Cerebrospinal Fluid Rhinorrhea

[Guest guest]

Okay, I've posted the entire article. It's really good and there's

some really important info here about the causes of Cerebrospinal

Fluid Rhinorrhea as well as how to diagnose and test for it.

This sounds way too much like so many of us, myself in particular.

Headaches, clear post nasal drip (CFS?), salty taste. Can result

from trauma (whiplash/FMS?), cranial defects, or fistulas caused by

infection.

Hello? I think we should all be getting tested for this.

penny

http://www.bcm.edu/oto/grand/120398.html

Cerebrospinal Fluid Rhinorrhea

Jayson Greenberg, M.D.

December 3, 1998

CSF rhinorrhea results from a breakdown of the dura and supporting

structures of the skull base resulting in a connection between the

subarachnoid space and the nose. It may be a complication of trauma,

tumor ablation, paranasal sinus disease, or surgery. Regardless of

etiology, the mechanism is essentially the same. There is a

disruption of the arachnoid and the dura, coupled with an osseous

defect, and a CSF pressure gradient that is either continuously or

intermittently greater than the healing tensile strength of the

disrupted tissue. This causes separation of the dural fibers and CSF

leakage.

CSF rhinorrhea may occur directly through the anterior cranial fossa

or indirectly from the middle or posterior fossa via the Eustachian

tube. More specifically, these portals of entry may take place

across the frontal sinus, cribriform plate of the ethmoid, the

sphenoid sinus, the sella, or via the temporal bone from the middle

ear and through the Eustachian tube.

CSF rhinorrhea is not a common entity, but a very important one,

because its complications can be catastrophic. Most fistulas will

heal with conservative management and without complications, but

pneumocephalus, meningitis, and hydrocephalus are potential fatal

complications. Meningitis occurs in 25%-50% of untreated cases, in

10% of cases within the first week of trauma, and 10% of cases

following spontaneous cessation of leakage.

Galen first described CSF rhinorrhea in the second century AD. He

postulated that CSF was released into the nose by way of the

pituitary and ethmoid regions. In 1826, published the

first case of CSF rhinorrhea in a hydrocephalic child with an

intermittent discharge of nasal fluid. Autopsy revealed

communications between the nasal and cranial cavities. In 1899, St.

Clair reported the first series of patients with

spontaneous CSF leaks and coined the term rhinorrhea. He also

differentiated between cerebrospinal rhinorrhea and nasal

rhinorrhea. He did not recommend surgical intervention. Walter Dandy

performed the first successful intracranial repair in 1926. Gusta

Dohlman described the first extracranial approach for repair in 1948

using flaps taken from the middle turbinate for a fistula in the

cribriform region. Hirsch described the first endonasal

approach in 1952 for CSF rhinorrhea in 2 acromegalic patients.

Finally, in 1981, Wigand described the first transnasal endoscopic

visualization and repair of a CSF leak.

CSF rhinorrhea is generally classified as traumatic or atraumatic.

Traumatic CSF leaks can be subdivided into those caused by head

injury versus those that are iatrogenic or post surgical. Eighty

percent are traumatic in origin, as a result of head injury. CSF

rhinorrhea occurs in 2% of all head injuries and in up to 15% of all

base of skull fractures. The majority of these occur through the

anterior cranial fossa. In this area, the dura is tightly adherent

to the thin bone of the cribriform plate and roof of the ethmoid.

Temporal bone fractures may also lead to passage of CSF from the

middle ear via the Eustachian tube manifesting as rhinorrhea.

Traumatic cases are more common in males. 80% present within the

first 48 hours, and 95% present within 3 months of the accident.

Fortunately, most of these fistulas close spontaneously. In these

instances, the dura may be torn, but there is no significant loss of

dural tissue.

Sixteen percent of leaks are traumatic and iatrogenic in origin.

They can be further subdivided into intra-operative and delayed

onset leaks. In the past, the majority of iatrogenic fistulas were

the result of neurosurgical procedures. However, endoscopic sinus

surgery has replaced neurosurgery as the leading cause of this

problem. CSF rhinorrhea is a well-recognized complication of ESS and

occurs in 0.5% of cases. The diagnosis can often times be made intra-

operatively by the presence of clear fluid draining from the roof of

the nasal cavity. Likewise, CSF rhinorrhea is a complication of

neurootology procedures, occurring in 12% of acoustic neuroma

resections. However, complications of acoustic neuroma resection has

been the subject of a prior grand rounds and will not be dealt with

today.

The remaining 4% of cases of CSF rhinorrhea are atraumatic or

spontaneous leaks. These can be further subdivided into high

pressure and normal pressure leaks. High-pressure leaks include

tumors and hydrocephalus. They are due to long-standing increases in

intracranial pressure and account for 45% of spontaneous leaks.

Increased pressure in the subarachnoid space forces CSF through a

weak or potential pathway. There is no direct invasion of the skull

base. Eighty-four percent are due to slow growing tumors, most

commonly originating in the pituitary. However, by the time a tumor

produces an increase in ICP great enough to precipitate a leak,

other neurologic signs are present. The remaining 16% are due to

hydrocephalus.

