Atlantoaxial Instability in Collagen Type 2 Skeletal Dysplasias: Surgical Indications and Recommendations

Objective: Cervical spine atlantoaxial instability (AAI) is frequently encountered in skeletal dysplasias (SD), particularly in patients with collagen type 2 (COL2) mutations. The values of clinical examination, radiographic measurements, and magnetic resonance imaging (MRI) findings in determining when to intervene on COL2 AAI are unknown. The goal of this study is to compare these tools and report the utility and significance of each in surgical decision-making. Methods: A retrospective case-control series from a single center's skeletal dysplasia database, 2007 to 2023, included COL2 patients that had documented history, examination, and flexion-extension (F-E) cervical spine radiographs and MRI. Radiographic measurements included dens morphology, anterior atlanto-dens interval (AADI), and posterior atlanto-dens interval (PADI). MRI measurements included cervical stenosis, C1 space available for the cord (SAC), and presence of myelomalacia. Criteria cutoffs were defined by receiver operating characteristic analysis and a scoring rubric was generated by internal validation. Results: Of all SD patients, 78 of 547 (14%) had COL2 syndromes, and 53 (68%) met all inclusion criteria. The most common diagnosis was spondyloepiphyseal dysplasia (65%). Patients were majority 54% female and 80% white. Ten (19%) underwent surgery at an average age of 4.3 years. There were no differences in AAI clinical histories (P = 0.18). Physical examination revealed higher rates of hyperreflexia in the surgical group (P = 0.0002). On radiographs, surgical patients had more os odontoidea (P = 0.0001) and greater change in AADI and PADI (P = 0.01 and P < 0.0001). On MRI, surgical patients had myelomalacia more frequently (P < 0.0001), more severe stenosis (P = 0.02), and greater change in SAC (P = 0.01). receiver operating characteristic analysis defined 7 surgical cutoffs as follows: (1) presence of hyperreflexia, (2) radiographic os odontoid, (3 and 4) F-E radiographs with ≥5 mm change in AADI and/or PADI, (5) myelomalacia on MRI, (6) C1-C2 stenosis ≥80%, and (7) ≥1.5 mm of C1-C2 SAC change on F-E MRI. Applying those cutoffs, it was determined that 100% of patients with 0 to 1 criteria were able to be managed nonoperative. Conversely patients with 4 or more criteria present were indicated for surgery in 80% of cases. Conclusion: This is the largest case series on AAI in SD patients with COL2-Associated conditions. This study identified 7 criteria that could be used to indicate surgical intervention. Patients who had ≤1 of these factors did not undergo surgery, while those with ≥4 had a high propensity to be recommended for surgical stabilization. Level of Evidence: Level III.