We agree with Beck and colleagues that it would be interesting to standardize the protocols for HHV-6 detection and consider sample exchange between laboratories because it is too difficult to compare different studies made in varying conditions. Their study and ours analyzed a very different number of samples (27 vs 103), with 2 different techniques for the extraction of DNA (phenol/chloroform vs Qiagen columns), with 2 different PCR assays (nested vs real-time quantitative PCR), to detect 2 different HHV-6 genes (glycoprotein gene U47 vs the immediate-early 1 region). In the study by Beck and colleagues, they did not find any sample positive for HHV-6 when they examined serum and CSF in 27 patients with MS. This is not surprising. In our study only 15 patients had a positive result for HHV-6 in the serum, and 88 patients tested negative for this virus. Perhaps the number of cases studied by Beck and colleagues was not high enough to reach statistical conclusions given that the expected prevalence was as low as 14.6%. Although there are no appreciable differences in the sensibilities of the 2 assays (2 copies for nested PCR and 1 copy for real-time quantitative PCR), we do not know the initial amount of DNA that these authors put into each reaction tube. Was this amount superior or inferior to the 50 ng of DNA used in our assay?
Álvarez-Lafuente R, de las Heras V, Arroyo R. Human Herpesvirus 6 in Serum and Spinal Fluid of Patients With Multiple Sclerosis?—Reply. Arch Neurol. 2003;60(4):639-640. doi:10.1001/archneur.60.4.639-a