TY - JOUR
T1 - Midlife smaller and larger infarctions, white matter hyperintensities, and 20-year cognitive decline
AU - Gwen Windham, B.
AU - Griswold, Michael E.
AU - Wilkening, Steven R.
AU - Su, Dan
AU - Tingle, Jonathan
AU - Coker, Laura H.
AU - Knopman, David
AU - Gottesman, Rebecca F.
AU - Shibata, Dean
AU - Mosley, Thomas H.
N1 - Funding Information:
Financial Support: The ARIC study is carried out as a collaborative study supported by contracts HHSN268201100005C, HHSN268201100006C, HHSN268201100007C, HHSN268 201100008C, HHSN268201100009C, HHSN268201100010C, HHSN268201100011C, and HHSN268201100012C from the National Heart, Lung, and Blood Institute (NHLBI). Neurocog- nitive data are collected under contracts 2U01HL096812, 2U01HL096814, 2U01HL096899, 2U01HL096902, and 2U01 HL096917 from the National Institutes of Health (NIH) (NHLBI, National Institute of Neurological Disorders and Stroke, National Institute on Aging, and National Institute on Deafness and Other Communication Disorders), and previous brain MRI examinations were funded by grant R01-HL70825 from the NHLBI.
Publisher Copyright:
© 2019 American College of Physicians.
PY - 2019/9/17
Y1 - 2019/9/17
N2 - Background: Smaller (<3-mm) infarctions are associated with stroke and stroke mortality, but relationships with cognitive decline are unknown. Objective: To characterize the relationships of smaller, larger, and both smaller and larger infarctions in middle age with 20-year cognitive decline. Design: Longitudinal cohort study. Setting: Two ARIC (Atherosclerosis Risk in Communities) study sites with magnetic resonance imaging data (1993 to 1995) and up to 5 cognitive assessments over 20 years. Participants: Stroke-free participants aged 50 years or older. Measurements: Infarctions were categorized as none, smaller only, larger only (3 to 20 mm), or both smaller and larger. Global cognitive Z scores were derived from 3 cognitive tests administered up to 5 times. Mixed-effects models estimated adjusted associations between infarctions and cognitive decline. Results are the average difference in standardized cognitive decline associated with infarctions versus no infarctions. Results: Among 1884 participants (mean age, 62 years; 60% women; 50% black), 1611 (86%) had no infarctions, 50 (3%) had smaller infarctions only, 185 (10%) had larger infarctions only, and 35 (2%) had both. Participants with both smaller and larger infarctions had steeper cognitive decline by more than half an SD (difference, -0.57 SD [95% CI, -0.89 to -0.26 SD]) compared with those who had no infarctions. Amounts of cognitive decline associated with only smaller infarctions and only larger infarctions were similar and were not statistically different from that associated with no infarctions. Limitation: Few participants had only smaller infarctions or both smaller and larger infarctions, and the data lacked counts of smaller infarctions and volumes of white matter hyperintensities. Conclusion: The substantial cognitive decline from middle age associated with having both smaller and larger infarctions, but not larger infarctions alone, suggests that the combination of smaller and larger infarctions may escalate risk for cognitive decline later in life in stroke-free persons.
AB - Background: Smaller (<3-mm) infarctions are associated with stroke and stroke mortality, but relationships with cognitive decline are unknown. Objective: To characterize the relationships of smaller, larger, and both smaller and larger infarctions in middle age with 20-year cognitive decline. Design: Longitudinal cohort study. Setting: Two ARIC (Atherosclerosis Risk in Communities) study sites with magnetic resonance imaging data (1993 to 1995) and up to 5 cognitive assessments over 20 years. Participants: Stroke-free participants aged 50 years or older. Measurements: Infarctions were categorized as none, smaller only, larger only (3 to 20 mm), or both smaller and larger. Global cognitive Z scores were derived from 3 cognitive tests administered up to 5 times. Mixed-effects models estimated adjusted associations between infarctions and cognitive decline. Results are the average difference in standardized cognitive decline associated with infarctions versus no infarctions. Results: Among 1884 participants (mean age, 62 years; 60% women; 50% black), 1611 (86%) had no infarctions, 50 (3%) had smaller infarctions only, 185 (10%) had larger infarctions only, and 35 (2%) had both. Participants with both smaller and larger infarctions had steeper cognitive decline by more than half an SD (difference, -0.57 SD [95% CI, -0.89 to -0.26 SD]) compared with those who had no infarctions. Amounts of cognitive decline associated with only smaller infarctions and only larger infarctions were similar and were not statistically different from that associated with no infarctions. Limitation: Few participants had only smaller infarctions or both smaller and larger infarctions, and the data lacked counts of smaller infarctions and volumes of white matter hyperintensities. Conclusion: The substantial cognitive decline from middle age associated with having both smaller and larger infarctions, but not larger infarctions alone, suggests that the combination of smaller and larger infarctions may escalate risk for cognitive decline later in life in stroke-free persons.
UR - http://www.scopus.com/inward/record.url?scp=85072512403&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85072512403&partnerID=8YFLogxK
U2 - 10.7326/M18-0295
DO - 10.7326/M18-0295
M3 - Article
C2 - 31450238
AN - SCOPUS:85072512403
SN - 0003-4819
VL - 171
SP - 389
EP - 396
JO - Annals of internal medicine
JF - Annals of internal medicine
IS - 6
ER -