TITLE: DIAGNOSIS AND TREATMENT OF CSF LEAKS
SOURCE: Dept. of Otolaryngology, UTMB, Grand Rounds
DATE: 19 April 1989
RESIDENT PHYSICIAN: Amy Coffey, M.D.
FACULTY: F. B. Quinn, Jr., M.D.
DATABASE ADMINISTRATOR: Melinda McCracken, M.S.
"This material was prepared by resident physicians in partial
fulfillment of educational requirements established for the
Postgraduate Training Program of the UTMB Department of
Otolaryngology/Head and Neck Surgery and was not intended for
clinical use in its present form. It was prepared for the purpose
of stimulating group discussion in a conference setting. No
warranties, either express or implied, are made with respect
to its accuracy, completeness, or timeliness. The material
does not necessarily reflect the current or past opinions of
members of the UTMB faculty and should not be used for purposes
of diagnosis or treatment without consulting appropriate
literature sources and informed professional opinion."
Causes of CSF fistulas:
80% head injuries with skull base fractures- M>F
16% surgery - nasal and paranasal cavities, skull base
4% spontaneous fistulas- F>M
2-3% of all head injuries; >11% with skull base fractures.
Occurs in areas of thinness of bone and adherence of the
dura to the skull: cribiform plate, posterior wall of frontal
sinus, sphenoid, sella floor or the great sphenoidal wing (large
lateral development of sinus).
Compound fractures more commonly associated with persistent leaks
than linear fractures
Late appearance of a fistula provoked by an elevation of
the intracranial pressure and by an enlargement of the frontal
ventricular horn as consequence of a frontal lobe contusion.
Appearance occurs by the 1st or 2nd day in 55%, one week
in 15%, and up to 20 years later in 10%. The remainder present
Treatment- stops spontaneously or under treatment usually in 3-5
days when detected on the day of or after the injury; treated
usually by "head-up" position, avoidance of coughing sneezing, nose
blowing and straining, laxitives, restricting oral intake, use of
steroids, diuretics or osmotically active medications and by
repeated lumbar CSF drainage with removal of about 150ml daily
(should not be done with marked brain edema, increased ICP).
Spontaneous cure not likely if persist longer than one week;
all should be treated operatively. If leak stops within one week,
no fracture or only a small linear fracture on X-ray, and no
anosmia then operative treatment is probably not necessary but
patient should be followed closely for at least one year for signs
of recurrent or occult rhinorrhea and warned about the increased
risk for meningitis. If anosmic, surgery is probably better
performed. Some authors feel all cases with compound fractures,
a defect within the posterior wall of the frontal sinus, the
cribiform plate and ethmoidal region, or fractures of the sphenoid
should be operated upon even if rhinorrhea has stopped spontaneously
due to high risk of recurrence and meningitis. All bullet wounds
of the anterior cranial fossa with rhinorrhea require operative
Most frequently from transsphenoidal approach to pituitary
adenomas due to intraoperative damage of the pituitary tentorium
Following removal of tumors of the base of the skull.
During ENT operations: endoscopic sinus surgery(bone
often dehiscent in long standing or severe polypoid disease),
septoplasty (must not crack the cribiform plate- most often occurs
using the Jansen-Middleton forceps and a rocking motion)
Fistulas produced unintentionally during ENT operations
can usually be closed from the same approach. Repair should be
undertaken at once; use temporalis fascia tucked in through the bony
defect after dural edges carefully detached from the bone-covered
with second piece of fascia and possibly a mucoperiosteal flap from
the nasal septum rotated and allowed to cover the second piece of
fascia then packed with Gelfoam and antibiotic-impregnated gauze
left in place 10-14 days. Transseptal sphenoidectomy offers a
limited approach for repairs in this area requiring radiologic
control for orientation. Treatment of choice for CSF leaks
associated with empty sella syndrome. The external ethmoid approach
to the sphenoid may give better visualization.
Posterior ethmoidal lesions are better closed via a transfrontal
If leak diagnosed during the immediate postoperative
period, a subarachnoid lumbar drain may be used to decompress;
kept in place 1 to 2 weeks; may use prophylactic antibiotics.
