Introduction
Background
Cerebrospinal fluid (CSF) leak may occur from the nose (rhinorrhea), from the external auditory canal (otorrhea), or from a traumatic or operative defect in the skull or spine. The fluid leak is a result of meningeal dural and arachnoid laceration with fistula formation. Blunt trauma is the most common cause.
Related eMedicine topics:
CSF Rhinorrhea
Skull Base, CSF Otorrhea
Related Medscape topics:
Resource Center Otitis Media
Resource Center Vascular Surgery
Resource Center Spinal Disorders
Resource Center Stroke/Cerebrovascular Disease
Resource Center Headache
CME Prediction Rule Identifies Risk for Bacterial Meningitis in Children in the Emergency Department
CME Severe Traumatic Brain Injury: Evolution and Current Surgical Management
Pathophysiology
Normal adult subarachnoid fluid has a circulating volume of 90-150 mL. Approximately 500 mL of cerebrospinal fluid (CSF) is produced daily, primarily from the ventricular choroid plexuses. Circulating CSF is absorbed into the venous circulation, mainly through the cranial arachnoid granulations and spinal arachnoid villi.
Normal CSF pressure is 100-200 mm of water. The normal CSF protein content is 20-45 mg/dL, and the normal CSF glucose range is 50-100 mg/dL, which is 60% of the measured serum glucose value. However, nasal mucous secretions and tears also have detectable glucose content. Therefore, tests used to identify CSF by its glucose content are often false positive (in 45-75% of cases). The absence of glucose tends to exclude CSF as the leaking fluid.
The enzyme beta-2 transferrin (B2Tr) is produced in the brain by neuraminidase activity and is present in CSF, perilymph, and ocular aqueous humor but not in sinonasal mucous secretions and tears.1 This feature is the basis for a specific test for CSF based on immunoelectrophoresis. B2Tr may be found in blood serum in liver disease, such as in chronic alcoholism and in patients with inborn errors of glycoprotein metabolism or genetic variants of transferrin. Beta-trace protein is prostaglandin D2 synthase. It is produced in epithelial cells of the choroid plexus and meninges and is found in CSF, perilymph, seminal fluid, and urine. It is approximately 35 times more concentrated in CSF than in blood serum. Immunoelectrophoretic assay of beta trace protein has been reported to have high specificity and sensitivity for CSF detection.2,3
CSF rhinorrhea
Causes of CSF rhinorrhea include (1) blunt head trauma; (2) sequelae of skull-base surgery,4 commonly functional endoscopic sinus surgery (FESS), transsphenoidal pituitary surgery, translabyrinthine acoustic schwannoma, and mastoid surgery with intact tympanic membrane; (3) destructive skull-base lesions, including neoplasms (both benign and malignant), and empty sella; (4) developmental defects of the ethmoid, sphenoid, frontal, or petrous temporal bones with the formation of a meningocele or meningoencephalocele (with an intact tympanic membrane); and (5) fracture of the petrous temporal bone or other destructive processes in which CSF in the middle ear drains to the nose in the presence of an intact tympanic membrane.5
Less than 5% of all cases of CSF rhinorrhea are spontaneous. Most cases of CSF rhinorrhea begin soon after a head injury and cease spontaneously within 7-180 days.
CSF otorrhea
CSF otorrhea occurs in the presence of a perforated tympanic membrane in the following settings: (1) fractures of the petrous temporal bone,6 (2) translabyrinthine and/or mastoid surgery, (3) developmental defects of the tegmen tympani or petrous apex with meningocele formation and spontaneous or posttraumatic meningeal laceration, (4) perilymphatic fistula from trauma with stapes fracture and torn round or oval window membrane, (5) translabyrinthine fistula due to the Mondini developmental defect of the cochlear modiolus and/or lamina cribrosa,7 and (6) wide cochlear aqueduct syndrome.
With a translabyrinthine fistula, CSF mixes with perilymph in the cochlea or vestibule and forms perilymphatic hydrops with displacement or perforation of the maldeveloped stapes footplate; the fluid leaks into the middle ear.8 Wide cochlear aqueduct syndrome is a controversial and doubtful entity in adult patients because the aqueduct is filled with fibrous tissue and not functional beyond childhood.
