Key Points
Question
Is reduced left atrial strain (LAε) associated with new-onset ischemic stroke?
Findings
In this community-based cohort of 806 adults aged 55 years or older without a history of stroke or atrial fibrillation followed up for a mean of 10.9 years, new-onset ischemic stroke occurred in 53 participants (7%). Reduced LAε and left atrial strain rate (LASR) were independently associated with new ischemic stroke after adjustments for pertinent covariates including left atrial volumes, left ventricular global longitudinal strain, and incident atrial fibrillation.
Meaning
This study suggests that LAε and LASR are independently associated with ischemic stroke risk and may improve stroke risk stratification in older adults.
Abstract
Importance
The risk of ischemic stroke is higher among patients with left atrial (LA) enlargement. Left atrial strain (LAε) and LA strain rate (LASR) may indicate LA dysfunction when LA volumes are still normal. The association of LAε with incident ischemic stroke in the general population is not well established.
Objective
To investigate whether LAε and LASR are associated with new-onset ischemic stroke among older adults.
Design
The Cardiovascular Abnormalities and Brain Lesions study was conducted from September 29, 2005, to July 6, 2010, to investigate cardiovascular factors associated with subclinical cerebrovascular disease. A total of 806 participants in the Northern Manhattan Study who were aged 55 years or older without history of prior stroke or atrial fibrillation (AF) were included, and annual follow-up telephone interviews were completed May 22, 2022. Statistical analysis was performed from June through November 2022.
Exposures
Left atrial strain and LASR were assessed by speckle-tracking echocardiography. Global peak positive longitudinal LAε and positive longitudinal LASR during ventricular systole, global peak negative longitudinal LASR during early ventricular diastole, and global peak negative longitudinal LASR during LA contraction were measured. Brain magnetic resonance imaging was used to detect silent brain infarcts and white matter hyperintensities at baseline.
Main Outcomes and Measures
Risk analysis with cause-specific Cox proportional hazards regression modeling was used to assess the association of positive longitudinal LAε and positive longitudinal LASR with incident ischemic stroke, adjusting for other stroke risk factors, including incident AF.
Results
The study included 806 participants (501 women [62.2%]) with a mean (SD) age of 71.0 (9.2) years; 119 participants (14.8%) were Black, 567 (70.3%) were Hispanic, and 105 (13.0%) were White. During a mean (SD) follow-up of 10.9 (3.7) years, new-onset ischemic stroke occurred in 53 participants (6.6%); incident AF was observed in 103 participants (12.8%). Compared with individuals who did not develop ischemic stroke, participants with ischemic stroke had lower positive longitudinal LAε and negative longitudinal LASR at baseline. In multivariable analysis, the lowest (ie, closest to zero) vs all other quintiles of positive longitudinal LAε (adjusted hazard ratio [HR], 3.12; 95% CI, 1.56-6.24) and negative longitudinal LASR during LA contraction (HR, 2.89; 95% CI, 1.44-5.80) were associated with incident ischemic stroke, independent of left ventricular global longitudinal strain and incident AF. Among participants with a normal LA size, the lowest vs all other quintiles of positive longitudinal LAε (HR, 4.64; 95% CI, 1.55-13.89) and negative longitudinal LASR during LA contraction (HR, 11.02; 95% CI 3.51-34.62) remained independently associated with incident ischemic stroke.
Conclusions and Relevance
This cohort study suggests that reduced positive longitudinal LAε and negative longitudinal LASR are independently associated with ischemic stroke in older adults. Assessment of LAε and LASR by speckle-tracking echocardiography may improve stroke risk stratification in elderly individuals.
This cohort study investigates whether positive longitudinal left atrial strain and positive longitudinal left atrial strain rate are associated with new-onset ischemic stroke among older adults.
