Elucidating the pathophysiology of syringomyelia.

OBJECT Syringomyelia causes progressive myelopathy. Most patients with syringomyelia have a Chiari I malformation of the cerebellar tonsils. Determination of the pathophysiological mechanisms underlying the progression of syringomyelia associated with the Chiari I malformation should improve strategies to halt progression of myelopathy. METHODS The authors prospectively studied 20 adult patients with both Chiari I malformation and symptomatic syringomyelia. Testing before surgery included the following: clinical examination; evaluation of anatomy by using T1-weighted magnetic resonance (MR) imaging; evaluation of the syrinx and cerebrospinal fluid (CSF) velocity and flow by using phase-contrast cine MR imaging; and evaluation of lumbar and cervical subarachnoid pressure at rest, during the Valsalva maneuver, during jugular compression, and following removal of CSF (CSF compliance measurement). During surgery, cardiac-gated ultrasonography and pressure measurements were obtained from the intracranial, cervical subarachnoid, and lumbar intrathecal spaces and syrinx. Six months after surgery, clinical examinations, MR imaging studies, and CSF pressure recordings were repeated. Clinical examinations and MR imaging studies were repeated annually. For comparison, 18 healthy volunteers underwent T1-weighted MR imaging, cine MR imaging, and cervical and lumbar subarachnoid pressure testing. Compared with healthy volunteers, before surgery, the patients had decreased anteroposterior diameters of the ventral and dorsal CSF spaces at the foramen magnum. In patients, CSF velocity at the foramen magnum was increased, but CSF flow was reduced. Transmission of intracranial pressure across the foramen magnum to the spinal subarachnoid space in response to jugular compression was partially obstructed. Spinal CSF compliance was reduced, whereas cervical subarachnoid pressure and pulse pressure were increased. Syrinx fluid flowed inferiorly during systole and superiorly during diastole on cine MR imaging. At surgery, the cerebellar tonsils abruptly descended during systole and ascended during diastole, and the upper pole of the syrinx contracted in a manner synchronous with tonsillar descent and with the peak systolic cervical subarachnoid pressure wave. Following surgery, the diameter of the CSF passages at the foramen magnum increased compared with preoperative values, and the maximum flow rate of CSF across the foramen magnum during systole increased. Transmission of pressure across the foramen magnum to the spinal subarachnoid space in response to jugular compression was normal and cervical subarachnoid mean pressure and pulse pressure decreased to normal. The maximum syrinx diameter decreased on MR imaging in all patients. Cine MR imaging documented reduced velocity and flow of the syrinx fluid. Clinical symptoms and signs improved or remained stable in all patients, and the tonsils resumed a normal shape. CONCLUSIONS The progression of syringomyelia associated with Chiari I malformation is produced by the action of the cerebellar tonsils, which partially occlude the subarachnoid space at the foramen magnum and act as a piston on the partially enclosed spinal subarachnoid space. This creates enlarged cervical subarachnoid pressure waves that compress the spinal cord from without, not from within, and propagate syrinx fluid caudally with each heartbeat, which leads to syrinx progression. The disappearance of the abnormal shape and position of the tonsils after simple decompressive extraarachnoidal surgery suggests that the Chiari I malformation of the cerebellar tonsils is acquired, not congenital. Surgery limited to suboccipital craniectomy, C-I laminectomy, and duraplasty eliminates this mechanism and eliminates syringomyelia and its progression without the risk of more invasive procedures.