Spontaneous Intracranial Hypotension Due to Skull‐Base Cerebrospinal Fluid Leak

Orthostatic headaches developed in a 2-year-old boy. Magnetic resonance imaging (MRI) was normal (Fig, A) except for a subcranial cerebrospinal fluid (CSF) collection. At age 9 years, headaches worsened and MRI showed pachymeningeal enhancement and brain sagging (Fig, B). MRI and cisternography revealed a skull-base CSF leak with contrast filling the petrous apex air cells and tracking extracranially into the soft tissues of the neck (Fig, D–F) and eventually into the subclavian vein. The left skull base showed marked thinning of bony structures. The patient underwent a temporal craniotomy. The lateral squamous temporal bone was found to contain multiple bony defects leaking CSF, but no dural defects were identified. The bony defects were packed with autologous fat grafts. Headaches were unchanged, and he underwent a subcranial percutaneous n-butyl cyanoacrylate injection. Improvement of brain sagging was noted on follow-up MRI scan (Fig, C). High-pressure–type headaches and papilledema developed. CSF opening pressure was elevated (65cm H2O, normal = 6–25cm H2O). A ventriculoperitoneal shunt was placed and the high-pressure headaches and papilledema resolved. However, severe orthostatic headaches reoccurred 1 year later, and MRI showed worsening of the brain sagging. Digital subtraction myelogram then showed rapid flow of CSF entering the posterior cranial fossa from the cervical spinal canal and escaping extracranially at the level of the porus acusticus into the soft tissues of the neck (Fig, G–I; Video S1). No spinal CSF leak was demonstrated. The patient underwent a suboccipital craniotomy. Intraoperatively, the defect in the ventral wall of the porus acusticus could be palpated and dural substitute was placed within and around the defect. Headaches were unchanged and postoperative MRI showed mild improvement of brain sagging. If this procedure fails to correct the CSF leak long-term, then a subtotal petrosectomy and mastoidectomy is the next step. Only CSF leaks at the level of the spine have been shown to cause spontaneous intracranial hypotension (SIH).1 Previously, we have stated that CSF leaks originating from the skull base do not cause SIH.1 We now report a unique patient with SIH that was caused by a spontaneous posterior fossa skull-base CSF leak, representing a new mechanism for the development of SIH. This patient had a CSF leak through a defect in the posterior cranial fossa. The anatomy of the CSF leak was well visualized on digital subtraction myelography (DSM), which showed rapid flow of CSF entering the posterior cranial fossa from the cervical spinal canal and escaping extracranially at the level of the porus acusticus into the soft tissues of the neck, eventually draining into the subclavian vein. This is similar to a recently described type of spontaneous CSF leak in patients with SIH, the spinal CSF-venous fistula.2 Brain sagging is seen in about half of patients with SIH on MRI. Brain sagging also is a common finding in patients with lumboperitoneal shunts3 and also can be seen with CSF shunts from the posterior fossa to the peritoneum.4 This is a similar pathophysiologic mechanism to that seen in our patient, that is, bulk CSF flow from the posterior fossa diverting CSF from the spinal subarachnoid space. Nothing to report. VIDEO S1 Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.