Effects of Weather and Season on Human Brain Volume

Abstract

<jats:title>Abstract</jats:title><jats:p>We present an exploratory cross-sectional analysis of the effect of season and weather on Freesurfer-derived brain volumes from a sample of 3,279 healthy individuals collected on two MRI scanners in Hartford, CT, USA over a 15 year period. Weather and seasonal effects were analyzed using a single linear regression model with age, sex, motion, scan sequence, time-of-day, month of the year, and the deviation from average barometric pressure, air temperature, and humidity, as covariates. FDR correction for multiple comparisons was applied to groups of non-overlapping ROIs. Significant negative relationships were found between the left- and right-cerebellum cortex and pressure (t = −2.25, p = 0.049; t = −2.771, p = 0.017). Significant positive relationships were found between left- and right-cerebellum cortex and white matter between the comparisons of January/June and January/September. Significant negative relationships were found between several subcortical ROIs for the summer months compared to January. An opposing effect was observed between the supra- and infra-tentorium, with opposite effect directions in winter and summer. Cohen’s d effect sizes from monthly comparisons were similar to those reported in recent psychiatric big-data publications, raising the possibility that seasonal changes and weather may be confounds in large cohort studies. Additionally, changes in brain volume due to natural environmental variation have not been reported before and may have implications for weather-related and seasonal ailments.</jats:p>

A systematic review of previous MRI studies demonstrating the relationship between season and brain volume

“…Further, (Pantazatos, 2014) found significant annual periodicity in the left superior temporal gyrus (STG) and grey matter volume, with a peak towards the end of December. Finally, (Book et al, 2020) found positive relationships between left and right cerebellum cortex and white matter between seasons (the comparisons of January/June and January/September) and negative relationships between several subcortical ROIs for the summer months compared to January.…”

Interaction of surface type, temperature, and week of season on diagnosed concussion risk American football: Bayesian analysis of 8 Seasons of National Football League Data.

“…It has been theorized that prior history of repeated sub-concussive head impacts could change one's biological threshold for a concussion, 55 and there is mixed evidence to support this. [15][16][17][18][56][57][58][59] Additionally, cross-sectional data suggests that brain volume changes seasonally in the general population, 60 and this could theoretically influence concussion risk over the course of a season, regardless of game conditions or cumulative impact exposure.…”