiplash

MR: Degen. C5C6. Myelo: short C6 root

6.

62

2

M

7

R

R

N-O-Fr

Continuous fluctuating

S-R 93 C6-C7

Concussion MR: Small herniation C4-C5. Degen. C6-C7. Moderate changes C5C6, C7-TH1

7.

65

5

M

9

B

L>R

N-O-Fr

Continuous Fo+La 91 Car accident MR: Herniation with C3-C4 C4-C5. Impresexacerbations C4-C5 bilat. sion cord

18.

61

3

M

10

L

L

N-O-Fr

Continuous fluctuating

Fo 90 C5-C6 L

19.

75

3

F

9

R

R

N-O-Fr R eye cheek

Continuous

Fo+La 91 0 C5-C6 bilat.

MR: Degen. Impression cord C5C6, less C6-C7

10.

53

1

/2

F

9

B

R

N-O-Fr

Continuous fluctuating

S-R 91 C5-C7

Car accident childhood

Myelo: Degen. Protrusion C6-C7 R>L MR also degen. C5-C6

11.

63

10

F

9

L

L

N-O-eye

Continuous Fo 91 with C5-C6 L exacerbations

Concussion

MR: Degen. Pro trusion C5-C6 and C6-C7 L

12.

57

5

F

9

R>L? (B??)

R>L? (B??)

N-O-Fr

Continuous fluctuating

Car accident MR: Herniation C5-C6. Degen. also C4-C5, C6-C7

S-R 91 C5-C6

Accident**

Myelo, shortening C6 roots L>R

continued

20

Functional Neurology 2003; 18(1): 17-28

Lower cervical spine and headache

Table IV continued Patient Age Preoperative Sex in year duration of 2000 headache (years)

Postop. obs. period (years)

Laterality headache

Laterality neck/arm pain

Localization Temporal headache pattern

Operative Trauma, procedure. head/neck Year of operation. Level(s) operated

13.

60

>2

F

6

R or L, but R>L. After 1st operat: L

R>L After 1st operat.: L

N-O-Fr

Attacks

S-R 92 C6-C7 Fo 94 C5-C6 L

Car accident MR and myelo: Degen. C6-C7. Lesser degree also C5-C6 94 MR: Degen. C5-6 and C6-C7 L>R

14.

66

6

M

8

L

L>>>R

N-O-Fr

Continuous fluctuating

Fo 91 C5-C6 L C6-C7 L S-R 92 C5-C6

Skiing accident

15.

52

15-20

M

9

B

L>R

N-O-Fr

Continuous fluctuating (attacks)

Fo 91 Fall on the C4-C5 back C5-C6 bilat C6-C7

MR: Degen. with central impression C3-C4. Reduced root channel C4C7 roots.

S-R 90 C5-C6 S-R 93 C6-C7

Myelo: (90). Herniation C5-C6, MR (92): Protrusion C3-C4, C4-C5, C6-C7. Myelo: Short C7 root L

}

16.

58

10

F

7

B

B

N-O-Fr

Continuous fluctuating

17.

52

10

F

10

R

R

N-O-Fr Cheek

Continuous S-R 90 with C5-C6 exacerbations

0

Diagnostic work-up

MR: Degen. C5C6, C6-C7, mostly C5-C6. Changes also C3-C4

Car accident MR: Degen. C5Airplane C6, L>R trauma***

Abbreviations: Myelo: myelography; Degen.: degenerative changes; M: Male; F: Female; R: Right; L: Left; B: bilateral; N: Neck; O: Occipital; Fr: Frontal; T: Temporal; Fo: Foraminectomy; La: Laminectomy; S-R: Smith-Robinson procedure: disc evacuation and fixation. * Small syrinx, approximately 2 cm long, C6-C7, without widening of the cord (MR) and no dissociated sensibility loss. ** Probably direct trauma against symptomatic side brachial plexus area. Complaints in direct temporal association with this. *** Sudden, major loss of height during flight. For case 12: see footnote, Table V.

