Pseudotumor Cerebri Secondary to Minocycline Intake

Earl Robert G. Ang, MD, J. C. Chava Zimmerman, MD, Elissa Malkin, DO, MPH


J Am Board Fam Med. 2002;15(3) 

In This Article


The syndrome of intracranial hypertension with papilledema, no focal neurologic deficit, normal cerebrospinal fluid, and normal to small ventricles was described nearly a century ago by Quinke[2] as serous meningitis. Other names include otitis hydrocephalus, toxic hydrocephalus, sinus thrombosis causing intracranial hypertension, hypertensive meningeal hydrops, pseudoabscess, intracranial pressure without brain tumor, brain swelling of unknown cause, and pseudotumor cerebri.

Modified Dandy criteria[1] for the diagnosis of idiopathic intracranial hypertension include signs and symptoms of increased intracranial hypertension[2] and no localizing neurologic signs (other than abducens nerve paralysis[3]) in an awake and alert patient, normal imaging studies except for small ventricles or empty sella,[4] increased lumbar pressure (≥25 cm of water) with normal cerebral spinal fluid,[5] and no primary structural or systemic causes of intracranial hypertension.[3]

In a retrospective review of 120 patients by Weisberg,[4] 99% of patients had headaches, and 35% had visual changes. In a study of cases of idiopathic intracranial hypertension in an emergency department, Jones et al[5] found that the mean patient age was 27 +/= 8.9 years, 83% were women, and 67% were obese. Ninety-two percent complained of headache, 75% had nausea and vomiting, 71% reported dizziness, and 65% had disturbances of visual acuity. Other symptoms included photosensitivity, diplopia, stiff neck, paresthesias, myalgias, tinnitus, and vertigo. Bilateral papilledema was observed in 67%, 54% had a visual field defect, and 29% had sixth cranial nerve palsy. Four percent had seventh cranial nerve palsy. If ocular motility defects other than from the sixth cranial nerve are encountered, the diagnosis of idiopathic intracranial hypertension is less likely. Unilateral papilledema is also possible, although less common, and is not associated with the duration of disease or severity of symptoms. It should also be noted that idiopathic intracranial hypertension can occur in the absence of headache or papilledema.

CT and MRI findings are typically normal, although there might be nonspecific findings, such as empty sella, prominent cisterna magna, and dilated optic nerve sheaths. In a study of 29 male patients with idiopathic intracranial hypertension by Digre and Corbett,[6] an empty sella was noted in 55% of the CT scans.

Cerebrospinal fluid pressure should be elevated, and the protein levels should be low or normal, glucose levels normal, and cell counts normal. A diagnosis of idiopathic intracranial hypertension should not be made without performing a lumbar puncture. An opening pressure of greater than 250 cm H2O is diagnostic. When clinical and radiologic evidence is highly indicative of idiopathic intracranial hypertension, and the opening pressure is normal, a second lumbar puncture or continuous intracranial pressure monitoring might be necessary.

Intracranial sinus thrombosis, which is associated with head injury, otitis media, and hypercoagulable and hyperviscosity syndromes, most often causes intracranial hypertension without focal neurologic defect. Findings of a CT scan and cerebrospinal fluid are normal, although an MRI might show an abnormality, especially in the transverse sinus. MRA and digital subtraction confirm the diagnosis.

Epidemiologic studies have confirmed an association between female sex, reproductive age-group, menstrual irregularity, obesity, and recent weight gain. Other associated conditions, although unconfirmed by case-control studies, include adrenal insufficiency, Cushing disease, hypoparathyroidism, hypothyroidism, chronic renal failure, and systemic lupus erythematosus.

Of particular interest, especially in this case, is the association of intracranial hypertension with medication intake. Minocycline-related pseudotumor cerebri was first reported in 1978,[7] Since then 16 additional cases have been reported. Chiu et al,[8] in a retrospective study, reviewed 12 cases of minocycline-induced pseudotumor cerebri syndrome. Seventy-five percent of the patients developed pseudotumor cerebri within 8 weeks of starting minocycline. Six (50%) of the patients were not obese. Two patients developed symptoms after 1 year of minocycline use. Pseudotumor cerebri was diagnosed by finding papilledema on routine examination in 1 patient who was asymptomatic after taking minocycline for 1 year. After discontinuing minocycline, all patients recovered from pseudotumor cerebri syndrome, and after at least 1 year of follow-up, there were no recurrences.

