Idiopathic intracranial hypertension

Idiopathic intracranial hypertension is a condition in which intracranial pressure is substantially raised. The cause of the condition is unknown, but it often occurs in obese young women with menstrual irregularities. Rarely it is caused by drugs, including corticosteroid drugs. Symptoms may include headache, vomiting, and visual disturbance. The condition is not fatal but can damage the optic nerves, leading to permanent visual loss. Treatment is with diuretic drugs; occasionally, excess cerebrospinal fluid may be diverted by way of a surgically inserted shunt (artificial passage).

MRI scan showing benign intracranial hypertension

Idiopathic intracranial hypertension - technical article.

Topics covered:

  • Essentials
  • Incidence
  • Clinical features
  • Aetiology
  • Medication
  • Pathogenesis
  • Investigations
  • Management
  • Prognosis
  • Further reading


Idiopathic intracranial hypertension is a syndrome of raised intracranial pressure in the absence of an intracranial mass lesion, enlargement of the cerebral ventricles due to hydrocephalus, or venous thrombosis in the dural sinuses. Impaired cerebrospinal fluid absorption and raised cerebral venous pressure have both been implicated. Obese females of childbearing age are predominantly affected. Rarely, provocative factors include tetracycline, vitamin A derivatives, hypervitaminosis A, and vitamin A excess.

Clinical features

Characteristic presentation is with headache which may be typical of raised intracranial pressure but is frequently non-specific. Papilloedema is almost invariably present, but often not severe and may be unilateral; visual field defects are common. There are no focal cerebal signs, and a history of epilepsy is strongly against the diagnosis.

Diagnosis, treatment and prognosis

After brain imaging to exclude other causes of raised intracranial pressure, diagnosis depends on measurement of cerebrospinal fluid pressure higher than 200 mm. Composition of fluid is normal. Spontaneous remission and relapse is common, but treatments including the following aim to prevent symptoms and visual loss, which should be monitored by serial analysis of visual fields: (1) high-dose corticosteroids—for relief of acute crises, but not as long-term treatment; (2) drugs—the carbonic anhydrase inhibitors acetazolamide, and thiazide diuretics are used in mild cases; (3) surgery—some patients require lumboperitoneal shunts, optic nerve decompression, or other procedures; (4) intermittent lumbar puncture—particularly of use in pregnancy. Aside from visual loss, serious complications are rare.


Idiopathic intracranial hypertension has been widely described in the literature as pseudotumour cerebri or benign intracranial hypertension. These terms can be regarded as synonyms, with the caveat that known causes of the syndrome were included under their rubric. Idiopathic intracranial hypertension is comparatively rare in the general population with an annual incidence of approximately 1 in 100 000, but this figure rises to 19 in 100 000 in obese women of childbearing age. The disease is certainly more common in females, the preponderance over males ranging from 3:1 to 15:1. Although idiopathic intracranial hypertension is primarily a disease of young women, it may occur in childhood. Very rarely, it is familial and may occur in more than one generation.

Clinical features  

It is the hallmark of idiopathic intracranial hypertension that presenting symptoms and signs are those of raised intracranial pressure alone. The diagnosis should not be entertained in the presence of neurological features that suggest a focal lesion. Furthermore, there is a remarkable preservation of consciousness and intellectual function, rarely encountered in patients with mass lesions causing raised intracranial pressure. A history of epilepsy, either generalized or focal, virtually excludes the diagnosis of idiopathic intracranial hypertension, whereas seizures may occur in patients with venous sinus thrombosis. Preservation of cerebral function also aids the distinction of idiopathic intracranial hypertension from acute viral or bacterial meningoencephalitis. Patients with idiopathic intracranial hypertension routinely present to outpatient departments and become a medical emergency when papilloedema is seen.


