Clinical manifestations

Presentation and course

According to the most current criteria (44), the following are required for the diagnosis of idiopathic intracranial hypertension:

The diagnosis is considered “definite” if the patient fulfills criteria A through E.

Patients meeting criteria A through D whose CSF pressure is below the limit specified in criterion E have “probable” idiopathic intracranial hypertension.

If papilledema is not present, the diagnosis of idiopathic intracranial hypertension should not be made unless criteria B through E are satisfied and the patient has a unilateral or bilateral sixth cranial nerve palsy.

If papilledema and a sixth cranial nerve palsy are not present, idiopathic intracranial hypertension is “suggested” when criteria B through E are fulfilled and at least 3 of the following imaging findings are apparent:

Idiopathic intracranial hypertension should be differentiated from intracranial hypertension without ventriculomegaly from an identifiable cause. Causes of “secondary” intracranial hypertension include a variety of medications, cerebral venous sinus thrombosis, cerebral arteriovenous fistula, spinal cord tumors, gliomatosis cerebri, and other conditions. Clinical features of idiopathic and secondary intracranial hypertension may be identical. However, because the treatment of secondary forms may be different, it is extremely important to investigate for secondary causes before concluding that intracranial hypertension is idiopathic. In this article, the term “pseudotumor cerebri syndrome” is used to describe aspects of both forms, unless otherwise specified.

Most patients with idiopathic intracranial hypertension are obese females. Patients with idiopathic intracranial hypertension and a normal body mass index (BMI) are likely to have a secondary cause (ie, medication-induced) of intracranial hypertension (13). Headache is present in 85% to 90% of patients and is the initial manifestation in most patients. The headache phenotype is variable but is usually characterized as severe, daily, and pulsatile, and it is often accompanied by migrainous features such as photophobia, phonophobia, and nausea (82). Visual signs and symptoms are frequent. Transient visual obscurations occur in three fourths of patients described as a blurring or complete loss of vision lasting from seconds to minutes. Papilledema is the hallmark of the disorder and is present in almost all patients with idiopathic intracranial hypertension.

Signs of optic disc swelling include elevation of the disc margins, a peripapillary halo, venous congestion and tortuosity, retinal exudates, nerve fiber layer hemorrhages, and retinal infarcts (also known as “cotton wool spots”). Extension of the disc swelling into the retina may produce choroidal folds and macular edema. In one fourth of patients, a sixth nerve palsy is present as a non-localizing sign of increased intracranial pressure. Nonvisual symptoms include back and shoulder pain, weakness, numbness, and pulsatile tinnitus; pulsatile tinnitus, when present, is highly suggestive of this disorder. Some patients are asymptomatic and are diagnosed when papilledema is discovered on a routine ophthalmic examination (48).

Idiopathic intracranial hypertension occurring after 44 years of age is uncommon. Obesity, female predominance, and headache are less common in this population (08; 13).

Although papilledema at the time of diagnosis is typical of idiopathic intracranial hypertension, the disc edema may be asymmetrical or, rarely, absent. A variant of “idiopathic intracranial hypertension without papilledema” may be considered in patients with a chronic daily headache that does not respond to standard headache therapy (124). However, unless a sixth nerve palsy is present, neuroimaging evidence of increased intracranial pressure is required to suggest this diagnosis (10; 44). Compartmentalization of the subarachnoid space of the optic nerves with impeding bidirectional CSF flow from the brain may explain this finding (67). Alternatively, the patient may have had mild papilledema months to years prior that subsequently resolved. The entity of idiopathic intracranial hypertension without papilledema is uncommon and not well defined. A prospective study in 62 patients with chronic migraine from a headache center found elevated CSF pressure in 6 patients, 5 of whom were obese (BMI over 25) (120). Idiopathic intracranial hypertension without papilledema was estimated to account for approximately 5% of patients with idiopathic intracranial hypertension seen in a neuroophthalmologic practice (28); more recently, a prospective observational study in Italy reported only a 2.5% prevalence of idiopathic intracranial hypertension without papilledema among refractory chronic headache patients (36). Compared to patients with papilledema, patients tend to be diagnosed later in their course, have higher rates of nonorganic visual loss, and respond less favorably to medical therapy (28).

Idiopathic intracranial hypertension in infants and young children produces a different constellation of signs and symptoms. Vomiting, lethargy, stiff neck, and focal neurologic deficits may be prominent. Strabismus from sixth nerve palsy is a common presenting sign (70). An open fontanelle in infants may preclude the development of papilledema. In contrast to adults with idiopathic intracranial hypertension, the gender ratio in prepubescent children is nearly equal, and obesity is less likely to be present (96). Behavioral changes, including disrupted attention and concentration, may be an early sign in school-age children and generally remit with treatment (94).

