Existing techniques of posterior multi-point C1/2 stabilisation are technically demanding and can be hazardous. The coauthors have recently reported successful atlantoaxial fusion using a novel C1/2 stabilisation technique employing C1 multi-axial posterior arch screws (MA-PAS) in a clinical series of three patients where anatomical anomalies precluded established techniques.

The technically less demanding nature of this new technique, and possible wider application in patients with normal anatomy, led the authors to investigate its biomechanical stability compared to other established techniques.

Twenty-four human fresh-frozen cadaveric spines were harvested C0-C5. Motion was restricted to between C0 and C4. Each spine was non-destructively tested in flexion/extension, lateral bending and axial rotation, firstly in the intact state and then after Type 2 odontoid fracture destabilisation and insertion of Magerl-Gallie, Unicortical Harms, Bicortical Harms or MA-PAS instrumentation. ROM between C1 and C2 was monitored using two digital cameras. Results for each technique were compared statistically compared using ANOVA.

The C1-C2 joint of the intact spines demonstrated high flexibility in flexion/extension (16.5deg). After instrumentation all specimens showed significantly reduced ROM in flexion/extension (Magerl-Gallie FE = 4.2deg, Unicort Harms FE = 7.2deg, Bicort Harms FE = 4.4deg). Lateral bend ROM of instrumented specimens (Magerl-Gallie LB =3.8deg, Unicort Harms LB = 3.8deg, Bicort Harms LB =2.3 deg) was, however, similar or slightly greater than intact (2.7 deg) . MA-PAS showed similar ROM in flexion/extension (4.2 deg) as the Magerl-Gallie and Harms techniques but was slightly higher in lateral bend (5.3 deg).

The MA-PAS technique was shown to have similar biomechanical stability to the Magerl-Gallie and Harms techniques. Given the demonstrated biomechanical stability of the MA-PAS technique, it may be a suitable alternative to the existing technically demanding, and possibly more hazardous, multi-point fixation techniques in patients with normal, as well as anomalous, C1/2 segmental anatomy.

Introduction: The complex anatomy and biomechanics of the atlantoaxial motion segment impose technical challenges in the achievement of safe and successful surgical stabilization and fusion. The coauthors have recently reported successful clinical results using a novel C1-C2 stabilization technique employing C1 multi-axial posterior arch screws (MA-PAS). This study compares biomechanical stability of MA-PAS with two established multi-point fixation techniques (Magerl-Gallie and Harms) using non-destructive and destructive testing.

Methods: 15 human fresh-frozen cadaveric occipital-C5 cervical spines (average age 77.4 [51–95], sourced from ScienceCare, USA) were randomly allocated to 3 equal groups. Screws were passed up through adjacent end vertebrae such that motion was limited to between C0 and C4. Each spinal column was non-destructively tested in flexion/extension (±1.5Nm), lateral bend (±1.5Nm) and axial rotation (±1.5Nm), firstly in their INTACT state and then after Type 2 odontoid fracture destabilization combined with MAGERL-GALLIE (n=5), HARMS (n=5) or MA-PAS (n=5) instrumentation. All 15 reconstructed spines were finally loaded to failure in forward flexion only.

Results: Non-destructive testing: The C1-C2 joint of the INTACT spines all demonstrated high flexibility in flexion/ extension (ave 16.5deg) and axial rotation (ave 52.6 deg) while lateral bending (ave 2.7deg) was less compliant (see Fig.3). After instrumentation all specimens showed significantly reduced ROM in flexion/extension (MAGERL-GALLIE=4.2deg, HARMS=4.4deg, MA-PAS=4.2deg) and axial rotation (MAGERL-GALLIE=4.05deg, HARMS=0.59deg, MA-PAS=3.7deg) while lateral bend ROM of all instrumented specimens was similar or slightly greater than INTACT (HARMS=2.3deg, MAGERL-GALLIE=3.8deg, MA-PAS=5.3deg). There was no significant difference between the instrumented groups in each loading direction.

Destructive testing: MAGERL-GALLIE was the strongest requiring an average of 13.5Nm to cause failure while HARMS was the weakest requiring 7.8Nm of torque. MA-PAS technique averaged 12.2Nm of torque to cause failure.

Conclusions: The MA-PAS technique was shown to have similar ultimate strength in flexion to the MAGERL-GALLIE and HARMS techniques and stability in flexion-extension, axial rotation and lateral bend. The MA-PAS failure load in flexion was greater than the HARMS technique, and nearly as high as the MAGERL-GALLIE. Given the biomechanical stability of the MA-PAS technique, it is proposed that this technique is an alternative to the technically demanding, and possibly more hazardous, conventional multi-point fixation techniques in patients with normal, as well as anomalous, C1/2 segmental anatomy.