her mother had had migraine. In the past 5 years the patient had also developed an ipsilateral shoulder/arm pain, radiating to fingers 1-3. Compression of the left C6 root and minor compression of the C7 root were diagnosed on MR. Facetectomy/foraminectomy was carried out in September 1991 at the level C5-C6, left side. An organized, herniated disc with encroachment upon the root frontally was not removed during the operation. The headache disappeared after the operation, only a slight shoulder pain remaining. This case shows “excellent” results as regards the headache, in spite of the fact that it – originally – was diagnosed as “migraine” (See Tables IV and VI). Functional Neurology 2003; 18(1): 17-28

Patient 13. A female, born 1940, consulted us at 52, because she had had head/neck/arm pain for more than 2 years. The headache was mainly right-sided and spread from the back of the head to the forehead; it could sometimes be most prominent on the other side. The pain radiated to right-side fingers 3-5; and left fingers 1-3. The pain could be provoked mechanically. She complained of reduced strength in both arms. Examination showed slightly reduced strength for flexion/extension at the elbow, right more than left. Sensation was reduced corresponding to the biceps and deltoid areas and also on the radial side of the forearm – right more than left. MR demonstrated bilateral C7

21

T.A. Fredriksen et al.

Table V - Various combinations of uni-/bilateral headache/ arm pain

Discussion Methodological shortcomings

Category

I.

II.

III.

IV.

V.

VI.

Combination

No. of patients

Unilateral head pain - ipsilateral arm pain

17

Unilateral head pain - ipsilateral arm pain (marked asymmetry of arm pain)

12

Somewhat less asymmetrical head pain - arm pain*

12

Unilateral head pain (change of side) bilateral arm pain**

11

Bilateral head pain - unilateral arm pain ***

13

Bilateral head pain - bilateral arm pain

12

Total

17

* Patients 5 and 12. As regards patient 12, there is some doubt over the correctness of laterality of the head pain. The head pain has been labeled as mainly unilateral; it may have been bilateral (in which case an error of 6% has been introduced). ** Patient 13, clear ipsilaterality of head and arm pain after 1st operation. *** In 2 of the 3 patients (7, 15), there was also bilaterality for arm pain, but with side preponderance. Category I. Cases 1, 4, 6, 8, 9, 11, 17; Category II. Cases 3, 14; Category III. Cases 5, 12; Category IV. Case 13; Category V. Cases 7, 10, 15; Category VI. Cases 2, 16

root affection that was confirmed by cervical myelography. There was also slight affection of the C6 root bilaterally. A S-R procedure was carried out at the C6-C7 level in December 1992, initially with good results. From spring 1993, progressive, strictly left-sided symptoms reappeared, with headache and pain radiating to the 1st finger. A repeat MR, April 1994, still showed a possible affection of the C7 root, left side, but also encroachment on the C6 root, and this was also most prominent on the left side. Clinically, reduced strength of the biceps muscle was found, as well as reduced C6 dermatome sensation, left side. A left-sided foraminectomy C5-C6 was performed, June 1994. After this, the headache was less strong and less frequent, and there was also improvement of neck and arm pain. Still, a residual, bilateral “heaviness” in the arms persisted. Analgesics were discontinued. This case tends to demonstrate that operations at two levels may be necessary. And if unilateral pain persists, the second invasive approach should be on that side (see Tables IV and VI). Functional Neurology 2003; 18(1): 17-28

The inherent weakness of the study (i.e., the mainly retrospective nature of it) makes it necessary to draw conclusions with a certain degree of reservation. Memory for pain is known to be far from perfect. This weakness is actually accentuated by an otherwise particularly strong aspect of the study, namely the long observation period. Telephone interviews may in many contexts be far from ideal. However, in the present situation they were probably adequate, given that one of our main ob jectives was to investigate the patients’ subjective feel ing of pain. It is unlikely that intramural questioning would have yielded a much more reliable response in this regard. The following should, nevertheless, be pointed out: In three of the patients (no.s 13, 16, and 17) there was only minor/partial relief after the first operation. After the second operation, the results were better (“moderate”, “excellent”, and “excellent”). The original operative results in this mini-group tend to show that operations not directed to the correct focus have little chance of success. Operations in this area must be directed towards the source of the evil to be successful. Can head pain be generated in the lower part of the neck? According to Kerr (15-18), cervical roots, via interneurons, connect with the nucleus of the spinal tract of the trigeminal nerve in the upper part of the cervical cord. Transmission of nuchal nociceptive stimuli via a given cervical root may in this way create the illusion that the pain stems from the anterior part of the head. The theo retical basis for why pathological processes at levels as far down as C7 putatively may give rise to frontal headache has not been worked out. Even if, as may possibly be the case, the sensory nucleus does extend caudally below the second cervical segment, this might not furnish a satisfactory basis for understanding head pain arising in these low, cervical areas. As demonstrated previously (9-11), and also herein, head pain stemming from the lower cervical segments seems to be a reality: headache history of long duration and with features characteristic of CEH, with maximum head pain in the temporo-frontal area (all of our subjects except one). The headache was clearly modified and even – in a considerable number of cases – abated after cervical operative procedures. And the follow-up time was substantial. Do these results have any bearing upon underlying pathology? With the foraminectomy procedure, principally only the compression of the root is addressed. With the S-R procedure, the disc is removed and the segment stabilized. Thus, the latter procedure may influence nociceptive signals generated in the circumference of the disc. It may also correct an instability in the segment as such, thereby influencing indirectly the facet joint. The fact that S-R seems to give the best results regarding headache could indicate that factors anterior to the root