Prognosis and complications

The most frequently reported complication of idiopathic intracranial hypertension is visual loss. The incidence of visual loss in idiopathic intracranial hypertension cannot be estimated with certainty due to the lack of prospective studies. Permanent visual field loss (excluding blind spot enlargement) occurs in approximately 25% to 30% of patients (20). Visual acuity loss occurs in approximately 10% of patients, with severe or complete loss in 7% (20). Adolescent females with idiopathic intracranial hypertension may be at higher risk of permanent visual loss than adults and young children (111). A retrospective study of 450 patients in the United States found that African Americans were more likely than those of other ethnic groups to have severe visual loss in at least 1 eye (14).

With scanning laser polarimetry, retinal ganglion cell axon loss may be detected in patients after regression of their papilledema even in the absence of optic disc pallor (76). Patients with idiopathic intracranial hypertension caused by cerebral venous thrombosis may deteriorate rapidly with a poor visual outcome. Systemic hypertension, renal failure, and increased intraocular pressure are also associated with a poor visual prognosis (25). Psychogenic visual loss in patients with idiopathic intracranial hypertension presents an additional management challenge, as practitioners aim to avoid unnecessary surgical intervention (88). Of importance, many patients will continue to have headaches despite optimal treatment of their intracranial pressure (46).

The Idiopathic Intracranial Hypertension Treatment Trial, which studied patients with mild visual loss, is discussed in detail in the Management section. In conjunction with a supervised weight loss diet, acetazolamide was superior to placebo in improving perimetric mean deviation, papilledema grade, and weight reduction. Among study patients, reductions in vision-specific quality of life at baseline were comparable to those of patients with multiple sclerosis and a history of optic neuritis. Headache was a significant contributor to reduced quality of life, in addition to visual acuity, perimetric mean deviation, diplopia, and transient visual obscurations. Greater improvements in quality of life assessments at 6 months occurred in the group receiving active treatment with acetazolamide (26). Risk factors among the 7 participants who met criteria for treatment failure in the study were male sex, high-grade papilledema at baseline, and decreased visual acuity at baseline (122).

A small percentage of patients have a fulminant course of idiopathic intracranial hypertension (115). Symptoms develop rapidly with an early rapid decline in vision. Aggressive intervention is required to preserve vision, although some patients are refractory to treatment.

There are few long-term data regarding the prognosis of idiopathic intracranial hypertension. One report of 20 patients followed for over 10 years at a neuroophthalmology clinic found that 11 patients had a stable course without worsening, and 9 patients had worsening after a stable course. Six of the 9 patients worsened between 28 and 135 months after their initial presentation and had recurrent papilledema 12 to 78 months after their initial remission (102). Weight gain may provoke a recurrence (69). Although idiopathic intracranial hypertension may be a chronic disorder in some patients, these findings are likely influenced by ascertainment and referral bias, as patients who are doing well are less apt to return for follow-up visits.

Most children with idiopathic intracranial hypertension have a favorable clinical outcome, although 1 study showed an overall recurrence rate of 23.7% (109). Most recurrences occurred within 18 months of diagnosis, and no specific risk factors for recurrence or predictors of outcome were identified in this cohort. A retrospective study of pediatric idiopathic intracranial hypertension completed in Israel indicated that the majority of patients responded well to acetazolamide therapy, with complete resolution of symptoms and papilledema in 76.6% of patients; 23.4% required surgical interventions (117). Twenty-six percent of patients relapsed after discontinuation of acetazolamide; nonresponders to medical therapy and those who relapsed were significantly younger in age (117).

Clinical vignette

A 12-year-old girl developed headaches that were intermittent and periocular. They were throbbing and pressure-like with accompanying dizziness, phonophobia, nausea, and vomiting. There was no photophobia or osmophobia. The pain worsened with routine activity, and she missed 6 days of school. She heard intracranial noises “like air rushing through my head” that were pulse-synchronous. One week later, she had double vision. A CT scan of the brain was normal, and she was treated with amoxicillin and decongestants for a presumed sinus infection. When she developed a noticeable esotropia, she came to the emergency department where a lumbar puncture revealed a high CSF pressure and normal CSF constituents. Examination found normal visual acuity, bilateral blind spot enlargement, and bilateral grade 3 optic disc edema. The patient had taken minocycline for acne for 6 months, which was discontinued. Acetazolamide was prescribed. The diplopia resolved shortly after the lumbar puncture, and she was asymptomatic with normal optic nerves within 1 month.

This case illustrates several important aspects of diagnosis and management. The patient had secondary idiopathic intracranial hypertension, and her symptoms were identical to those of the idiopathic form. The patient was appropriately treated with acetazolamide in addition to discontinuing the minocycline.

Biological basis

Etiology and pathogenesis

The exact pathophysiology of idiopathic intracranial hypertension remains unknown (85). Often documented is a dysfunction of CSF dynamics (CSF hypersecretion or decreased CSF absorption) or increased intracranial venous pressure. Case-control studies show that obesity, female gender, and recent weight gain are significantly more common in adults with idiopathic intracranial hypertension than in controls.