Normal pressure leaks account for 55% of cases. They are thought to

be due to slow erosion of the skull base secondary to normal

fluctuations in intracranial pressure leading to point erosion and

CSF rhinorrhea. Ninety percent originate from a congenital or

potential pathway such as a persistent craniopharyngeal canal. The

remaining 10% are due to direct erosion of the skull base by tumor

or infections. Examples include osteomas of the frontoethmoids,

nasopharyngeal angiofibromas, nasopharyngeal carcinomas, and

osteolytic erosions secondary to sinusitis, syphilis, or mucoceles.

In either case, the initial leak is frequently precipitated by

coughing, sneezing or straining in 30%.

The pathogenesis behind traumatic leaks is easily understood, while

there have been several theories behind the pathogenesis of

spontaneous leaks. Some have proposed that they are the results of

normal anatomic and physiologic factors. The majority of these

fistulous communications involve dehiscences in the region of the

fovea ethmoidalis, cribriform plate, and sphenoid sinus. Studies

have shown the fovea ethmoidalis is dehiscent in 14% of bones.

Similar dehiscences have been described in the sphenoid and frontal

sinus. Why these bony dehiscences occur is unknown. Excessive

pneumatization of the sinuses may be one explanation. Sphenoid

recesses are common, and inferolateral recesses may be seen in 25%-

36% of cases.

This increased pneumatization increases sinus surface area making an

association with a bony dehiscence more likely. Pneumatization of

the middle fossa floor as well as the sinuses along with normal

pulsatile CSF and brain pressures can cause thinning of the bone

from the intracranial surface. Small pits and holes can result in

elongation of the dura, arachnoid and brain through holes, creating

meningoceles or meningoencephaloceles, which may rupture. It remains

unproven whether those patients with spontaneous CSF rhinorrhea have

a higher incidence of sphenoid pneumatization versus control

populations.

Ommaya's theory of focal atrophy hypothesizes that the normal

contents of the cribriform plate or sella turcica areas can become

reduced in bulk because of ischemia. The resulting empty space

becomes a pouch filled with CSF. The normal CSF pressure pulse

causes this pouch to exert a focal continually erosive effect.

The persistence of the embryonic lateral craniopharyngeal canal has

been another proposed route. One analysis of 138 sphenoidal bones

revealed 18% had remnants of the craniopharyngeal canal, with 5%

having defects connecting the sphenoid sinus and the cranial vault.

It is pertinent to remember that maximum CSF pressure is attained in

an adult at a pressure that is 3times greater than in an infant.

This could partially explain why an underlying congenital defect

would become manifest only at an adult age.

The diagnosis of CSF rhinorrhea revolves around three principles:

determining the fluid is from the nose, determining the fluid is

CSF, and localizing the fistula. The patient's history and physical

exam coupled with a high index of suspicion should prompt the search

for a leak.

The majority of cases present with a complaint of clear, watery

nasal discharge with a recent history of head trauma or predisposing

surgery. Specifically a persistent clear nasal discharge that is

unilateral would suggest a diagnosis of CSF rhinorrhea. The flow may

change with alterations in posture. When supine, the patient may

complain of a postnasal drip. A salty taste may also be noted.

Cessation of flow is frequently associated with headache, which is

relieved by the onset of flow. Confirming the diagnosis is more

challenging in those with spontaneous leaks. The initial onset is

insidious and may occur after an episode of coughing or sneezing. It

may be mistaken for vasomotor rhinitis or rhinosinusitis. A patient

with repeated episodes of meningitis should also prompt further

investigation of a dural tear.

Physical examination is usually unremarkable, except for rhinorrhea,

which may or may not be present at the time of examination.

Intranasal exam may reveal an encephalocele, but is also

unremarkable. Most fistulas occur high in the nose and are difficult

to visualize without magnification. Jugular compression or different

head positions may help stimulate flow through the fistula.

Specifically, asking the patient to lean forward and strain with a

closed glottis may stimulate flow.

Distinguishing between CSF and serous nasal secretions may be

difficult. If a sufficient sample of nasal drainage can be obtained,

it can be sent to the lab and analyzed. CSF is usually clear unless

associated with trauma. Blood from head injured patients may mix

with CSF and mask the recognition of a leak. Most of us have read

about the halo sign. CSF will separate from blood when the mixture

is placed on filter paper resulting in a central area of blood with

an outer ring or halo. Dula et al studied this ring sign and found

that mixtures of CSF and blood will produce a clinically detectable

ring with CSF:blood ratios of 30%-90%. Blood alone does not produce

a ring. The best ring was obtained with a 50: 50 mix of blood and

CSF. More importantly, they found that the presence of a ring was

not exclusive for CSF. Blood mixed with tap water, saline, and

rhinorrhea fluid also produced a ring. The halo sign does occur, but

clearly does not clinch the diagnosis.

