Article Text

C1/C2 rotary subluxation due to spasmodic torticollis
  1. S DALVIE,
  2. A P MOORE,
  1. Walton Centre for Neurology and Neurosurgery, Lower Lane, Fazakerley, Liverpool L9 7LJ, UK

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Rotary subluxation of the atlantoaxial complex is encountered more often in children than in adults. It is usually associated with a clear history of cervical trauma, upper respiratory infection, recent head or neck surgery, or rheumatoid arthritis. This paper reports on an adult presenting with C1/C2 rotary subluxation in whom all of these causes were absent. The subsequent clinical course showed that the patient in fact had spasmodic torticollis which we think was the cause of the rotary subluxation at C1/C2.

    A 37 year old man presented in November 1997 with a history of an insidious onset of progressive deformity of the neck associated with spasm of the left sternocleidomastoid and trapezius muscles and occipital pain from April of that year. He had had mild neck pain for several years. The occipital pain was left sided and became increasingly severe. Electromyography was not performed. He had. been off work for 6 months and found it increasingly difficult to sleep in a comfortable position. The only relevant history was one of “anxiety attacks”.

    On examination there were no signs of rheumatoid arthritis. The patient had a classic “cock robin “ deformity with his head tilting to the left and turning to the right. This was associated with spasm and tenderness, but no obvious hypertrophy, of the left sternocleidomastoid and trapezius which was thought to be voluntary as it subsided when he was relaxed. Plain radiography of the atlantoaxial region was essentially normal and thus CT was obtained under a general anaesthetic as the patient was extremely anxious. After the induction of the general anaesthetic, the element of spasm and tilting of his neck which was thought to be due to psychological overreaction disappeared and examination showed persistence of the “cock robin” deformity. Spinal CT in the neutral position confirmed a C1/C2 rotary subluxation which reduced with the head turned towards the left but was exaggerated by turning to the right. There was no history of recent neck injury, rheumatoid arthritis, or pharyngeal infection and thus a cause for the C1/C2 rotary subluxation was not apparent at that stage. In view of the severity of the pain surgical stabilisation of C1/C2 was suggested.

    The patient was placed in halo-traction for a week, and further CT was performed. This showed significant improvement but not total correction of the rotary subluxation. As reduction was not total, it was decided not to perform transarticular screw fixation of C1/C2 but a posterior modified Gallie fusion was performed in the position of maximum reduction. A halo-vest was applied. Check radiography was satisfactory and the patient reported a very pleasing relief of the pain and spasm which he had preoperatively. The halo-vest was maintained for 10 weeks during which his pain and spasm had completely resolved. However, shortly after removing the halo, he had a recurrence of the pain and spasm in the left sternocleidomastoid al though the severe occipital pain was still completely relieved. His neck posturing was now variable and typical of spasmodic torticollis with no residual “cock robin” deformity. There was rotation to the right, tilt to the left, and left shoulder elevation. Check radiography showed no loss of the previous position.

    At this point a diagnosis of spasmodic torticollis was made with a torticollis severity score as described by Tsuiet al of 15 indicating moderately severe deformity.1 Five hundred units (200 mU/ml) of botulinum toxin A (Dysport—Ipsen) were injected into the left sternocleidomastoid (250 mU) and trapezius (250 mU). The first injection itself produced a 50% reduction in pain and spasm. Three further injections of the same dose over a year produced further improvement and at latest follow up 16 months after surgery his torticollis score had fallen to eight. His posture was similar in nature but much less pronounced and variable. The occipital pain remains absent and he has achieved a solid fusion.

    Atlantoaxial rotary subluxations were first described by Wortzman and Dewar in 19682 and further clarified by Fielding and Hawkins in 1977.3 The maximum normal rotation of the atlantoaxial joints is 45–47 degrees. Beyond this the lateral inferior facet of the atlas rocks over the lateral superior articular facet of the axis.3

    Rotary atlantoaxial subluxation can be caused by severe twisting injury of the neck, usually associated with violent sport and vehicular accidents.4 It has been reported as a result of a greater than 90 degree rotation of the neck under general anaesthesia.4 5

    In this case the common causes of C1/C2 rotary subluxation were absent and we suggest that the subluxation was caused by the spasmodic torticollis over time. Surgery and application of the halo abolished the occipital pain and the spasm was reduced, possibly due to the extra somatosensory input from the halo ( a mechanical geste antagonistique). Removal of the halo was followed by recurrence of the dystonic spasm but the occipital neuralgia remained absent due to the stabilisation of the atlantoaxial complex.

    In this case it was not possible to determine at what stage the rotary subluxation occurred. It is possible that the subluxation was the primary event leading to malposition of the neck and muscle spasm—a type of “post-traumatic” dystonia. However, in patients with atlantoaxial rotary subluxation, the normal neck deformity is the classic “cock-robin” deformity and activation of sternocleidomastoid and trapezius does not occur. The surgical treatment in this case resolved the “cock robin” deformity and occipital pain but the typical clinical findings of spasmodic torticollis reappeared once the halo was removed. It is most likely that the halo provided sufficient somatosensory input to inhibit the sternocleidomastoid spasm during the length of time the halo was in position. The evolution of the clinical findings with relief of occipital pain and “cockrobin” deformity followed by a more typical appearance of spasmodic torticollis strongly suggest that it was the dystonia which caused the rotary subluxation.

    Spasmodic torticollis is a focal and usually idiopathic dystonia with cervical muscle spasm causing involuntary neck posturing and movement. It can occur at any age. Chemical denervation of the overactive muscles with botulinum toxin is now the usual treatment and is effective in most patients.6

    Dystonia can cause subluxation or dislocation at various joints. For instance, the temporomandibular joints can undergo recurrent or chronic dislocation in idiopathic or tardive oromandibular dystonia.7-9 Angelini et aldescribed subluxation of the subaxial cervical spine resulting in a cervical myelopathy in a child with spastic dystonic cerebral palsy10 and Tunkel et alreported cervical subluxation causing improvement in the dystonia in a patient with longstanding idiopathic torsion dystonia.11To our knowledge adult onset spasmodic torticollis causing and presenting as a rotary atlantoaxial subluxation has never been reported in the literature. Prolonged rotation beyond the physiological limit is likely to be the cause of this subluxation.

    When atlantoaxial subluxation appears after prolonged involuntary neck posturing an underlying diagnosis of dystonia should be considered. Botulinum toxin will not resolve the subluxation, but was necessary in this case to control the underlying dystonia. External braces and collars rarely control the forceful movements of cervical dystonia, and the toxin may take some days or even weeks to work, so we recommend treatment as soon as dystonia is diagnosed. In theory botulinum toxin might enhance the effect of or interfere with recovery from acute muscle paralysing agents used in anaesthesia. However, no such reactions have been reported in 12 years of extensive experience worldwide with botulinum toxin, and therefore it is probably safe to give injections even before neck surgery. When a patient with spasmodic (variable posturing) torticollis develops a fixed and sufficiently extreme “cock robin” posture, the clinician should consider investigation by plain radiography and CT to exclude rotary subluxation, even if the muscle spasm is intermittent.