Spontaneous cerebrospinal fluid rhinorrhea represents a distinct clinic entity that is likely a variant of idiopathic intracranial hypertension (IIH). Patients with spontaneous cerebrospinal fluid (CSF) leaks are generally middle-aged obese women with radiographic evidence of skull base defects, associated meningoencephaloceles, and empty sella syndrome, a common sign of increased intracranial pressure. Significant overlap exists in the characteristics of patients with spontaneous CSF leak and IIH. Endoscopic repair of the CSF fistula is the gold standard treatment for this condition, but emerging evidence supports the reduction of CSF pressure as an important adjuvant treatment in this patient population.
| EBM Question | Level of Evidence | Grade of Recommendation |
|---|---|---|
| Does spontaneous CSF leaks represent a distinct clinical entity and a variant of idiopathic intracranial hypertension? | 2b | B |
| Does a reduction in CSF pressure alone lead to resolution of CSF leaks? | 5 | D |
| Does decreasing ICP, transiently or long-term, improve outcomes of endoscopic closure of spontaneous CSF leaks? | 4 | C |
Spontaneous nasal cerebrospinal fluid (CSF) leaks likely represent a distinct clinical entity in which CSF rhinorrhea occurs in the absence of any inciting event. Historically, the term spontaneous CSF leaks included CSF leaks of multiple causes such as delayed CSF leaks after trauma, congenital skull base malformations, and skull base defects from tumors. By separating CSF leaks with a discernable cause, spontaneous CSF leaks can be uniformly evaluated and studied. Most of these patients show clinical signs and radiographic features of increased intracranial pressure (ICP). Accurately diagnosing patients with spontaneous CSF leaks is critical for the successful repair of these patients because multiple studies have identified increased ICP as a negative risk factor for successful repair. Although repair of nasal CSF leaks using endoscopic techniques has high success rates, patients with spontaneous CSF leaks have historically had a significantly lower success rate. This may be explained, in part, by the strong association between spontaneous CSF leaks and idiopathic intracranial hypertension (IIH), a disease associated with increased ICP without other cause. This article examines the evidence correlating spontaneous CSF leaks and IIH and the role of decreasing ICP in the treatment of spontaneous CSF leaks.
Normal CSF physiology
Understanding normal CSF physiology is useful in diagnosing and treating patients with CSF rhinorrhea. CSF is produced in the choroid plexus within the lateral, third, and fourth ventricles at a rate of 0.35 mL/min. CSF is produced by the choroid plexus and flows from the ventricular system into the subarachnoid space. CSF absorption occurs at the arachnoid villi along the cerebral convexities. The villi project into the dural sinuses and act as one-way valves that typically require a pressure gradient of 1.5 to 7 cm H 2 O for antegrade flow. At lower pressure differentials, the villi close, preventing retrograde flow. The total volume of CSF in adults is approximately 90 to 150 mL and is turned over 3 to 5 times daily.
Normal CSF pressure is between 5 and 15 cm H 2 O in the lumbar cistern with the patient lying in the decubitus position. The pressure varies depending on the time of day, patient age, activity level, and cardiopulmonary cycles. Neurologic symptoms can occur when ICP exceeds 15 to 20 cm H 2 O.
Clinical presentation and evaluation of spontaneous CSF leak
Patients are generally obese middle-aged women who present with spontaneous clear rhinorrhea. Although men and normal-weight women may also present with similar symptoms, the incidence is much lower. The initial diagnostic step is to confirm the presence of CSF in the clear rhinorrhea by laboratory testing. In the United States, β-2 transferrin is the most commonly used laboratory test to confirm the presence of CSF. Cerebral neuraminidase produces β-2 transferrin from β-1 transferrin by desialization. β-2 Transferrin is only found in CSF, perilymph, and the vitreous humor of the eye. The identification of β-2 transferrin in nasal secretions is highly sensitive and specific for CSF rhinorrhea.
