The Relationship between Cytomegalovirus Infection and Guillain–Barré Syndrome

Abstract

The association between cytomegalovi- rus (CMV) and Guillain-Barresyn- drome (GBS) was first noted in 1967 (1) and was soon endorsed by other reports and case series. The largest previous case series showed that CMV infection was associated with GBS in 8% of 229 pa- tients from a European and North American trial (2). In this issue of Clinical Infectious Diseases, Orlikowski et al (2) confirmed this finding in a larger population of people with GBS by using the resources of the French reference centre in Garches, Paris, which has been a major source of useful research into the disease. Of 506 patients with GBS, 63 (12.5%) had serological evidence of primary CMV infection. It is difficult to distinguish between primary and reactivation of infection, an important problem, because 50% of the nonimmunosuppressed people in de- veloped countries have serological evi- dence of exposure to CMV. The serological methods used by Orlikowski and colleagues incorporated a test for high immunoglobulin (Ig) G avidity to exclude positive results due to re- activation. Low-avidity IgG antibody is found in primary infection before affinity maturation has occurred. The serological results were supported by the finding of CMV DNA in the plasma by poly- merase chain reaction (PCR) in 36 (62%) of 58 subjects who had positive serological test results. The sensitivity of CMV PCR nears 100% for acute in- fection, but it would be expected to be reduced in the late infectious phase, when GBS occurs because virus is being cleared. Unfortunately no conva- lescent-phase samples were available for analysis because of the intervening treat- ment and quantitative PCR was not performed for enough patients to publish the result. The authors used contemporary French epidemiological data to calculate, for the first time, that the risk of GBS following CMV is in the order of 0.6-2.2 cases per 1000 persons. There were no control subjects in the Orlikowski et al (2) study, but 2 earlier studies had shown that the frequency of recent CMV infection preceding GBS significantly exceeded that of control subjects: in the United Kingdom, 11% of 99 case patients and only 1% of control patients had evi- dence of recent CMV infection (3); in the Netherlamds, 13% of 154 case patients and only 2% of control subjects (4) had evidence of recent CMV infection. The study by Orlikowski et al (2)