Purpose: Several techniques have been proposed for C1-C2 fusion. The anterior transoral technique is the most direct approach but at the cost of major risk of infection. The posterior approach allows wiring (Gallie technique) or direct atloidoaxoid screw fixation (Magerl technique). The retrosternomastoid bilateral approach (Du Toit technique) allows direct screwing in the lateral masses. The rate of nonunion reported in the literature is high for wiring techniques. Biomechanical studies have demonstrated the mechanical superiority of trans-articular screwing which has been confirmed in clinical series. The purpose of the present study was to describe a modification of the Du Toit technique and describe results in a short series.

Material and methods: This modification of the Du Toit technique consists in an abrasion of the C1 lateral mass at its origin enabling the penetration of a Cloward curette to create a stable introduction point for the drill bit and thus avoid slippage forward as can occur with the conventional technique. The screw is directed towards C2, in a strict frontal plane. The obliquity depends on the room allowed by the mastoid. The drill bit should cross both corticals of the inferior facet of C1 and the superior facet of C2. The screws must cross in a coronal plane just under the odontoid. Fusion of the C1-C2 lateral masses is achieved by abrasion and grafting.

We have used the modified Du Toit technique for C1-C2 arthodesis in four patients with rheumatoid arthritis, fracture of the odontoid, an odontoid bone, and isolated degeneration. The procedure was a first intention treatment for the patient with primary degeneration, and a second intention procedure for the others who had developed nonunion after wiring.

Results: Mean hospitalisation was six day. Operative time was 2 h 10 min. Mean blood loss was 200 ml peroperatively and 120 ml postoperatively. None of the patients had developed nonunion or mechanical failure at a mean follow-up of 2.7 years (range 1 – 5 years). Operative complications included one case of venous bleeding which was treated with vascular clips and two cases were the lateral mass of C1 was weakened requiring cementing. There was one early postoperative neurological complication with hypoaesthesia of the hemitongue that regressed spontaneously. None of these complications produced sequelae.

Conclusion: The advantages of the modified technique for lateral screw fixation of C1-C2 is the improvement in the entry point for the drill bit allowing an optimal screw position and a stable drilling to achieve good mechanical fixation and certain union.

Purpose of the study: Missing a cervical stenosis in patients with lumbar canal stenosis can lead to an inadequate surgical strategy and delay in treatment of overt cord compression.

Material and methods: Among 100 patients with lumbar canal stenosis, we identified patients with symptoms related to cervical stenosis. These four patients had to undergo surgical decompression of the cervical and lumbar spine to achieve full symptom relief. Careful analysis of the clinical expression is essential to identify cervical stenosis with few or no signs. Presence of gait disorders related to balance disorders, widening of the balance polygone, or use of a crutch, are suggestive of an associated lesion. MRI facilitates diagnosis of cervical compression. EMG and somesthetic evoked potentials are the gold standard examinations to confirm clinical and radiographic suspicion.

Results: Gait disorders, other than simple claudication and/or radiculalgia warranted MRI and electric explorations in these patients. In these four patients, compression of the posterior cord explained well the gait disorders via a deep sensorial mechanism.

Discussion: MRI evidence of cervical osteoarthritis is not sufficient to confirm the origin of the patient’s complaints. There must be a perfect correlation with the electrical results, particularly evoked potentials associated with MRI to confirm the organic origin of disorders resulting from cervical stenosis. Positive diagnosis of such an association requires a specific treatment algorithm as was used in three of our four patients. In the fourth case, the lumbar compression appeared to predominate over the cervical compression leading to decompression of the lumbar canal followed later by decompression of the cervical canal. Between the two procedures, the patient’s status had considerably declined. This strategy which had appeared adequate was thus found to be quite inadequate and even dangerous.

Conclusion: Presence of gait disorders other than simple claudication or single-level or multiple-level radiculalgia in patients with lumbar canal stenosis should lead to search for an associated cervical stenosis. The perfect correlation between the radiographic and electrical findings is indispensable to establish certain diagnosis. The cervical stenosis should be treated before the lumbar stenosis.