Pathological evidence of extracellular brain edema has been refuted (121). Occult venous sinus thrombosis may be present in patients with otherwise typical idiopathic intracranial hypertension, suggesting a vascular etiology (21; 64). The lack of ventriculomegaly may be attributed to cerebral venous hypertension (68). Any condition that decreases flow through the arachnoid granulations or obstructs the venous pathway from the arachnoid granulations to the right heart may elevate intracranial pressure without producing ventriculomegaly. A retrospective case series of 3 patients found that in susceptible persons with dural venous sinus stenoses on venous imaging, prolonged and intense coughing can trigger idiopathic intracranial hypertension (17). Such aggressive and prolonged Valsalva maneuvers triggering clinical manifestations of idiopathic intracranial hypertension support a hydraulic model of pathogenesis where patients with high venous mural compliance may be more susceptible to resurgences of idiopathic intracranial hypertension (17). Considerable attention has been focused on abnormalities in cerebrospinal fluid absorption at the level of the arachnoid villi, resulting from increased resistance to drainage across the arachnoid membrane. However, evidence suggests that the extracranial lymphatics, rather than the arachnoid villi, may be the primary site of CSF absorption (61).

Obstructive sleep apnea and idiopathic intracranial hypertension may coexist, and apneic episodes have been linked to a rise in intracranial pressure. A case-control study of men with idiopathic intracranial hypertension found that obstructive sleep apnea and symptoms associated with testosterone deficiency were more common in patients than in controls (39). Polycystic ovary syndrome, also associated with marked obesity, androgen excess, and thrombophilia, may be associated with intracranial hypertension (51). Ongoing investigation is underway regarding the role of adipose tissue, androgens, and gut peptides. Obesity has long been associated with idiopathic intracranial hypertension. Adipose tissue serves as an endocrine organ, secreting numerous proinflammatory cytokines, chemokines, adipokines, and hormones that may be pathogenically linked to the development of idiopathic intracranial hypertension (54). Leptin is an adipokine that mediates hypothalamic regulation of satiety and weight control, and investigators have found elevated CSF leptin levels in idiopathic intracranial hypertension patients, even when controlled for body weight (07). Such findings raise the possibility of leptin resistance among idiopathic intracranial hypertension patients, as leptin levels are high, yet the affected patients remain obese. Leptin receptors are located within the choroid plexus, which is the principal site of CSF production, representing another possible connection to the development of idiopathic intracranial hypertension (54).

Obesity can also be viewed as a chronic inflammatory condition, which leads to a theory of pathogenic inflammation causing idiopathic intracranial hypertension. Cytokine antibody arrays and ELISA assessment showed significant elevations of CSF chemokine CCL2 in idiopathic intracranial hypertension patients compared to controls (24). Natriuretic propeptide levels in the CSF of idiopathic intracranial hypertension patients are similar to those in control subjects, although plasma pro-B-type natriuretic peptide levels are lower in patients with idiopathic intracranial hypertension than in controls; plasma concentrations of B-type natriuretic peptide are inversely associated with body mass index and may increase during weight loss (107). Significant elevations of IL-2 and IL-17 in the CSF of idiopathic intracranial hypertension patients have been documented (54).

Additionally, obesity is linked to conditions characterized by an excess secretion of glucocorticoids, such as Cushing syndrome. Cortisol is regulated by the enzyme 11β-hydroxysteroid dehydrogenase, type 1 of which is abundant within adipose tissue. This enzyme has also been found within the choroid plexus, where it fuels cortisol, and possibly CSF, production. A novel 11β-hydroxysteroid dehydrogenase type 1 inhibitor is being tried as a possible treatment modality for idiopathic intracranial hypertension through a phase 2, double-blind, randomized, placebo-controlled study (83).

Supporting a pathogenic role of androgens are some case reports detailing the development of idiopathic intracranial hypertension among patients undergoing gender reassignment with testosterone therapy (reassigning gender from female to male) (55). Idiopathic intracranial hypertension has been reported among hypogonadal males, particularly in men after androgen deprivation therapy for prostate cancer. There may be a “pathophysiological window” of testosterone levels in humans, with androgen excess in women and androgen deficiency in men that causes metabolic disturbances such as increased visceral fat deposition and insulin resistance, thus creating the possible neurometabolic syndrome of idiopathic intracranial hypertension.

The role of gut peptides in regulating intracranial pressure also remains at the forefront of idiopathic intracranial hypertension research. Glucagon-like peptide-1 (GLP-1) is a gut peptide secreted by the small intestine in response to food consumption. It stimulates glucose-dependent insulin secretion, inhibits glucagon release, and triggers satiety and weight loss. GLP-1 receptors have been localized to the choroid plexus; their role in modulating CSF production is speculative (54). Exendin-4, a GLP-1 mimetic currently utilized in the treatment of obesity and diabetes mellitus, was shown to reduce intracranial pressure in rats (11). The possible treatment application to humans with idiopathic intracranial hypertension requires further investigation.

