Radiological and anatomical evaluation of the atlantoaxial transarticular screw fixation technique

Sixty-one patients treated with C1-2 transarticular screw fixation for spinal instability participated in a detailed clinical and radiological study to determine outcome and clarify potential hazards. The most common condition was rheumatoid arthritis (37 patients) followed by traumatic instability (15 patients). Twenty-one of these patients (one third) underwent either surgical revision for a previously failed posterior fusion technique or a combined anteroposterior procedure. Eleven patients underwent transoral odontoidectomy and excision of the arch of C-1 prior to posterior surgery. No patient died, but there were five vertebral artery (VA) injuries and one temporary cranial nerve palsy. Screw malposition (14% of placements) was comparable to another large series reported by Grob, et al. There were five broken screws, and all were associated with incorrect placement. Anatomical measurements were made on 25 axis bones. In 20% the VA groove on one side was large enough to reduce the width of the C-2 pedicle, thus preventing the safe passage of a 3.5-mm diameter screw. In addition to the obvious dangers in patients with damaged or deficient atlantoaxial lateral mass, the following risk factors were identified in this series: 1) incomplete reduction prior to screw placement, accounting for two-thirds of screw complications and all five VA injuries; 2) previous transoral surgery with removal of the anterior tubercle or the arch of the atlas, thus obliterating an important fluoroscopic landmark; and 3) failure to appreciate the size of the VA in the axis pedicle and lateral mass. A low trajectory with screw placement below the atlas tubercle was found in patients with VA laceration. The technique that was associated with an 87% fusion rate requires detailed computerized tomography scanning prior to surgery, very careful attention to local anatomy, and nearly complete atlantoaxial reduction during surgery.