Biomechanical evaluation of cervical lateral mass fixation: a comparison of the Roy-Camille and Magerl screw techniques.

OBJECT: The purpose of this study was to assess human cervical spine pullout force after lateral mass fixation involving two different techniques: the Roy-Camille and the Magerl techniques. Although such comparisons have been conducted previously, because of the heterogeneity of results and the importance of this procedure in clinical practice, it is essential to have data derived from a prospective and randomized biomechanical study involving a sufficient sample of human cervical spines. The authors also evaluated the influence of the sex, the vertebral level, the bone mineral density (BMD), the length of bone purchase, and the thickness of the anterior cortical purchase.

METHODS: Twenty-one adult cervical spines were harvested from fresh human cadavers. Computerized tomography was performed before and after placing 3.5-mm titanium lateral mass screws from C-3 to C-6. Pullout forces were evaluated using a material testing machine. The load was applied until the pullout of the screw was observed. A total of 152 pullout tests were available, 76 for each type of screw fixation. The statistical analysis was mainly performed using the Kaplan-Meier survival method. The mean pullout force was 266 +/- 124 N for the Roy-Camille technique and 231 +/- 94 N for the Magerl technique (p < 0.025). For the C3-4 specimen group, Roy-Camille screws were demonstrated to exert a significantly higher resistance to pullout forces (299 +/- 114 N) compared with Magerl screws (242 +/- 97 N), whereas no difference was found between the two techniques for the C5-6 specimen group (Roy-Camille 236 +/- 122 N and Magerl 220 +/- 86 N). Independent of the procedure, pullout strengths were greater at the C3-4 level (271 +/- 114 N) than the C5-6 level (228 +/- 105 N) (p < 0.05). No significant correlation between the cancellous BMD, the thickness of the anterior cortical purchase, the length of bone purchase, and maximal pullout forces was found for either technique.

CONCLUSIONS: The difference between pullout forces associated with the Roy-Camille and the Magerl techniques was not as significant as has been previously suggested in the literature. It was interesting to note the influence of the vertebral level: Roy-Camille screws demonstrated greater pullout strength (23%) at the C34 vertebral level than Magerl screws but no significant difference between the techniques was observed at C5-6.