If unrecognized until patient develops meningitis,
external ethmoidectomy approach indicated. If this approach fails,
may need anterior craniotomy (morbidity much higher with this
approach due to invariable damage to the olfactory nerve). May
need to identify site of lesion to help with planning using
Nasal CSF fistula with no evidence of traumatic origin,
are not the result of operation, or for no clear reason escapes
through or along the olfactory nerves.
Almost always located in the ethmoidal region.
Seen with high pressure hydrocephalus of markedly raised
intracranial pressure from other reasons, tumours of the base of
the skull, ethmoidal encephalomeningocele and sphenoidal defects
such as the empty sella syndrome (10% associated with CSF rhinorrhea)
A. Most frequently following temporal bone fractures.
B. Second most frequently seen after intracranial surgery
involving the temporal bone.
C. Spontaneous CSF otorrhea - two different subgroups
72% childhood type with congenital defects of the otic
1. median age 4 years old.
2. meningitis occurs in 92% (usually pheumococcal,
3. follows myringotomy for a presumed serous effusion
4. Mondini deformity; uni- or bilateral absence of
cochlear and vestibular function usually
5. CSF usually enters inner ear through a dural
defect in the lateral aspect of the internal auditory canal and
exits through the oval window.
Treatment- most effective operation is stapedectomy and
packing of the vestibule with a muscle graft. If hearing is present,
subtotal petrosectomy including a radical mastoidectomy with
exenteration of the hypotympanic and pericarotid air cells,
obliteration of the eustachian tube, middle ear and mastoid, and
blind sac closure of the external auditory canal.
28% adult type with bony dehiscenses, most commonly of the
tegmen tympani or tegmen mastoideum (floor of the middle cranial
fossa), less commonly of the posterior fossa plate
1. range from 27 to 72 years old; median age 56.
2. Women > men.
3. present after myringotomy for effusion with
conductive hearing loss; CSF rhinorrhea; meningitis.
4. meningeal defects are either meningoencephaloceles
(associated with middle cranial fossa floor defects, none found
in posterior defects) or holes in the dura.
5. 6% of routine autopsies have small, often multiple, bony
defects in the floor of the middle fossa yet incidence of
spontaneous CSF otorrhea is low. Other factors are size of the
defect, the length of time that it is present, and the presence of
increased intracranial pressure.
Treatment- combined mastoid and transtemporal
supralabyrinthine approach. Tiny defects can then be repaired
through this approach. If a meningoencephalocele is protruding
through the tegmen, it is safer to perform a middle fossa
craniotomy. Also possible to do a combined mastoid and
mini-craniotomy approach for repair of tegmen defects.
A. Used for prophylaxis - controversial.
1. infection rates do not differ significantly
comparing treated to nontreated using penicillin (8 and 5%).
2. few antibiotics penetrate the normal meninges in
sufficient concentration to be of value
3. increase the risk of antimicrobial resistance
4. may mask the appearance of an acute infection
B. Pneumococcus (Streptococcus pheumoniae) frequent cause of
meningitis secondary to temporal bone fractures; may be solated on
culture in up to 50% of cases.
A. Observation of clear watery fluid
look for hidden leak: comatose- place in lateral position
and bend the face downward; conscious-pressure test, valsalva, etc.
B. Handkerchief test
moistened with nasal secretions dries stiff; with CSF dries soft.
C. Nasal rhinorrhea usually thin, watery, and opalescent and
throws down a stringy precipitate secondary to mucin when tested
with alcohol or acetic acid; CSF is clear and watery and does not
precipitate when tested with alcohol or acetic acid.
D. Halo test
great amount of CSF discharge is necessary
E. Glucose oxidase test
Clinistix or Tes-tape introduced into the nares; positive
indicates CSF rhinorrhea, negative indicates rhinal or allergic
many false positives and false negatives documented.
F. Chemical analysis of glucose and protein content
collect several cubic centimeters in test tube; positive
when the glucose content exceeds 0.4 g/l and the protein content is
from less than 1 g/l to a maximum of 2 g/l; not reliable when CSF
is contaminated with blood or wound secretion or excessive nasal
or lacrimal secretions (>20%).
G. Dyes- Fluorescein
can use intaoperatively with darklight
some complications (i.e. status epilepticus, etc.)
