BACKGROUND: Accumulating evidence links air pollution exposure to late-life cognitive deterioration. Whether air pollution alters brain structure remains poorly understood. Thus, we aimed to quantify the association between long-term exposure to particulate matter ≤2.5 µm and ≤10 µm (PM 2.5 and PM 10 , respectively) and late-life brain structural changes. METHODS: In the Swedish National Study on Aging and Care in Kungsholmen, Stockholm, 555 participants free from dementia underwent brain magnetic resonance imaging (MRI) scans at baseline and after 6 years (cohorts <78 years) or after 3 and 6 years (cohorts aged ≥78 years). After the exclusion of participants with neurological conditions (including previous stroke) and suboptimal MRI quality, we had 457 participants with available repeated MRI examinations, where total brain tissue volume, ventricles, hippocampus, and white matter hyperintensities volumes were assessed. PM 2.5 and PM 10 have been assessed since 1990 using dispersion models at residential addresses. Brain volumes have been standardized using baseline mean and SD. Long-term exposure to PM 2.5 and PM 10 in relation to the baseline and longitudinal brain MRI volumes were tested through multiadjusted (age, sex, educational level, smoking, socioeconomic status, and neighborhood household mean income) linear regression models. RESULTS: At study entry, the mean (SD) age of the participants was 70 (SD, 8.9) years and 41% were males. Individuals who before baseline had been exposed to levels of PM 2.5 or PM 10 above the median (8.5 and 14.9 μg/m 3 , respectively) had smaller total brain tissue volume (β, −0.20 [95% CI, −0.33 to −0.06] and β, −0.14 [95% CI, −0.28 to −0.01], respectively) at baseline than those with lower PM 2.5 and PM 10 levels. Participants exposed during the follow-up to PM 2.5 >8.7 μg/m 3 had on average an annual shrinkage of total brain tissue volume of 0.22 (95% CI, −0.43 to −0.01) and an annual increase of 0.25 (95% CI, 0.07–0.43) of the white matter hyperintensities as compared with participants exposed to PM 2.5 <8.7 μg/m 3 . No association was detected between PM 10 and an annual rate of change in brain MRI volumes. CONCLUSIONS: Long-term exposure to comparatively low levels of PM 2.5 was associated with a greater load of structural brain changes, encompassing brain atrophy and vascular pathology. These findings, in a dementia- and cerebrovascular disease-free sample, underscore the importance of addressing air pollution as a modifiable risk factor for late-life cognitive decline, and highlight the need for targeted interventions to prevent its detrimental effects on brain integrity.