AbstractBackgroundEnlarged perivascular spaces (ePVS) and sleep disturbances have been associated with incident dementia. While healthy sleep is suggested to promote glymphatic clearance in the brain, poorer sleep may be associated with higher ePVS burden (representing impaired perivascular drainage), which may contribute to sleep’s role in dementia risk. We assessed the association between sleep characteristics measured with polysomnography and subsequent ePVS burden.MethodWithin the Framingham Heart Study, 522 participants (age: 46.1±9.0 years; 50.4%M) underwent a full‐night in‐home polysomnography. Sleep variables measuring sleep architecture (total sleep time, wake after sleep onset, sleep efficiency), sleep stages (N1, N2, N3 and REM sleep in absolute minutes and percent of total sleep time), and obstructive sleep apnea severity (apnea‐hypopnea index, obstructive sleep apnea categories [mild, moderate, severe], sleep time <90% SaO2) were extracted. On average 16.7 ± 2.6 years later, participants underwent magnetic resonance imaging. ePVS were rated in the basal ganglia and centrum semiovale, and dichotomized for analyses: low burden (<20 counts, grade 1 and 2) or high burden (>20 counts, grade 3 and 4). Logistic regressions analyses relating sleep characteristics to ePVS burden were used, adjusted for age, sex, time interval between polysomnography and magnetic resonance imaging, APOE4 allele carrier status, hypertension, and smoking.Results21 participants had high ePVS burden in basal ganglia, and 68 participants in the centrum semiovale. Longer time spent in N1 (minutes and % of total sleep time; OR[95%]: 1.02[1.00‐1.04], p = 0.026; 1.11[1.03‐1.19], p = 0.008 respectively), shorter time spent in N3 (minutes; OR[95%]: 0.99[0.98‐0.99], p = 0.029), and longer time spent <90% SaO2 (OR[95%]: 1.03[1.00‐1.05], p = 0.025) were associated with higher ePVS burden in the centrum semiovale. No other associations between sleep characteristics and ePVS ratings were observed.ConclusionLighter sleep, as characterized by longer N1 sleep and shorter N3 sleep, as well as the hypoxemia component of obstructive sleep apnea are associated with higher ePVS burden in subsequent brain magnetic resonance imaging. If replicated, these findings suggest a role for sleep disturbances in glymphatic clearance and cerebral small vessel disease, which may be implicated in the association between sleep and dementia risk.