22

Lower cervical spine and headache

or stabilization factors are important in many of our patients. “Excellent” result as regards headache may also follow foraminectomy (e.g., case 11). This could indicate that abnormalities around the root might be of importance in the occasional case. However, part of the facet joint is also removed during this procedure, and that may also have been important. The categorization dilemma and the current criteria The extent of the congruity of the symptomatology in our patients with the diagnostic scheme of cervical headache (IHS (3)) and the current CHISG criteria for CEH (1) is detailed in Table VIII (p. 26). The current CEH criteria (1) were not developed with CEH associated with upper extremity radiculopathy in mind. As for point I (1) (see Table VIII) (where only I a needs to be positive), both subjective precipitation of attacks (a1) and reduced range of motion in the neck (I b), neck/shoulder/ arm pains (I c) were invariably present (the latter being indications for operation) (Tables IV and V). Thus, this combination of symptoms and signs points strongly towards CEH, even in cervicobrachialgia cases. In addition, there is a unilaterality of the head pain in the majority of the cases. The first-version CHISG criteria (12) are fulfilled by the present cases, if one applies the rule of “unilaterality on two sides”. Like the CHISG criteria, the IHS criteria were not constructed with a view to headache combined with radicu lar arm pain. This special situation represents a challenge to any diagnostic scheme. The IHS criteria (D3, Table VIII) state that spondylosis and osteochondrosis are not criteria for headache stemming from the neck in general. Such roentgenological changes are common phenomena that appear with age, in patients and controls alike. Such signs can hardly generally be allotted a major, stereotyped pathogenetic role in this type of headache. Such X-ray signs were, nevertheless, claimed to be typical of what was termed “migraine cervicale” by Bärtschi-Rochaix (19) (such as the “signe des casserolles entassées”). We found only slightly higher incidence of such degenerative changes in a CEH series, compared to healthy controls (20). The situation may differ essentially when a localized, C5-C7 “spondylosis/osteochondrosis” is associated with corresponding radicular arm symptoms. The suspicion of a causal relationship between them may be accentuated if refined neuroimaging backs up the findings on plain X-ray. The term in D3: “- or other distinct pathology” (3) is open to interpretation. The rather clear and localized pathological cervical spine findings on neuroimaging in our cases – findings that actually substantiate the clinical picture – might fit with this proviso if interpreted in not too rigid a way. The findings in our cases may almost – but not quite – fulfil the IHS criteria. And even according to the IHS criteria, other specified headaches are unlikely to have this combination of symptoms and signs. The laterality problem Some degree of unilateral headache was present in 11 cases. And there was ipsilaterality of the arm pain. In other words, in 11 of 17 patients (65%, or 71% if case 13 is included, Table V), there was unilaterality of Functional Neurology 2003; 18(1): 17-28