Familial intracranial hypertension may occur in up to 10% of patients. It is most often associated with obesity and suggests a dominant inheritance, but no genetic defect has been identified (19). The first genome-wide association study was conducted utilizing chromosomal DNA from 95 patients enrolled in the Idiopathic Intracranial Hypertension Treatment Trial, compared to controls. Of over 300,000 single nucleotide polymorphisms evaluated, 3 were identified as associated with idiopathic intracranial hypertension (rs2234671 on chromosome 2, rs79642714 on chromosome 6, and rs200288366 on chromosome 12), as well as 3 candidate regions with multiple associated single nucleotide polymorphisms on chromosomes 5, 13, and 14 (73). Candidate genes LINC00359 and FOXN3 were identified for further studies.

Case reports implicate a variety of medications, vitamin A supplementation, and diseases including renal failure, vitamin D-dependent rickets, and HIV infection, as secondary causes of idiopathic intracranial hypertension (25; 89). Associated medications include antibiotics (ofloxacin, tetracycline and related compounds, trimethoprim-sulfamethoxazole, nalidixic acid, nitrofurantoin), vitamins (retinoids, vitamin A, isotretinoin), psychotropics (lithium), miscellaneous medications (cyclosporine, benzene hexachloride, indomethacin), and endocrine medications (beta-human chorionic gonadotropin hormone, growth hormone, thyroid hormone, synthetic LH-RH, anabolic steroids, and corticosteroids) (25; 40). A case-control study found the adjusted rate ratio for current users of fluoroquinolone antibiotics (obtained prescription within 15 days of idiopathic intracranial hypertension diagnosis) was 5.67, suggesting an increased risk among current users of fluoroquinolones (108). To date, there is no compelling evidence implicating either implantable or oral contraceptives as a cause, although a retrospective case series found an increased risk of developing idiopathic intracranial hypertension among women with levonorgestrel-eluting intrauterine devices (119). The association may be explained by the fact that idiopathic intracranial hypertension occurs more often among women with obesity, headache, and/or polycystic ovarian syndrome, the same population that is more likely to be intolerant to oral contraceptives, requiring such an intrauterine contraceptive device (119). Medication-induced intracranial hypertension is frequently identified in children (70).

Secondary intracranial hypertension in adults may result from impairment of cranial venous outflow and subsequent venous hypertension. Chronic ear disease, head injury, intracranial arteriovenous fistulas, meningeal tumors, surgical ligation of extracranial veins, cardiac failure, chronic respiratory disease, and jugular paragangliomas can all contribute to cranial venous outflow obstruction (60). Transverse sinus stenosis has been widely implicated, although whether it is the cause or the result of elevated intracranial pressure remains debatable (will be discussed further in the Management section). Venous sinus stenosis can resolve when the intracranial pressure is lowered by lumbar puncture or shunting procedure. However, direct retrograde cerebral venography and manometry may demonstrate morphological and functional venous outflow obstruction. The presence of ophthalmoparesis, other than sixth nerve palsy, should prompt the search for venous sinus thrombosis or a hypercoagulable state (41).

Case control studies have not been performed in the pediatric population to identify those conditions associated with secondary intracranial hypertension. Case reports suggest a link to parameningeal infections, endocrinopathies, Lyme disease, vitamin A usage, human growth hormone, Down syndrome, venous sinus thrombosis from mastoiditis, otitis media, obesity in adolescents (but not in preadolescents), endocrine disorders (such as hypoparathyroidism), medications, and malnutrition (25; 70; 12). Nonetheless, even when a potential inciting exposure is identified, obesity should be considered in all children with idiopathic intracranial hypertension. A study of 60 children with secondary pseudotumor cerebri found no difference in body mass index compared to children with primary (idiopathic) pseudotumor cerebri, with 79% being overweight or obese (93).

Epidemiology

Several epidemiologic studies found an average annual age-adjusted incidence rate of 0.9 per 100,000 for the total population, with a rate of 3.3 per 100,000 for females aged 15 to 44 years. Obesity (20% above ideal body weight) increased the incidence rate to approximately 19 per 100,000 among women ages 20 to 44 years (31). The female-to-male ratio is approximately 8 to 1 (31; 65). The mean age at time of diagnosis is 30 years.

The incidence of idiopathic intracranial hypertension seems to reflect the prevalence of obesity in the population. There is evidence that the incidence of idiopathic intracranial hypertension in the United States has at least doubled since 1990, commensurate with the rising incidence of obesity (58). A study from Israel found an overall annual incidence of 2.02 per 100,000 (3.17 per 100,000 for women and 0.85 per 100,000 for men), among whom 83.4% of patients were obese. However, routine screening of morbidly obese individuals for signs of idiopathic intracranial hypertension has an exceedingly low yield (72).

