7.7% Prevalence of neural axis abnormalities on routine magnetic resonance imaging in patients with presumed adolescent idiopathic scoliosis scheduled for spine surgery: a consecutive single surgeon retrospective cohort of 182 patients

Purpose The purpose of this study was to use a Delphi analysis to identify a clinically relevant threshold for the prevalence of neural axis abnormalities (NAAs) that would warrant routine preoperative screening. Methods A panel of experienced physicians specializing in pediatric spine surgery, pediatric neurosurgery, and pediatric neuroradiology was formed to establish consensus using a Delphi process to identify a minimum prevalence of NAAs that would initiate the implementation of preoperative MRIs as standard of care. Following the Delphi analysis, patients scheduled for PSF (Posterior Spinal Fusion) from 2010 to 2018 were retrospectively identified. Patients were included based on the following criteria: (1) 10–18 years old at time of MRI (inclusive), (2) AIS diagnosis prior to preoperative MRI (no concerning curve pattern, rate of progression, or neurologic signs/symptoms to suggest alternative diagnosis to AIS), and (3) standard preoperative MRI of the cervical, thoracic, and lumbar spine undergone. The prevalence of NAAs on preoperative MRI was recorded for all patients. Results There were 182 eligible patients. 14 had NAAs on MRI. The prevalence of NAAs was 7.7% [95% CI 4.27–12.57%]. This prevalence was significantly [ p < 0.0001] higher than the clinically relevant threshold of 1.3% established by the Delphi panel. Of the 14 patients with NAAs noted on preoperative MRI, neurosurgical intervention was recommended for 4 patients, 2.2% [95% CI 0.6–5.5%] of the total cohort. Conclusions Delphi panelists reported a low tolerance for NAAs among patients undergoing PSF for presumed AIS. Group consensus recommended routine screening should be implemented if the prevalence of NAAs is greater than 1.3%. The prevalence of NAAs in our cohort as well as related studies was significantly higher than this threshold. Level of evidence Diagnostic—level III.