Background and purpose: Phase-contrast MR imaging (PCMR) has only partially characterized cyclic CSF flow and pressure, which, hypothetically, have a role in the pathogenesis of syrinx and symptoms in the Chiari I malformation. Our goal was to use computational flow analysis (CFA) to better understand CSF hydrodynamics.
Materials and methods: High-resolution MR images were obtained in a healthy volunteer and a patient with Chiari I malformation. With standard segmentation and discretization techniques, 3D models of the subarachnoid space, cerebellum, and spinal canal were created. CSF flow during systole and diastole were simulated with the boundary element method in the models. CSF velocities and pressures computed in the patient with Chiari I malformation were compared with those in the healthy volunteer. Flow patterns were also compared with PCMR results for validation of the technique.
Results: The CFA and PCMR results agreed well. Inhomogeneous flow patterns characterized by fluid jets anterior and lateral to the spinal cord were demonstrated in both the Chiari I and volunteer models by CFA. Significant circumferential velocities were evident, suggesting swirling flow in the spinal canal. Higher magnitude jets were found in the patient with Chiari I than in the healthy volunteer. Relatively even pressure gradients were found along the spinal canal in both cases, with a 50% steeper gradient in the patient with Chiari I malformation.
Conclusions: Circumferential velocities and pressure gradients in the spinal canal, which may be clinically relevant to Chiari I and other malformations, can be obtained by CFA in patient-specific geometries.