The impact of lifestyle factors on disease risk and long-term disability progression in multiple sclerosis

Abstract

Background: Multiple sclerosis (MS) is a disabling inflammatory disease of the central nervous system (CNS) likely caused by genetic susceptible variants and environmental triggers. Low vitamin D levels and smoking are already established risk factors for MS, while obesity and physical activity may also influence the risk. In addition, some of these factors are associated with disease course in MS, but less is known about their potential long-term effects on MS.

Objectives: In this thesis, we examined (i) the association between body size and MS risk across different geographical areas (Paper 1), (ii) whether frequency and intensity of physical activity in adolescence may be an independent risk factor for MS (Paper 2) and (iii) whether vitamin D levels, tobacco use and body mass index (BMI) can influence long-term disability progression in MS (Paper 3).

Methods and materials: In Paper 1 and 2, we used retrospective self-reported data from a large multinational population-based case-control study on environmental and lifestyle factors in MS (the EnvIMS study). The study on body size and MS risk in Paper 1 was based on self-reported body sizes on a 9-figure scale, at 5-year intervals, from age 5 to age 30 years in Norway and Italy. The study on physical activity (PA) and MS risk in Paper 2 was based on reported average weekly amounts of light and vigorous PA during adolescence in Norway, Sweden and Italy. We used logistic regression models to examine the associations between lifestyle factors and the risk of MS, with adjustment for relevant covariates. For Paper 3, we had available baseline and 10-year follow-up data from 80 patients who initially participated in a randomized study on omega-3 fatty acids treatment in MS (the OFAMS study). In linear regression models, we examined the association between mean baseline levels of serum 25-hydroxyvitamin D (25(OH)D), serum cotinine (a nicotine metabolite) and BMI, and 10-year disability progression given by the 10-year change in Expanded Disability Status Scale (EDSS) score. We also examined the importance of seasonal fluctuations of 25(OH)D on this association.

Results: In Paper 1, a large body size (body figure 6-9) was significantly associated with increased MS risk in Norway from age 15- 25 years. The association was strongest at age 25, with an age-adjusted odds ratio (OR) of 2.10 (95% confidence interval (CI): 1.08-4.09) for men and 1.48 (95% CI: 0.94-2.32) for women, compared to a “normal weight” body size 3. Further adjusting for smoking and outdoor activity gave similar estimates. In Italy we found no clear association between body size and the risk of MS, but after disease onset, the controls in both countries reported larger body sizes relative to the cases. In Paper II, the pooled analyses for Norway, Sweden and Italy showed that vigorous PA ≥ 3 hours compared to < 1 hour per week was associated with a reduced risk of MS with an age- and sex-adjusted OR of 0.74 (95% CI: 0.63-0.87). We found similar estimates in country-specific analyses, also after adjusting for other established risk factors. No clear evidence of reverse causation explaining this association was observed in a subgroup analysis, excluding participants with disease onset within 10 years from reported PA. In Paper 3, one standard deviation (SD; 18.7 nmol/L) increase in seasonally adjusted 25(OH)D levels during the OFAMS baseline study was associated with 0.45 point (95% CI: -0.75 to -0.16) less change in EDSS score after 10 years, in a model adjusting for sex, age and baseline EDSS score. There was a significant dose-response relationship across quartiles of 25(OH)D levels (p for trend = 0.024). The association was mainly driven by low 25(OH)D levels during spring and seasonally adjusted levels below 80 nmol/L. For BMI and tobacco use, no significant associations were observed, but we found a trend towards less progression with higher BMI.

Conclusions: A large body size during childhood and young adulthood was associated with increased risk of MS among men and women in Norway, but less so in Italy. Higher amounts of regularly vigorous PA were associated with lower MS risk across different geographical areas, also after adjustment for potential confounders. Higher levels of 25(OH)D during a two-year period were associated with less 10-year disability progression, which appeared to be driven by low spring levels. Our findings suggest that healthy lifestyle changes during young ages may influence the risk of developing MS in a beneficial way, and that better long-term outcomes can be achieved by maintaining 25(OH)D levels above 80 nmol/L throughout the year.