This is the most common symptom and is present to some degree in almost every case. The headache may be typical of raised intracranial pressure. It is then generalized, throbbing, worse on waking, and aggravated by factors that temporarily increase cerebrospinal fluid pressure such as straining, coughing, or changing posture. Not infrequently, however, headache is mild and nonspecific so that the distinction from common tension headache can be difficult. At presentation, headache has usually been present for weeks, although sometimes for months. Although up to 50% of patients complain of nausea, typical early-morning projectile vomiting is rare. Headache may be accompanied by tinnitus, which is often pulsatile in type.


Among the medical conditions associated with idiopathic intracranial hypertension, obesity is sufficiently common to be a characteristic feature. About 90% of patients in adult series are overweight. The association with both obesity and gender is less marked in childhood.


This is a virtually universal finding and the importance of fundus examination in every patient with headache cannot be overemphasized. Papilloedema is often moderate and may be unilateral. Occasionally the appearance of the optic discs may be equivocal, and fluorescein angiography is indicated to demonstrate the characteristic leakage of dye in true papilloedema.

The classic symptom of papilloedema, which is not specific to idiopathic intracranial hypertension, is a transient obscuration of vision, often described as a fleeting greyness, a halo, or a more vivid episode of ‘Catherine wheels’ lasting for a few seconds. Obscurations may be provoked by straining or a change in posture, but may also occur spontaneously. Persistent blurring of vision may also occur and patients may describe scotomas in the field of vision associated with optic nerve damage. Occasionally, sudden and permanent loss of vision results from infarction of the optic nerve.

Visual obscurations, persistent blurring, or scotomas are reported by 30 to 70% of patients. A history of obscurations is often elicited only by direct questioning.


Above: papilloedema in patient with idiopathic intracranial hypertension

Above: Papilloedema in patient with idiopathic intracranial hypertension


Fundoscopi examination revealed bilateral papilledema

Visual field defects

Together with measurement of best-corrected visual acuity, visual field analysis is an essential investigation in the examination and follow-up of patients with idiopathic intracranial hypertension. The most common defects are enlargement of the blind spots, generalized constriction of the fields, and scotomas caused by optic nerve damage. There may be a predilection for visual field loss in the inferior nasal quadrants.


About 30% of patients complain of horizontal diplopia due to cranial nerve VI palsy, which may be bilateral. The cause is a false localizing sign of raised intracranial pressure.

convergent strabismus, and bilateral sixth nerve paresis


In most patients with idiopathic intracranial hypertension no trigger can be identified. Many clinical associations have been reported, but these may have occurred by chance. Preceding minor head injury and intercurrent infections come into this category. Furthermore, the known associations are rare with the exceptions of obesity and the predilection for females. A positive family history, vitamin deficiency, and drugs are a factor in less than 2% of cases.

Dural sinus thrombosis

Before the advent of antibiotics, benign intracranial hypertension was frequently described in chronic middle-ear disease complicated by dural sinus thrombosis. The term ‘otitic hydrocephalus’ was coined to describe this syndrome on the erroneous assumption that ventricular enlargement was present.

Although true ‘otitic hydrocephalus’ is now rare, dural sinus thrombosis remains an important differential diagnosis of idiopathic intracranial hypertension. Sinus thrombosis may complicate pregnancy, the use of oral contraceptives, head injury, venous occlusive disease due to hypercoagulability states, dehydration due to any cause, or mediastinal obstruction. Sinus thrombosis should be suspected clinically when the onset of headache is sudden and accompanied by focal signs or impaired consciousness. The clinical presentation may, however, mimic idiopathic intracranial hypertension and sinus thrombosis must be excluded radiologically.

Menstrual disorders

An association of idiopathic intracranial hypertension with pregnancy has been reported but has not been confirmed by small epidemiological studies. It is not clear whether an association with menstrual irregularity is more than would occur by chance in obese young women. An association with menarche has been reported.

In spite of the clinical associations that suggest an underlying disorder of female endocrinology, hormonal studies have not shown a consistent abnormality. The pituitary–adrenal axis is intact and occasional abnormal responses may be due to obesity rather than idiopathic intracranial hypertension. Thyroid function and prolactin secretion both appear to be normal. Vasopressin levels in cerebrospinal fluid are raised, but this is not specific and may occur in a variety of neurological diseases. Reports of a specific increase of oestrone in cerebrospinal fluid, which might link idiopathic intracranial hypertension with obesity because adipocytes are the major source of oestrone, have not been confirmed.