Prevention

At present, there is no method to prevent idiopathic intracranial hypertension, although maintaining a normal weight is likely useful.

Differential diagnosis

Confusing conditions

Brain tumors. Although the diagnosis of idiopathic intracranial hypertension may seem apparent, particularly when it affects an obese female of childbearing age, it is a diagnosis of exclusion. Most importantly, a mass lesion causing intracranial hypertension must be ruled out. The initial manifestations of space-occupying lesions include headaches, other focal neurologic deficits, behavioral changes, cognitive decline, seizures, fatigue, and visual manifestations. Such signs and symptoms of brain tumors overlap with those of the idiopathic intracranial hypertension. A study of children with brain tumors who were first evaluated by an ophthalmologist found that the most common visual findings were decreased vision with disc pallor or papilledema (04). Less common presentations were strabismus with either disc pallor or swelling, acquired esotropia or exotropia, nystagmus, and papilledema with headache.

Headaches due to other causes. Clinicians should be aware of the striking similarities between the headache phenotypes of brain tumors, migraine, and idiopathic intracranial hypertension. Tension headache, migraine headache, medication overuse headache, and depression may coexist in patients with idiopathic intracranial hypertension (46). Therefore, it is essential to perform a careful funduscopic examination in patients with long-standing headache, particularly in those who are female or obese; infrequently, chronically increased intracranial pressure very rarely exists without papilledema, as discussed previously (idiopathic intracranial hypertension without papilledema) (28).

Secondary intracranial hypertension. Secondary intracranial hypertension in adults may result from impairment of cranial venous outflow and subsequent venous hypertension. Chronic ear disease, head injury, intracranial arteriovenous fistulas, meningeal tumors, surgical ligation of extracranial veins, cardiac failure, chronic respiratory disease, and jugular paragangliomas can all contribute to cranial venous outflow obstruction (60).

Other disorders that may be mistaken for idiopathic intracranial hypertension include cerebral venous sinus thrombosis, intracranial infection, and intracranial malignancy. Cerebral venous sinus thrombosis was detected in approximately 9% of idiopathic intracranial hypertension patients seen in tertiary referral centers (81). Most of the patients were “atypical,” and an underlying cause for the venous sinus thrombosis was identified in all but 2 patients. Headache may precede the onset of focal deficits and seizures in patients with cerebral venous sinus thrombosis, and it is frequently refractory to common pain medications (57). Transverse sinus stenosis has been widely implicated. The presence of ophthalmoparesis, other than sixth nerve palsy, should prompt the search for venous sinus thrombosis or a hypercoagulable state (41).

Associated or underlying disorders

The Biologic basis section (etiology and pathogenesis) discusses associated and underlying disorders in detail. The most relevant conditions, including both those associated with IIH and those responsible for secondary intracranial hypertension, can be viewed from the Associated Disorders link in the article menu.

Diagnostic workup

Neuroimaging studies and a lumbar puncture are essential to the diagnosis of idiopathic intracranial hypertension. A careful measurement of the opening pressure in the lateral decubitus position reveals spinal fluid pressures above 250 mm water. Pressures between 200 to 250 mm water are nondiagnostic in adults. The upper limit of normal CSF pressure for diagnosing idiopathic intracranial hypertension in children is 280 mm CSF (06). The spinal fluid contents are normal. Papilledema is expected at the time of diagnosis, although it may be absent with preexisting optic atrophy, recurrent disease, or very early in the course of the disease.

Imaging studies are used to exclude a secondary cause of intracranial hypertension. CT and CTV exclude many such causes, including dural venous sinus thrombosis, but are much less sensitive than MRI and MRV in detecting fistulas, gliomatosis cerebri, meningeal disorders, and the subtle signs of idiopathic intracranial hypertension. MRI and MRV are particularly applicable when evaluating “atypical” patients (eg, men or non-obese women who would be unlikely to have idiopathic intracranial hypertension, those with severe manifestations, and those with rapid worsening) (66; 99). In idiopathic intracranial hypertension, CT and MRI often show an empty sella, flattening of the posterior sclerae, distention and/or tortuosity of the optic nerve complex, protrusion of the optic nerve head into the vitreous cavity (papilledema), and tortuosity of the optic nerve. The empty sella results from bony enlargement of the sella from chronically raised intracranial pressure (75). CTV and MRV often reveal smooth-walled transverse sinus stenosis or sinus flow gaps, often mistakenly interpreted as thromboses (22). Although MRI findings are helpful for diagnostic purposes, their presence or absence does not predict visual outcome (100).

