C2 versus C3 or C4 as the upper instrumented vertebra for long-segment cervical fusions: a systematic review and meta-analysis

OBJECTIVE Selecting C2 versus C3 or C4 (i.e., C3/C4) as the rostral anchoring level in long -segment cervical fusions is a common clinical conundrum. The data regarding proximal failure in long constructs of the cervical spine is scarce. The objective of this study was to systematically review the published literature and perform a meta -analysis of the incidence for proximal adjacent -segment disease (ASD) in the context of long cervical fusions and cervicothoracic fusions ending in C2 versus those ending in the subaxial spine (C3 or C4). METHODS Using the PRISMA guidelines, the authors performed a search of the PubMed/MEDLINE, Embase/Ovid, and Cochrane Central databases to identify all full-text articles in the English -language literature with the following inclusion criteria: 1) studies including patients with the upper instrumented vertebra (UIV) at C2 versus C3/C4; 2) patients undergoing >= 3 -level posterior cervical fusion; and 3) indication for surgery of degenerative disc disease, cervical spon- dylotic myelopathy, or cervical deformity. Studies that were not published in the English language, case reports, review articles, letters to the editor, and meeting abstracts were excluded. A meta -analysis was conducted using a fixed -effects model when I2 values were below 70%. Conversely, when I2 values were equal to or greater than 70%, a random-effects model was used. A funnel plot was used to assess the presence of publication bias. RESULTS Seven studies consisting of 1215 patients were included in the meta -analysis. There were 403 (32.8%) patients in the C2 UIV group and 812 (67.2%) patients in the C3/C4 UIV group. When the 7 studies were analyzed, the overall rate of reoperation was comparable between the C2 (9.2%) and C3/C4 (9.4%) UIV groups (p = 0.93) but the rate of surgical ASD due to proximal pathology was 1.2% and 3%, respectively (OR 0.36, 95% CI 0.15-0.86; p = 0.02). When comparing between groups, no statistical difference was found regarding the rate of reoperation due to distal pathology or surgical infection. CONCLUSIONS Long-segment cervical or cervicothoracic constructs that anchor into C2 may have similar complication rates but lower revision rates for proximal ASD than constructs that anchor into the subaxial spine.