headache and ipsilaterality of arm pain, with side preponderance varying somewhat. The laterality was more complex in the remaining patients. Bilateral head pain was combined with either uni- or bilateral arm pain. An overt discrepancy between laterality of head and arm pain is a possibility that, not only theoretically, should be reckoned with: the headache could in fact have been on one side only and the arm pain exclusively on the opposite side. There were actually no cases with this combination in our series. The combinations observed in our series do not, therefore, in our opinion, necessarily overtly violate the rule of ipsilateral arm pain (Table V). To us, CEH is – until proven otherwise – a secondary headache, with the primary abnormality in the cervical area (21,22). The abnormality enabling the generation of headache will frequently be on one side – or be most marked on one side – and the headache can then be the typical, recognizable CEH (1) form. But the situation can become more complex. To typify the situation (manifestation of pain), one may think in terms of degenerated disc(s). Spurs may initially be present mainly on one side of a degenerated disc; later, the other side of the disc and its circumference may be coinvolved. One may then get a “unilaterality on two sides”, both as far as headache and – if in the mid/lower cervical spine – arm pains are concerned. The chance of such a coinvolvement is probably as great as or even greater than the chance that another disc might be affected in the same way. Still more complex situations may arise: Disc degeneration/spurs on one side and at one level may give rise to unilateral headache + arm pain. Disc degeneration/ spurs on the other side may give rise to only one of these symptoms, either arm or head pain. One may then get the following constellations: bilateral arm pain – unilateral head pain, or: unilateral arm pain – bilateral head pain. To further complicate the situations, discs at two (or more) levels may be implicated. The one disc may cause headache on one side, the other disc causing arm pain on the other side. Understandably, if this were the situation, and the disc causing arm pain were treated surgically, head pain might persist, because the disc causing it has been left untreated. The latter case would be an exception to the rule of ipsilaterality of headache/arm pain, but it would be an understandable exception. Such a situation would represent a major challenge to the concept of CEH. Solving the problem would take the following: one intervention ipsilateral to the headache should more or less take away the headache. Another intervention on the opposite side (and at another level) should more or less remove the arm pain! The results of the present study should also be viewed in this light: the aim of the surgery was to combat the neck/arm pain, not to abate the headache. The singling out of the correct level for arm/neck pain intervention may not have led to the selection of the correct level as regards headache. Assessment of the individual patient in order to single out the pain-generating disc is not easy, as recently shown by Schellhas et al. (6). Differential diagnostic considerations The natural diagnostic alternatives in these cases are: migraine/tensiontype headache (T-TH). These alterna-

23

T.A. Fredriksen et al.

Table VI - Description of pain and operative results, based on record and patients’ own evaluation (HA = headache) Patient Hospital record

Patients’ own description of preoperative situation. Telephone interview (T)

Result (patients’ evaluation)

Authors’ evaluation

11

Head/neck/ shoulder pain R. Moderate head/neck Last 2 y, reduced control R arm; pain R. Stronger pain R numbness and periodically radi- arm. Loss of strength. ating pain fingers 4-5 with reduced sensation. EMG: C6 root.

12

Increasing, constant neck pain, radiating occipito-frontally bilat. Recent years, diffuse pain, numbness R arm.

“Belt around the head”. Pain neck/arm, bilat. Normal daily activity on painkillers.

Head, neck/arm pain: still pre- Head: moderate sent, but to a lesser degree. Neck: moderate First 6 weeks postoperatively Arm: moderate almost no HA.

13

HA L from neck to frontotemporal area (maximum); recent years, worsening with pressing HA L>R. Attacks 3 days duration, phono- and photophobia, vomiting. Interval pain at 10%. Radiation fingers 3-4 L; reduced sensibility.

Severe head/neck pain, nearly continuous: less arm pain, L>R. Vomiting. Numbness of arm, radiation to fingers 1-2.

Foraminectomy (I) followed by Head: good S-R (II). I: Pain unchanged.II: Neck: good Much improved head/neck. Un- Arm: unchanged changed arm; possibly, even slight deterioration over the years.

14

Neck pain radiating to: L occipi- Strong pain head/neck/ tal region/arm and fingers 3-5. arm L. Inability to work. Noturnal pain + pain related to neck movement.

15

Pain in neck + L shoulder/arm. Arm pain almost continuous, neck pain intermittent (patient’s impression: the 2 pain regions are unrelated). Recently, even R pain. HA L; only mentioned postoperatively.

