Objective: A high-riding vertebral artery (HRVA) has been defined as a C2 isthmus height of ≤5 mm and/or internal height of ≤2 mm measured 3 mm lateral to the border of the spinal canal. Its reported prevalence has varied widely. If overlooked during the approach for craniocervical fusion, injury to the vertebral arteries can occur, affecting the outcome. The present meta-analysis aimed to provide the pooled prevalence of HRVAs.

Methods: A comprehensive database search was conducted by 3 of us. Peer-reviewed studies that had followed the strict definition for HRVAs and had reported its prevalence were included. The risk of bias was assessed using the anatomical quality assessment tool. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were followed. The pooled prevalence was calculated using a random effects model.

Results: The data from 20 studies with 3126 subjects (7496 sides) were analyzed. The overall pooled prevalence of ≥1 HRVA was 25.3% (95% confidence interval [CI], 19.6%-31.5%). The prevalence in those without the most important confounding factor, rheumatoid arthritis (RA), was 20.9% (95% CI, 16.5%-25.8%). Patients with RA had a prevalence of 42.9% (95% CI, 23.8%-63.1%). The difference between the non-RA and RA groups was statistically significant (P < 0.001, test of homogeneity, χ²). No geographical differences were noted (P = 0.20, test of homogeneity, χ²). Among those with HRVA, unilateral HRVA was present in 70.3% (95% CI, 65.2%-75.2%) and bilateral in 29.7% (95% CI, 24.8%-34.8%). No left or right side predilection was found (left, 50.8%; 95% CI, 33.8%-67.6%; right, 49.2%; 95% CI, 32.4%-66.2%).

Conclusions: Craniocervical fusion should be preceded by examination of the vertebral arteries at the level of C2 because the presence of HRVAs is common and might preclude the safe insertion of transarticular or transpedicular screws.