Meningitis due to Aspergillus is an uncommon clinical entity with approximately 40 cases reported. In all cases it is defined by culture of Aspergillus from the CSF or biopsy of the meninges. The affected host groups are slightly different from those with most other forms of invasive aspergillosis. A number of cases have resulted from extension of disease from the paranasal sinuses. At least 3 cases have occurred in intravenous drug users and other cases have followed neurosurgery or administration of intrathecal antibiotics. Other underlying disorders include leukaemia, systemic lupus erythematosus, tuberculosis, sarcoidosis and diabetes mellitus. At least 3 cases have occurred in otherwise healthy individuals.

In the CSF most patients had elevated protein, reduced glucose and an elevated white cell count with a neutrophil predominance. Aspergillus precipitins may be demonstrable in the CSF and these may fall (to zero) with treatment. A recent report describes the sequence of appearance of microscopy, showing hyphae, a positive culture for Aspergillus, antigen, antibody and PCR positivity in CSF in a patient with Aspergillus fumigatus meningitis (Verweij 1999). Galactomannan antigen was detectable in the earliest specimen, 15 days before microscopy and culture became positive and 12 days before PCR became positive. Furthermore antigen was detectable in lumber CSF, whereas microscopy, culture and PCR were only positive on ventricular CSF. Galactomannan antigen titres were useful in following response to therapy.

Once the diagnosis of Aspergillus meningitis is made, the therapeutic outcome is usually good. A combination of amphotericin B and flucytosine is appropriate or possibly itraconazole if the pace of disease is slow. Use of repeat CSF samples with CSF antigen titre monitoring may assist in evaluating response. Reviews of treatment of Aspergillus meningitis are based on very few evaluable patients (Denning 1990). In patients with impaired mental function, amphotericin B (0.8mg/kg) and flucytosine are appropriate. In less ill patients loading doses of itraconazole (at least 600mg/d in adults) and then 400-800mg/d thereafter is appropriate (Mikolich, 1996). Reduction of immunosuppression and removal of infected tissue or neurosurgical hardware (Casey 1994) may be important in certain cases. There are insufficient data to know whether lipid preparations of amphotericin hold any advantage over amphotericin deoxycholate for meningitis but would be appropriate for patients with toxicity and/or renal impairment. Also promising is voriconazole which in very high doses was successful in one case (Verweij, 1999).