Introduction
Stroke is one of the most common causes of death worldwide, with a prevalence greater than 3% in the general population.1 Because of the increasing life expectancy of the population, this number is expected to increase as the stroke rate doubles with each decade after 55 years of age.2 Despite improvements in the diagnostic workup of stroke, up to one-fourth remain without a clear etiology; identifying stroke mechanisms and early indicators of increased risk is therefore of critical importance.3
Abnormalities of the left atrium (LA) have been associated with increased stroke risk, especially ischemic stroke. Assessment of LA volume and reservoir function has shown them to be associated with cardiovascular outcomes.4 Left atrial strain (LAε) and LA strain rate (LASR) assessed by speckle-tracking echocardiography were reported to provide additional prognostic information and detect LA dysfunction at an earlier stage, despite normal LA volumes.5 Left atrial strain was associated with cerebrovascular events in a retrospective study of patients with cryptogenic stroke and no history of atrial fibrillation (AF).6 However, it is unclear whether reduced LAε and LASR are independently associated with an increased risk of ischemic stroke in the general population and whether the association may be present even with normal or mildly increased LA volumes. Therefore, the present study aimed to examine the association of LAε and LASR measured by speckle-tracking echocardiography with the risk of incident ischemic stroke in a community-based cohort of predominantly elderly individuals.
Methods
Study Population
The Cardiovascular Abnormalities and Brain Lesions (CABL) study investigated cardiovascular factors associated with subclinical cerebrovascular disease in a community-based cohort. The study cohort was derived from the Northern Manhattan Study (NOMAS), an observational, prospective, population-based cohort of stroke-free participants enrolled in the Northern Manhattan neighborhood in New York, New York, between 1993 and 2001. The details and design of the study of NOMAS have been described previously.7 Eligibility criteria for participants required them to be without a prior stroke history, to be 40 years of age or older, and to have been a resident of Northern Manhattan for at least 3 months in a household with a telephone. From 2003 to 2008, participants who were 50 years of age or older who had no previous diagnosis of stroke and no contraindications to magnetic resonance imaging (MRI) were invited to participate in a brain MRI substudy, the NOMAS-MRI; 1290 participants were enrolled. Starting September 29, 2005, NOMAS-MRI participants aged 55 years or older who voluntarily agreed to undergo a more extensive cardiovascular evaluation were included in CABL, whose enrollment ended in July 6, 2010. Of the total 1004 CABL participants, 26 were excluded for prior cardiac events, 62 for history of AF or evidence of AF on baseline electrocardiography, and 110 for missing or inadequate echocardiographic images, leaving a final study cohort of 806 participants (eFigure in Supplement 1). Data on race and ethnicity were collected to factor possible race-ethnic differences into the results. The classification of races and ethnicities was based on self-identification and modeled after the US Census. Categorization was non-Hispanic White (White), non-Hispanic Black (Black), and Hispanic; the “other” category included all other races and ethnicities, including American Indian or Alaska Native, Hawaiian or Pacific Islander, Asian, or more than 1 race. Written informed consent was obtained from all study participants. The study complied with the Declaration of Helsinki8 and was approved by the institutional review boards of Columbia University Irving Medical Center and the University of Miami. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Risk Factor Assessment and Follow-up
Cardiovascular risk factors were assessed through direct examination and interview by trained research assistants. Hypertension was defined as systolic blood pressure of 140 mm Hg or higher or diastolic blood pressure of 90 mm Hg or higher at the time of the visit (mean of 2 readings) or a patient’s self-reported history of hypertension or use of antihypertensive medications. Diabetes was defined as a fasting blood glucose level of 126 mg/dL or higher (to convert to millimoles per liter, multiply by 0.0555) or patient’s self-reported history of diabetes. Hypercholesterolemia was defined as a total serum cholesterol concentration of 240 mg/dL or greater (to convert to millimoles per liter, multiply by 0.0259) or use of lipid-lowering medications.
