For the detection of Varicella zoster virus (VZV) in CSF, PCR is the method of choice. Relation between the severity of illness and quantity of virus in CSF is unknown.

Method:

Quantitative real-time PCR was performed on Light Cycler instrument (Roche, UK) using Real Art VZV LC kit (Artus, Germany). Standards for quantitative curve assessment were a part of the kit. DNA was isolated by High Pure Viral Nucleic Acid Kit (Roche) according to the manufacturer protocol. In limited number of samples the interassay variability of the method and variability in yield of isolation were tested. Interassay difference was maximum 163 copies per millilitre of sample. The difference in DNA isolation yield was higher, 188–881 copies per millilitre of sample.

Patients:

Samples for quantitative testing were selected by qualitative PCR. VZV DNA was detected in 17 CSF samples from 16 patients with Herpes zoster (age 21–88 years) and in one sample from a child with Varicella, complicated by meningitis. All patients had clinical signs of meningeal irritation and biochemical markers of neuroinfection in CSF. One additional sample of vesicular fluid and blood sample from an adult patient with acute severe varicella without meningitis was tested.

Results:

CSF virus load ranged from 10² to 5 × 10⁴ copies per millilitre. In comparison the virus load in vesicular fluid was 3 × 10⁶ copies per millilitre. The highest virus loads (5 × 10⁴ and 2 × 10⁴) were detected in a patient with paresis of facial nerve and a young patient with relatively mild disease. The lowest virus load (10 × 10² copies per millilitre) had a child with varicella meningitis and an old patient with severe Herpes zoster of the trunk. Quantitative PCR has good reproducibility and is useful for assessment of viral load in CSF samples. However, the correlation between virus load and severity of illness remains uncertain.