CSF can also be confirmed in the laboratory. It is odorless, salty,

and has a specific gravity of 1.006. The protein level is much less

than nasal fluid, while the chloride level is greater. More

importantly, CSF has a greater concentration of glucose than mucus

or lacrimal secretions. The quantitative determination of a glucose

level in nasal fluid not contaminated by blood can be diagnostic of

CSF rhinorrhea if the nasal fluid contains more than 30mg/dl.

Negative test results for glucose virtually eliminate the

possibility of CSF.

Glucose oxidase paper or dextrose sticks have historically been used

to identify CSF. However, they have been shown to be unreliable

because lacrimal gland secretions and nasal mucus have reducing

substances that may cause a positive reaction with glucose

concentrations as low as 5 mg/dl.

When the leakage is profuse and clear, the diagnosis can be

unmistakable. Intermittent leaks, especially when mixed with blood

or nasal secretions may be overlooked. The presence of beta2

transferrin was first used to diagnose a CSF leak in 1979.

What is beta2 transferrin? Transferrin is a polypeptide involved in

ferrous ion transport. Beta1 transferrin is present in serum, nasal

secretions, tears, and saliva. Beta2 transferrin has only been

demonstrated in CSF, perilymph, and aqueous humor. Beta2 transferrin

accounts for 15% of the total transferrin content in CSF. The

structural differences between the 2 forms results in a slower

migration of beta2 transferrin towards the cathode. The result is 2

distinct bands produced during electrophoresis. This is a non-

invasive test requiring only .5cc of fluid for analysis. Beta 2

transferrin can be detected in 3 hours.

How useful is this test? Skedros et al (1993) at the University of

Pittsburgh studied 68 patients that had specimens submitted for

beta2 transferrin to evaluate for a CSF leak. The results

facilitated appropriate intervention in 7 patients, and a negative

test result avoided any invasive procedures in 33 patients. Eighteen

patients had test results that were unavailable at the time of

intervention. In 6 patients clinical management was contrary to the

test results. Five underwent invasive procedures despite a negative

test. Three had operative procedures during which no leak was found.

Two received lumbar drains and continued to have rhinorrhea

diagnosed as post surgical vasomotor rhinitis. The sixth patient was

discharged with a positive test and later brought back for repair of

a post surgical leak. Overall they found no false positive tests in

23/23 patients confirmed by subsequent management, and 2 surgically

confirmed false negative results (43/45).

However, when the clinical impression sharply contrasts with the

beta2 transferrin results, repeat testing or the use of other

methods to detect CSF rhinorrhea should be used. These other

methods, besides, diagnosing CSF rhinorrhea, also help to localize

the leak.

Now that we know the fluid is CSF, how do we localize the leak?

Radiologically, CT scan alone is an insensitive method for

confirming a leak. CT alone can yield indirect evidence such as a

bony defect or fracture or fluid in the paranasal sinuses. CT can

also show any causative intracranial lesions including tumors, bony

erosions, or hydrocephalus. Thin cut CT can be particularly useful

in identifying bony abnormalities within the sinuses or skull base.

CT cisternography with metrizamide introduced intrathecally is more

accurate, and its reliability and usefulness in localizing CSF

fistulas has been documented in numerous studies. Metrizamide is a

nonionic, tri-iodinated, water-soluble compound that was first used

in the early seventies as a contrast agent for myelography and

ventriculography. Chow et al (1989) studied 13 patients with

Metrizamide CT cisternography (MCTC) to localize the site of a CSF

leak. Seventeen studies were performed, and 13 identified the site

of the leakage. Nine scan results were confirmed surgically and the

other four were performed on patients who refused surgery. Fifteen

scans were performed in patients with active leaks, and 13 were

positive. Visualization of metrizamide passing through a bony defect

is irrefutable evidence of a CSF leak. The combination of a bony

defect with extracranial metrizamide adjacent to the bony defect can

also adequately define the leakage site. Extracranial metrizamide

within one sinus or in a focal area at the base of the skull may not

precisely delineate the leak site, but does help to localize it.

Intra nasal pledgets can also be placed in the nose and examined for

the presence of contrast.

Studies have shown MCTC to be successful in anywhere from 76% to

100% of cases. Side effects are minimal and include primarily

headache and nausea. MCTC has been shown to be much more accurate

and useful in active versus inactive leaks. Sensitivity drops to

less than 60% with inactive leaks. However, repositioning and

elevation of intracranial pressure by Valsalva, coughing, or

infusion of saline intrathecally have been used to promote CSF

leakage and demonstrate the presence of a fistula.

Radionuclide cisternography is similar to CT cisternography, but it

involves intrathecal administration of radioactive agents, either

technetium 99 or Indium 111. Intranasal pledgets are also placed.

Radioactive counts of the pledgets are compared and scintigrams of

the skull are obtained. Placement of nasal pledgets by differential

radiocontamination may add some localizing ability. However, the

presence or absence of a leak at the time of diagnosis influences

the test result. Eljamel in 1994 reviewed 325 patients from the

literature with CSF rhinorrhea. He found a 70% accuracy rate for

active leaks with radionuclide cisternography. Accuracy decreased to

28% with inactive leaks.

