In the April 15 issue, Dr. Diab and colleagues presented the largest series of pre-operative MRIs in idiopathic scoliosis (IS) ever published. Their aim was to determine who should be screened with MRI prior to surgery for idiopathic scoliosis. Of over 2,200 pediatric patients undergoing fusion for IS, 923 underwent a pre-operative MRI. This is nearly three times the number of MRIs in the second largest series published, and a study of this size—if not even larger—is needed in order to make sufficiently powered conclusions about risk factors for neural abnormalities given the rarity of these findings. 1 This study confirmed that juvenile IS patients and those with thoracic hyperkyphosis were at increased risk for having a syrinx, though only 2.7% of these “higher risk” patients had syringes (compared to 1% of all other patients). Surprisingly, male patients and those with left thoracic curves—groups suggested as being at higher risk of neural abnormality in the past—did not have a higher rate of abnormal MRIs. Unfortunately, the authors did not show the actual number of patients in these subgroups with abnormal findings. It is certainly possible that the study was underpowered to detect differences among these subgroups since both the risk factors (male gender, left thoracic curves) and the neural abnormalities were uncommon in this study.
The major limitation of this study is its risk for selection bias on two fronts. The first concern is that the patients undergoing MRI were pre-selected for being at risk by the treating surgeon, and the MRI cohort had a higher rate of thoracic hyperkyphosis, juvenile IS, immature Risser stage, severe curve magnitude, three major curves, male patients, and AP surgery compared to the group that did not get an MRI. If the goal is to determine risk factors for neural abnormalities to determine who should get a pre-operative MRI, all patients should be screened. Otherwise, the relative risk of neural abnormality for the various characteristics can only be determined for a population already selected for being at risk rather than the entire population of IS patients. Another aspect of this study that introduces selection bias is that all patients undergoing surgery for a neural abnormality—presumably diagnosed with MRI--were excluded from the study, and the authors did not report how many such patients were excluded. Excluding these patients could have certainly skewed the results if certain risk factors (i.e. left thoracic curve) were associated with needing to undergo a neurosurgical procedure.
How should this paper affect our practice? It certainly suggests that MRI is probably not necessary for all patients undergoing surgery for IS. Unfortunately, it does not offer much guidance in terms of who does need an MRI. Should we get MRIs on all hyperkyphotic patients with normal neurological examinations just because they have a 2.7% rate of syrinx? Despite this being the largest study on this topic ever published, we still do not have answers to these questions. This paper underscores the huge number of patients needed to be studied in cases where the outcome of interest is rare. Whether or not a massive study that obtains pre-operative MRIs on all IS patients will ever be performed remains to be seen. Until then, the indications for pre-operative MRI in IS will remain controversial. Please read Dr. Diab’s article and commentary. Let us know if this will affect your practice by posting a comment on The Spine Blog.
Adam Pearson, MD, MS
1. Do T, Fras C, Burke S, Widmann RF, Rawlins B, Boachie-Adjei O. Clinical value of routine preoperative magnetic resonance imaging in adolescent idiopathic scoliosis. A prospective study of three hundred and twenty-seven patients. J Bone Joint Surg Am 2001;83-A:577-9.