H. Radioisotope cisternography
radioactive tracers injected into the lumbar subarachnoid
space followed by scintigraphy
compare count of cotton pledgets and blood; normal maximum
ratio is 0.3; Indium or TC pertechnetate most frequently used;
choice for scanning CSF fistulae when identifying questionable,
very small, or intermittent leaks. Very sensitive-
false-positives (detected in patients with no CSF leak by
circulation to nasal mucosa through blood stream after
absorption via the arachnoid villi), large amount of technical
I. Radiopaque contrast material
metrizamide- water soluble; used in conjunction with
computerized axial and sagittal tomography; may be performed
in "head up" position
relatively free of neurotoxicity
scan of choice in localizing active CSF leaks
J. Immunofixation techniques- identify B2-transferrin
pathognomonic for CSF fluid; has not been found in any
other body fluid
collected in test tube, on Merocel sponges, or by
aspiration through the TM
procedure: measure the protein concentration - if >5g/l,
precipitation with ammonium sulfate; precipitate dissolved in
distilled water; electrophoresis; immunofixation by soaking sample
in monospecific antiserum against human transferrin; rinsed,
fixed and stained with silver staining solution(more sensitive
than earlier used Coomassie brilliant blue dye).
must examine the patient's serum also therefore avoiding
false positives (liver cirrhosis, genetic protein variance);
if high levels of contamination with blood or wound secretions,
Merocel sponge is centrifuged; if blood contamination is above
3g/l hemoglobin, the hemoglobin content is reduced by column
chromatography. High sensitivity (4ul sample volume sufficient),
high level of accuracy, speed of testing (in biochemical lab),
can be repeated and it is not hazardous to the patient.
Adkins, WY, et.al. Mini-Craniotomy for Management of CSF Otorrhea
from Tegmen Defects, LARYNGOSCOPE August 1983, 93:1038-1040.
Bret, P., et.al. Treatment of Cerebrospinal Fluid Rhinorrhea by
Percutaneous Lumboperitoneal Shunting: Review of 15 Cases,
NEUROSURGERY 1985, 16:44-47.
Cannon, C.R., & R.A. Jahrsdoerfer Temporal Bone Fractures: Review of
90 Cases, ARCH OTOLARYNGOL May 1983, 109:285-288.
Curnes, J.T. et.al. CSF Rhinorrhea: Detection and Localization Using
Overpressure Cisternography with Tc-99m-DTPA. RADIOLOGY 1985,
Hasegawa, M. et.al. Transfer of Radioisotope from CSF to Nasal
Secretion. ACTA OTOLARYNGOL 1983, 95:359-364.
Hubbard, J.L. et.al. Spontaneous Cerebrospinal Fluid Rhinorrhea:
Evolving Concepts in Diagnosis and Surgical Management Based on the
Mayo Clinic Experience from 1970 through 1981. NEUROSURGERY 1985,
Katz, R.T. and P.E. Kaplan Glucose Oxidase Sticks and Cerebrospinal
Fluid Rhinorrhea. ARCH PHYS MED REHABIL 1985, 66:391-393.
Loew, F. et.al. Traumatic, Spontaneous and Postoperative CSF
Luotonen, J. et.al. Localisation of a CSF Fistula by Metrizamide CT
Cisternography. 1986, 100:955-958.
Maniglia, A.J. Fatal and Major Complications Secondary to Nasal and
Sinus Surgery. LARYNGOSCOPE 1989, 99:276-283.
Nishihira, S. and T.V. McCaffrey The Use of Fibrin Glue for the
Repair of Experimental CSF Rhinorrhea. LARYNGOSCOPE 1988, 98:625-627.
Oberascher, G. Cerebrospinal Fluid Otorrhea-New Trends in Diagnosis.
AM J OTOLOGY 1988, 9:102-108.
Oberascher, G. and E. Arrer Efficiency of Various Methods of
Identifying Cerebrospinal Fluid in Oto- and Rhinorrhea. ORL 1986,
Park, J. et.al. Current Management of Cerebrospinal Fluid Rhinorrhea.
LARYNGOSCOPE 1983, 93:1294-1300.
Wetmore, S.J. et.al. Spontaneous Cerebrospinal Fluid Otorrhea. AM J
OTOLOGY 1987, 8:96-101.
Yokoyama, K. et.al. Diagnosis of CSF Rhinorrhea: Detection of
tau-Transferrin in Nasal Discharge. OTOLARYNGOL H & N SURG 1988,