Introduction to the virological diagnosis of the most frequent infections of the central nervous system.

The introduction of polymerase chain reaction techniques (PCR) has made major progress in the aetiological diagnosis of viral infections of the central nervous system (CNS) possible. Although this method is becoming widely accepted as the reference technique, technical problems persist. Amplicon contamination, leading to false-positive results, has been greatly reduced (separate manipulation sites, amplicon destruction, PCR carried out in sealed tubes). However, further improvements are still required in the standardisation of the various techniques and their reproducibility. Positive initiatives, such as that of the European Union Concerted Action on Virus Meningitis and Encephalitis, which has proposed the setting up of a panel for controlling the techniques used, should be welcomed (van Vliet et al., 1998). In current practice, in many hospital structures, the viruses most frequently screened for in cases of neurological diseases are the enteroviruses and the principal human herpes viruses (HSV 1 and 2, CMV, VZV, EBV and HHV-6). In certain regions, tests to detect viruses transmitted by arthropods (ticks) are also carried out. The viruses currently thought to be responsible for lymphocyte meningitis, such as the mumps virus, are rarely screened for: vaccination is probably responsible for this epidemiological change. Infections of the CNS involving adenoviruses, paramyxoviruses such as measles and the influenza A and B viruses, seem to be rare. Very few prospective, controlled studies have been carried out on these viral infections of the CNS. For obvious ethical reasons, it is not possible to carry out lumbar puncture systematically, or on healthy control subjects. In cases in which lumbar puncture is justified, only a limited amount of cerebrospinal fluid is collected, particularly if the subject is a young child. The traditional cytological, biochemical and bacteriological examinations are therefore still a matter of priority. Thus we are forced to rely on the accumulation of retrospective experience by diverse teams, despite methodological reservations, identify correlations between clinical and virological diagnostic tests. These considerations led us to ask various teams to recount their experience in certain aspects of this domain. For enteroviruses, the sensitivity and specificity of PCR, and the comparison of PCR with the Abbre iations: CMV, cytomegalovirus; CNS, central nervous system; CSF, cerebrospinal fuid; EBV, epstein-barr virus; HHV-6, human herpesvirus type 6; HHV-7, human herpesvirus type 7; HHV-8, human herpesvirus type 8; HSV-1, herpes simplex virus type 1; HSV-2, herpes simplex virus type 2; PCR, polymerase chain reaction; VZV, varicella zoster virus. * Tel.: +33-299284276; fax: +33-299284159 E-mail address: ronald.colimon@univ-rennes1.fr (R. Colimon).