Use of prone position magnetic resonance imaging for detecting the terminal filum in patients with occult tethered cord syndrome

Abstract

Object The pathophysiology of occult tethered cord syndrome (OTCS) with no anatomical evidence of a caudally shifted conus and a normal terminal filum is hard to understand. Therefore, the diagnosis of OTCS is often difficult. The authors hypothesized that the posterior displacement of the terminal filum may become prominent in patients with OCTS who are in a prone position if filum inelasticity exists, and they investigated prone-position MRI findings. Methods Fourteen patients with OTCS and 12 control individuals were examined using T2-weighted axial MRI with the patients in a prone position on a flat table. On each axial view, the distance between the posterior and anterior ends of the subarachnoid space (A), the distance between the posterior end of the subarachnoid space and the terminal filum (B), the distance between the posterior end of the subarachnoid space and the dorsal-most nerve among the cauda equina (C), and the distance between the posterior end of the subarachnoid space and the ventral-most nerve (D) were measured. The location ratios of the terminal filum, the dorsal-most nerve, and the ventral-most nerve were calculated by the ratio of A to B (defined as TF = B/A), A to C (defined as DN = C/A), and A to D (defined as VN = D/A), respectively. Patients underwent sectioning of the terminal filum with the aid of a surgical microscope. The low-back pain Japanese Orthopaedic Association score was obtained before surgery and at the final follow-up visit. Results On prone-position axial MRI, the terminal filum was separated from the cauda equina and was shifted caudally to posterior in the subarachnoid space in all patients with OTCS. The locations of the caudal cauda equina shifted to ventral in the subarachnoid space. The TF values in the OTCS group were significantly lower than those in the control group at the L3–4 (p = 0.023), L-4 (p = 0.030), L4–5 (p = 0.002), and L-5 (p < 0.001) levels. In contrast, the DN values in the OTCS group were significantly higher than those of the control group at the L-2 (p = 0.003), L2–3 (p = 0.002), L-3 (p < 0.001), L3–4 (p < 0.001), L-4 (p = 0.007), L4–5 (p = 0.003), and S-1 (p = 0.014) levels, and the VN values in the OTCS group were also significantly higher than those of the control group at the L2–3 (p = 0.022), L-3 (p = 0.027), L3–4 (p = 0.002), L-4 (p = 0.011), L4–5 (p = 0.019), and L5–S1 (p = 0.040) levels. Sections were collected during surgery for histological evaluation, and a decreased elasticity within the terminal filum was suggested. Improvements in the Japanese Orthopaedic Association score were observed at the final follow-up in all patients. Conclusions The authors' new method of using the prone position for MRI shows that the terminal filum is located significantly posterior and the cauda equina is located anterior in patients with OTCS, suggesting a difference in elasticity between the terminal filum and cauda equina.