Pneumocephalus
Pneumocephalus can occur in up to one third of all patients with posttraumatic or spontaneous CSF leak.9 This condition is likely the result of the pressure gradient created during respiration, sneezing, or nose blowing.
Spinal CSF leak
Spinal CSF leaks can occur as a result of (1) blunt or penetrating trauma10 ; (2) postoperative sequela with leakage through a dural tear or incision; (3) lumbar puncture11 ; (4) inadvertent meningeal puncture during epidural anesthesia; (5) spontaneous leakage from 1 or more spinal nerve root sleeves, particularly in the thoracic and lumbar areas; and (6) Valsalva maneuver during excessive weightlifting.
Increased intracranial pressure facilitates the development of CSF leaks. Meningeal dysplasia (as in Marfan syndrome) may also contribute to the development of CSF leak in some patients.
Spontaneous intracranial hypotension syndrome
Spontaneous intracranial hypotension syndrome (SIHS) can result from a persistent CSF leak. SIHS is usually spinal and seldom originates from the skull base (eg, ethmoidal defects). Frequently, SIHS and persistent orthostatic headache after lumbar punctures can be successfully treated by lumbar epidural blood patch.12,13,14
Frequency
United States
Cerebrospinal (CSF) rhinorrhea occurs in 2-6% of patients with head injury. Rhinorrhea or otorrhea occurs in up to 30% of patients with a skull-base fracture. Head trauma accounts for 50-80% of all cases of CSF leak, and up to 16% are iatrogenic.
Postoperative CSF leak has been noted in 0.5-15.0% of patients with transsphenoidal surgery, particularly after reparative operations.15 CSF leak has been reported in 5.0-12.5% of translabyrinthine acoustic schwannoma surgeries. Functional endoscopic sinus surgery (FESS) is a common procedure, with CSF leak occurring in 1.0-2.5%, but 90% of leaks are detected and repaired intraoperatively.
About 4% of CSF leaks are of spontaneous and nontraumatic causes (eg, developmental skull-based defects with meningocele, skull-base tumor, empty sella,16 osteomyelitis).
Mortality/Morbidity
Meningitis occurs in 25-50% of untreated traumatic cerebrospinal fluid (CSF) fistulas and in 10% of patients in the first week after trauma with head injury. The incidence of meningitis up to several years after spontaneous cessation of posttraumatic CSF leak is 10%. Meningitis-related mortality rates up to 20% have been reported, particularly when meningitis is due to antibiotic-resistant organisms.
- 50-85% of traumatic CSF leaks resolve spontaneously within 7 days, and almost all leaks cease within 6 months.
- CSF fistula of spontaneous origin is often intermittent, persisting in at least 60% of cases if untreated. The risk of meningitis is higher with spontaneous CSF fistula than with traumatic CSF fistula.
- CSF otorrhea after trauma ceases spontaneously more often than traumatic rhinorrhea and seldom recurs.
Sex
- No sexual predilection is reported for traumatic cerebrospinal fluid (CSF) fistula.
- For postsurgical and spontaneous CSF leaks, the female-to-male ratio is 2:1.
Age
- No age predilection is reported for traumatic cerebrospinal fluid (CSF) fistula.
- Postsurgical and spontaneous CSF leaks are most common in adults older than 30 years.
Anatomy
Cerebrospinal fluid (CSF) leak resulting from trauma occurs usually with fractures of ethmoid, sphenoid, or petrous temporal bones.17 The ethmoid bones are particularly vulnerable to trauma. The orbital plates of the frontal bone do not cover the ethmoid bones completely; therefore, the thin and perforated cribriform plates are partially unprotected.
The dura is thinnest at and adherent to the cribriform plates and adjacent ethmoid sinus medial segments. The anterior ethmoidal arteries course in grooves on the surface of the ethmoid bones. In addition, multiple developmental defects occasionally occur in the sphenoid bone and in the floor of the middle cranial fossa. Because of these vascular grooves in the ethmoid bones and cribriform plates and because multiple anatomic defects are frequently present, it is sometimes difficult to demonstrate a fracture or developmental defect of the ethmoid bone in association with a CSF fistula.