Echocardiographic Assessment
Two-Dimensional Echocardiography
Transthoracic echocardiography was performed using a commercially available system (iE 33; Philips) by a trained, registered cardiac sonographer according to a standardized protocol. Left ventricular (LV) dimensions were measured from a parasternal long-axis view according to the recommendations of the American Society of Echocardiography (ASE),9 and LV mass was calculated with a validated method10 and indexed by body surface area. Left ventricular ejection fraction was assessed using the biplane modified Simpson rule.
Speckle-Tracking Strain Imaging
Speckle-tracking analysis was performed offline using commercially available software (Tomtec Image Arena, version 4.6; Tomtec Imaging Systems). Left atrial strain analysis was performed from the apical 4-chamber view. Speckle-tracking analysis of LA myocardial deformation was performed from 2-dimensional grayscale loops by automatically tracking myocardial speckles using R-R gating after manually selecting landmark points. The atrial wall was divided into 6 segments by the software (2 segments each from the atrial septum, lateral wall, and roof of the LA). Left atrial strain measures obtained were (1) LA reservoir function, expressed as global peak positive longitudinal LAε and positive longitudinal LASR during ventricular systole; (2) LA conduit function, expressed as global peak negative longitudinal LASR during early ventricular diastole; and (3) LA pump function, expressed as global peak negative longitudinal LASR during LA contraction. At least 2 cardiac cycles were recorded, at a frame rate of 45 frames per second. Studies with inadequate loop images were excluded.
Analysis of LV global longitudinal strain (GLS) over the longitudinal axis was performed from 2-dimensional grayscale loops by automatic tracking of myocardial speckles after manual selection of landmark points as previously described11; GLS was measured from 3 apical views. Abnormal GLS was defined as GLS of −14.7% or greater, representing the cutoff identifying the lowest 5% of the GLS distribution in a subgroup of healthy participants, as previously reported.12
Other Echocardiographic Measurements
Real-time 3-dimensional echocardiography was used to measure LA volumes using 5 anatomical landmarks (septal, lateral, anterior, inferior mitral annulus, and posterior wall of the LA). A pyramidal full-volume data set was obtained from the acquisition of 4 subvolumes from 4 consecutive cardiac cycles triggered to the R wave of the electrocardiogram, allowing volume rates between 15 and 25 mm per second. Left atrial minimum volume, LA maximum volume, and LA emptying volume were measured and indexed by the body surface area. A detailed description of the technique has been reported previously.13 Echocardiograms were interpreted blinded to the results of the brain MRI.
Brain MRI
A detailed description of the assessment of subclinical cerebrovascular lesions at baseline has been published previously.14 In brief, brain imaging was performed on a 1.5-T MRI system (Philips Medical Systems). The interval between MRI and echocardiographic examination was less than 3 months for 519 participants (64.4%), and more than 2 years for the remainder. Silent brain infarcts were defined as cavitation on the fluid-attenuated inversion recovery sequence of at least 3 mm, distinct from a vessel (due to the lack of signal void on T2 sequence). Interobserver agreement for silent brain infarct detection was 93.3%.14 White matter hyperintensity (WMH) analysis was based on a fluid-attenuated inversion recovery image, was expressed as a proportion of total cranial volume to correct for differences in head size, and log transformed to achieve a normal distribution. The upper quartile (WMH4) was compared with the other quartiles combined. All measurements were performed blinded to the participant’s clinical information.
Follow-up and Detection of Outcomes
Annual telephone interviews were conducted for follow-up of participants, with the most recent contact on May 22, 2022. During the standardized interview, any vascular event or acknowledgment of neurologic or cardiac symptoms triggered an in-person assessment. Active hospital surveillance of admission and discharge International Classification of Diseases, Ninth Revision, codes was performed. Stroke was defined by the first symptomatic occurrence of any stroke type. Ischemic strokes were defined by the TOAST (Trial of ORG 10172 in Acute Stroke Treatment) criteria.15 Two neurologists (M.S.V.E. and R.L.S.) diagnosed stroke independently, and the NOMAS principal investigators adjudicated disagreements. New-onset AF was ascertained through a questionnaire administered by trained research assistants at the annual follow-up, then confirmed by analysis of electrocardiograms or medical records; active hospital surveillance for AF as admission or discharge diagnosis was performed; and the participant’s primary physician was contacted when appropriate.
