All articles containing the search term “multiple sclerosis” from 1997–2007 were obtained from Scopus. This approach yielded some 24 746 titles and the abstracts of those related to the broadest definition of epidemiology were read. On the basis of abstract content, relevant papers were read specifically to provide context for this review rather than to comprehensively catalogue all publications. This review summarises the epidemiology of MS with emphasis on population-based studies.
ReviewEnvironmental factors and multiple sclerosis
Introduction
Nature versus nurture arguments about the pathogenesis of multiple sclerosis (MS) have given way to evidence that both are important. Independent of seminal South African migration studies by Dean,1 the low prevalence of the disorder among people of east Asian descent in Canada compared with that in people of north European descent has long been recognised. A similar relative dearth of cases in Queensland is obvious to local practising neurologists familiar with the prevalence of the disorder in Sydney and Melbourne. And these impressions have been confirmed by real data.2 It is more timely to ask if and how nature and nurture interact rather than debate their relative importance. Identification of environmental factors might pinpoint biological pathways, and vice versa.
The limited ability of case–control studies to identify differences led to a temporary de-emphasis of the study of environmental factors. The environment does not necessarily differ for affected and unaffected individuals in the same family, and geographically separate regions have the same risk. Several recent studies discussed here refocus attention on environment, which is so clearly important in pathogenesis.
The study of genetic factors, perceived a decade ago to be more tractable than that of environmental influences, has been slow in fulfilling initial promise. Claims for linkage on nearly every chromosomal arm suggest systematic methodological failures.3 Two recent small but confirmed effects leave half of the genetic contribution to MS unexplained.4, 5 Re-examination of the conceptual and practical importance of the environment in the pathogenesis of MS seems timely.
Given that the effect of the environment in MS is not necessarily mediated by singular or discrete events, it may be useful to separate susceptibility, triggering, and outcome. Here we focus on the first of these. Although there is convincing evidence that environment affects outcome and clinical features,6 few studies have examined differences in long-term outcome by geographic location.
Non-specific viral (and perhaps bacterial) infections seasonally precede both onset and relapses (and MRI activity) in perhaps a quarter of MS cases7 despite common disparity between biological and clinical onset. The relation of relapses to long-term disability is surprisingly unclear8 and the dissociation of the two is evident in early studies of interferon.9
Section snippets
Geography
The uneven geographical distribution of MS is central to understanding the role of environment and has influenced epidemiologists since Davenport's report that ethnicity affected risk of MS in veterans of World War I.10 US Veteran's Administration data enabled Kurtzke's pivotal studies of the geographic gradients noted by Davenport.
US data revealed north–south and east–west gradients.11 Distribution of MS in the USA resembles that of Scandinavian immigration; thus the geography reflects both
Migration studies
In Australia, prevalence in Tasmania is five to six times that in Queensland, a difference far too large to result from a genetic cline, mirroring findings in British migrants to South Africa. Conversely, migrants from east Asia to the west coast of North America have a low prevalence and large populations in Vancouver, Canada, retain low risk; although small recent increases in prevalence have been reported in these migrant populations and in Japan.
In the UK and North America, the risk of MS
Family studies
Twin studies have been influential in separating genetic from environmental influences on MS susceptibility. Studies of other family relationships provide useful epidemiological insights; each type of relative pair illuminates potentially different features related to whether or not relatives are colineal or vertically related, and to differential common exposure. Comparisons of risks in relatives more distant than half siblings (eg, first cousins, nieces, and nephews) have limitations. The
Paediatric MS
The vanguard of any change in demographics, phenotype, or incidence should first appear in children. Collaborative studies of paediatric populations such as Kids with Multiple Sclerosis (KIDMUS)76, 77 that collect important baseline material should yield valuable harbingers of temporal trends in MS. Comprehensive studies of class II MHC alleles in this population would help address uncertainties in disease homogeneity.
Evolutionary considerations
Dobzhansky said that nothing in biology makes sense except in the context of evolution. We speculate here on an adult-onset and largely post-reproductive disease of northern European populations. The relation between skin colour and environment and how it arose has been reviewed by Jablonski.78 This relatively recent story has played out over the past 50 000 years. MS and other diseases of white northern Europeans, such as type 1 diabetes and inflammatory bowel disease, might result from
Conclusions
Much is known about the characteristics of environmental effects in MS. The most influential risk factor operates early in life and determines geographical gradients in ethnically homogeneous populations. Perhaps distinct from this, or primed by early exposure, are later influences that seem also to be geographically distributed. The possibility of more than one effect leaves undiminished strong implications that genes determine familial risk. Major environmental influences, directly or
Search strategy and selection criteria
References (80)
- et al.
Microsatellites and genome scans—a GAMES postscript
J Neuroimmunol
(2007) - et al.
Genes for multiple sclerosis
Lancet
(2008) - et al.
Clinical viral infections and multiple sclerosis
Lancet
(1985) - et al.
Effect of low saturated fat diet in early and late cases of multiple sclerosis
Lancet
(1990) - et al.
Association between microchimerism and multiple sclerosis
J Neuroimmun
(2006) - et al.
Increasing sex ratio of multiple sclerosis in Canada
Lancet Neurol
(2006) - et al.
Parent-of-origin effect in multiple sclerosis: observations in half-siblings
Lancet
(2004) - et al.
Offspring recurrence rates and clinical characteristics of conjugal multiple sclerosis
Lancet
(1997) Infections and autoimmune diseases
J Autoimmunity
(2005)- et al.
Multiple sclerosis and birth order: a longitudinal study
Lancet Neurol
(2005)