Deficiency states

A rare cause of intracranial hypertension in children is hypovitaminosis A due to generalized nutritional deficiency or malabsorption. In such cases the condition responds specifically to vitamin A supplements.

Vitamin A excess

Poisoning with vitamin A due to excessive consumption of fish or animal liver may also cause intracranial hypertension. The hypothesis that a disorder of vitamin A metabolism is an underlying cause of idiopathic intracranial hypertension continues to be debated. An increased concentration of retinol in cerebrospinal fluid has been reported.


Both tetracycline and the retinoids isotretinoin and etretinate, which are vitamin A derivatives, may cause intracranial hypertension during long-term treatment for acne. These drugs should not be used in combination. Intracranial hypertension due to tetracycline has also been reported during prophylaxis for malaria. All-trans-retinoic acid in treatment of acute promyelocytic leukaemia is also a cause in this category. Other drugs occasionally responsible for the syndrome include nalidixic acid, nitrofurantoin, lithium, danazol, and synthetic growth hormone. Corticosteroids may lead to intracranial hypertension during their withdrawal after chronic treatment and the syndrome may occur in Addison’s disease.

Empty sella

It has been suggested that this association in about 4% of cases is caused by raised intracranial pressure in combination with incompetence of the diaphragma sellae. The theory is supported by the finding of raised pressure at lumbar puncture in some patients with empty sella, suggesting chronic idiopathic intracranial hypertension as the underlying cause. Clinical hypopituitarism does not occur, but occasionally the empty sella may harbour a prolactinoma.


The mechanism by which intracranial pressure rises is poorly understood and the contribution of various factors controversial. As the intracranial contents are housed in a rigid container, an increase in cerebrospinal fluid pressure may result from an increase in blood volume, swelling of the brain parenchyma, or an increase in the cerebrospinal fluid volume due to overproduction or malabsorption. There is little evidence to suggest that increased blood volume or cerebrospinal fluid production is an important factor. An association with thrombophilia has been reported but the significance of the finding is unclear. Recent studies have drawn attention to the possible contribution of raised cerebral venous pressure in pathogenesis.

Swelling of the brain parenchyma

Direct evidence of swelling due to cerebral oedema is slight and a single report of oedematous changes in brain biopsies has not been confirmed. However, the tendency for the ventricles to be small may indicate an increase in cerebral volume secondary to leakage from the cerebral vascular bed rather than transudation of cerebrospinal fluid from the ventricular system. Brain imaging in idiopathic intracranial hypertension does not show the periventricular leakage of cerebrospinal fluid that occurs in hydrocephalus. Although there is currently no direct evidence for vasogenic cerebral oedema, this factor cannot be ignored, because it is one mechanism for raised pressure that does not anticipate some degree of hydrocephalus. MRI studies of brain water content have been equivocal.

Decreased absorption of cerebrospinal fluid

A defect of cerebrospinal fluid absorption is widely regarded as the important factor in the pathogenesis of idiopathic intracranial hypertension. It is assumed that the defect lies in the arachnoid villi of the superior sagittal sinus where the bulk of the cerebrospinal fluid absorption takes place. The delayed clearance of radio-iodinated human serum albumin from the ventricular system after injection into the lumbar subarachnoid space is indirect evidence of reduced cerebrospinal fluid absorption. Simultaneous cannulation of the superior sagittal sinus and the subarachnoid space has shown increased resistance to cerebrospinal fluid absorption in most cases. Finally, vitamin A deficiency in rats and cows produces a rise in intracranial pressure associated with diminished absorption of cerebrospinal fluid and histological changes in the arachnoid villi, which are reversible with vitamin A supplements.

The absence of hydrocephalus in idiopathic intracranial hypertension has been cited as an objection to the theory of reduced cerebrospinal fluid absorption. It is more significant, however, that sinus thrombosis may cause intracranial hypertension by preventing cerebrospinal fluid absorption, and is similarly associated with normal or small ventricles.