Ophthalmologic evaluation of visual acuity, color vision, formal visual field testing, extraocular motility and alignment, as well as a fundoscopic examination is requisite. Serial perimetry (visual field testing) is important to detect and prevent visual loss. Common visual field defects include enlargement of the physiologic blind spot, nasal step defects, arcuate defects, and generalized peripheral constriction (62). Echography (A scan and B scan) and optical coherence tomography may be helpful in differentiating idiopathic intracranial hypertension from pseudopapilledema related to optic disc drusen (80). Spectral domain optical coherence tomography was studied prospectively in the Idiopathic Intracranial Hypertension Treatment Trial; it was more sensitive than fundus photography in detecting retinal and choroidal folds in study participants (104). Retinal nerve fiber layer thickness, total retinal thickness, and optic nerve head volume measurements showed greater improvement with acetazolamide and dietary management than with placebo and dietary management (Optical Coherence Tomorgraphy Substudy Committee 2015). Optical coherence tomography measurements were more strongly correlated with papilledema grade at baseline than at 6 months.

Early visual acuity loss in patients with idiopathic intracranial hypertension arises from optic neuropathy and/or retinal changes, such as subretinal fluid, chorioretinal folds, and peripapillary choroidal neovascularization (18). The outer retinal changes are mostly reversible, and the degree of optic neuropathy, as determined by an algorithm based on optic nerve and macular optical coherence tomography, best predicted the visual outcome.

Management

Medication and diet. Treatment of this condition is predicated, to some extent, on determining its cause. Medical therapy has included a supervised weight loss program, repeated lumbar punctures, salt and fluid restriction, carbonic anhydrase inhibitors, diuretics, and medical treatment of headaches (47; 59; 74; 105).

Discontinuation of a causative agent is mandatory but may not be sufficient. Patients with intracranial hypertension from a secondary cause often need additional treatment to prevent visual loss (42). Treatment strategies are similar in adults and children.

The Idiopathic Intracranial Hypertension Treatment Trial, sponsored by the National Eye Institute, studied the effect of acetazolamide and a supervised diet (aimed at losing at least 6 percent of body weight within 6 months) compared to placebo and a supervised diet in patients with mild visual loss (45). Perimetric mean deviation as measured by the Humphrey Visual Field Analyzer ranged from -2 to -7 dB at study entry, and the primary outcome measure was change in the perimetric mean deviation at 6 months. The starting dose of study medication was 500 mg twice daily and was gradually increased to 2 grams twice daily, or the maximum tolerated dose (123). At 6 months, participants randomized to acetazolamide therapy had improved perimetric mean deviation, papilledema grade, vision-related quality of life, and a greater reduction in weight than patients randomized to placebo and weight loss. The overall treatment effect was an improvement of 0.71 dB, with a more robust effect (2.27 dB) in participants with more severe papilledema at baseline (Frisén papilledema grade of 3 to 5). Acetazolamide was not superior to placebo for reducing headache disability. Based on the study results, treatment with acetazolamide up to 4 grams daily combined with a weight loss regimen is recommended for patients with mild visual loss. There are no evidence-based guidelines for treating patients with severe visual field loss at presentation (90), although corticosteroids are often utilized for acute and severe visual loss. But some experts contend that patients worsen as the steroids are withdrawn, precipitated by a rebound rise in intracranial pressure. The side effects associated with long-term steroid use are undesirable in these patients.

Therapeutic lumbar punctures may be useful as a temporizing measure in cases of fulminant idiopathic intracranial hypertension, but repeated lumbar punctures as a long-term therapeutic option pose an infection risk and postprocedure overdrainage with resultant low-tension headache (62).

Based on the posited metabolic pathogenesis of idiopathic intracranial hypertension, an 11β-hydroxysteroid dehydrogenase type 1 inhibitor is currently being investigated as a treatment through a phase 2, double-blind, randomized, placebo-controlled trial (83). The primary outcome will assess the effect of this novel medication on intracranial pressure over a 12-week time frame. Secondary outcomes will include symptoms, visual function, papilledema, and safety.

Surgery. Surgical options for idiopathic intracranial hypertension are generally reserved for patients with moderate-to-severe visual loss when medical therapy alone is inadequate. Surgical options include optic nerve sheath fenestration, cerebrospinal fluid diverting procedures (lumboperitoneal shunting, ventriculoperitoneal shunting), temporizing cerebrospinal fluid diversion [lumbar or ventricular drain], and cerebral venous sinus stenting. No prospective randomized trials have been conducted to compare 1 type of procedure to the other, or to maximal medical therapy; the choice of procedure often depends on available resources at various institutions (118). At times, more than 1 type of procedure is necessary.

