Article Text

Ethambutol related headache in HIV infection: sensitisation to normal CSF pressure? A case report
  1. Department of Clinical Neurology, Institute of Neurology, The National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK
  2. Department of Neurology, University of Essen, Hufelandstr. 55, 45122 Essen, Germany
  1. Dr Holger Kaube holgerk{at}
  1. Department of Clinical Neurology, Institute of Neurology, The National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK
  2. Department of Neurology, University of Essen, Hufelandstr. 55, 45122 Essen, Germany
  1. Dr Holger Kaube holgerk{at}

Statistics from

It is well recognised that headache can be a key symptom in intracranial hypertension, often accompanied by lesions of the optic and other cranial nerves. Typical characteristics of this headache are postural dependence, worsening during exercise, Valsalva's manoeuvre, and relief of headache after lumbar puncture.

Here, we report on a patient with a headache that bears all of the above symptoms including relief after lumbar puncture but with normal CSF opening pressures.

A 48 year old white man with a long history of HIV infection (first diagnosis 1991, actual staging CDC/WHO C3, CD4 79/ml, CD8 179/ml, CD4/CD8 0.4) was referred to the neurological department because of new onset progressive intractable headache that had started 3 months previously. The headache was described as generalised but with highest intensity retro-orbitally and of a constricting and pulsating character. Further characteristics were a pronounced aggravation when resting supine that forced the patient to sit upright at night and extreme exacerbations during coughing or sneezing. The patient complained of photophobia associated with severe headaches, but nausea and vomiting were absent.

In the previous weeks he had received several unsuccessful treatments including high doses of intravenous acetylsalicylic acid (1000-3000 mg daily), other non-steroidal anti-inflammatory drugs such as ibuprofen and naproxen, metamizol, and tramadol, and short term treatment with prednisolone over 1 week. Concomitant medication related to HIV infection consisted of antiviral triple therapy with lamivudin, stavudin, and nelfinavir for 2 months. Other medication included foscavir and ganciclovir for previous cytomegalovirus retinitis, pentamidine for sustained pneumocystic pneumonia prevention, and pyrimethamine/sulfadoxine for previous cerebral toxoplasmosis for 12 months. Six months previously he also had been started on ethambutol when atypical mycobacteria were detected during bronchial lavage although without clinical manifestation.

On neurological examination there was mild meningism but no papilloedema. Vision was permanently impaired bilaterally after cytomegalovirus retinitis 3 years previously and fundoscopy showed only old lesions and no florid inflammatory signs. Proximal reflexes were normal but were reduced peripherally due to HIV and medication related sensorimotor neuropathy.

On admission cranial CT and MRI (including contrast enhancement) showed no signs of local or generalised oedema. Ventricle size and configuration was normal and identical with previous scans. Neuroimaging showed no evidence of intracerebral tumours nor opportunistic CNS infection. White blood cell count was 2100/ml; routine blood chemistry including antinuclear antibody screening was normal.

Diagnostic lumbar puncture showed normal CSF opening pressure (16 cm H2O), cell count, and protein. Extended CSF analysis was normal including microbiology, antigen screening for HIV, cytomegalovirus, and polymerase chain reaction for mycobacteria. However, within minutes after the lumbar puncture the patient reported substantial relief from his headache and meningism disappeared. Two days after the lumbar puncture the headache worsened again and the procedure was repeated on day 5. Because of the lack of response to any analgesic medication the procedure had to be repeated every 4 to 5 days for 3 weeks. The CSF opening pressure was always in the range 10-18 cm H2O. Altogether, six lumbar punctures were performed and the headache resolved each time.

To exclude a placebo effect of the lumbar puncture the patient gave consent to a blinded protocol under continuous monitoring of pressure via a three way connector. Lumbar puncture without withdrawal of CSF had no beneficial effect on the headache. Two withdrawals of 10 cm3 CSF resulted in a significant decrease of the headache when the pressure changed from 12 (10) to 4 (4) cm H2O. By contrast, a slow intrathecal injection of sterile saline increasing the pressure to 16 cm H2O produced an immediate exacerbation of the headache. The treatment option of lumboperitoneal shunting was denied by the local neurosurgeons because of lack of objective signs of increased intracranial pressure.

Eventually, ethambutol was stopped because of its general neurotoxicity and the headaches stopped 5 days afterwards without further need for lumbar puncture.

Postural headaches that are worse in the supine position can be a diagnostic challenge. They may be present in Chiari-type I malformation, as a symptom of intracranial hypertension with or without papilloedema and with papilloedema but without increased intracranial pressure. 1-3

However, to our knowledge this is the first case report of a secondary headache syndrome without papilledoema but with strong dependence on intracranial pressure with relief after lumbar puncture although ntracranial pressure was within normal limits. Under normal circumstances the trigeminal intracranial sensory system is insensitive to small changes in intracranial pressure and even high pressures may not result in headache. The nociceptive sensory threshold for pressure differences can be lowered by inflammatory processes as seen in meningitis, after histamine administration, and during acute migraine attacks.4 In our patient none of these conditions were met.

It has been demonstrated in experimental animals that trigeminal sensory fibres can change their transduction and transmission properties due to their chemosensitivity and lead to sensitisation inducing increased mechanical intracranial hypersensitivity.5 This lowered threshold for the activation of nociceptive trigeminal fibres may have been the case in our patient. From the clinical presentation during the detailed lumbar puncture it seemed evident that a drop in pressure was beneficial, whereas an increase was barely tolerable; pressures below 6 cm H2O produced no headache, whereas pressures above 10 cm H2O did.

On ethical grounds the patient was not re-exposed to ethambutol. Therefore, despite the temporal link between the ethambutol treatment and the new onset headache no further evidence for a causative relation could be established. Because of the many other drugs coadministered with ethambutol a noxious direct drug synergism or accumulation of metabolites cannot be excluded. It also remains unclear to what extent an HIV or cytomegalovirus related change of the cranial sensory nociceptive system may have contributed to the described headache. Ethambutol, apart from its known neurotoxicity particularly involving the optic nerve, can cause headaches as an adverse event.6How ethambutol may cause headache is not known but induction of trigeminal sensitisation seems a plausible explanation. A postural headache after ethambutol has not been described previously.

In conclusion a clearly postural headache clinically suggestive of intracranial hypertension can occur with normal intracranial pressure.


Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.