Study Design/Setting. Retrospective study at a single academic tertiary care center Objective. To evaluate osseous and vascular structures at C1 on preoperative MRIs and CTAs. Background Data. No prior studies have compared the diagnostic accuracy between magnetic resonance imaging (MRI) and computed tomography angiography (CTA) to assess C1 and internal carotid artery (ICA) anatomy prior to surgical instrumentation. Methods. We retrospectively reviewed adult patients who underwent both cervical spine CTA and MRI between 2007 and 2018. Patients with prior cervical surgery or MRIs not extending to the atlas were excluded. Ten standardized osseous and vascular measurements were performed on both modalities using anatomical landmarks at C1. Paired t-tests and intraclass correlation coefficients (ICCs) assessed differences and agreement. A sub-analysis normalized measurements to anterior-posterior C1 length to account for potential modality-based scaling differences. Results. Of 209 patients reviewed, 119 met inclusion criteria (mean age 65.1 years; 56% female). The agreement between CTA and MRI across 10 anatomical measurements was low, with ICCs ranging from 0.006 to 0.427. All measurements except the distance from the end of the ideal screw trajectory to the anterior plane of C1 demonstrated statistically significant differences between CTA and MRI ( P <0.05). After standardizing values to anterior-posterior C1 length, ICCs for the nine measurements remained low (0.012 to 0.305), with only standardized measurements, distance from the end of the ideal screw trajectory to the anterior plane of C1 and the distance from the origin of the ideal screw trajectory to the ICA showing no statistically significant differences. Conclusions. This study evaluated the accuracy of preoperative CTA and MRI in assessing C1 and ICA anatomy for atlantoaxial fusion planning, revealing variability in measurement agreement between the two. The findings highlight the need to tailor imaging choices to each clinical scenario, balancing diagnostic value, radiation exposure and surgical risk.
Arzani et al. (Mon,) studied this question.