Optic nerve sheath fenestration. The mechanism by which optic nerve sheath fenestration resolves papilledema and preserves visual function is not entirely elucidated. It is believed to serve as an additional pathway for CSF egress, or cause subarachnoid scarring around the nerve, serving as a mechanical blockade to shift pressures away from the retrolaminar optic nerve. Success of optic nerve sheath fenestrations may also be related to a cyst-like structure that develops contiguous to the fenestration site, diverting CSF away from the confines of the optic nerve sheath, or by improving blood perfusion to the optic nerve head (50). Fenestrations can be completed via a medial or lateral orbitotomy or superomedial lid crease incision (62). A comprehensive review has documented improvement in papilledema and vision but less effectiveness in relieving headache (50). A retrospective study of 41 patients undergoing optic nerve sheath fenestration or shunting showed similar improvements in visual acuity but worse results in perimetric mean deviation in the group treated with optic nerve sheath fenestration (38). However, the treatments were not randomized and the baseline visual function may have influenced the surgical procedure chosen. This procedure appears to have similar efficacy in both children and adults (116). Unfortunately, it may fail at any time after surgery (37), and if the papilledema is chronic, this modality may be less successful.

Lumboperitoneal and ventriculoperitoneal shunting. Shunting offers the benefit of treating the underlying problem of elevated intracranial pressure. CSF shunting procedures have been proven effective in improving visual acuity, visual field, and papilledema (106; 56; 98). However, shunting is limited by fairly high failure rates, requiring 1 or more shunt revisions in most cases (118; 106), and it is also not an effective long-term treatment for idiopathic intracranial pressure-associated headaches (106). A programmable shunt valve may help diminish the complications of ventriculoperitoneal shunting (126), particularly when used in conjunction with a ventricular access device. This device is inserted stereotactically into the frontal horn of the lateral ventricle, providing an access point for measuring CSF pressure in the setting of suspected shunt failure (87). Stereoscopic surgical navigation is helpful to achieve cannulation of normal-sized ventricles when standard maneuvers are unsuccessful (49). Stereotactic frameless procedures have been employed with good initial results but with a 50% failure rate at 1 year (125). Direct comparison studies of the various surgical procedures have not been performed, although ventriculoperitoneal shunts may require revision less frequently than lumboperitoneal shunts (86).

Of note, a multicenter, prospective study comparing the efficacy of optic nerve sheath fenestration and ventriculoperitoneal shunting for the treatment of idiopathic intracranial hypertension is currently enrolling patients (Surgical Idiopathic Intracranial Hypertension Treatment Trial – SIGHT – Clinical Trials Identifier: NCT03501966).

Dural venous sinus stenting. Venous sinus stenting has emerged as a possible minimally invasive treatment option for medically refractory idiopathic intracranial hypertension. Improved MR venographic technology has identified dural venous sinus stenoses. Whether this venous narrowing is pathogenic or the result of elevated intracranial pressure remains unresolved. Case reports and case series document normalization of CSF opening pressures after endovascular stenting of transverse sinus stenoses (30; 127), whereas others have detailed resolution of venous sinus narrowing after CSF drainage by cervical/lumbar drainage or CSF diversion procedures (68; 09; 23; 112; 53).

Numerous uncontrolled interventional studies and case series have indicated that transverse sinus stenting is a potentially safe and effective therapeutic option, with high rates of symptomatic improvement (114; 32; 02; 29; 101; 16; 71). An extensive literature review including 8 systematic reviews/metaanalyses and 29 published patient series revealed 78% to 83% improvement in headache, 74% to 85% improvement in visual symptoms, and 87% to 97% improvement in papilledema following venous sinus stenting (35). Average reduction in CSF opening pressure was 16.8 cm of water in a single cohort of idiopathic intracranial hypertension patients after venous sinus stenting, independent of acetazolamide use or weight loss (95). A retrospective chart review documented that the duration of acetazolamide treatment among patients who underwent venous sinus stenting decreased from 571+/-544 days in medically-treated patients to 188+/-209 days among stented patients (103). Although the difference failed to reach statistical significance due to the small sample size, the results suggest that venous sinus stenting may improve medical compliance and decrease the incidence/severity of adverse effects.

The criteria for patient selection for stenting remain unclear. Noninvasive neuroimaging alone (magnetic resonance or CT venography) is unsatisfactory when identifying possible venous sinus stenoses amenable to stenting; rather, pressure gradients are generally measured across stenotic areas utilizing cerebral catheter angiography/venography and manometry. Pressure gradients of at least 8 mmHg are commonly considered necessary to prompt stenting. Patients most often undergo unilateral transverse sinus stenting, even if bilateral stenoses are present, generally along the dominant or higher pressure gradient side (62). At this time, stenting is largely considered for patients who are symptomatically refractory to or unable to tolerate conventional medical treatments, particularly including those with fulminant visual field loss. Recommendations for the selection and treatment of idiopathic intracranial hypertension patients with venous sinus stenting include (35):

Markedly elevated lumbar puncture opening pressures (> 50 cm CSF) prior to stenting and persistent papilledema after stenting were associated with a refractory course, requiring later shunting (52). Highest degrees of improvement in maximum mean intracranial venous pressure and pressure gradient after stenting have been documented among the more obese patients (97). Patients with higher mean pressure gradients and higher changes in pressure gradients after stenting tend to achieve more favorable outcomes (84). A metaanalysis of optic nerve sheath fenestration, shunting, and stenting evaluating 30 studies containing class 3 evidence (77) disclosed that visual acuity improved with all 3 modalities, with shunting and stenting demonstrating only a modest improvement in headache. The authors concluded that there is insufficient evidence to recommend or reject any of the treatment modalities. Future prospective, randomized trials including venous sinus stenting are needed to provide a more vigorous assessment of the efficacy of this newer treatment modality.

