Embed Size (px)
Citation preview
Scand J Rheumatolow 14: 355-363. 1985
Subaxial Antero-Posterior Dislocation of the Cervical Spine in Rheumatoid Arthritis
INGA REDLUND-JOHNELL and HOLGER PETTERSSON Department of Diagnostic Radiology, Malmo General Hospital, University of Lund, S-214 01 Malmo, Sweden
Redlund-Johnell, I. and Pettersson. H. Subaxial antero-posterior dislocation of the cervical spine in rheumatoid arthritis. (Submitted for publication December 12, 1984 and in revised form February 27, 1985.) Scand J Rheumatology 14: 355-363, 1985.
A subaxial (below C2) antero-posterior dislocation of the cervical spine was present in 18 % of 407 patients with rheumatoid arthritis. The dislocation was as common at only one level as at two or more levels and was more severe if the height of the spinous process was reduced. Most of these same patients also had an atlanto-axial dislocation. At follow-up examinations, the frequency of a subaxial dislocation was the same in patients with a severe atlanto-axial dislocation who were receiving conservative treatment as in patients with a cervico-occipital fusion. Neurological symptoms occurred more frequently in pa- tients with an encroached spinal canal and in patients with a reduced height of the entire cervical spine. Inga Redlund-Jo~nell, Department of Diagnostic Radiology, Malmo General Hospital, S- 21401 Malmo, Sweden.
In rheumatoid arthritis (RA), various types of dislocation of the cervical spine may occur, and since the first reported fatal case reported in 1951 (9) thee dislocations have attracted increasing attention (1, 7, 8, 11, 14, 15, 22).
The most common and obvious type of dislocation is the anterior atlanto-axial, which is also the one most studied (7, 11, 13). The second most common is the subaxial (serial or staircase) dislocation (7, 11, 13). This type is a forward or backward dislocation of the cervical vertebrae below C2 (21) and is usually associated with arthritic changes in the adjacent vertebral end-plates.
The purpose of the present investigation was to examine the frequency, severity and course of subaxial dislocation in RA and its relation to age, sex, duration of disease, neurological symptoms and to other arthritic changes in the cervical spine. An additional aim was to investigate whether cervico-occipital fusion promotes the development of a subaxial dislocation.
MATERIAL AND METHODS All consecutive cervical radiographs performed during the period 1978-80 at Malmo General Hospital and V h h e m Hospital, also in Malmo, in southern Sweden (population about 230000) were reviewed. In these consecutive radiographs there were 407 patients with RA, 96 men and 311 women, and they constitute Material I. Their mean age was 63 (range 34-92) years. Of these 407 patients, 177 had changes compatible with RA of the cervical spine.
To enlarge the material of pathological cervical radiographs, all patients with RA who were not examined during the above-mentioned period, and who had a radiological diagnosis of arthritis of the cervical spine at one of the hospitals during the period 1973-77 were included. Thus, 43 patients constituting Material I1 were added, 3 men and 40 women with a mean age of 68 (range 47-84) years.
Most of these RA patients had been routinely examined before surgical intervention for some reason, or to determine the progression of the disease irrespective of symptoms from the neck. About 20 patients were examined solely because of local symptoms.
All radiographs of the 407 patients have been preserved; also, all of their cervical radiographic examinations from 1950 onward were reviewed for subaxial dislocations, as well as for other arthritic changes.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
356 I . Redlund-Johnell and H . Pettersson Scand J Rheumatology 14
Fig. I
I I I
10 10 20 men
(number) women
(number)
MATERIAL I1
number Fig. 2 of disbcations 11 ' 60
10 10 men women
(number) (number)
level of disbcation
Fig. 1 . Age- and sex-related distribution of patients with a subaxial dislocation in Materials I and 11. Fig. 2. Distribution of dislocations according to cervical level and severity. W , Severe dislocation (4 mm or more); M, mild dislocation.
The cervical radiographic examinations were performed using a standard technique: the patient was in an erect position and the film-focus distance was 150 cm. The lateral radiographs in flexion and extension were studied and all measurements were made directly on the film without correction for magnification.
