Crespo-Cuevas AM, Canento T, Hernández-Perez M, et al. - In a longitudinal population-based study, researchers analyzed if incident long-term cognitive impairment could be predicted by non-invasive vascular markers. This population-based study included a random sample of 933 Caucasian individuals (mean age 66 years, 64% male) with a moderate-high vascular risk and with no history of stroke or dementia. At the baseline visit, subclinical carotid and intracranial stenosis were evaluated via cervical and transcranial color-coded duplex (TCCD) and validated via magnetic resonance angiography. Ninety-one participants (9.7%) had cognitive impairment after a median of 7.16 [6.91-7.75] years of follow-up, of whom 27 had mild cognitive impairment and 64 had dementia. Non-invasive neurosonological markers of incident long-term cognitive impairment in the population are subclinical cervico-cerebral stenosis and higher MCA-PI. Both of these measures were independently linked with incident cognitive impairment with a HR of 2.07, in multivariate Cox regression analyses.