Bariatric surgery. Bariatric surgery may be helpful in the long-term management of idiopathic intracranial hypertension in morbidly obese individuals.

Bariatric surgery is not useful in the setting of acute visual loss. The first randomized controlled trial of bariatric surgery in idiopathic intracranial hypertension is underway (92). Study participants will be randomized to either bariatric surgery or a dietary weight loss program (Weight Watchers for 12 months), and followed for 5 years. The primary outcome measure will be intracranial pressure by lumbar puncture at 12 months; secondary outcome measures include visual function, papilledema, headache, quality of life, and cost-effectiveness. A cost comparison of all idiopathic intracranial hypertension treatment options has been published (62).

Outcomes

Complication rates of optic nerve sheath fenestration. These range from 15.6% to 40% (62). Such complications include: diplopia (3.4%-26%), anisocoria (tonic pupil; 9.7%-13.4%), corneal dellen (0.8%-2.2%), cranial nerve palsies (7.2%), orbital hematoma (7.2%), perilimbal conjunctival filtering blebs (less than 0.4%), conjunctival abscess (less than 0.4%), orbital apex syndrome (less than 0.4%), and traumatic optic neuropathy (less than 0.4%) (50).

Complications of cerebrospinal fluid shunting. The main complication of cerebrospinal fluid shunting is shunt failure, often defined as the need for shunt revision, replacement, or removal. Failure rates of ventriculoperitoneal shunts among idiopathic intracranial hypertension patients were 20% at 1 year, 35% at 2 years, and 52% at 3 years in a retrospective cohort; 1 out of 19 patients required removal of the ventriculoperitoneal shunt due to meningitis and gangrenous small bowel (56). A 27% rate of shunt obstruction is reported among lumboperitoneal shunts, the most common cause being migration of the peritoneal catheter (33). Shunt overdrainage also occurs more readily among lumboperitoneal shunts, with a published rate of 13%, contributing to low pressure headaches (33). Karabatsou and colleagues published a 1% infection rate per lumboperitoneal shunting procedure, or 33% patient (63). Such shunt obstruction, overdrainage, and infections among lumboperitoneal shunts contribute to failure rates.

Other complications of ventriculoperitoneal shunting include intracranial hemorrhage and difficulty with shunt placement within small ventricles of idiopathic intracranial hypertension patients, resulting in catheter malposition.

Further complications of lumboperitoneal shunting include subdural hematoma formation and cerebellar tonsillar descent/acquired Chiari malformation (up to 70% incidence in a pediatric population; 14% among idiopathic intracranial hypertension adults) (49).

More unusual complications of both types of shunts include intraabdominal pain, retroperitoneal hematoma, and cerebrospinal fluid leak (110; 50).

Due to the complications listed above, shunt revision rates can be quite high, upwards of 60% for lumboperitoneal shunts and 30% for ventriculoperitoneal shunts (01).

Complications of dural venous sinus stenting. These include restenosis proximal to the stent or along a contralateral nonstented stenosis, which may occur in up to 20% of procedures (05), subdural in-stent thrombosis, stent migration, epidural/subdural/subarachnoid hemorrhage, uncontrolled intracranial hypertension, development of an intracranial dural arteriovenous fistula, femoral pseudoaneurysm, and retroperitoneal hematoma. Two deaths have been reported, 1 from a severe cerebellar hemorrhage (03; 113; 15; 78). Stent survival was 87.8% at 120 days (05). Restenting is necessary in 9.8% to 12% of cases (114; 05).

Special considerations

Pregnancy

Although pregnancy is not a risk factor per se, idiopathic intracranial hypertension may develop or worsen in pregnancy. Most patients can be successfully managed conservatively during pregnancy with serial lumbar punctures. If vision is threatened, acetazolamide, corticosteroids, optic nerve sheath decompression, and lumboperitoneal shunting are alternatives (79). Vaginal delivery is not contraindicated in these patients. Venous sinus thrombosis must be considered when idiopathic intracranial hypertension develops in the puerperium after a miscarriage or an ectopic pregnancy (34).

Anesthesia

Anesthesia is not contraindicated with idiopathic intracranial hypertension.