We have adopted the following criteria and definitions. A subaxial dislocation is said to exist if the distance between the posterior edges of two adjacent vertebral bodies exceeds 2 mm (18), and a distance of 4 mm or more is defined as a severe dislocation. When there is a mobile dislocation, the greatest distance is recorded. An anterior atlanto-axial dislocation is defined as a distance of >3 mm between the anterior arch of the atlas and the dens (14) and a distance of 8 mm or more is regarded as a severe dislocation. A vertical dislocation of the C1 and C2 vertebrae exists if the centre of the lower end-plate of C2 is located at a distance of less than 34 mm, in men, and 29 mm, in women, from the palato-occipital or McGregor's line (17).
All the patients who had undergone cervico-occipital fusion had had an anterior atlanto-axial dislocation of 8 mm or more. The development of a subaxial dislocation in these patients was compared with the development of the same dislocation in patients with an equally severe anterior atlanto-axial dislocation but who were receiving conservative treatment.
The occurrence of a subaxial dislocation was related to age, sex and duration of disease at the time of detection. The course of the dislocation was analysed using available previous examinations. A bony bridge between the two dislocated vertebral bodies was not regarded as an improvement unless the dislocation had diminished.
The sagittal diameter of the spinal canal was measured at the level of dislocation directly on the radiograph.
The height of the cervical spine from the base of the skull above the atlas to the lower end-plate of the body of the C7 vertebra w a s measured in the neutral position. If the height was at least 20 mm less compared with an earlier examination, it was regarded as decreased and when less than 20 mm, preserved. This wide limit w a s chosen to exclude patients with evident reduced cervical height caused by small changes in cervical flexion. Earlier radiographs were available in all the cases where a diminished height was suspected.
Diminished height of the spinous process was regarded as existing if its height was at least 1 mm less than on earlier radiographs. Thus, a shortening in the antero-posterior direction of the spinous process w a s not considered in this investigation.
The charts of the patients were reviewed for neurological disturbances, such as paresthesia, paresis and micturition disturbances. The autopsy charts of the cedeased patients were also studied.
The X*-test was used for the statistical analyses.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
Scand J Rheumatoloav 14 Subaxial dislocation in RA 357
Fig. 3. Woman, born in 1906. (a) Lateral radiograph from 1976 with dislocations at C3-C5. (b) Lateral radiograph a year later, showing deterioration. At the C K 5 level the dislocation is 8 mm and the sagittal bony diameter 12.5 mm. The patient had no neurological symptoms. There are only minor changes at the Cl-CZ level.
RESULTS Frequency and severity of subaxial dislocation according to age, sex and disease duration In Material I there were 72 patients (8 men and 64 women) with a subaxial dislocation, which is 18 % of all the patients in this material, and 41 % of those presenting with arthritic changes of the cervical spine. Their mean age was 65 (range 34-92) years.
In Material I1 there were a further 21 patients with a subaxial dislocation (46%), whose mean age was 67 (range 48-82) years. In all, there were 93 patients in the combined material (1 1 men and 82 women). The two materials with subaxial dislocation are present- ed in Fig. 1. At detection, the mean duration of disease was 20 (range 3-50) years in the 82 patients with a known duration.
The distribution of dislocations at different cervical levels is given in Fig. 2. A disloca- tion at only one level was found in 48 patients (52 %). A severe dislocation (4 mm or more) was found in 30 patients (30%) and 8 of them had a severe dislocation at two or more levels. It was not possible to assess the status at the level of C7-Th1 in all cases due to technical limitations; the true number of dislocations at this level might therefore be higher.
Course of subaxial dislocations Among the 49 patients with an examination 2 years or more after the detection of a dislocation, 15 (31 %) had progression, while only one patient (2%) showed improvement. An example of the progression is given in Fig. 3.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
358 I. Redlund-Johnell and H . Pettersson Scand J Rheumatology 14
Relation between subaxial dislocation and horizontal andlor vertical atlanto-axial dislocation A pre-existing horizontal anterior atlanto-axial dislocation was found in 31 patients (33 %) and a co-existing dislocation of this kind, in another 37 patients (40%). Thus, coexistence of subaxial and horizontal atlanto-axial dislocation occurred in 73 %.