Like CT cisternography, radionuclide cisternography is an invasive

procedure. The radioisotope can be rapidly absorbed into the

circulation and distributed in turbinate tissue making

interpretation difficult or leading to false positive results. For

the study to be abnormal, readings should be impressively high.

Borderline or slightly elevated readings are not reliable. Care must

also be taken to avoid contamination of neighboring pledgets, which

can confound test results. With MCTC and radionuclide

cisternography, lumbar puncture may conceal an active CSF fistula by

reducing CSF pressure. Overall, radionuclide cisternography is a

limited localizing technique secondary to less spatial resolution

and anatomic detail. Some clinicians advocate radionuclide

cisternography as more of a diagnostic rather than localization test.

MRI is another useful radiological tool and demonstrates cranial

anatomy in detail. The most important benefit of MRI is

overestimation of the bright signal from trapped CSF or herniated

arachnoid in the dural-bone defect on T2 weighted images. The signal

magnifies the fistulous tract, making it more conspicuous. Studies

have shown the sensitivity of MRI to be comparable to CT

cisternography. In addition, there is no significant decrease with

inactive leaks. Eljamel was able to localize 100% of inactive leaks

in 11 patients using MRI. It may be difficult to differentiate CSF

from mucosal thickening and fluid that may be observed within a

sinus. If in doubt, IV gadolinium will show enhancement of

inflammatory tissue. Diagnosing a CSF fistula should include the

presence of a high signal that is continuous with the subarachnoid

space in an extradural location on T2 weighted images. As we all

know, MRI is noninvasive and does not expose the patient to ionizing

radiation, but lacks the bony detail of CT.

Injecting intrathecal dye is another method used to diagnose and

localize the fistula site. Fluoroscein is the predominant dye used.

This study can provide precise information regarding the location of

the leak by direct intranasal examination or cottonoid pledget

staining. It is most useful if injected immediately pre-operatively

to assist in intraoperative identification of the CSF leak.

Transient neurologic complications have been reported, but are

minimal when the agent is used in its proper dose.

Treatment can be divided into medical and surgical. Conservative

medical management is advocated to allow the body's reparative

processes a chance to heal. Most protocols involve head elevation

and bed rest. The patient is to avoid any coughing, straining and

nose blowing. Laxatives are given and fluid is restricted. Some

advocate the use of steroids or diuretics.

Lumbar drains can also be used to reduce CSF pressure, allowing the

dura to approximate and heal. The usefulness of lumbar drains in

post surgical patients has been well documented in the neurosurgical

literature. Controlled continuous spinal drainage has proven to be a

useful adjunct to surgical attempts to seal persistent CSF fistulas.

As I stated earlier, the majority of post-traumatic CSF leaks, up to

80%, are associated with a high probability of early spontaneous

closure.

Prophylactic antibiotic use is controversial. The goal of

antibiotics is to decrease the risk of meningitis, a potentially

fatal complication. Many advocate use for cases of traumatic origin.

The traumatic wound is contaminated by CSF exposure to potentially

pathogenic organisms. Other researchers have found no significant

difference in meningitis rates in retrospective studies.

Brodie et al in 1997 performed a meta-analysis of 6 studies over the

last 25 years involving post traumatic CSF leaks, including otorrhea

and rhinorrhea. This included 324 patients. Two hundred thirty-seven

patients received prophylactic antibiotics and 87 did not. The

incidence of meningitis was significantly lower in the patients who

received prophylactic antibiotics. Two and one-half percent of the

patients (6/237) in the antibiotic group developed meningitis versus

9 of 87 or 10% of the no antibiotic group.

The risk of meningitis depends on various factors, including

duration of CSF leakage, delayed onset of CSF leakage, site of

fistula, and concomitant infection. Prolonged duration of CSF

leakage has been shown to be associated with meningitis in many

studies. Brodie also found that patients with post traumatic leaks

lasting longer than 7-10 days have and 8-10 fold increased risk of

developing meningitis.

The benefit of antibiotic prophylaxis in spontaneous fistulas has

not been adequately studied, as the number of patients in this

subset is significantly smaller. Some have advocated a 4-6 week

trial of antibiotics to observe for a spontaneous closure, while

others claim that antibiotic prophylaxis in this setting is

ineffective and can select out resistant organisms.

Surgical management is the other treatment option. In addition to

surgical repair of the CSF fistula, many patients, especially those

with traumatic leaks, may have additional facial fractures requiring

operative fixation. A dural tear is not a contraindication to

reduction of a midface fracture. Early repair is advocated.

Reduction of facial fractures in traumatic fistulas provides a

strong bony support for the repair and approximation of torn dural

edges. Delayed reduction is more difficult and may re-open a closed

fistula.

Surgical repair is indicated for open wounds, intracranial

hemorrhage, pneumocephalus, cases of recurrent meningitis, and those

leaks not responding to conservative management. Patients with

persistent post-traumatic CSF rhinorrhea despite conservative

management require surgical repair because of the increasing

incidence of meningitis,. This subset also includes delayed onset

and spontaneous leaks that fail conservative management. Some feel

that non-operative management in these cases is rarely permanent.