Statistical Analysis
Statistical analysis was performed from June through November 2022. Data are presented as mean (SD) values for continuous variables and as frequencies for categorical variables. The risk analysis with cause-specific Cox proportional hazards regression model was used to assess the association of LAε and LASR parameters with incident ischemic stroke; 95% CIs and hazard ratios (HRs) were calculated comparing the lowest (ie, worst) quintile of each LAε variable against the other quintiles combined, with “worst” being intended as “closest to zero.”16 Death was considered a competing event. Incident AF was treated as a time-dependent covariate. Variables associated with incident stroke in univariate analysis with a P value of .10 or less were entered as covariates in multivariable models. Sensitivity analyses in participants with echocardiography and MRI performed at the same time and after exclusion of history of anticoagulation for other clinical indications were performed. Cumulative incidence curves for ischemic stroke onset were obtained comparing the lowest quintile of LAε variables vs the other quintiles combined. Analyses were conducted using SAS, version 9.4 (SAS Institute Inc), and figures were generated using R, version 4.0.3 (R Group for Statistical Computing). All P values were from 2-sided tests and results were deemed statistically significant at P < .05.
Results
Characteristics of the Study Population
Left atrial strain and LASR were assessed in 806 participants (305 men [37.8%] and 501 women [62.2%]; mean [SD] age, 71.0 [9.2] years). Clinical and demographic characteristics of the study population are shown in Table 1. A total of 119 participants (14.8%) were Black, 567 (70.3%) were Hispanic, 105 (13.0%) were White, and 15 (1.9%) were other race or ethnicity. Table 2 shows the echocardiographic variables in the study population. The mean (SD) positive longitudinal LAε was 28.0% (5.3%), the mean (SD) positive longitudinal LASR during ventricular systole was 0.94 (0.23) second−1, the mean (SD) negative longitudinal LASR during early ventricular diastole was −0.53 (0.23) second−1, and the mean (SD) negative longitudinal LASR during LA contraction was −1.00 (0.29) second−1. Silent brain infarcts were present in 90 individuals (11.2%); the WMH4 subgroup consisted of 201 individuals.
Table 1. Demographic and Clinical Characteristics of the Study Population.
| Characteristic | No. (%) (N = 806) |
|---|---|
| Age, mean (SD), y | 71.0 (9.2) |
| Sex | |
| Female | 501 (62.2) |
| Male | 305 (37.8) |
| BMI, mean (SD) | 28.3 (4.7) |
| Heart rate, mean (SD), beats/min | 69.2 (10.7) |
| BP, mean (SD), mm Hg | |
| Systolic | 136.1 (17.6) |
| Diastolic | 78.5 (9.5) |
| Hypertension | 633 (78.5) |
| Diabetes | 237 (29.4) |
| Hypercholesterolemia | 535 (66.4) |
| Coronary artery disease | 35 (4.3) |
| Cigarette smoking history | 415 (51.5) |
| Antihypertensive medication | 578 (17.7) |
| Lipid-lowering medication | 393 (48.8) |
| Aspirin | 507 (62.9) |
| History of warfarin use | 29 (3.6) |
| CHA2DS2-VASC score | |
| 0 | 31 (3.8) |
| 1 | 95 (11.8) |
| ≥2 | 680 (84.4) |
| Race and ethnicity | |
| Black | 119 (14.8) |
| Hispanic | 567 (70.3) |
| White | 105 (13.0) |
| Othera | 15 (1.9) |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); BP, blood pressure; CHA2DS2-VASC, congestive heart failure, hypertension, age ≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74 years, and sex category (female).