Four cases of minocycline-induced intracranial hypertension were documented in Australia by Lander.[9] Durations of therapy ranged from 25 days to 18 months. All had severe headaches and papilledema, and visual disturbance was reported in two cases. Cessation of minocycline reversed the intracranial hypertension, although 1 patient had persistent lower nasal quadrantic field-of-vision loss 6 months later. In all cases, the diagnosis was missed by the primary care physician. The ability of minocycline to decrease cerebrospinal fluid absorption is the postulated mechanism for minocycline-induced pseudotumor cerebri. Isotretinoin, tetracycline, trimethoprim-sulfamethoxazole, cimetidine, corticosteroids, tamoxifen, lithium, nitrofurantoin, and levothyroxine have been implicated in addition to minocycline.

The only serious complication of idiopathic intracranial hypertension is vision loss, which can be sudden or gradual and can occur at any time during the course of the disease. Appropriate treatment can prevent vision loss, however. Risk factors for vision loss include duration of related symptoms before diagnosis, systemic hypertension, anemia, older age, and high degrees of myopia. African American men also appear to be at higher risk. Studies have shown that perimetry, using either the Goldman manual perimeter or a computed automated perimeter, is the best test to detect and monitor vision loss associated with idiopathic intracranial hypertension.

Treatment is directed at preventing vision loss and treating the cephalgia. Initially, predisposing factors should be corrected, such as stopping possible inciting medications and or treating any underlying medical condition. Lumbar puncture, aside from being diagnostic, is therapeutic. If the symptoms resolve after the initial lumbar puncture, no further action is warranted. If the symptoms are unresolved, repeated lumbar punctures, up to four in the first 2 to 4 weeks, are recommended. Beyond the first 4 weeks, repeated lumbar punctures are unlikely to be of benefit.

Acetazolamide (Diamox), a carbonic-anhydrase inhibitor, is frequently used for idiopathic intracranial hypertension. Treatment is started at a dosage of 250 mg/d and gradually increased until target doses of 500 mg four times a day is reached or until side effects are encountered. Adverse reactions include paresthesias, drowsiness, nausea, malaise, metabolic acidosis, altered taste, and renal calculi. For those who are intolerant or unresponsive to acetazolamide, a short course of oral corticosteroids might be of benefit. Prednisone, 40 to 60 mg/d, should resolve symptoms in 10 to 14 days, at which point the medication is tapered for the next 2 weeks.

Surgery should be considered when vision loss does not respond to medical treatment, when initial vision loss is severe, or when the patient response is unreliable at visual field testing and there is an increased delay in the major positive peak of visual evoked response. Surgical options include lumboperitoneal shunt or optic nerve decompression. Fenestration of the optic nerve sheaths is becoming the treatment of choice. It provides immediate decompression of the optic nerve, as well as long-term filtration of cerebrospinal fluid. Complications include papillary dysfunction, peripapillary hemorrhage, chemosis, and chorioretinal scarring. Although lumboperitoneal shunting has been shown to fail to prevent the progression of vision loss,[10] this procedure is appropriate in those who do not respond to optic nerve fenestration.

In summary, pseudotumor cerebri, or idiopathic intracranial hypertension, is a syndrome in which patients most commonly complain of headache and visual disturbance and have signs and symptoms typical of increased intracranial pressure, no focal neurologic signs, normal neuroimaging studies, increased opening pressure on a lumbar puncture with a normal cerebrospinal fluid findings, and no structural or systemic cause for intracranial hypertension. It is most often seen in reproductive-age women who are obese, who have recently gained weight, and who have irregular menses. Although an abducens nerve palsy might be observed, any other ocular motility disorder makes the diagnosis of idiopathic intracranial hypertension unlikely. Idiopathic intracranial hypertension can be associated with endocrine dysfunction, systemic lupus erythematosus, chronic renal failure, and with some medications.

A history of minocycline use should be determined in cases of pseudotumor cerebri, especially when the patient is not obese. Minocycline is among the top four oral antibiotics most commonly prescribed for the treatment of acne, with more than 374,000 prescriptions written a year.[11] Family physicians who prescribe minocycline should be vigilant, as this potential complication is not entirely benign. Patients must be aware of the symptoms and seek medical attention should they arise. Additionally, physicians should be knowledgeable of this complication, screen their patients with questions regarding the symptoms of intracranial hypertension, and routinely perform ophthalmologic examinations of their patients who are taking this medication. Treatment, which is directed toward the prevention of visual loss by correcting predisposing factors, includes lumbar puncture, serial if necessary, acetazolamide, corticosteroids if there is no response to acetazolamide, or surgery when visual loss is severe or unresponsive to medical therapy.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.