Hubbard et al from the Mayo clinic managed 3 of 28 patients

conservatively because of spontaneous remission. Within 24 months,

all 3 were again symptomatic and underwent surgical repair. The main

surgical approaches are intracranial and extracranial; the latter of

which also includes endoscopic repair.

Intracranial repair or craniotomy involves unilateral or bilateral

frontal bone flaps. Advantages include direct visualization,

simultaneous repair and inspection of adjacent cortex, and a better

chance of tamponading a leak precipitated by increased ICP.

Drawbacks include increased morbidity, extended operative time,

prolonged hospitalization, poor view of communicating fistulas from

the sphenoid sinus, and increased risk of anosmia. Craniotomy also

has a high incidence of persistent leak. Series report anywhere from

a 20%-40% failure rate. Ten percent have persistent leaks despite

multiple repair attempts. Current thought is that unless there is a

coexisting indication for intracranial exploration, the most

appropriate initial approach is extracranial, with craniotomy

reserved for those that fail or persist despite extracranial repair

attempts.

CSF rhinorrhea from the frontal sinus is best approached with an

osteoplastic flap technique. An eyebrow or coronal incision can be

used. After the sinus is entered and mucosa removed, the bony defect

in the posterior table is identified. Lowering the patient's head

and performing a Valsalva maneuver may help localize the leak.

Direct repair of the dura can be accomplished using interrupted silk

sutures. For larger defects, homograft dura or fascia lata grafts

can be sewn into place medial to the bony defect. The frontal

sinuses are then obliterated with fat, and the bone flap secured

into place with miniplates.

The ethmoid roof and cribriform plate region are most commonly

involved in traumatic leaks. Repair in this region can be

accomplished via an external ethmoidectomy approach. The orbital

contents are dissected posteriorly in a subperiorbital plane. The

ethmoid labyrinth is entered by perforating the lacrimal bone and

lamina papyrecea. A complete ethmoidectomy is performed. Then the

dural defect is exposed in the ethmoid or cribriform plate region. A

mucoperiosteal flap is then constructed using either nasal septum,

middle turbinate or lateral nasal wall as the donor site. The flap

is then rotated into position and secured into place. Some recommend

a free fascial graft from either the temporalis fascia or tensor

fascia lata to reinforce the flap into place. Free grafts of middle

turbinate mucosa can also be used. Once the graft is secured into

place, it is covered with Gelfoam and buttressed with nasal packing.

If the fistula is the result of a prior transphenoidal procedure or

is confined to the sphenoid sinus based on pre-operative studies, a

transseptosphenoid approach can be used. The surgical approach is

via a sublabial or transnasal route. The sinus is stripped of

mucosa. The fistula can be repaired by placing fascia directly over

the defect. The graft is held in place with strips of abdominal fat.

Extracranial repairs are associated with decreased morbidity,

decreased incidence of anosmia, and superior exposure of the

sphenoid, parasellar, and posterior ethmoid regions. Studies show

success rates from 80%-90% with an extracranial approach.

Limitations include an inability to examine the underlying cortex,

poor visualization of frontal or sphenoid sinuses with prominent

lateral extensions, and lack of success with repairing high-pressure

leaks. Those patients with intracranial hypertension require CSF

shunting as an adjunct to repair. Complications include facial

numbness, septal perforation, and various orbital complications.

Several surgeons have described endoscopic closure of CSF fistulas.

Some surgeons prefer an operating microscope versus endoscopy. The

microscope allows magnification, depth perception, and freedom of

motion with a two handed technique. Nevertheless, increased

experience with endoscopic sinus surgery has led to expansion of

surgical indications.

The endoscopic approach provides the most versatility for

visualization and exact localization of the fistula site. Extended

visualization with angled telescopes has made a variety of anterior

and middle fossa defects accessible. All walls of the sphenoid sinus

can be accurately visualized. The surgeon can precisely clear mucosa

off the defect without increasing defect size. An endoscopic

approach avoids an external incision, minimizes intranasal trauma,

and decreases OR time.

There have been 2 large studies that have examined the success of

endoscopic repair of CSF fistulas for all etiologies. Dodson et al

treated 29 cases of CSF rhinorrhea with endoscopic techniques.

Seventy-five percent had resolution after their initial repair.

Duration of follow-up ranged from 3 to 43 months. Lanza et al

reviewed 36 patients that underwent endoscopic repair of CSF

fistulas. During the first attempt, successful endoscopic repair was

achieved in 94%. The mean duration of follow-up was 24.6 months,

with a range of 2 to 57 months. Failure of an endoscopic approach

may relate to inability to successfully localize the defect, graft

displacement, insufficient graft size, incomplete apposition of the

graft to the skull base defect, and patient non-compliance with post-

operative instructions.

The most important consideration in endoscopic repair is exposure.

If the septum is deviated to such an extent that it compromises

exposure, a septoplasty may be necessary. Likewise, if the middle

turbinate compromises exposure, a portion may need to be removed. In

this instance, the middle mucosa and bone can be stripped and used

as a graft. Angled telescopes are most useful in visualizing the

area of the leak. Any debris or blood clot should be removed. For

cribriform or fovea ethmoidalis repairs, the dura is usually very

adherent and cannot be elevated. Dural elevation is then best left

alone to prevent further tearing.