Raised cerebral venous pressure

MR venography (MRV) in idiopathic intracranial hypertension may show narrowing of the lateral and sigmoid sinuses and manometry has shown consistent venous hypertension. It is not yet clear whether these findings are a primary abnormality or secondary to raised intracranial pressure. A rise in central venous pressure due to obesity has also been implicated in pathogenesis, although the significance of the finding remains unclear.


The diagnosis of idiopathic intracranial hypertension only be confirmed can by measurement of cerebrospinal fluid pressure, but in suspected cases brain imaging is essential before proceeding to lumbar puncture.


CT scanning may show small and slit-like cerebral ventricles that increase in volume as intracranial hypertension resolves. Sagittal sinus thrombosis may be visualized on CT scanning as the characteristic ‘empty delta’ sign due to clot within the sinus. MRI is far superior to CT and frequently provides graphic images of sinus thrombosis. As a result of the implications for treatment, MRV or CT venography is recommended to exclude thrombotic disease when plain imaging is normal.

cranial mri scan shows intracranial hypertension

Above: Cranial MRI scan showing intracranial hypertension in an 11 year old male patient with no signs of venous sinus thrombosis in the MRI venography. After neuroimaging procedures, a lumbar puncture was performed revealing an opening pressure of 80 cm H2O with normal CSF composition with no evidence of pleocytosis, elevated protein or low glucose.

Segmentation of T1-weighted MR images into GM, WM, and CSF regions. Examples of T1-weighted MR images in sagittal (A, E) and coronal (C, G) planes are shown without and with color overlay of the tissue segmentation into GM (blue), WM (yellow), and CSF (red) from a control subject (top row) and a patient with IIH (bottom row)

Above: MRI scan - Segmentation of T1-weighted MR images into GM, WM, and CSF regions. Examples of T1-weighted MR images in sagittal (A, E) and coronal (C, G) planes are shown without and with color overlay of the tissue segmentation into GM (blue), WM (yellow), and CSF (red) from a control subject (top row) and a patient with IIH (bottom row). Relatively larger subarachnoid CSF spaces can be visualized in the floor of the anterior cranial fossa and in the extra-axial spaces overlying the convexities in IIH.

Cerebrospinal fluid pressure

At lumbar puncture the opening pressure is greater than 200 mmCSF, but in simple obesity the cerebrospinal fluid pressure may be as high as 250 mm. The diagnostic significance of cerebrospinal fluid pressure must therefore be correlated with the clinical picture. It is important that cerebrospinal fluid pressure is measured in the lateral decubitus position to obtain a valid reading. In the few patients whose cerebrospinal fluid pressure is equivocal, continuous monitoring may demonstrate intermittent peaks of raised pressure.

Cerebrospinal fluid analysis

The composition of the cerebrospinal fluid in idiopathic intracranial hypertension is entirely normal and the diagnosis should not be accepted in the presence of an increase in white cells or protein concentration. The unexpected finding of a pleocytosis requires further investigation to exclude meningitic infiltration or infection, including HIV. A rare syndrome resembling idiopathic intracranial hypertension occurs in association with postinfective polyneuropathy and spinal tumours. Both conditions may lead to intracranial hypertension with papilloedema and normal-size ventricles but a marked rise in cerebrospinal fluid protein.


Patients given a diagnosis of idiopathic intracranial hypertension are usually bewildered and frightened. It is important to provide a simple explanation of the nature of the condition and the rationale for treatment. Improvement has been reported with weight reduction and obese patients sholuld be offered dietary advice. In practice, however, weight loss in difficult to achieve and to maintain. With the exception of rare cases due to drug treatment, the management of idiopathic intracranial hypertension is aimed at the symptomatic reduction of intracranial pressure to protect vision and relieve headache. The methods available are difficult to evaluate because of the high spontaneous remission rate and the lack of controlled trials. Choice of treatment is further complicated by the absence of reliable risk factors for visual loss. In particular, the height of the cerebrospinal fluid pressure at diagnosis is of no prognostic significance.

