Detection of herpes viruses in the cerebrospinal fluid of adults with suspected viral meningitis in Malawi

This study was a sub-analysis of cerebrospinal fluid (CSF) from a cohort of adults (≥16 years of age) with clinically suspected meningitis admitted to Queen Elizabeth Central Hospital (QECH), Blantyre, Malawi, from September through to December 2007 [10]. Demographics and clinical information were collected using a standardised case record form. All persons gave their informed consent prior to inclusion in the study. Routine microscopy and culture of the CSF was performed and patients with microbiologically confirmed bacterial, cryptococcal or tuberculous meningitis were excluded, as were those with probable bacterial meningitis (CSF white cell polymorphs >100 cells per mm³). Patients with a diagnosis of a CNS infection within the last month were also excluded; this was primarily done to ensure that our cohort of patients had new clinical events and to avoid selecting those with a relapse of past CNS infection. Patients with ≥5 white cells per mm³ of CSF were defined as having an aseptic meningitis, whilst those with <5 white cells per mm³ were defined as not having meningitis using a laboratory diagnostic criteria [11]. A CD4 count of <200 cells/µl was used to define advanced HIV disease.

The study was approved by the local ethics committee at the College of Medicine, University of Malawi, Blantyre, Malawi (P05/06/471) and the ethics committee medical faculty, University Heidelberg, Heidelberg, Germany (S 026/2007).

CSF samples were aliquoted and stored at −70 °C prior to molecular diagnostics for HSV, VZV, EBV and CMV. The samples were tested as a batch at the time of this study.

DNA was extracted from 200 µl of CSF using a QIAamp DNA kit as per the manufacturer’s instructions. DNA recovery was confirmed by spectroscopic analysis. We used 5 µl of DNA for each polymerase chain reaction (PCR) for the HSV, VZV and EBV assays, and 10 µl for the CMV assay.

Primers and probes targeting conserved regions of the DNA polymerase gene were used for HSV-1 (primer: forward-GAC AGC GAA TTC GAG ATG CTG, reverse-ATG TTG TAC CCG GTC ACG AAC T; probe: 5′-FAM-CAT GAC CCT TGT GAA ACA-MGB-3′-BHQ1); HSV-2 (primer: same as HSV-1; probe: 5′-VIC-TGA CCT TCG TCA AGC AG-MGB-3′-BHQ1) and VZV (primer: forward-GCG CGG TAG TAA CAG AGA ATT TC, reverse-ACG TGC ATG GAC CGG TTA AT; probe: 5′-FAM-ACC ATG TCA TCG TTT CAA-MGB-3′-BHQ1). Similarly, the tegument protein (BNFR1 gene) for EBV and UL123/UL 55 genes for the CMV genome were targeted [12, 13].

Applied Biosystems (ABI, UK) AmpliTaq Gold (2×) Master Mix was used for all the real-time PCR assays. All reactions were carried out in the ABI 7300 real-time PCR machine using 96-optical-well plates, with a total volume of 25 µl per well. The PCR thermocycler conditions for all assays included three stages; activation of UNG at 50 °C for 2 min, activation of Taq at 95 °C for 15 min, followed by amplification using 45 cycles at 95 °C for 15 s and 60 °C for 1 min.

A positive control (provided by the National Institute for Biological Standards and Control, UK) for each herpes virus was run with unknown specimens to ensure DNA recovery. Similarly, specimens were processed in parallel with aliquots of molecular-grade water to monitor for false-positive results; the run was discarded if this was the case. Each individual sample was processed in triplicate and the sample was considered to be positive if two or more of the three test results were positive. When only one out of the three test results was positive, the PCR test was repeated (n = 3) and the sample was then considered to be positive if the test result was positive (n = 1). Quantification was only performed with EBV and shown as genome copies per millilitre of CSF. Serial dilution of 10⁰–10⁸ copies/ml of virus stock was used to obtain a standard curve with an efficiency performance of 92 %. The concentration of EBV genome in the samples was interpolated from this standard curve.

VZV is known to have high genotype diversity [14]. To ensure that the PCR primers were appropriate for VZV strains circulating in Malawi, swabs (stored in viral medium) taken from the vesicles of four patients with classical shingles (vesicular rash in a dermatomal distribution) were also analysed. The vesicular samples of all four patients with shingles gave positive results.

A sensitivity analysis was performed to determine the limit of detection using QCMD panels (Qnostics Ltd., UK). We found the limit of detection to be 10³ copies/ml or less with each virus.

The data were analysed using PASW Statistics 18, Release Version 18.0.0 (Ó SPSS, Inc., 2009, Chicago, IL, http://​www.​spss.​com). Quantitative variables were expressed as medians with interquartile ranges and qualitative variables as frequencies and percentages. Fisher’s exact test was used to compare categorical data and the Mann–Whitney U test for continuous data; a p value of <0.05 was taken to be statistically significant.