Role of spinal MRI for pre-operative work up in patients with adolescent idiopathic scoliosis: a retrospective case study and narrative review

There remains variability in the acquisition of whole-spine MRI prior to surgical correction in patients with adolescent idiopathic scoliosis (AIS). In this study, we take a retrospective look at the clinical impact of uniformly obtaining spinal MRI on all patients with a diagnosis of AIS.

Three hundred thirty four patients with presumed AIS who underwent surgery between 2017 and 2022 were identified; 283 of these patients who were asymptomatic and had a preoperative MRI in the database were included. We investigated radiographic, demographic, and clinical risk factors for the presence of neural axis anomalies. Radiologists’ reports were reviewed to determine the level of the conus medullaris and the presence of any intra-spinal dysraphisms. The utility of known risk factors for neural axis anomalies and the utility of MRI identifying anomalies on clinical decision-making, intra-operative neuromonitoring (IONM) alerts, and postoperative neurologic insult in asymptomatic AIS patients were investigated.

There were 283 patients with a mean age of 14.1 years, 67 males (22.6%) and 26 (9.2%) left-sided thoracic curves. MRI identified nine patients with central cord dilations, four patients with low cerebellar tonsils, four patients with Chiari Malformations, five patients with low-lying conus, one patient with a tethered cord, and five patients with arachnoid cysts. Six (2.1%) of the 283 patients underwent a neurosurgical intervention and 3 (50%) of the 6 went on to have IONM alerts. Eighteen (6.4%) of the 283 patients had IONM alerts and 5 (28%) of the 18 had neural axis anomalies. None of the patients with IONM alerts had a permanent neurologic deficit post-operatively.

In asymptomatic patients presumed to have AIS, 9.9% had a neuro-axis anomaly, 2.1% needed a neurosurgical intervention, and 6.4% of the patients had an IONM alert. We did not find known risk factors for identifying neural axis anomalies to have statistically significant application in asymptomatic AIS patients. Presence of neuroaxis anomalies increased the chance of having IONM alerts, however, we were unable to demonstrate if neurosurgical interventions in these patients with anomalies were preventative for IONM alerts or have statistically significant protection against clinical complications.