Other race and ethnicity consisted of all other races and ethnicities, including American Indian or Alaska Native, Hawaiian or Pacific Islander, Asian, or more than 1 race.
Table 2. Echocardiographic Data.
| Data | Mean (SD) |
|---|---|
| 2-Dimensional echocardiography | |
| LV septal thickness, mm | 11.3 (1.8) |
| LV end-diastolic diameter index, mm/m2 | 25.5 (3.0) |
| LV posterior wall thickness, mm | 11.1 (1.5) |
| LV mass index, g/m2 | 102.1 (25.2) |
| LV end-diastolic volume index, mL/m2 | 53.3 (14.7) |
| LV end-systolic volume index, mL/m2 | 19.6 (8.4) |
| LV ejection fraction, % | 63.9 (6.2) |
| LV global longitudinal strain, % | −17.3 (3.0) |
| Mitral regurgitation (>mild), No. (%) | 59 (7.3) |
| LAε and SR | |
| Positive longitudinal LAε, % | 28.0 (5.3) |
| Positive longitudinal LASR during ventricular systole, s−1 | 0.94 (0.23) |
| Negative longitudinal LASR during early ventricular diastole, s−1 | −0.53 (0.23) |
| Negative longitudinal LASR during LA contraction, s−1 | −1.00 (0.29) |
| 3-Dimensional echocardiography | |
| Left atrial maximum volume index, mL/m2 | 24.5 (7.3) |
| Left atrial minimum volume index, mL/m2 | 13.5 (5.6) |
| Left atrial emptying volume index, mL/m2 | 11.0 (3.8) |
Abbreviations: LAε, left atrial strain; LASR, left atrial strain rate; LV, left ventricular; SR, strain rate.
Clinical Variables and Incidence of Ischemic Stroke
The mean (SD) follow-up duration was 10.9 (3.7) years. Ischemic stroke occurred in 53 participants (6.6%). Incident AF was observed in 103 participants (12.8%), of whom 19 (18.4%) experienced an ischemic stroke. Clinical variables associated with ischemic stroke using univariable Cox proportional hazards regression analyses are shown in Table 3. Age, LV ejection fraction, LV mass index, β-blocker use, LA minimum volume index, left atrial emptying volume index, abnormal LV GLS, and incident AF were significantly associated with new-onset ischemic stroke in univariable analysis. Compared with patients without ischemic stroke, patients with incident ischemic stroke had significantly lower mean (SD) values of positive longitudinal LAε (25.7% [4.9%] vs 28.1% [5.3%]; P = .001), positive longitudinal LASR during ventricular systole (0.90 [0.25] second−1 vs 0.94 [0.23] second−1; P = .04), negative longitudinal LASR during early ventricular diastole (−0.47 [0.21] second−1vs −0.53 [0.23] second−1; P = .03), and negative longitudinal LASR during LA contraction (−0.89 [0.31] second−1 vs −1.01 [0.29] seconds−1; P = .002).
Table 3. Demographic and Clinical Variables Associated With Incident Ischemic Stroke–Univariable Analysis.