Eighteen patients (19 %) had a pre-existing vertical atlanto-axial dislocation and, at the time of detection of a subaxial dislocation, an additional 31 patients (33%) also had a vertical atlanto-axial dislocation.
Neither a horizontal nor a vertical atlanto-axial subluxation was found in 10 patients (11 %), but 4 of these patients had a severe subaxial dislocation. Subaxial dislocation occurred significantly more often in conjunction with an atlanto-axial dislocation (p<O.001; Table I), but the severity of subaxial dislocation was not correlated with the occurrence of an atlanto-axial dislocation (Table I).
Relation between subaxial dislocation and severe atlanto-axial dislocation, treated conservatively or by ceruico-occipital fusion In the entire material there were 47 patients with an advanced anterior atlanto-axial dislocation (>8 mm). In 10 of them the symptoms had become so severe that an operation, a cervico-occipital fusion, was performed. A follow-up examination after 2-9 years had been made in 7 patients. In the non-operated group, examinations were available in 19 patients 1-8 years after the appearance of the advanced atlanto-axial dislocation. In Table I1 the development and deterioration of the subaxial dislocation in the two groups are compared. It is obvious that the cervico-occipital fusion did not promote development or aggravation of a subaxial dislocation.
Relation between a subaxial dislocation and diminished height of spinous process Subaxial dislocation occurred significantly more often in patients with a diminished height of the spinous process ( p ~ O . 0 0 1 ) than in patients with preserved heights (Table 111). An affection of the spinous process occurred significantly more often together with a severe subaxial dislocation than with a mild dislocation @<O.OOl; Table 111).
Relation between degree of subaxial dislocation, sagittal diameter of spinal canal, and neurological symptoms Neurological symptoms were noted in 20 patients. There was no connection between the degree of dislocation and occurrence of neurological symptoms. However, the sagittal diameter of the spinal canal at the level of dislocation was less than 13 mm in 26 patients and there were significantly more patients with neurological disturbances among those with a narrow spinal canal (<13 mm) than among those with a wide canal (>13 mm; Table
Table I. Relation between occurrence and severity of an atlanto-axial dislocation and a subaxial dislocation
No subaxial Mild subaxial Severe subaxial dislocation dislocation dislocation (no. of pats.) (no. of pats.) (no. of pats.)
No atlanto-axial dislocation 230 6 2 Atlanto-axial dislocation 127 61 22
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
Scand J Rheumatology 14 Subaxial dislocation in RA 359
Table 11. Relation between occurrence or deterioration of a subaxial dislocation in patients with a severe anterior atlanto-axial dislocation, treated conservatively or with cervico-occipital fusion, at a follow-up examination 1-9 years later No statistical difference was found
No or unchanged New or deteriorated subaxial dislocation subaxial dislocation
Conservatively treated patients Fusion-operated patients
1 1 4
8 3
IV). Six of the 7 patients with neurological symptoms, which occurred despite a wide spinal canal (Table IV), had atlanto-axial dislocations with encroachment at that level, and the seventh patient had only minor symptoms (paresthesia in the arms).
Among the 13 patients with a narrow spinal canal and neurological symptoms, both mild and severe disturbances occurred. Four patients had tetraparesis. Two of these patients got their symptoms suddenly when lifted by the nursing staff. One was treated with skull traction, giving relief of the symptoms. A posterior fusion operation was therefore per- formed, which was successful. The other patient had a dislocation of only 3.5 mm, but had
Table 111. Relation between occurrence of a subaxial dislocation and diminished height of the spinous process at the dislocation level
Degree of subaxial dislocation
None Mild Severe (no. of pats.) (no. of pats.) (no. of pats.)
Preserved height of spinous process 350 50 10 Diminished height of spinous process 7 13 20
Table IV. Relation between neurological symptoms, sagittal diameter of the spinal canal and height of the cervical spine (between base of skull and lower border of vertebral body of C7)
Sagittal diameter Sagittal diameter 313 mm <I3 mm (no. of pats.) (no. of pats.)