16

Severe neck pain (main com- Strong neck pain, radiat- Head/neck: pain-free. Arm: im- Head: excellent plaint), radiating to R occipital ing to temporal area, R. provement, but still pain; re- Neck. excellent area/shoulder/arm (ulnar side). Less arm pain, R. duced strength. Arm: moderate

17

Pain and reduced movement, Severe frontal HA. Pain, Head/neck/arm: pain-free. Head: moderate neck, radiating to frontal region neck/hands: L>R. Painfree 4 y. Last 5 y slowly Neck: moderate (main problem: “lid over the growing pain, especially last 1 Arm: unchanged 1 head”). Vomiting. Radiation to /2 y, and especially arms. fingers 4-5. Myelopathia.

18

Head/shoulder/arm pain radiat- Strong L head, neck, HA: possibly less intense, peri- Head: unchanged ing to all fingers (mostly 4-5). and arm pain. odically. Neck. unchanged Reduced strength, atrophy. Neck: no change. Arm: unchanged Arm: no change.

19

Neck/head/arm pain, finger 1, Continuous strong pain R. Reduced strength R arm. head/face, R; disturbing Neck pain, radiating to maxillar sleep. area. Strong pain during night, disturbing sleep. Reduced sensibility fingers 1-3.

10

Constant HA, bilateral, occipital, vertex, but also frontoocular area. Neck/arm pain, R, fingers 2-3. Nausea.

Severe occipital HA. Neck pain, radiating shoulder/arm, bilaterally L>R.

HA: slight improvement? Neck: Head: Unchanged unchanged. Neck: Unchanged Arm: periodically, slight dis- Arm: Excellent comfort; no real pain.

No pain head/neck/arm and fin- Head: excellent gers. Operation: arthrosis L Neck: excellent shoulder. Pain free, even shoul- Arm: excellent der. Improvement head/neck. Even Head: moderate arm better; still shoulder pain. Neck: moderate Operated R shoulder: improve- Arm: unchanged ment. Last 5 y, slowly worsening pain same region, but R.

Head/neck: pain-free. Still “stiff- Head: excellent ness” neck. Arm: much im- Neck: good proved, but still joint pain (vari- Arm: moderate ous joints).

Pain neck/shoulder/arm, Pain-free. fingers 2-3, R. Did not, initially, remember headache.

Head: excellent Neck: excellent Arm: excellent continued

24

Functional Neurology 2003; 18(1): 17-28

Lower cervical spine and headache

Table VI continued Patient Hospital record

Patients’ own description of preoperative situation. Telephone interview (T)

Result (patients’ evaluation)

Authors’ evaluation

11

Neck/head pain: occipital region, to eye L. Pain L arm to fingers 1 and - lesser degree 2-3.

Severe constant pain Pain-free. head/eye L. Attacks with (Slight shoulder pain) vomiting (ca. 2 days). Less disturbing neck/arm pain L.

12

Pain neck/both arms and fingers 1-2. R>L. “Chronic HA”, only mentioned postoperatively. Reduced sensation, fingers 1-2.

Severe (global?) head/ neck pain. Vomiting, worst exacerbations; phono- photophobia. Sat up at night to obtain optimal neck position. Radiating pain, fingers 1-2; “desperate pain”.

Head: clear improvement (no Head: good vomiting). Neck: moderate Neck: improvement. Arm: moderate Arm: slight pain.

13

Neck pain, radiating both arms, R: fingers 3-5. L: fingers 1-3 (R>L). Head pain R, from occipital to frontal area, but varying side preponderance. Worse with movement/exercise. Head/ arm pain, L only, after 1st operation.

Occipito-temporal h e a d a c h e , both sides and transitory facial pain. Constant pain in neck/arms. Reduced strength, arms.

Operation I: 1992 (S-R) and II: Head: moderate 1994 (foraminectomy). Im- Neck: moderate provement I; still more im- Arm: moderate provement II. Less intense HA. Reduced neck/arm pain.

14

Nearly continuous fronto-temporal HA. L>R. R only at pain maximum. Pain, neck/L arm, diffuse.

Severe pain, most L, neck/arm: radiation, finger 1. HA continuous, but superimposed attacks.

Improvement after operation I: Head: moderate (foraminectomy). After 4-5 Neck: moderate months, back to status before Arm: moderate operation. Much improvement after S-R: same pain pattern, but to a lesser degree.

