A Novel Classification of Cervical Spine Trauma in Ankylosing Spondylitis and Corresponding Surgical Outcomes

OBJECTIVE: Currently, there are no reports on the specific classification of cervical spine trauma (CST) in ankylosing spondylitis (AS) based on the trauma mechanism. In this study, we aimed to describe a novel classification of CST in AS with more details, and put forward the corresponding surgical outcomes related to different types, hoping to provide a practical reference for clinical decision‐making and academic communication. METHODS: From January 2008 to December 2021, AS patients who experienced CST were retrospectively reviewed and included. Clinical data including gender, age, reason of trauma, time interval between AS diagnosis and trauma were collected. The American Spinal Injury Association (ASIA) grade system was used to describe patients' neurological status. Based on the combination of surgical experience and follow‐up observation, the lower cervical spine trauma in AS patients was divided into three main types, namely single level fracture‐dislocation (type 1), spinal cord injury without fracture‐dislocation (type 2), and Andersson lesion (type 3). Furthermore, we performed detailed subtypes according to whether cervical spine was completely fused and the location of injury. Meanwhile, according to different approaches, surgical methods mainly included Anterior Cervical Discectomy and Fusion (ACDF), Anterior Cervical Corpectomy and Fusion (ACCF), Posterior Expansive Open‐door Cervical Laminoplasty (PEOLP), Posterior Cervical Laminectomy Decompression and Fusion (PCLDF), and their combination. Postoperative general and surgery‐related complications were also recorded. RESULTS: A total of 102 patients were enrolled, including 91 males and 11 females, with an average age of 51.9 years. Their average interval time between AS diagnosis and injury was 27.8 years. Patients with high‐energy and low‐energy trauma were 54 and 48 respectively. There were 79 patients suffering spinal cord nerve impairment after trauma. With regard to the distribution of different types, the number of patients in type 1, type 2, and type 3 were 86, 14, and two, respectively. For different types, PCLDF was the most commonly used surgical method, accounting for 55.9%, while ACCF was only applied for one time. In type 1, the frequencies of ACDF, ACCF, PCLDF, and ACDF+PCLDF were 10.5%, 1.2%, 55.8%, and 32.5%. In type 2, the frequencies of ACDF, PCLDF, ACDF+PCLDF, and PEOLP were 7.1%, 50.0%, 7.1%, 35.8%. Postoperatively, 21 patients achieved neurological function improvement. The incidences of general and surgery‐related complications were 19.6% and 5.9%, respectively. All patients achieved bone fusion and durable decompression at the last follow‐up. CONCLUSIONS: Our novel classification could enrich the scope of CST in AS patients and provide valuable references to the corresponding clinical management. Besides, there are strict indications of different surgical methods, factors like patient's physical condition, trauma type, surgical purpose, and expected efficacy were all required to consider before making a clinical decision.