No neurological symptoms With preserved height of cervical spine With diminished height of cervical spine
Neurological symptoms With preserved height of cervical spine With diminished height of cervical spine
50 10
4 3
10 3 (1)"
3 (2)" 10
~~
a Denotes a patient with a narrow spinal canal at the CI-C2 level.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
360 I . Redlund-Johnell and H . Pettersson Scand J Rheumatology 14
Fig. 4. Woman, born in 1899. (a) Lateral hypopharyngeal radiograph from 1956 free from arthritic changs but with a narrow spinal canal of 12.5 mm at the C 3 4 4 level. (b ) Lateral radiograph from 1977 after the appearance of tetraparesis. The dislocation is only 3.5 mrn, but this means a sagittal diameter of the spinal canal of 9 rnrn. There is also a vertical atlanto-axial dislocation.
a congenitally narrow canal (Fig. 4). Because her general condition was poor, no operation was performed. She died a year later and the autopsy revealed cord compression.
In at least 5 patients the neurological symptoms subsided without treatment. In patients with neurological symptoms, the dislocation occurred as often in the upper as in the lower levels of the subaxial part of the cervical spine.
Relation between height of cervical spine and neurological symptoms There were 13 patients with a decreased height of the cervical spine among the patients with a subaxial dislocation. The relation between neurological symptoms and height of the cervical spine in the 26 patients with a narrow spinal canal is presented in Table IV. There were significantly more patients with neurological disturbances among those with a decreased height of the cervical spine than among those with a preserved height (p<O.Ool).
DISCUSSION Subaxial cervical dislocation may be found in cervical spondylosis (18) and is then usually mild. In traumatic cases the dislocation may be mild or severe, and even if the dislocation
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
Scand J Rheurnatology 14 Subaxial dislocation in RA 361
at the radiographic examination is slight, it may have been worse at the time of the accident. In RA the development of a dislocation is insidious and the degree of dislocation may be at its hitherto maximum at the examination, provided that this includes lateral radiographs in flexion and extension.
In cervical spondylosis a subaxial dislocation is said to be present if the dislocation exceeds 2 mm (18). In RA the dislocation has also been expressed as a percentage of the antero-posterior diameter of the upper end-plate of the lower vertebra, and Smith et al. (19) regarded a 15% dislocation as a significant displacement, corresponding to about 2 mm. The possible advantage of expressing the dislocation as a percentage in comparison with absolute values in mm was not obvious in that study, so we prefer a simple measure.
Some reports have required a dislocation at two or more levels in order to satisfy the definition of a subaxial dislocation and to be able to use the terms ‘serial’ or ‘staircase’ dislocation (1, 7, 13). The risk of including RA patients with a spondylotic dislocation (false-positive) has then disappeared, but at the same time patients with a dislocation caused by RA will be excluded. However, most studies, including ours, accept a disloca- tion at only one level as an expression for a rheumatoid affection (10, 11, 14, 15, 16, 22).
In earlier investigations a subaxial dislocation frequency of 6-41 % has been reported (7, 10, 11, 15, 19, 22), and our frequency of 18% falls within this range. The mean age of our patients, 65 years, also accords with earlier reports (19), as does the preponderance of women (11, 15, 19) and the long duration of rheumatoid disease (13, 19). But in a newly published prospective investigation by Winfield et al. (22) a frequency of 24% was found as early as within 5 years of onset of the disease. Their criterion for a dislocation was only 1 mm, which would possibly explain the high incidence so early in the disease. However, their incidence would be only 8 % when using the same criterion for a dislocation as in our study.
A subaxial dislocation is said to occur mostly in the middle part of the cervical spine in RA (1 1, 14) and our investigation confirms this, but a severe dislocation may appear as high up as the C Z C 3 level or as low as the C7-Thl level.
Our findings of about the same frequency of a subaxial dislocation at one level as at two or more levels are in agreement with Winfield et al. (22), as are the findings of a progression in 31 % of our patients at a follow-up examination. In the Winfield’s investiga- tion there was, as in most other studies, no report of improvement, but in our study one patient did show improvement. Only one earlier investigation reported improvement in a subaxial dislocation (19): such was observed in nearly half of the patients, an astonishingly high figure. Their criterion for improvement was not given and bony bridges might have been regarded as improvement.