| Variable | HR (95% CI) | P value |
|---|---|---|
| Age, per 1-y increase | 1.06 (1.03-1.09) | <.001 |
| Male sex | 1.15 (0.66-1.99) | .63 |
| BMI | 1.01 (0.95-1.07) | .81 |
| Hypertension | 1.96 (0.88-4.33) | .10 |
| Diabetes | 0.94 (0.51-1.73) | .84 |
| Cigarette smoking | 0.72 (0.42-1.24) | .24 |
| Hypercholesterolemia | 1.28 (0.71-2.33) | .42 |
| Heart rate, beats/min | 0.99 (0.96-1.01) | .28 |
| Coronary artery disease | 2.05 (0.74-5.68) | .17 |
| Blood pressure, mm Hg | ||
| Systolic | 1.01 (1.00-1.03) | .09 |
| Diastolic | 0.99 (0.96-1.02) | .46 |
| LV ejection fraction, per each percentage point increase | 0.96 (0.92-0.99) | .01 |
| LV mass index, per 1-g/m2 increase | 1.01 (1.00-1.02) | .01 |
| Mitral regurgitation (>mild) | 2.09 (0.89-4.91) | .09 |
| Left atrial maximum volume index, per 1-mL/m2 increase | 1.01 (0.97-1.06) | .53 |
| Left atrial minimum volume index, per 1-mL/m2 increase | 1.06 (1.01-1.11) | .01 |
| Left atrial emptying volume index, per 1-mL/m2 increase | 0.89 (0.81-0.98) | .02 |
| Silent brain infarcts | 1.75 (0.85-3.58) | .13 |
| White matter hyperintensities | 1.75 (0.98-3.12) | .06 |
| Antihypertensive medications | 1.83 (0.92-3.64) | .09 |
| Lipid-lowering medications | 1.33 (0.78-2.29) | .30 |
| Aspirin | 1.60 (0.88-2.91) | .12 |
| History of warfarin use | 0.59 (0.08-4.28) | .60 |
| β-Blocker use | 1.87 (1.07-3.26) | .03 |
| New-onset AF during follow-up | 3.45 (1.44-8.29) | .01 |
| Abnormal LV GLS (≥−14.7%)a | 3.50 (1.75-6.99) | <.001 |
| CHA2DS2-VASC score | ||
| 0 | 1 [Reference] | |
| 1 | 0.65 (0.12-3.55) | .62 |
| ≥2 | 1.25 (0.30-5.13) | .76 |
Abbreviations: AF, atrial fibrillation; BMI, body mass index; CHA2DS2-VASc, congestive heart failure, hypertension, age ≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74 years, and sex category (female); GLS, global longitudinal strain; HR, hazard ratio; LV, left ventricular.
Abnormal GLS of −14.7% or more indicates the 5th percentile of the distribution of GLS.
LAε and Incident Ischemic Stroke
Univariable and multivariable Cox proportional subdistribution hazards regression analyses were performed to identify whether LAε and LASR were associated with incident ischemic stroke (Table 4). In unadjusted analysis, positive longitudinal LAε (HR, 3.33; 95% CI, 1.91-5.79; P < .001), negative longitudinal LASR during LA contraction (HR, 3.23; 95% CI, 1.86-5.59; P < .001), positive longitudinal LASR during ventricular systole (HR, 2.12; 95% CI, 1.19-3.78; P = .01), and negative longitudinal LASR during early ventricular diastole (HR, 1.93; 95% CI, 1.06-3.52; P = .03) were all associated with ischemic stroke. In multivariable analysis, positive longitudinal LAε (adjusted HR for lowest vs all other quintiles, 3.12; 95% CI, 1.56-6.24) and negative longitudinal LASR during LA contraction (HR for lowest vs all other quintiles, 2.89; 95% CI, 1.44-5.80) remained significantly associated with incident ischemic stroke after adjusting for covariates (for the individual associations of covariates, see eTable 1 in Supplement 1). The Figure shows the cumulative incidence curves for incident ischemic stroke stratified by lowest vs all other quintiles for positive longitudinal LAε and negative longitudinal LASR during LA contraction.
Table 4. Association of LAε and LASR With Incident Ischemic Stroke.
| Variable | Univariable model | Multivariable modela | ||
|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |
| Positive longitudinal LAε | 3.33 (1.91-5.79) | <.001 | 3.12 (1.56-6.24) | .001 |
| Positive longitudinal LASR during ventricular systole | 2.12 (1.19-3.78) | .01 | 1.24 (0.59-2.60) | .57 |
| Negative longitudinal LASR during early ventricular diastole | 1.93 (1.06-3.52) | .03 | 1.52 (0.73-3.15) | .26 |
| Negative longitudinal LASR during LA contraction | 3.23 (1.86-5.59) | <.001 | 2.89 (1.44-5.80) | .003 |
Abbreviations: HR, hazard ratio; LAε, left atrial strain; LASR, left atrial strain rate.