15

Neck pain/HA, occipital to frontal area. Pain both arms (1/2 y), radiating to fingers 4-5. Reduced strength.

Severe HA entire head Initially pain- free; then slow re- Head: moderate → frontal area. Severe currence. HA: reduced in inten- Neck: good neck pain radiating to sity. Neck/arm: much less pain. Arm: good arms, L>R.

16

HA/neck pain. Pain both arms, Severe bilateral, continuous HA, occipital to and fingers 2-3. frontal. Neck pain, radiating to fingers 2-3.

17

Head/neck/arm pain R, fingers 1-4. HA most prominent complaint; pain radiation: vertex, fronto-temporal area.

Functional Neurology 2003; 18(1): 17-28

Operations I/II (S-R) 1990/ Head: excellent 1993. Initial, good results I. Af- Neck: good ter 1 y, HA as before I. II: HA Arm: good disappeared. Much less pain neck/arm.

Severe HA, continuous, Pain-free. but with attacks. Occasional facial pain. Neck/ arm pain less prominent.

tives will, therefore, be scrutinized. For this purpose, we have summarized (Table IX) the number of solitary migraine/T-TH features (as specified by the IHS) in our patients. As regards the migraine alternative, points B through D are the most pertinent ones. Only 2 of our patients (12%) had clear attacks with pain freedom in-between

Head: excellent Neck: excellent Arm: excellent

Head: excellent Neck: excellent Arm: excellent

(point B). The remainder had permanent pain associated with exacerbations, a pattern differing essentially from the migraine attack pattern. Point C of the IHS migraine criteria will naturally be fulfilled by many CEH patients, since unilaterality (point C1) and a moderate to severe intensity of pain (point C3) pertain to both migraine and CEH. Clearly, therefore, the fact that these

25

T.A. Fredriksen et al.

two criteria are present in our cases (71% of the cases) does not indicate – and certainly does not prove – that our patients are migraineurs. It tends, rather, to show that the current migraine criteria are insufficient as a separating tool in this respect. If the parameter “unilaterality” were qualified, using the subgroups “with” (being typical for migraine) and “without” side alternation, one could arrive at quite different conclusions as regards the fulfillment of migraine criteria in this series. In a migraine without aura series (no.=32), only 16% exhibited unilaterality without sideshift (23). The unilaterality without sideshift figure in the present series (65%, Table VIII) is, of course, widely different from 16%. Only 5 patients (29%) fulfilled point D (Table IX). The nausea/vomiting combination (5 out of 17) is significantly more rare in our subjects than in Ekbom’s (24) series of migraine patients, i.e., 28 out of 40 (p=0.003). Also the figures for phonophobia (2 out of 17) differ significantly from those of Ekbom, i.e., 33 out of 40 patients (p=0.0000007). Only 12% of our cases satisfied the IHS criteria for migraine (point B is a must). Age at onset differs: in Ekbom’s series (mostly migraine without aura), the mean age at onset was 15.6 years ±1.2 (S.E.M.) and the range 3-35 years. In our series, the mean age at onset was 45.8 years (range 28-63). These figures are statistically vastly different (p=6x10–17, Student’s t-test). Accordingly, the sympto-

Table VII - Operative results: headache, neck, and arm pain S-R

Foraminectomy

Total

4 2 2 0

3 0 4 2

7 2 6 2

3 2 3 0

2 2 3 2

5 4 6 2

2 1 4 1

3 1 2 3

5 2 6 4

HEADACHE Excellent Good Moderate No improvement NECK PAIN Excellent Good Moderate No improvement ARM PAIN Excellent Good Moderate No improvement

Excellent = Pain free; Good = Marked improvement. Pain only a mi nor problem; Moderate = Less improvement; considerable residual pain; No improvement = The same problem as before operation, even though a minor improvement may have been achieved.