In the prospective study by Winfield et al. (22), with early RA and a mean follow-up period of 7 years, a co- or pre-existing anterior atlanto-axial dislocation was reported in 17%, whereas in another prospective study with patients in different stages of their disease (with a mean follow-up period of 6 years) a co- or pre-existing anterior atlanto- axial dislocation was reported in about 60% (15). This latter study is in better agreement with our investigation (72 %).
The coincidence of a vertical atlanto-axial dislocation seems to have been studied in only one report earlier, with a reported incidence of 12% (15). Our frequency is much higher (53 %), which might be due to different methods of measuring the vertical disloca- tion (20). It is noteworthy that in our material only 11 % of the patients did not have an atlanto-axial dislocation. Thus, the subaxial dislocation will usually be preceded by an atlanto-axial dislocation or will appear concurrently, but a severe subaxial dislocation may still be the first and only sign of cervical arthritis. Consequently, lateral radiographs of the neck should include the entire cervical spine in both flexion and extension.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
362 I . Redlund-Johnell and H . Pettersson Scand J Rheumatology 14
It has been suggested that patients will be more liable to injury of the rest of the cervical spine after a cervico-occipital fusion (20). Our results show that there is no difference as regards the development of a subaxial dislocation in patients with conservative treatment or cervico-occipital fusion.
There appear to be only a few earlier investigations mentioning a subaxial dislocation and its possible relation to changes of the spinous processes (8, 14, 16). One of these found no connection between shortening of the spinous process and degree of dislocation (I@, but no earlier reports have correlated changes in the height of the spinous processes with the degree of subaxial dislocation. In our investigation, severe dislocation occurred significantly more often at a level with diminished height of the spinous process than at a level with preserved height. This might have been due to Baastrup’s syndrome (5) in combination with osteoporosis and/or interspinous RA, because bursae with arthritic changes have been found between the spinous processes in adult RA patients (4). Irrespec- tive of the mechanism, resorption of the spinous process may indicate a severe disloca- tion.
Neurological symptoms secondary to RA of the cervical spine were first observed in atlanto-axial dislocations (9) but they have also been reported in subaxial dislocations (8, 10, 21). Neurological disturbances are said to appear more often with a subaxial disloca- tion at a lower level where the spinal canal is narrower (1). Other investigators have reported that cord compression may be unpredictable, or may depend on the width of the cervical canal (6, 12). In spondylotic stenosis it has been reported that with a sagittally measured spinal canal of 10-13 mm, cord compression may or may not be present (23). In our study, when the spinal canal had these dimensions, neurological symptoms were present in about half of the cases with RA, and there were thus signficantly more neurological symptoms among patients with a narrow spinal canal than among those with a wide canal. There was only one patient with neurological symptoms who had no encroach- ment on the bony spinal at any level, but she had only minor symptoms (paresthesia of the arm), which are not unusual even in the spondylotic patient without cord compression (3).
The possible relation of a decreased height of the cervical spine to neurological symp- toms does not seem to have been investigated earlier in RA, but that a measurable decrease in height may appear has been observed (22). In one laboratory investigation with measurement of the cervical spinal cord in flexion and extension, it was shown that when the cord became shorter the cross-sectional area increased, or in other words, the spinal cord adapted itself to varying length with plastic deformation (2). With decreased height of the cervical spine, the cross-sectional area of the spinal cord should increase, and thus be more vulnerable to a diminishing spinal canal space. This may explain our finding that among patients with a narrowed canal, neurological symptoms occurred significantly more often when there also was a decreased height of the cervical spine.
REFERENCES 1. 2.
3.
4.
5.
6.
