Staging Lumbar Spondylolysis in Adolescents: Can Magnetic Resonance Bone Imaging Replace Computed Tomography?

Retrospective, single-center, observational study. To evaluate the diagnostic reliability of magnetic resonance (MR) bone imaging in stage classification of adolescent lumbar spondylolysis, assessing its clinical applicability as an alternative to computed tomography (CT). The diagnosis of adolescent lumbar spondylolysis typically requires CT; however, concerns regarding medical radiation exposure have been raised. With recent advancements in magnetic resonance imaging (MRI) technology, MR bone imaging has been developed, enabling the acquisition of CT-like images. This study included 116 adolescent patients diagnosed with lumbar spondylolysis between August 2022 and April 2024. In addition to conventional MRI, MR bone imaging and CT scans were performed on the affected vertebrae. Two independent raters assessed sagittal images and classified the fractures using the Oba classification system. MRI examinations were performed using either a 1.5T or 3T scanner. For statistical analysis, CT classification was used as the reference standard. Interrater and intermodality agreements for both CT and MR bone imaging were evaluated using the weighted kappa coefficient (linear weights). Among the 116 patients, 70 cases (140 pars) underwent 3T MRI, while 46 cases (92 pars) underwent 1.5T MRI. In the 3T MRI group, the interrater agreement was κ = 0.98 for CT and κ = 0.91 for MR bone imaging, while intermodality agreement was κ = 0.89 for rater A and κ =0.83 for rater B. In the 1.5T MRI group, interrater agreement was κ = 0.94 for CT and κ = 0.92 for MR bone imaging, while intermodality agreement was κ = 0.83 for rater A and κ = 0.84 for rater B. These results indicate that MR bone imaging provides stage classification accuracy comparable to that of CT in the evaluation of adolescent lumbar spondylolysis. Furthermore, similar findings were observed for both 1.5T and 3T MRI.