After diagnosing CSF rhinorrhea by β-2 transferrin, localization of the skull base defect and site of CSF fistula follows. High-resolution computed tomography (CT) is the initial radiographic test of choice, allowing for evaluation of the bony integrity of the skull base and paranasal sinuses. Patients with spontaneous CSF leaks have characteristic CT findings that support the diagnosis. The bone of the skull base is broadly attenuated and thin. Arachnoid pits secondary to the bony impressions from the arachnoid villi in the skull base are present in 63% of patients with spontaneous CSF leaks. The most common sites of skull base dehiscence are the lateral recess of the sphenoid and the ethmoid roof ( Figs. 1 and 2 ). Pneumatization of the lateral recess of the sphenoid is reported in 91% of patients with spontaneous CSF leaks, in comparison with 23% to 43% in normal patients. Dehiscence of the ethmoid roof or cribiform plate is seen in 14% of patients with spontaneous CSF leaks. The presence of multiple skull base defects is common and may be present in 31% of these patients.
Magnetic resonance imaging (MRI) is a useful adjunct in the evaluation of patients with CSF rhinorrhea. Patients with spontaneous CSF leaks have the highest rate of meningoencephalocele formation, ranging from 50% to 100%. MRI is effective in assessing the contents of meningoencephalic sacs. Another benefit of MRI in the evaluation of patients with spontaneous CSF leaks is the recognition of the empty sella. Empty sella syndrome is a common radiographic finding seen in both spontaneous CSF leaks and IIH ( Fig. 3 ). Increased ICP are exerted on sites of inherent structural weakness including the fascia of the sellar diaphragm. The resulting herniation of the meninges and CSF through the sellar diaphragm produces the appearance of an empty sella on MRI. The presence of empty sella syndrome has been associated with both increased ICP and spontaneous CSF leaks. One group has suggested that the reversal of empty sella in patients with IIH is a radiographic marker of successful reversal of increased ICP.
Multiple radiographic signs support the theory that spontaneous CSF leaks and IIH are linked. The attenuated skull base with resulting skull base defects, large encephaloceles, and the presence of arachnoid pits support the hypothesis that increased ICPs are contributing to spontaneous CSF leaks. The high prevalence of empty sella syndrome in both IIH and spontaneous CSF leaks further strengthens this correlation.
Clinical presentation and evaluation of spontaneous CSF leak
Patients are generally obese middle-aged women who present with spontaneous clear rhinorrhea. Although men and normal-weight women may also present with similar symptoms, the incidence is much lower. The initial diagnostic step is to confirm the presence of CSF in the clear rhinorrhea by laboratory testing. In the United States, β-2 transferrin is the most commonly used laboratory test to confirm the presence of CSF. Cerebral neuraminidase produces β-2 transferrin from β-1 transferrin by desialization. β-2 Transferrin is only found in CSF, perilymph, and the vitreous humor of the eye. The identification of β-2 transferrin in nasal secretions is highly sensitive and specific for CSF rhinorrhea.
After diagnosing CSF rhinorrhea by β-2 transferrin, localization of the skull base defect and site of CSF fistula follows. High-resolution computed tomography (CT) is the initial radiographic test of choice, allowing for evaluation of the bony integrity of the skull base and paranasal sinuses. Patients with spontaneous CSF leaks have characteristic CT findings that support the diagnosis. The bone of the skull base is broadly attenuated and thin. Arachnoid pits secondary to the bony impressions from the arachnoid villi in the skull base are present in 63% of patients with spontaneous CSF leaks. The most common sites of skull base dehiscence are the lateral recess of the sphenoid and the ethmoid roof ( Figs. 1 and 2 ). Pneumatization of the lateral recess of the sphenoid is reported in 91% of patients with spontaneous CSF leaks, in comparison with 23% to 43% in normal patients. Dehiscence of the ethmoid roof or cribiform plate is seen in 14% of patients with spontaneous CSF leaks. The presence of multiple skull base defects is common and may be present in 31% of these patients.