Bland, J . H.: Rheumatoid arthritis of the cervical spine. J Rheumatol 1:319, 1974. Breig, A. : Dehnungsverschiebungen von Dura und Riickenmark in Spinalkanal. Fortschr Neurol Psychiat 32: 195, 1964. Burrows, E. H.: The sagittal diameter of the spinal canal in cervical spondylosis. Clin Radio1 14: 77, 1%3. Bywaters, E. G . L. Rheumatoid and other diseases of the cervical interspinous bursae and changes in the spinous processes. Ann Rheum Dis 41: 360, 1982. Bywaters, E. G . L.: The lumbar interspinous bursae and Baastrup’s syndrome. An autopsy study. Rheumatol Int 2: 87, 1982. Cabot, A. & Becker, A.: The cervical spine in rheumatoid arthritis. Clin Orthop 131: 130, 1978.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
Scand J Rheumatology 14 Subaxial dislocation in RA 363
7. Conlon, P. W., Isdale, I. E. & Rose, B. S.: Rheumatoid arthritis of the cervical spine. An analysis of 333 cases. Ann Rheum Dis 25: 120, 1966.
8. Crellin, R. Q., Maccabe, J. J. & Hamilton, E. B. D.: Severe subluxation of the cervical spine in rheumatoid arthritis. J Bone Joint Surg 52 B: 244, 1970.
9. Davis, F. W. & Markley, H. E.: Rheumatoid arthritis with death from medullary compression. Ann Intern Med 35: 451, 1951.
10. Kataoka, O., Hirohata, K. & Kurihara, A.: The surgical treatment of myelopathy secondary to rheumatoid arthritis of the lower cervical spine. Int Orthop 3: 103, 1979.
1 1 . Meikle, J. A. & Wilkinson, M.: Rheumatoid involvement of the cervical spine. Ann Rheum Dis 30: 154, 1971.
12. Nakano, K. K.: Neurological complications of rheumatoid arthritis. Orthop Clin North Am 6:861, 1975.
13. Nakano, K. K., Schoene, W. C., Baker, R. A. & Dawson, D. M.: The cervical myelopathy associated with rheumatoid arthritis: Analysis of 32 patients, with 2 postmortem cases. Ann Neurol3: 144, 1978.
14. Park, W. M., O’Neill, M. & McGall, I. W.: The radiology of rheumatoid involvement of the cervical spine. Skeletal Radiol 4: 1 , 1979.
15. Pellici, P. M., Ranawat, C. S., Tsairis, P. &Bryan, W. J.: A prospective study of the progression of rheumatoid arthritis of the cervical spine. J Bone Joint Surg 63 A: 342, 1981.
16. Ranawat, C. S., O’Leary, P., Pellici, P., Tsairis, P., Marchisello, P. & Dorr, L.: Cervical spine fusion in rheumatoid arthritis. J Bone Joint Surg 61 A: 1003, 1979.
17. Redlund-Johnell, I. & Pettersson, H.: Radiographic mesurements of the cranio-vertebral region designed for evaluation in rheumatoid arthritis. Acta Radiol Diagn 25: 23, 1984.
18. Rube, W. & Schulte, G. A.: Die degenerativen Erkrankungen der Wirbelsaule. In Encyclopedia of Medical Radiology (ed. L. Diethelm), vol. VI, part 2, p. 9. Springer Verlag, Berlin, Heidelberg and New York, 1974.
19. Smith, P. H., Benn, R. T. & Sharp, J.: Natural history of rheumatoid arthritis. Ann Rheum Dis 31:431, 1972.
20. Sweetnam, R.: Atlantoaxial instability in rheumatoid arthritis. Ann Rheum Dis 29: 333, 1970. 21. Vogelsang, H., Zeidler, H., Wittenborg, A. & Weidner, A,: Rheumatoid cervical luxations with
22. Winfield, J., Cooke, D., Brook, A. S. & Corbett, M.: A prospective study of the radiological
23. Wolf, B. S., Khilnani, M. & Malis, L.: The sagittal diameter of the bony cervical spinal canal and
fatal neurological complications. Neuroradiology 6: 87, 1973.
changes in the cervical spine in early rheumatoid arthritis. Ann Rheum Dis 40: 109, 1981.
its significance in cervical spondylosis. J Mt Sinai Hosp 23: 283, 1956.
Scan
d J
Rhe
umat
ol D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
New
cast
le U
pon
Tyn
e on
12/
19/1
4Fo
r pe
rson
al u
se o
nly.