Surgical repair involves obtaining a watertight closure until

fibrosis produces a permanent seal. To obtain a watertight seal, the

dura can be directly sutured. This is often impossible due to

inadequate tissue. In this instance various types of grafts have

been described to obliterate the defect and obtain a seal. Types of

grafts include mucoperiosteal/ mucoperichondrial free grafts,

cartilage, bone, temporalis myofascial grafts, and fat. Others have

described pedicled turbinate or nasal septal flaps. However, these

grafts may fold, tent, or contract, resulting in an inadequate seal.

For this reason, others prefer free grafts from the same sites. Some

recommend free grafts from remote sites, like tensor fascia lata or

temporalis fascia, to eliminate any mucosal defects and risk of

synechiae formation.

In either case, the size of the graft should exceed the defect size

by 30% to compensate for post-operative graft shrinkage. The graft

should then be bolstered into place. Gauze, gel foam, merocel

sponges, Foley balloons, and nasal trumpets have all been described.

As endoscopic approaches are used to repair CSF fistula, endoscopy

can also be the cause of CSF rhinorrhea as a complication of sinus

surgery. This is an accepted and known complication. One-third of

Lanza's and Dodson's patient population were CSF fistulas after

endoscopic sinus surgery. Treatment involves immediate repair of

intraoperative leaks. Delayed onset leaks, if constant and

unresponsive to conservative management should be repaired 1-2 weeks

after diagnosis. Of course, the best treatment is prevention, the

scope of which extends beyond this presentation.

There have been few studies that have clearly documented success in

relation to types of grafts. Sample sizes are small, and many have

had success with different types of grafts. Several surgeons have

used fibrin glue or an Avitene slurry to provide better adhesion of

the graft and improve the initial seal during healing. Fibrin glue

is made by combining topical thrombin and 10% calcium chloride and

mixing this combination simultaneously with cryoprecipitate.

Nishihira and McCaffrey studied the use of fibrin glue in

experimentally induced CSF rhinorrhea in an animal model. CSF

rhinorrhea was produced by creating a defect in the anterior cranial

fossa in 36 rats. There were 4 treatment groups: 1) a no treatment

control, 2) fibrin glue alone, 3) muscle packing alone, 4) fibrin

glue with muscle packing. CSF leaks were evaluated at 3 weeks post-

treatment. Persistent CSF leakage was noted in 89% of the control

group, 55% of the second grou2, 33% of group three, and 22% with

muscle and fibrin glue. Their results suggest that fibrin glue can

optimally be used to stabilize tissue grafts and autologous tissue

to obliterate skull base defects.

Researchers also differ in their use of lumbar drains. Seventy-eight

percent of Lanza'a patients underwent pre-operative lumbar drain

placement. None of Dodson's patients had a lumbar drain placed. In

general, a patient with a fresh fistula probably does not need a

lumbar drain, because there is no hypersecretion of CSF. Patients

with longstanding fistulas have increased the amount of CSF produced

to compensate for what is lost. Without a drain in these patients,

an overabundance of CSF will collect intranasally and put pressure

on any repair, increasing the risk of failure. Lumbar drains are not

without complications and need to be monitored closely. Draining off

too much CSF can be fatal.

Post-operative instructions, and these apply to any type of

approach, include head-of-bed elevation, complete bed rest for 3-5

days, stool softeners, anti-hypertensives and analgesics, lumbar

drain for 2-5 days, and no strenuous activity for 4-6 weeks.

Antibiotic prophylaxis for nasal packing is also initiated.

In closing, CSF rhinorrhea is not a common entity, but it is an

important diagnosis to make since its complications can be

catastrophic. Most are due to trauma, and most close with

conservative management. Surgery is then reserved for those patients

that fail conservative therapy.

Case Presentation

MR is a 30-year-old Hispanic female who presents with complaints of

clear, watery drainage from her left nostril for several months,

since she returned from Mexico for her grandmother's funeral. The

drainage has not subsided since that time. She does not recall any

precipitating trauma. She does not have any alterations of taste or

smell. She denies any changes in vision or hearing. She does

complain of occasional frontal pressure type headaches. She

initially went to her primary care physician, who prescribed a nasal

spray, which did not improve her symptoms after one month of

treatment. She sought no further treatment for 7 months, until she

saw another internist, who diagnosed CSF rhinorrhea based on

elevated glucose levels in the nasal fluid. She was then referred

for further evaluation.

Past Medical History: positive PPD. Past surgical history was a

tonsillectomy in 1982. Currently she is taking a prescribed anti-

tuberculin medication. All: NKDA.

On Physical Examination of the head and neck, her TM's were clear,

intact bilaterally. There were no effusions present. Her nose was

clear, with no drainage or lesions noted on rigid endoscopy. OC/OP -

clear. No lymphadenopathy or masses were observed in the neck.

Cranial Nerves II - XII are intact.

Bibliography

Anand VK, Murali RK, Glasgold MJ. Surgical decisions in the

management of cerebrospinal fluid rhinorrhoea. Rhinology 1995;33:212-

218.