Presentation
CSF leak after trauma
Cerebrospinal fluid (CSF) leak from the fistula occurring after head trauma consists of watery, blood-stained fluid that abruptly leaks from 1 or both nostrils or an external auditory canal. Two thirds of patients with these fistulas present within 48 hours of their head injury. Almost all of these fistulas occur within 3 months of injury. Occasionally, the fistula appears many months or years after injury, with a sudden gush of fluid or meningitis; these episodes are sometimes recurrent. CSF leak occurs often without nasal congestion, sneezing, lacrimation, or aural discharge.
Rhinorrhea may occur intermittently and can increase on bending forward, with a Valsalva maneuver or jugular vein compression. Nasal vasoconstrictor or antihistamine therapy does not affect the leak. Headache is sometimes but not always present. The patient may have physical signs of skull-base fracture, including periorbital ecchymosis or edema, mastoid-area skin ecchymoses, and cranial nerve deficits.
Frontal trauma may result in anosmia from an injury to the olfactory nerves, tracts, or orbitofrontal cortex; visual deficit from an injury to the optic nerve, optic globe, or extraocular muscle; or fractures of the orbit medial wall and floor. A fracture of the temporal bone may be associated with a blood clot and hemorrhagic fluid in the external auditory canal, a perforated tympanic membrane, and conductive or sensorineural hearing loss. Transverse fractures of the petrous temporal bone result in injury to cranial nerves VII and VIII in 50% of patients. In addition, deafness may result from ossicular disruption. Longitudinal fractures of the temporal bone result in injury to cranial nerve VII in 25% of patients.
Labyrinthine injury may result in vertigo. Otitis media and meningitis may occur. Increased intracranial pressure and hydrocephalus may prolong a CSF leak that might otherwise cease spontaneously.
A ventriculostomy catheter in a patient with CSF leak is associated with an increased incidence of meningitis.18
Pneumocephalus
Pneumocephalus can sometimes give an audible succussion splash with shaking of the head. Tension pneumocephalus occurs in rare cases and can result in an emergency with an acute change in the level of consciousness. The air must then be drained by inserting a needle through a twist drill hole in the cranial bone.
Spontaneous intracranial hypotension syndrome
Patients with spontaneous intracranial hypertension syndrome (SIHS) typically present with orthostatic headaches, which are maximal in intensity in the upright position and diminished in the recumbent position. Other symptoms include neck pain and stiffness, nausea, diplopia, dizziness, hearing loss, photophobia, visual field defects, facial numbness, and, occasionally, radicular pain in the arms. Patients may have a history of recent lumbar puncture, spinal trauma, or surgery. In SIHS, CSF pressure is usually 40-60 mm of water, but it is sometimes normal.
Preferred Examination
A suggested algorithm for the diagnosis of a cerebrospinal fluid (CSF) fistula follows.
- Confirm or exclude the presence of CSF in leaking fluid by means of an immunoelectrophoretic study of the fluid for beta-2 transferrin (B2Tr) or, where available, beta-trace protein. For this specialized laboratory study, 0.5-1.0 mL of the fluid may be required. An absorptive sponge pad placed at or near the presumed site of fluid leak can facilitate the collection of the fluid.
- Perform high-resolution, thin-section axial and coronal cranial and facial CT. Include all of the paranasal sinuses and petrous temporal bones in the scans.
- Perform magnetic resonance (MR) cisternography. This study may also be useful for detecting inactive fistulas.
- CT cisternography or radionuclide cisternography may be useful if CT and MR cisternography do not show the CSF fistula.
- Radionuclide cisternography may be useful to detect an intermittently active CSF fistula.
- Cisternography with an intrathecal injection of radioisotope or nonionic iodinated myelographic contrast medium or noninvasive MRI cisternography usually localizes the CSF leak.
- Brain and spinal MRI is useful in demonstrating meningocele and meningoencephalocele when associated with CSF leak, as well as for examining patients with spontaneous intracranial hypotension syndrome.
- On occasion, the methods listed above do not help in localizing the CSF fistula, and surgical exploration is necessary.
Fluid leaking from the nose or external auditory canal must first be positively identified as CSF. Drops of fluid from a CSF leak placed on absorbent filter paper may result in the double-ring sign, which is a central circle of blood and an outer clear ring of CSF. Results of glucose, chloride, and total protein tests of the fluid are not specific or conclusive for CSF.
