PurposeDermal sinus tract (DST) is a challenging clinical diagnosis in children. The purpose of our study was to analyze the added value of magnetic resonance imaging (MRI) in the diagnosis of DST involving the lumbosacral spine. We also sought to establish an MRI-based scoring system to simplify the diagnosis of DST.MethodsMRI images of 20 patients with clinically suspected DST were retrospectively assessed by two neuroradiologists blinded to the surgical results. The MRI studies were performed from July 2003 to July 2013. Institutional Review Board (IRB) approval was obtained. All MRI studies were assessed with respect to five imaging signs: A) visualization on both sagittal and axial images, B) dural penetration, C) associated tethered cord, D) presence of tumor or inflammation, and E) attachment to conus medullaris. The frequency of each imaging sign in the study population was calculated. For the 20 patients who underwent surgery, sensitivity and specificity of each neuroradiologist’s interpretation of the MRI data were calculated using operative findings as the gold standard.ResultsTwelve of the 20 had confirmed DSTs. The incidences of the five imaging signs were as follows: A) visualization on both sagittal and axial images (12/12, 100%), B) dural penetration (10/12, 83.3%), C) associated tethered cord (7/12, 58.3%), D) presence of tumor or inflammation (4/12, 33.3%), and E) attachment to the conus medullaris (4/12, 33.3%). The best combination of findings predictive of DST was simply the appearance of DST on both axial and sagittal imaging, which resulted in a sensitivity of 100% and a specificity of 75-100%.ConclusionVisualization of DST on both axial and sagittal imaging is the best marker for pathology proven DST. Using a higher threshold score on the five-point scoring system that we proposed did not increase sensitivity or specificity in the diagnosis of DST; however, it may still prove clinically helpful in standardizing reporting leading to a more accurate and detailed assessment.