Ambient Air Pollution and Occurrence of Multiple Sclerosis Relapses

Abstract

Background/Aim: Triggers of multiple sclerosis (MS) relapses are essentially unknown. Relapses' incidence varies across seasons, suggesting a possible role of season-dependent factors such as meteorological variables and ambient air pollution. Relatively few studies have investigated this hypothesis, but the role of ambient air pollution is of growing interest; a recent study from our group observed an increased risk of relapses with ambient PM10 exposures until 3 days before the relapse onset (cold season). Methods: We conducted a time-stratified case-crossover design to further explore separately the short-term associations between other pollutants (e.g., NO2, Benzene (C6H6), O3, and CO) and the odds of MS relapses occurrence. Our study population consists of 536 relapsing MS patients living in Strasbourg area (France) who experienced 2,052 relapses (2000-2009). Control days are chosen to be ±35 days relative to the case (relapse) day. We performed a conditional logistic regression coupled with a distributed-lag model (considering lags 0 to 3) using the "dlnm" R package, stratified by season (hot and cold), and adjusted on all lagged meteorological variables (daily maximum temperature, maximum relative humidity and maximum atmospheric pressure), lagged pollen count, lagged influenza-like epidemics, and holidays. Results: An interquartile range increase in NO2 (lags 1 to 3 cumulated) was associated with MS relapses incidence (ORNO2=1.28 [95%CI: 1.05-1.55]) during cold months (i.e., October to March). C6H6 and CO were not significantly related to MS relapse incidence (ORC6H6=1.13 [95%CI: 0.95-1.35] and ORco=1.09 [95%CI: 0.89-1.34], respectively). We also observed non-significant association between O3 during hot months and MS relapse incidence (ORO3=1.12 [95%CI: 0.88-1.44]). Conclusions: We found a significant association between NO2 and the occurrence of MS relapses using a case-crossover design, relevant to studies with rare event outcomes. Strengths of our study include a precise exposure spatio-temporal model and a clinically-diagnosed outcome from a fairly exhaustive MS registry.