All methods of cisternography—radionuclide, CT, and MR—provide improved or optimal CSF fistula detection when the fistula is active and when a Valsalva maneuver or jugular venous compression is added to the imaging protocol. CSF fistula can usually be demonstrated by using some method of cisternography, but localization of the leak to the right or left nasal cavity may be difficult because of the tendency of the fluid to cross sides and flow from both nostrils.
Methods for detecting CSF fistulas with intrathecal injections of dye pose a risk of chemical meningitis. Methylene blue, indigo carmine, and phenolsulfonphthalein (PSP) dyes are no longer in use. Some otolaryngologists use a dilute solution of fluorescein to localize CSF fistulas both preoperatively and during surgery. Typically, 0.5 mL of a 10% fluorescein solution is injected into the lumbar subarachnoid space over more than 1 minute. Cotton pledgets are placed in the nose, as for radionuclide cisternography. The dye reaches the skull base in 6 hours and is present over the cerebral convexities in 24 hours. The pledgets are examined for green fluorescence in a dark room with ultraviolet light 6 hours after the intrathecal PSP injection.
Limitations of Techniques
Skull radiographs are of limited diagnostic use in cerebrospinal fluid (CSF) leaks, but they may show a skull fracture or suggest the presence of empty sella.
Computer-reconstructed coronal images are less accurate and are acceptable only until direct coronal images can be obtained.
Differential Diagnoses
Other Problems to Be Considered
Acute or chronic rhinitis (allergic, infectious, vasoactive)
Perforated otitis interna (serous, catarrhal)
Otitis externa
Foreign body in the external auditory canal
More on Cerebrospinal Fluid, Leak |
 Overview: Cerebrospinal Fluid, Leak |
| Imaging: Cerebrospinal Fluid, Leak |
| Follow-up: Cerebrospinal Fluid, Leak |
| Multimedia: Cerebrospinal Fluid, Leak |
| References |
| Next Page » |
References
Warnecke A, Averbeck T, Wurster U, Harmening M, Lenarz T and Stover T. Diagnostic Relevance of Beta-2 Transferrin for the Detection of Cerebrospinal Fluid Fistulas. Arch Otolarygol Head and Neck Surg. Oct 2004;130:1178-1194.
Arrer E, Meco C, Oberascher G, Piotrowski W, Albegger K and Patsch W. B-Trace Protein as a Marker for Cerebrospinal Fluid Rhinorrhea. Clinical Chemistry. 2002;48:939-941.
Bachmann G, Nekic M and Michel O. Clinical Experience with Beta Trace Protein as a Marker for Cerebrospinal Fluid. Ann Otol Rhinol Laryngol. 2000;109:1099-1102.
Snow JB Jr, Ballenger JJ. Complications of skull base surgery. In: Ballenger's Otorhinolaryngology. 16th ed. Hamilton, Ontario: Decker;. 2003: 508-36; 684-7; 799.
Pearson BW. Cerebrospinal fluid rhinorrhea. In: Otolaryngology. 3rd ed. Philadelphia: W. B. Saunders Co;. 1991: 1899-909.
May JS, Mikus JL, Matthews BL, et al. Spontaneous cerebrospinal fluid otorrhea from defects of the temporal bone: a rare entity?. Am J Otol. Nov 1995;16(6):765-71. [Medline].
Burton EM, Keith JW, Linden BE, Lazar RH. CSF fistula in a patient with Mondini deformity: demonstration by CT cisternography. AJNR Am J Neuroradiol. Jan-Feb 1990;11(1):205-7. [Medline].
Rajkumar KT, Orabi AA, Timms MS. Spontaneous cerebrospinal fluid leak presenting as unilateral (left-sided) middle ear effusion. Ear Nose Throat J. Feb 2008;87(2):79-80. [Medline].
Ommaya AK. Cerebrospinal fluid fistula and pneumocephalus. In: Wilkins RH, Rengachary SS, eds. Neurosurgery. 2nd ed. New York: McGraw-Hill;. 1996: 2773-82.