Table VIII - Headache characteristics in the present series, as fitting into the demands of 2 different diagnostic schemes for headache stemming from the neck. IHS criteria: cervical headache (3)

CHISG criteria: cervicogenic headache (1)

A Pain localized to neck and occipital region. May project to forehead, orbital region, temples, vertex or ears

I no.=17

B Pain is precipitated or aggravated by special neck movements or sustained neck posture

no.=17

C At least one of the following: 1) Resistance to or limitation of passive neck movements 2) Changes in neck muscle contour, texture, tone or response to active and passive stretching and contraction 3) Abnormal tenderness of neck muscles D Radiological examination reveals at least one of the following: 1) Movement abnormalities in flexion/extension 2) Abnormal posture 3) Fractures, congenital abnormalities, bone tumours, rheumatoid arthritis or other distinct pathology (not spondylosis or osteochondrosis)

Neck involvement a) precipitation of attacks 1) subjectively 2) iatrogenically b) reduced range of motion neck c) ipsilateral shoulder/arm pain

II Anesthetic blockade effect no.=17

N.S.T. N.S.T.

III A Unilaterality of head pain without sideshift B “Unilaterality on two sides”, see text

no.=17** N.S.T. no.=17 no.=17*** N.S.T.

no.=11 (65%) no.=6

Point I a suffices as the sole criterion. Otherwise, at least two of the points under I should be present

N.S.T.* N.S.T.*

no.=0*

Abbreviations and symbols: N.S.T.: Not systematically tested; * Plain X-rays: not available; Localized, degenerative changes were, however, invariably demonstrable on neuroimaging; ** Provocation of pain by sustained, awkward head positioning and/or physical activities, like hoovering, mowing the lawn, washing, etc.; *** In some patients; “unilaterality on two sides”, see Discussion; no. = number of patients.

26

Functional Neurology 2003; 18(1): 17-28

Lower cervical spine and headache

Table IX - Headache characteristics in the present series as satisfying the IHS criteria for migraine and tension-type headache Migraine without aura

Chronic tension-type HA

A. At last 5 attacks fulfilling B-D

A. Average HA frequency ≥15 days/months/180 days/year) for ≥6 months fulfilling criteria B-D listed below

B. Headache attack lasting 4-72 hours

no.=2

C. HA has at least two of the following characteristics: 1. Unilateral location no.=12* 2. Pulsating quality – 3. Moderate or severe intensity no.=17 4. Aggravation by walking stairs or similar routine physical activity no.=9** D. During headache, at least one of the following: 1. Nausea and/or vomiting no.=5 2. Photophobia and phonophobia no.=2

B. At least 2 of the following pain characteristics: 1. Pressing/tightening quality no.=0? 2. Mild or moderate severity no.=4 3. Bilateral location no.=5*** 4. No aggravation by walking stairs or similar routine physical activity no.=17 (?) C. Both of the following: 1. No vomiting 2. No more than one of the following: nausea photophobia phonophobia

no.=13 no.=15

* When including strong unilateral preponderance and changing of side; ** Nine experienced exacerbation with more than only light exercise. None experienced exacerbations on mild physcial activity. Nine is, therefore, probably the maximal, conceivable figure; 0 is probably the correct one; *** Patient no. 13 is not included in this context.

matology of the present cases, also in this respect, fits far better with CEH than with migraine. Furthermore, most headache researchers would probably agree that it is unlikely that migraine as such would have been abated by the operations employed in the present study. With regard to chronic T-TH, points B 2 through 4 (Table IX) seem to be particularly important. Only 29% of our cases satisfied two out of the four criteria.

neous one from appropriate neck/occipital structures. These variables were found to discriminate vs migraine and T-TH (25). To substantiate the findings of the present study, a prospective study would be most welcome, provided that such a study adheres to up-to-date diagnostic principles.

Consequences?

Acknowledgment

Should we be more aware of the lower neck segments as possible pain generators in this type of headache – and even in CEH in general? In several of the present patients, headache was a dominant symptom, and it might be the first appearing one. In such cases, one may be in an intriguing diagnostic situation prior to the appearance of overt arm symptoms. We are somewhat concerned about the IHS criteria for “cervical headache” (3). These are some of the shortcomings, as we see them: Concerning what has been stated: Point C3: “Abnormal tenderness of neck muscles”. This has in our view little discriminative power vs not only T-TH, but also probably vs migraine without aura; Point D3: “Fractures …..”: has little relevance to ordinary cases of CEH. Concerning what has not been stated: the IHS should incorporate the following CHISG criteria (1): i) Unilaterality of pain without sideshift; ii) Ipsilateral, non-radicular – or even radicular – shoulder/arm pain; iii) Iatrogenic provocation of pain similar to the sponta-

We are indebted to Professor Trond Sand for his help with the statistical analyses.

