Translaminar versus pedicle screw fixation of C2: comparison of surgical morbidity and accuracy of 313 consecutive screws.

Objective: C2 translaminar (TL) screws rigidly capture the posterior elements of C2, avoid risk of vertebral artery injury, and are less technically demanding than C2 pedicle (PD) screws. However, a C2-TL screw breach places the spinal cord at risk, and the durability of C2-TL screws remains unknown. It is unclear if TL versus PD screw fixation of C2 is truly associated with less operative morbidity, greater accuracy of screw placement, or equivalent durability.
Methods: We retrospectively reviewed the records of 167 consecutive patients undergoing posterior cervical fusion with either PD or TL screw fixation of C2. Perioperative morbidity, breach of the C2 lamina or pedicle on postoperative computed tomographic scans, and rates of operative revision were compared between PD and TL screw constructs in axial (C1-C2 or C1-C3) and subaxial (C2 and caudal) cervical fusions.
Results: In total, 152 C2-TL screws and 161 C2-PD screws were placed in 167 patients. Thirty-one (19%) cases of axial cervical fusion (C1-C2 or C1-C3) were performed (mean age, 63.8 +/- 20.6 years) with either C2-TL (16 [52%]) or C2-PD (15 [48%]) screw fixation. One hundred thirty-six (81%) cases of subaxial cervical fusion (C2-caudal) were performed (mean age, 57.9 +/- 14.7 years) with either C2-TL (66 [49%]) or C2-PD (70 [51%]) screw fixation. For both axial and subaxial cervical fusions, baseline patient characteristics and all measures of perioperative morbidity were similar between C2-TL and C2-PD screw cohorts. In total, 11 (7%) PD screws breached the pedicle (0 requiring acute revision) versus only 2 (1.3%) TL screws that breached the C2 lamina (1 requiring acute revision) (P = 0.018). By 1 year postoperatively, pseudoarthrosis or screw pullout requiring reoperation was required in 4 (6.1%) patients with C2-TL screws versus 0 (0%) patients with PD screws (P < 0.05 for subaxial constructs). No cases of C2-TL or C2-PD axial fusion required reoperation or screw pullout or pseudoarthrosis.
Conclusion: In our experience, radiographic breach of C2 pedicle screws occurred more frequently than C2 laminar screw breach. However, this was not associated with an increase in morbidity. By 12 months postoperatively, C2-TL screws were associated with a greater incidence of operative revision when used in subaxial constructs but similarly effective for axial cervical constructs. The 1-year durability of C2-TL screws might be inferior to C2 pedicle screws for subaxial fusions, but equally effective for axial cervical fusions.