Buchanan RJ, Brant A, Marshall LF. Traumatic cerebrospinal fluid fistulas. In: Winn HR, ed. Youmans Neurological Surgery. 5th ed. Philadelphia: W. B. Saunders Co;. 2004: 5265-72.
Clarot F, Callonnec F, Douvrin F, et al. Giant cervical epidural veins after lumbar puncture in a case of intracranial hypotension. AJNR Am J Neuroradiol. Apr 2000;21(4):787-9. [Medline].
Zaatreh M, Finkel A. Spontaneous intracranial hypotension. South Med J. Nov 2002;95(11):1342-6. [Medline].
Mokri B. Spontaneous intracranial hypotension. Curr Neurol Neurosci Rep. Mar 2001;1(2):109-17. [Medline].
Mokri B. Spontaneous intracranial hypotension. Curr Pain Headache Rep. Jun 2001;5(3):284-91. [Medline].
Shiley SG, Limonadi F, Delashaw JB, et al. Incidence, etiology, and management of cerebrospinal fluid leaks following trans-sphenoidal surgery. Laryngoscope. Aug 2003;113(8):1283-8. [Medline].
Schlosser RJ, Bolger WE. Spontaneous nasal cerebrospinal fluid leaks and empty sella syndrome: a clinical association. Am J Rhinol. Mar-Apr 2003;17(2):91-6. [Medline].
Burns BJ. Images in emergency medicine. Traumatic cerebrospinal fluid leak. Ann Emerg Med. Jun 2008;51(6):704, 706. [Medline].
Daudia A, Biswas D, Jones NS. Risk of meningitis with cerebrospinal fluid rhinorrhea. Ann Otol Rhinol Laryngol. Dec 2007;116(12):902-5. [Medline].
Shetty PG, Shroff MM, Sahani DV, Kirtane MV. Evaluation of high-resolution CT and MR cisternography in the diagnosis of cerebrospinal fluid fistula. AJNR Am J Neuroradiol. Apr 1998;19(4):633-9. [Medline].
Matsumura A, Anno I, Kimura H, et al. Diagnosis of spontaneous intracranial hypotension by using magnetic resonance myelography. Case report. J Neurosurg. May 2000;92(5):873-6. [Medline].
Jinkins JR, Rudwan M, Krumina G, Tali ET. Intrathecal gadolinium-enhanced MR cisternography in the evaluation of clinically suspected cerebrospinal fluid rhinorrhea in humans: early experience. Radiology. Feb 2002;222(2):555-9. [Medline].
Goel G, Ravishankar S, Jayakumar PN, Vasudev MK, Shivshankar JJ, Rose D. Intrathecal gadolinium-enhanced magnetic resonance cisternography in cerebrospinal fluid rhinorrhea: road ahead?. J Neurotrauma. Oct 2007;24(10):1570-5. [Medline].
El Gammal T, Sobol W, Wadlington VR, et al. Cerebrospinal fluid fistula: detection with MR cisternography. AJNR Am J Neuroradiol. Apr 1998;19(4):627-31. [Medline].
Albayram S, Kilic F, Ozer H, Baghaki S, Kocer N, Islak C. Gadolinium-enhanced MR cisternography to evaluate dural leaks in intracranial hypotension syndrome. AJNR Am J Neuroradiol. Jan 2008;29(1):116-21. [Medline].
Okizaki A, Shuke N, Aburano T, et al. Detection of cerebrospinal fluid leak by dual-isotope spect with In-111 DTPA and Tc-99m HMDP. Clin Nucl Med. Jul 2001;26(7):628-9. [Medline].
Lawrence SK, Delbeke D, Partain CL. Cerebrospinal fluid imaging. In: Diagnostic Nuclear Medicine. 4th ed. Baltimore: Lippincott, Williams & Wilkins. 2003: 835-9.
Duffy PJ, Crosby ET. The epidural blood patch. Resolving the controversies. Can J Anaesth. Sep 1999;46(9):878-86. [Medline].
Sencakova D, Mokri B, McClelland RL. The efficacy of epidural blood patch in spontaneous CSF leaks. Neurology. Nov 27 2001;57(10):1921-3. [Medline].
Abuabara A. Cerebrospinal fluid rhinorrhoea: diagnosis and management. Med Oral Patol Oral Cir Bucal. Sep 1 2007;12(5):E397-400. [Medline].
