RESUMEN
Background: Cardiac structure is an important determinant of ischemic stroke (IS) etiology; however, whether an association between cardiac structural markers and cognition post-IS exists is not yet established. The aim of this study is to examine the association between LAD and LVEF with cognitive performance among IS patients. Methods: IS patients admitted to the Johns Hopkins Hospital (2017-2019) underwent transthoracic echocardiography. IS was classified (TOAST) by a masked reviewer. Left atrial diameter (LAD) was evaluated as a non-linear continuous variable with one spline knot at 4 cm; left ventricle ejection fraction (LVEF) was dichotomized, then further evaluated as a non-linear continuous variable with spline knots at 50% and 70%. Patients were contacted by telephone on average 422 days post-stroke and administered the Six-Item Screener (SIS) to assess for dementia. SIS scores were dichotomized into low and high, imputing low scores for non-answerers. Multivariable logistic regression determined the association of SIS category with LAD or LVEF. A sensitivity analysis re-evaluated the association between SIS category and LAD, excluding participants with atrial fibrillation (AF). Results: Participants (N = 108) were on average 61 years old (range = 18-89 years), 55% male, and 63% Black. Among patients considered to have a normal LAD (≤ 4 cm), a 1 mm larger LAD was associated with 1.20 greater odds (95%CI = 1.05-1.38) of scoring in the high SIS category in the final adjustment model. This association remained significant when excluding participants with prevalent AF. There was no association between a 1 mm larger LAD and SIS category among patients with a LAD >4 cm in both the primary analysis and the sensitivity analysis. There was no association between LVEF and SIS category. Conclusions: In this prospective study, among ischemic stroke patients with a LAD within the normal range, a 1 mm increase in LAD was associated with higher scores on a telephone cognitive battery, without an association found among those with a LAD >4 cm.
RESUMEN
Objective: Cerebral microbleeds (CMB) are small accumulations of hemosiderin associated with cerebrovascular risk factors, but whether they are associated with atrial cardiopathy is not known. The goal of this study is to determine, among ischemic stroke patients, the association between study-defined atrial cardiopathy and CMB presence, location, and number. Methods: Ischemic stroke patients admitted to Johns Hopkins (2015-2019) with transthoracic echocardiography and electrocardiography were included. Cerebral microbleeds were defined as small, round hypo-intensities on T2* susceptibility weighted imaging or gradient recalled echo magnetic resonance imaging sequences. Atrial cardiopathy was defined as the presence of ≥1: left atrium diameter >4.0 cm (males) or >3.8 cm (females), PR interval >200 ms, or N-terminal pro-B-type natriuretic peptide >250 pg/ml. Binary/Ordinal logistic regression models were used to determine the association between atrial cardiopathy, and cerebral microbleed presence, location (lobar/deep), or number, each, adjusted for potential confounders. Results: Patients (N = 120) were mean age 60 years (range 22-98), 46% female, 62% black, and 39% were on anti-thrombotic medication at time of admission. 39 (32%) participants had ≥1 cerebral microbleeds. Forty-six (38%) patients had atrial cardiopathy. Atrial cardiopathy was associated with higher odds of having cerebral microbleeds (OR 2.50, 95% CI 1.02-6.15). Atrial cardiopathy was associated with lobar cerebral microbleeds (OR 2.33, 95% CI 1.01-5.37) in univariate analysis but not with deep cerebral microbleeds (OR 0.45, 95% CI 0.13-1.54), with neither association significant after adjustment. There was no difference in risk of having 1 vs. no cerebral microbleeds (RRR 2.51, 95% CI 0.75-8.37) and >1 cerebral microbleed vs none (RRR 2.57, 95% CI 0.87-7.60) among those with atrial cardiopathy. Conclusions: Atrial cardiopathy is associated with the presence, but not burden, of cerebral microbleeds in ischemic stroke patients. We cautiously suggest that atrial cardiopathy, either directly or through shared vascular risk, may contribute to the presence of CMB.