Determinants of enlarged perivascular spaces (ePVS) on MRI: The Florida Vascular Imaging Phenotypes (FL‐VIP) Study of AD risk. (December 2021)

• Record Type: Journal Article
• Authors: Rundek, Tatjana; Brutto, Victor Del; Goryawala, Mohammed; Saporta, Anita Seixas Dias; Merritt, Stacy S.; Camargo, Christian; Agudelo, Christian; Ariko, Taylor; Dong, Chuanhui; Loewenstein, David; Duara, Ranjan; Haq, Ihtsham U.
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 4
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.054617
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: PVS are fluid filled compartments surrounding small intracerebral vessels. They are conduits for fluid transport, exchange, and waste clearance. Their relationship to cardiovascular markers is unknown. We tested whether ePVS counts are affected by vascular risk factors and comorbidities. Method: We analyzed nondemented subjects' brains from the FL‐VIP (comprised of 1Florida ADRC participants with available high quality 3T MRI). ePVS were defined as parenchymal hypo/hyperintensities of <3 mm seen on a single axial T1, T2 or FLAIR section through the basal ganglia (BG) and centrum semiovale (CSO). ePVS were defined as parenchymal hyperintensities <3 mm and graded as 0 (none), 1 (<10), 2 (11‐20), 3 (21‐40), and 4 (>40). The side with highest ePVS count was graded. BG and CSO scores were summed to yield a 0‐8 total. Generalized linear models were used to estimate ePVS score associations with demographics, vascular risk factors, and vascular comorbidities (VASCom score). Result: Of 228 non‐demented participants (mean age 72±8 years; 61% women, 60% Hispanic, mean education 15±4 years, 33% ApoE4 carriers), 59% were hypertensive, 21% diabetic, 66% hypercholesteremic, and 30% obese. Mean VASCom score was 2.3±1.6. ePVS scores ranged from 0‐7 (mean 2.5±1.4), 0‐4 for BG (mean 1.3±0.7) and CSO (mean 1.2 ±0.9). Majority of subjects (73%) had a total ePVS score greater than 1. In unadjusted analysis, BG ePVS were correlated with age (r=0.27, P<0.001), hypertension (r=0.19, p=0.008), MI/CHF/Angina (r=0.14, P=0.03), and coronary bypass (r=0.20, P=0.004). CSO ePVS were correlated with age (r=0.23, P=0.0005), obesity (r=‐0.15, p=0.03), and dyslipidemia (r=‐0.14, P=0.04). In multivariable stepwise regression analyses, BG ePVS were associated with age (β=0.024 per year, p=0.0001), being Hispanic (OR=0.21, P=0.035), coronary bypass (OR=0.70, p=0.025), and hypertension (OR =0.17, p=0.079). CSO PeVS were associated with age (β=0.028 per year, P<0.001) and dyslipidemia (OR=‐0.31, P=0.02). Conclusion: ePVS were more prevalent with aging, greater vascular risk burden, and Hispanic origin. Hypertension, cardiovascular disease and prior coronary bypass surgery were associated with ePVS in BG, but not in CSO. BG ePVS may be an important marker of small vessel disease linking vascular burden and cognitive impairment.