Bluestone CD, Strol SE, Alper LM. Otorrhea. Pediatric Otolaryngology. 4th ed. Philadelphia, Pa: WB Saunders Co;. 2003: 297-9, 839-42.
Chiapparini L, Farina L, D'Incerti L, et al. Spinal radiological findings in nine patients with spontaneous intracranial hypotension. Neuroradiology. Feb 2002;44(2):143-50; discussion 151-2. [Medline].
Chung SJ, Kim JS, Lee MC. Syndrome of cerebral spinal fluid hypovolemia: clinical and imaging features and outcome. Neurology. Nov 14 2000;55(9):1321-7. [Medline].
Dillon WP. Spinal manifestations of intracranial hypotension. AJNR Am J Neuroradiol. Aug 2001;22(7):1233-4. [Medline].
Falcioni M, Romano G, Aggarwal N, Sanna M. Cerebrospinal fluid leak after retrosigmoid excision of vestibular schwannomas. Otol Neurotol. Apr 2008;29(3):384-6. [Medline].
Gubbels SP, Selden NR, Delashaw JB Jr, McMenomey SO. Spontaneous middle fossa encephalocele and cerebrospinal fluid leakage: diagnosis and management. Otol Neurotol. Dec 2007;28(8):1131-9. [Medline].
Hannallah D, Lee J, Khan M, Donaldson WF, Kang JD. Cerebrospinal fluid leaks following cervical spine surgery. J Bone Joint Surg Am. May 2008;90(5):1101-5. [Medline].
Hegarty SE and Millar JS. MRI in the Localization of CSF Fistulae: Is It of Any Value?. Clinical Radiology. 1997;52:768-770.
Limawararut V, Valenzuela AA, Sullivan TJ, McNab AA, Malhotra R, Davis G, et al. Cerebrospinal fluid leaks in orbital and lacrimal surgery. Surv Ophthalmol. May-Jun 2008;53(3):274-84. [Medline].
Sautter NB, Batra PS, Citardi MJ. Endoscopic management of sphenoid sinus cerebrospinal fluid leaks. Ann Otol Rhinol Laryngol. Jan 2008;117(1):32-9. [Medline].
Schlosser RJ, Bolger WE. Nasal cerebrospinal fluid leaks: critical review and surgical considerations. Laryngoscope. Feb 2004;114(2):255-65. [Medline].
Schnabel C, Di Martino E, Gilsbach JM, Riediger D, Gressner AM and Kunz D. Comparison of B2-Transferrin and B-Trace Protein for Detection of Cerebrospinal Fluid in Nasal and Ear Fluids. Clinical Chemistry. 2004;50:661-663.
Spelle L, Boulin A, Tainturier C, et al. Neuroimaging features of spontaneous intracranial hypotension. Neuroradiology. Aug 2001;43(8):622-7. [Medline].
Taivainen T, Pitkanen M, Tuominen M, Rosenberg PH. Efficacy of epidural blood patch for postdural puncture headache. Acta Anaesthesiol Scand. Oct 1993;37(7):702-5. [Medline].
Wax MK, Ramadan HH, Ortiz O, Wetmore SJ. Contemporary management of cerebrospinal fluid rhinorrhea. Otolaryngol Head Neck Surg. Apr 1997;116(4):442-9. [Medline].
Yousry I, Forderreuther S, Moriggl B, et al. Cervical MR imaging in postural headache: MR signs and pathophysiological implications. AJNR Am J Neuroradiol. Aug 2001;22(7):1239-50. [Medline].
Zuckerman JD, DelGaudio JM. Utility of preoperative high-resolution CT and intraoperative image guidance in identification of cerebrospinal fluid leaks for endoscopic repair. Am J Rhinol. Mar-Apr 2008;22(2):151-4. [Medline].
Further Reading
Keywords
cerebrospinal fluid leak, CSF leak, dural tear, dural leak, CSF rhinorrhea, CSF otorrhea, pneumocephalus, spinal CSF leak, intracranial hypotension, spontaneous intracranial hypotension syndrome, SIHS, traumatic CSF fistula, double-ring sign, lumbar extradural blood patch