Magnetic resonance imaging (MRI) is a useful adjunct in the evaluation of patients with CSF rhinorrhea. Patients with spontaneous CSF leaks have the highest rate of meningoencephalocele formation, ranging from 50% to 100%. MRI is effective in assessing the contents of meningoencephalic sacs. Another benefit of MRI in the evaluation of patients with spontaneous CSF leaks is the recognition of the empty sella. Empty sella syndrome is a common radiographic finding seen in both spontaneous CSF leaks and IIH ( Fig. 3 ). Increased ICP are exerted on sites of inherent structural weakness including the fascia of the sellar diaphragm. The resulting herniation of the meninges and CSF through the sellar diaphragm produces the appearance of an empty sella on MRI. The presence of empty sella syndrome has been associated with both increased ICP and spontaneous CSF leaks. One group has suggested that the reversal of empty sella in patients with IIH is a radiographic marker of successful reversal of increased ICP.
Multiple radiographic signs support the theory that spontaneous CSF leaks and IIH are linked. The attenuated skull base with resulting skull base defects, large encephaloceles, and the presence of arachnoid pits support the hypothesis that increased ICPs are contributing to spontaneous CSF leaks. The high prevalence of empty sella syndrome in both IIH and spontaneous CSF leaks further strengthens this correlation.
Pathophysiology: do spontaneous CSF leaks represent a distinct clinical entity and a variant of IIH?
Significant overlap exists in the demographic, clinical, and radiographic characteristics of patients with spontaneous CSF leak and IIH. The similarities between these 2 patient populations are striking and strengthen the theory that high-pressure spontaneous CSF leaks are likely a variant of IIH. Patients with IIH classically present with symptoms of headache, pulsatile tinnitus, and visual changes. In case-control trials, factors associated with IIH are female sex, reproductive age, obesity, and recent weight gain. Using the modified Dandy criteria to establish the diagnosis of IIH, 70% of patients with spontaneous CSF leaks meet the clinical, radiographic, and ICP criteria for IIH. In addition to clear rhinorrhea, patients often present with pressurelike headaches and pulsatile tinnitus. Case series of patients with spontaneous CSF leaks highlight the overlapping demographics, including female predisposition (70%–80%) and obesity (82%–92%). Case reports have described patients originally diagnosed with IIH and subsequently presenting with spontaneous CSF rhinorrhea months to years later. In most of these patients, the underlying IIH was poorly controlled, suggesting that persistent high intracranial pressure directly resulted in the subsequent CSF rhinorrhea.
The prevalence of obesity in both patient groups is remarkable and may provide some insight into the pathophysiology of both disease processes. For IIH, Radhakrishnan and colleagues calculated an increased incidence of 21.4 per 100,000 for obese women with a body mass index (BMI) of 30, approximately 10-fold to 20-fold higher than the general population. In a large retrospective review of endoscopic CSF leak repair, the average BMI between spontaneous CSF leak (35.4) and traumatic CSF leak (29.7) was statistically significant. These similarities suggest that obesity has a potential role in the development of spontaneous CSF leaks and further highlights this population as a likely variant of IIH. However, the pathophysiology of both disease processes remains unknown. If the prevalence of obesity is as high as 16%, a higher prevalence of both IIH and spontaneous CSF leaks would be expected if obesity were the primary factor for the development of this condition. Although the recent increased reporting of spontaneous CSF leaks may reflect the increased prevalence of obesity, it seems likely that an additional causal factor or event must be present to develop IIH.
Regardless of the cause of the increased CSF pressure, persistent pulsatile pressure exerted on the skull base at sites of inherent structural weakness results in bony erosion and the potential for CSF leak. Radiographic findings associated with IIH in a blinded case-control study include empty sella syndrome, abnormalities of the optic nerve, arachnoid pits, and skull base flattening, specifically posterior globe flattening. Schlosser and Bolger showed the presence of empty sella syndrome in 100% of patients with multiple spontaneous CSF leaks in comparison with 11% of patients with nonspontaneous CSF leaks. Additional radiographic features seen in spontaneous CSF leaks, such as arachnoid pits, encephaloceles, and dural ectasia, are consistent with increased ICP and support the association with IIH ( Fig. 4 ). The presence of skull base fistula and CSF leak in this patient population is not isolated to the anterior cranial fossa. Two retrospective studies of patients with spontaneous CSF otorrhea found an empty sella in 80% of patients as well as similar patient demographics such as increased BMI and female disposition.
