Epidemiologic studies have been part of the fabric of multiple sclerosis (MS) research for nearly a century. From the time when Bramwell 1 noted that the frequency of MS might vary by geographic location, environmental factors have been hypothesized as causing or triggering the onset of multiple sclerosis. Despite decades of research, however, the cause of MS remains elusive.
By far the most widely accepted hypothesis for the etiology of MS suggests that the disease is an immune-mediated disorder that occurs in response to a delayed childhood infection in a genetically susceptible individual. It further suggests that the disease is subsequently triggered by one or more insults (possibly environmental) shortly before disease onset. Cluster studies, migrant studies, case-control studies, and historical cohort studies have contributed supporting evidence for this hypothesis. 2 But which infection is responsible and what is the second trigger? It is challenging to answer these questions with epidemiologic studies. Notwithstanding the tremendous efforts of investigators, the results of the studies have often been inconclusive.
For many years, the measles virus was the candidate virus and several studies examined its possible role in the etiology of MS. Indeed, expectations were high in the research community that following the widespread use of the measles vaccination the incidence of multiple sclerosis would decline. This decline has not occurred, and recently attention has turned to the possible role of Epstein Barr virus (EBV) in the etiology of multiple sclerosis. Infection with EBV, a herpesvirus, is mild and generally asymptomatic in young children. In older children, adolescents, and adults, EBV presents as infectious mononucleosis (IM) in 50% of those who become EBV positive. Seroepidemiologic studies are consistent in reporting an increased prevalence of antibodies to EBV in MS cases (as high as 100% prevalence) as compared with healthy controls (85–95%). In a recent systematic review, Ascherio and Munch 3 combined the results of eight seroepidemiologic studies and reported an estimated odds ratio of 13.5 (95% CI = 6.3–31.4) indicating a strong association between markers of past infection with EBV and the risk of multiple sclerosis. Wandinger et al. 4 reported that antibodies toward EBV, as determined by the presence of IgG antibodies to the EBNA-1 antigen (a marker of latent infection), were found in 100% of MS patients as compared with 90.1% of healthy controls. In addition, the group reported no evidence of active EBV infection in the MS patients. This latter finding supports the assertion that EBV infection precedes MS rather than being a consequence of the disease. In addition, they showed that EBV, which is known to reactivate, was more likely to be reactivated in a subgroup of MS patients undergoing exacerbations (N = 11) than those with a stable disease course (N = 8) over a period of a year. These authors also showed the specificity of the EBV findings by revealing that the prevalence of antibodies to several other viruses was not increased in the MS patients relative to the controls. Several case-control studies have been conducted in which a history of infectious mononucleosis has been assessed. These studies have been inconclusive, with only three showing evidence of an association between a history of IM and MS (Marrie and Wolfson, Multiple Sclerosis and Epstein-Barr virus: A review, 2000, unpublished manuscript). The strongest and most consistent evidence to date, however, comes from historical cohort studies, all of which have been conducted in Scandinavia. Two published studies 5,6 and one, as yet, unpublished study (Riise T, 2000, personal communication) have shown a higher than expected number of cases of MS in a cohort of individuals with infectious mononucleosis derived from the diagnosis register of the Hospital of Infectious Diseases in Göteborg, Sweden;5 a cohort of individuals with a positive Paul Bunnel test from Denmark;6 and a cohort of subjects with a positive heterophile test in Hordaland, Norway (Riise T, 2000, personal communication).
In this issue of the journal, Hernán et al. 7 report the results of the first nested case-control study conducted to examine the relation between a history of infection with a variety of common viruses and the risk of multiple sclerosis. Using the data from two ongoing prospective studies, the Nurses’ Health Study (NHS) and the Nurses’ Health Study II (NHS II), Hernán et al. report moderate associations between both a reported history of infectious mononucleosis and mumps after age 15 and the risk of multiple sclerosis. No age at onset effect for IM was found. This elegant study was conducted with methodological rigor and attention was paid to possible sources of bias inherent in case-control studies. Because this case-control study is rooted in a well-defined cohort of medical professionals, opportunities for selection bias were few and the reliability of the reported information is high. The fact that this study is limited to women does not detract from the conclusions, as no data suggest that the relation between EBV or IM and MS is modified by sex.
Is EBV a cause of MS? The evidence is not sufficient to draw that conclusion. It might be that EBV is a marker of an as yet unidentified virus that is a necessary cause of MS or that EBV itself is that necessary cause. It is clear, however, that EBV is not sufficient to cause MS and it is likely that a variety of other environmental (perhaps viral) factors act together to produce the disease. If EBV is found to be a cause of MS, this result would prompt a variety of research activities. Research into viruses that may be the final trigger would become a major focus. There is already some suggestion that respiratory tract infections 8 may “turn on” the disease. Work is underway to develop a vaccine for EBV, but, as pointed out by Ascherio and Munch, 3 caution is warranted. If the vaccine serves only to delay the time of infection with EBV then the preventative agent may do more harm than good and could result in an increase in the incidence of MS by delaying the age at onset of EBV past early childhood. This finding might also prompt more work on treatment strategies involving anti-viral drugs, although current evidence for their efficacy in MS is limited. 9
Epidemiologists have a great deal to contribute to the understanding of the etiology of this enigmatic disease. The strongest and most consistent evidence to date of the role of EBV in MS comes from historical cohort studies and the nested case-control study in this issue. Hernán et al. are to be congratulated on their creativity in using these existing prospective studies. I encourage other researchers to seek out other cohorts where similar information may have been collected to replicate their finding.
1. Bramwell B. On the relative frequency of disseminated sclerosis in this country (Scotland and the North of England) and in America. Rev Neurol Psychiatr (Edinburgh). 1903; 1: 12–17.
2. Riise T, Wolfson C, (eds). The epidemiologic study of exogenous factors in the etiology of multiple sclerosis. Neurology 1997;49(suppl).
3. Ascherio A, Munch M. Epstein-Barr virus and multiple sclerosis. Epidemiology. 2000; 11: 220–224.
4. Wandinger K-P, Jabs W, Siekhaus A, Bubel S, Trillenberg P, Wagner H-J, Wessel K, Kirchner H, Hennig H. Association between clinical disease activity and Epstein-Barr virus reactivation in MS. Neurology 2000; 55: 178–184.
5. Lindberg C, Andersen O, Vahlne MD, Runmarker B. Epidemiological investigation of the association between infectious mononucleosis and multiple sclerosis. Neuroepidemiology 1991; 10: 62–66.
6. Haahr S, Koch-Hendricksen N, Moller-Larsen A, Eriksen LS, Andersen HMK. Increased risk of multiple sclerosis after late Epstein-Barr virus infection. A historical prospective study. Acta Neurol Scand 1997; S169: 70–75.
7. Hernán MA, Zhang SM, Lipworth L, Olek MJ, Ascherio A. Multiple sclerosis and age at infection with common viruses. Epidemiology 2001; 12: 301–306.
8. Marrie RA, Wolfson C, Sturkenboom MCJM, Gout O, Heinzlef O, Roullet E, Abenhaim L. Multiple sclerosis and antecedent infections. A case-control study. Neurology 2000; 54: 2307–2310.
9. Munch M. Hvas J, Christensen T, Møller-Larsen A, Haahr S. The implications of Epstein-Barr virus in multiple sclerosis–a review. Acta Neurol Scand 1997; S169: 59–64