Functional Neurology 2003; 18(1): 17-28

References 11. Sjaastad O, Fredriksen TA, Pfaffenrath V. Cervicogenic headache: diagnostic criteria. Headache 1998;38:442-445 12. Merskey H, Bogduk N. Classification of Chronic Pain. Seattle; IASP Press 1994: 94-95 13. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988;8(Suppl 7):1-96 14. Bovim G, Fredriksen TA, Stolt-Nielsen A, Sjaastad O. Neurolysis of the greater occipital nerve in cervicogenic headache. A follow-up study. Headache 1992;32:175-179 15. Bogduk N, Marsland A. The cervical zygapophyseal joints as a source of neck pain. Spine 1988; 13:610-617 16. Schellhas KP, Smith MD, Gundry CR, Pollei SR. Cervical discogenic pain: prospective correlation of magnetic resonance imaging and discography in asymptomatic subjects and pain sufferers. Spine 1996;21:300-312

27

T.A. Fredriksen et al.

17. Hunter CR, Mayfield FH. Role of the upper cervical roots in the production of pain in the head. Am J Surg 1949;48:743-751 18. Hildebrandt J, Jansen J. Vascular compression of the C2 and C3 roots - yet another cause of chronic intermittent hemicrania? Cephalalgia 1984;4:167-170 19. Kehr P, Lang G, Jung FM. Uncusektomie und Uncoforaminektomie nach Jung. Langenbecks Arch Chir 1976;341:111-126 10. Fredriksen TA, Salvesen R, Stolt-Nielsen A, Sjaastad O. Cervicogenic headache: long-term postoperative followup. Case report. Cephalalgia 1999; 19:897-900 11. Michler R-P, Bovim G, Sjaastad O. Disorders in the lower cervical spine. A cause of unilateral headache. A case report. Headache 1991;31:550-551 12. Sjaastad O, Fredriksen TA, Pfaffenrath V. Cervicogenic headache: diagnostic criteria. Headache 1990;30:725-726 13. Cloward R. The anterior approach for removal of ruptured cervical discs. J Neurosurg 1958;15:602-617 14. Smith G, Robinson R. The treatment of certain cervicalspine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg 1958;40A:607-624 15. Kerr FWL. A mechanism to account for frontal headache in cases of posterior-fossa tumors. J Neurosurg 1961;18:605-609 16. Kerr FWL. Structural relation of the trigeminal spinal tract to upper cervical roots and the solitary nucleus in the cat. Exp Neurol 1961;4:134-148

28

17. Kerr FWL, Olafson RA. Trigeminal and cervical volleys. Convergence on single units in the spinal grays at C-1 and C-2. Arch Neurol 1961;36:254-260 18. Kerr FWL. Atypical facial neuralgias: their mechanism as inferred from anatomic and physiologic data. Proceedings of the Staff Meetings of the Mayo Clinic 1961;36:254260 19. Bärtschi-Rochaix W. Migraine cervicale. Bern: Hans Huber 1949 20. Fredriksen TA, Fougner R, Tangerud Å, Sjaastad O. Cervicogenic headache. Radiological investigations concerning head/neck. Cephalalgia 1989;9: 139-146 21. Sjaastad O, Saunte C, Hovdal H, Breivik H. Grønbæk E. “Cervicogenic” headache. An hypothesis. Cephalalgia 1983;3:249-256 22. Fredriksen TA, Hovdal H, Sjaastad O. “Cervicogenic headache”: clinical manifestations. Ceph-alalgia 1987;7:147-160 23. Sjaastad O, Bovim G, Stovner LJ. Laterality of pain and other migraine criteria in common migraine. A comparison with cervicogenic headache. Funct Neurol 1992;7:289294 24. Ekbom K. A clinical comparison of cluster headache and migraine. Acta Neurol Scand 1970; 46 (Suppl 41):1-48 25. Vincent M, Luna RA. Cervicogenic headache: a comparison with migraine and tension-type headache. In: Sjaastad O, Milonas I, Antonaci F. eds Proceedings 4th European Headache Federation Congress. Cephalalgia 1998;19 (Suppl. 25):11-16

Functional Neurology 2003; 18(1): 17-28