Comparison of Cobb Angles on Radiographs With Magnetic Resonance Imaging in Idiopathic Scoliosis

Patients with idiopathic scoliosis are exposed to repetitive x-ray for angle measurement. Therefore, the discovery or development of alternative techniques with less radiation has continuously been a major concern.

In this study, we compared the Cobb angles on supine Magnetic Resonance Imaging (MRI) with those on standing plain radiographs to figure out how precisely the supine MRI can show the real curve.  Consequently, the need for certain exposures throughout the management of idiopathic scoliosis might be eliminated. 

A total of 103 patients with idiopathic scoliosis were included in this prospective study. The standing radiographs and supine MRI were obtained with less than a 1-month time lag. One senior author assessed Cobb angles of the major curves were on both standing radiographs and MR images. All the eligible patients had already signed the consent for diagnostic imaging, including MRI. The individuals, who were not requested for x-ray and MRI, were excluded from this study.

The Mean±SD Cobb angle was 55.5±11.2°‎‏ on the standing plain radiographs and  44.5±10°‎‏ on MR images (P<0.001). The Mean±SD difference between the Cobb angles on the standing plain radiographs and MR images was 11±1.4°‎‏. A significant positive correlation was found between the Cobb angles calculated on plain radiographs and MRI (r=0.996, P<0.001). Accordingly, Cobb angles on MRI could be converted to Cobb angles on plain radiographs under the formula of MRI=0.9* XRAY-5.31 (absolute error of 5.31°‎‏).

Cobb angles on supine MRI correlates with measured ones on standing radiographs with an acceptable range of error and could be used as a valuable alternative for radiographic Cobb angle measurement.