In the past, repeated therapeutic lumbar puncture every 2 to 5 days has been shown to reduce cerebrospinal fluid pressure temporarily and may occasionally lead to spontaneous remission. This approach may still be useful in pregnancy, because of the risk of diuretics to the fetus. Teratogenicity of acetazolamide has been shown in animal studies.

When acute medical treatment is indicated, prednisolone 40 to 60 mg daily may be effective in relieving headache and visual obscuration due to papilloedema. However, steroids are unsatisfactory as long-term treatment because of their complications, especially in obese young females. For this reason diuretics are widely used in patients with mild symptoms. Acetazolamide, thiazide diuretics, and furosemide may relieve headache, but their efficacy in preventing slowly progressive visual loss is unproven. Acetazolamide has been favoured on the basis that carbonic anhydrase inhibition may reduce cerebrospinal fluid production, but side effects often limit its use.

As a result of the limitations of medical treatment, surgical intervention may be required, progressive visual field loss and unrelieved headache being the indications for surgery. A lumboperitoneal shunt is usually favoured, being less invasive than a ventricular approach. Unfortunately the technical failure rate of lumboperitoneal shunt is usually high and surgical revision may be required in about 50% of patients. For this reason the alternative operation, in which the optic nerve sheath is decompressed, has recently been revived, particularly in North America. This procedure produces rapid improvement in papilloedema, occasionally in both eyes after unilateral surgery. It is not clear whether long-term improvement is due to the creation of a cerebrospinal fluid fistula in the orbit or fibrosis of the meninges preventing transmission of the high cerebrospinal fluid pressure to the optic nerve head. However, headache is often unrelieved by this procedure, and lumboperitoneal shunting may not be avoided. There is also a small risk of intraoperative visual loss as well as delayed visual loss postoperatively.

Currently it would seem reasonable to begin treatment with diuretics, reserving steroids as a temporary medical treatment in patients with severe symptoms. If surgery becomes necessary, lumboperitoneal shunting seems a logical procedure, with resort to optic nerve sheath decompression in the event of repeated shunt failure. Occasionally, subtemporal decompression may be required. Stenting of the venous sinuses in the light of MRV and manometry evidence for raised cerebral venous pressure has lead to improvement in some patients but the place of this procedure in management has yet to be established.


Idiopathic intracranial hypertension is a chronic condition in most patients, but spontaneous relapse and remission of symptoms are common. There is evidence that raised intracranial pressure may be found at follow-up lumbar puncture in patients whose symptoms have been in remission for several years. Serious complications of idiopathic intracranial hypertension, other than visual loss, are rare, whereas an underlying dural sinus thrombosis may lead to a fatal outcome. A degree of visual field loss persists in about 50% of patients and a loss of visual acuity, occasionally leading to blindness, in about 10%. As loss of central vision usually occurs late in the course of the disease, serial visual field analysis is the important yardstick of clinical progression.

Further reading


Ahlskog JE (1982). Pseudotumour cerebri. Ann Intern Med, 97, 249–56. Ball AK, Clarke CE (2006). Idiopathic intracranial hypertension. Lancet Neurol, 5, 433–42.

Binder BK, et al. (2004). Idiopathic intracranial hypertension. Neurosurgery, 54, 538–51.

Bruce BB, et al. (2009). Idiopathic intracranial hypertension in men. Neurology, 72, 304–9.

McComb JG (1983). Recent research into the nature of cerebrospinal fluid formation and absorption. J Neurosurg, 59, 369–83. 

Rush JA (1980). Pseudotumour cerebri: clinical profile and visual outcome in 63 patients. Mayo Clinic Proc, 55, 541–6.

Shah VA, et al. (2008). Long-term follow-up of idiopathic intracranial hypertension: the Iowa experience. Neurology, 70, 634–40.

Vieira DS, et al. (2008). Idiopathic intracranial hypertension with and without papilloedema in a consecutive series of patients with chronic migraine. Cephalgia, 28, 609–13.