Aarabi B, Leibrock LG. Neurosurgical approaches to cerebrospinal

fluid rhinorrhea. Ear Nose Throat J 1992;71:300-305.

Beckhardt RN, Setzen M, Carras R. Primary spontaneous cerebrospinal

fluid rhinorrhea. Otolaryngol Head Neck Surg 1991;104:425-432.

Brockbank MJ, Veitch DY, HG. Cerebrospinal fluid in the

rhinitis clinic. J Laryngol Otol 1989;103:281-283.

Brodie HA. Prophylactic antibiotics for posttraumatic cerebrospinal

fluid fistulae. A meta-analysis. Arch Otolaryngol Head Neck Surg

1997;123:749-752.

Brodie HA, TC. Management of complications from 820

temporal bone fractures. Am J Otol 1997;18:188-197.

Calcaterra T. Diagnosis and management of ethmoid cerebrospinal

rhinorrhea. Otolaryngol Clin North Am 1985;18:108-115

Choi D, Spann R. Traumatic cerebrospinal fluid leakage: risk factors

and the use of prophylactic antibiotics. Br J Neurosurg 1996;10:571-

575.

Chow JM, Goodman D, Mafee MF. Evaluation of CSF rhinorrhea by

computerized tomography with metrizamide. Otolaryngol Head Neck Surg

1989;100:99-105.

D, Bullock P, Hui F, Firth J. Benign intracranial

hypertension: a cause of CSF rhinorrhoea. J Neurol Neurosurg

Psychiatry 1994;57:847-849.

Clemenza JW, Kaltman SI, Diamond DL. Craniofacial trauma and

cerebrospinal fluid leakage: a retrospective clinical study. J Oral

Maxillofac Surg 1995;53:1004-1007.

Coiteiro D, Tavora L, Antunes JL. Spontaneous cerebrospinal fluid

fistula through the clivus: report of two cases. Neurosurgery

1995;37:826-828.

Colquhoun IR. CT cisternography in the investigation of

cerebrospinal fluid rhinorrhoea. Clin Radiol 1993;47:403-408.

Daly DT, Lydiatt WM, Ogren FP, GF. Extracranial approaches to

the repair of cerebrospinal fluid rhinorrhea. Ear Nose Throat J

1992;71:311-313.

Dodson EE, Gross CW, Swerdloff JL, Gustafson LM. Transnasal

endoscopic repair of cerebrospinal fluid rhinorrhea and skull base

defects: a review of twenty-nine cases. Otolaryngol Head Neck Surg

1994;111:600-605.

Dohlman G. Spontaneous cerebrospinal rhinorrhea. Acta Otolaryngol

Suppl 1948;67:20-23.

Dula DJ, Fales F. The 'ring sign': is it a reliable indicator for

cerebral spinal fluid? Ann Emerg Med 1993;22:718-720.

Eljamel MS. Antibiotic prophylaxis in unrepaired CSF fistulae. Br J

Neurosurg 1993;7:501-505.

Eljamel MS. Fractures of the middle third of the face and

cerebrospinal fluid rhinorrhoea. Br J Neurosurg 1994;8:289-293.

Eljamel MS, Pidgeon CN. Localization of inactive cerebrospinal fluid

fistulas. J Neurosurg 1995;83:795-798.

Eljamel MS, Pidgeon CN, Toland J, JB, O'Dwyer AJ. MRI

cisternography, and the localization of CSF fistulae. Br J Neurosurg

1994;8:433-437.

Friedman M, Venkatesan TK, Calderelli DD. Composite mucochondral

flap for repair of cerebrospinal fluid leaks. Head Neck 1995;17:414-

418.

Gjuric M, Goede U, Keimer H, Wigand ME. Endonasal endoscopic closure

of cerebrospinal fluid fistulas at the anterior cranial base. Ann

Otol Rhinol Laryngol 1996;105:620-623.

Gupta V, Goyal N, Gaikwad S, Sharma A. MR evaluation of CSF

fistulae. Acta Radiol 1997; 38:603-609.

Hao SP. Transnasal endoscopic repair of cerebrospinal fluid

rhinorrhea: An interposition technique. Laryngoscope 1996;106:501-

503.

Hirsch O. Successful closure of cerebrospinal fluid rhinorrhea by

endonasal surgery. Arch Otolaryngol 1952;56:1-12.

Hoffman RA. Cerebrospinal fluid leak following acoustic neuroma

removal. Laryngoscope 1994;104:40-58.

Hubbard JL, Mc TJ, Pearson BW, Laws ER. Spontaneous

cerbrospinal fluid rhinorrhea: evolving concepts in diagnosis and

surgical management based on the Mayo Clinic experience from 1970

through 1981. Neurosurgery 1985;16:314-321.

RGM, NS, on IJA. The endoscopic treatment of

cerebrospinal fluid rhinorrhoea: the Nottingham experience. J

Laryngol Otol 1997;111:125-128.

DBS, Brennan P, Toland J, O'Dwyer AJ. Magnetic resonance

imaging in the evaluation of cerebrospinal fluid fistulae. Clin

Radiol 1996;51:837-841.

Kelley TF, Stankiewicz JA, Chow JM, Origitano TC, Shea J. Endoscopic

closure of postsurgical anterior cranial fossa cerebrospinal fluid

leaks. Neurosurgery 1996;39:743-746.

Lanza DC, O'Brien DA, Kennedy DW. Endoscopic repair of cerebrospinal

fluid fistulae and encephaloceles. Laryngoscope 1996;106:1119-1125.

Lloyd MNH, Kimber PM, Burrows EH. Post-traumatic cerebrospinal fluid

rhinorrhoea: modern high-definition computed tomography is all that

is required for the effective demonstration of the site of leakage.

Clin Radiol 1994;49:100-103.

Mattox DE, Kennedy DW. Endoscopic management of cerebrospinal fluid

leaks and cephaloceles. Laryngoscope 1990;100:857-862.

McGuirt WF, Stool SE. Cerebrospinal fluid fistula: the

identification and management in pediatric temporal bone fractures.

Laryngoscope 1995;105:359-364.

McMormack B, PR, Persky M, Rothstein S. Extracranial repair

of cerebrospinal fluid fistulas: technique and results in 37

patients. Neurosurgery 1990;27:412-417.

Nandapalan V, ID, Swift AC. Beta-2-transferrin and

cerebrospinal fluid rhinorrhoea.

Clin Otolaryngol 1996;21:259-264.

Nishihira S, McCaffrey TV. The use of fibrin glue for the repair of

experimental CSF rhinorrhea. Laryngoscope 1988;98:625-627.

Ommaya AK. Spinal fluid fistulae. Clin Neurosurg 1976;23:363-392.

Papay FA, Maggiano H, Dominquez S, Hassenbusch SJ, Levine HL,

Lavertu P. Rigid endoscopic repair of paranasal cerebrospinal fluid

fistulas. Laryngoscope 1989;99:1195-1201.

Persky MS, Rothstein SG, Breda SD, Cohen NL, P, Ransohoff J.

Extracranial repair of cerbrospinal fluid otorhinorrhea.

Laryngoscope 1991;101:134-136.

Porter MJ, s GB, Zeman AZJ, Keir G. Use of protein

electrophoresis in the diagnosis of cerebrospinal fluid rhinorrhoea.

J Laryngol Otol 1992;106:504-506.

Ryall RG, Peacock MK, Simpson DA. Usefulness of beta 2-transferrin

assay in the detection of cerebrospinal fluid leaks following head

injury. J Neurosurg 1992;77:737-739.

Shapiro SA, Scully TS. Closed continuous drainage of cerebrospinal

fluid via a lumbar subarachnoid catheter for treatment or prevention

of cranial/spinal cerebrospinal fluid fistula. Neurosurgery

1992;30:241-245.

Skedros DG, Cass SP, Hirsch BE, Kelley RH. Beta-2 transferrin assay

in clinical management of cerebral spinal fluid and perilymphatic

fluid leaks. J Otolaryngol 1993;22:341-344.

Stankiewicz JA. Cerebrospinal fluid fistula and endoscopic sinus

surgery. Laryngoscope 1991; 101:250-256.

Swanson SE, Chandler WF, Kocan MJ, Bogdaserian RS. Flow regulated

continuous spinal drainage in the management of cerbrospinal fluid

fistulas. Laryngoscope 1985;95:104-106.

Tolley NS. A clinical study of spontaneous CSF rhinorrhoea.

Rhinology 1991;29:223-230.

Tolley NS, Lloyd GAS, HOL. Radiological study of primary

spontaneous CSF rhinorrhoea. J Laryngol Otol 1991;105:274-277.

Wax MK, Ramadan HH, Ortiz O, Wetmore SJ. Contemporary management of

cerebrospinal fluid rhinorrhea. Otolaryngol Head Neck Surg

1997;116:442-449.

Wormald PJ, McDonogh M. 'Bath-plug' technique for the endoscopic

management of cerebrospinal fluid leaks. J Laryngol Otol

1997;111:1042-1046.

Yerkes SA, DH, Fisher WS. Spontaneous cerebrospinal fluid

rhinorrhea. Ear Nose Throat J 1992;71:318-320.

Yessenow RS, McCabe BF. The osteo-mucoperiosteal flap in repair of

cerbrospinal fluid rhinorrhea: A twenty year experience. Otolaryngol

Head Neck Surg 1989;101:555-558.

Yoon JH, Lee JG, Kim SH, Park IY. Microscopical surgical management

of cerebrospinal fluid rhinorrhoea with free grafts. Rhinology

1995;33:208-211.

Zeitouni AG, Freinkiel S, Mohr G. Endoscopic repair of anterior

skull base cerebrospinal fluid fistulas: an emphasis on

postoperative nasal function maximization. J Otolaryngol 1994;

23:225-227.

Zlab MK, GF, Daly DT, Yonkers AJ. Cerebrospinal fluid

rhinorrhea: a review of the literature. Ear Nose Throat J

1992;71:314-317.