Adjusted for age, hypertension, systolic blood pressure, white matter hyperintensities, mitral regurgitation greater than mild, hypertension medications, left ventricular ejection fraction, left ventricular mass, β-blocker use, left atrial minimum volume index, left atrial emptying volume index, abnormal left ventricular global longitudinal strain; and new-onset atrial fibrillation during follow-up. The values are comparing the lowest (worst) quintile with all other quintiles.
Figure. Cumulative Incidence of Ischemic Stroke for Lowest Quintile vs All Other Quintiles for Positive Longitudinal Left Atrial Strain (LAɛ) and Negative Longitudinal Left Atrial Strain Rate (LASR) During LA Contraction.
Association of LAε With Incident Ischemic Stroke Among Participants With Normal LA Size
To investigate the association of LAε with incident ischemic stroke among participants with a normal-sized LA, we divided our cohort by the median value of the LA maximum volume index measured by 3-dimensional echocardiography (23.7 mL/m2 [IQR, 19.70-28.09 mL/m2]). Among participants with LA size below the median value, positive longitudinal LAε (HR for lowest vs all other quintiles, 4.64; 95% CI, 1.55-13.89) and negative longitudinal LASR during LA contraction (HR for lowest vs all other quintiles, 11.02; 95% CI, 3.51-34.62) remained associated with incident ischemic stroke even after multivariable adjustment (eTable 2 in Supplement 1).
Among participants with a larger (≥23.7 mL/m2) LA maximum volume index, positive longitudinal LAε (unadjusted HR for lowest vs all other quintiles, 3.93; 95% CI, 1.59-9.68) was associated with ischemic stroke. However, no association remained after multivariable adjustment (eTable 3 in Supplement 1).
Sensitivity Analyses
Adjustment for Time Between Echocardiography and MRI
We conducted a sensitivity analysis among the 64.4% of participants who had brain MRI and echocardiography within 3 months of each other. After multivariable adjustment, positive longitudinal LAε (HR for lowest vs all other quintiles, 3.35; 95% CI, 1.37-8.15) and negative longitudinal LASR during LA contraction (HR for lowest vs all other quintiles, 3.17; 95% CI, 1.30-7.74) remained significantly associated with incident ischemic stroke.
History of Anticoagulant Use
We also conducted a sensitivity analysis excluding the 29 participants with a history of anticoagulant use. After multivariable adjustment, positive longitudinal LAε (HR for lowest vs all other quintiles, 3.50; 95% CI, 1.73-7.10) and negative longitudinal LASR during LA contraction (HR for lowest vs all other quintiles, 3.09; 95% CI, 1.53-5.27) remained significantly associated with incident ischemic stroke.
Discussion
In the present study, we assessed the independent association of LAε variables with ischemic stroke occurrence in a community of older adults without known AF or stroke history. Decreased LAε and LASR were significantly associated with ischemic stroke development after adjustment for other risk factors, including LA volumes, LV GLS, and incident AF. The independent association of LAε with incident ischemic stroke was found even among participants with a normal LA size; the adjusted HR for incident ischemic stroke was higher among participants with an LA maximum volume index below vs above the median value. Abnormalities in LA structure and function were associated with cardiovascular events.
Electrocardiographic markers of LA enlargement have been associated with vascular brain injury in the absence of documented AF.17 Rates of ischemic stroke recurrence18 and first ischemic stroke19 were higher among participants with LA enlargement. A previous study by our group4 observed that LA phasic volumes and reservoir function were significantly associated with cardiovascular events. In the present study, we report on the independent association of LAε variables with incident ischemic stroke after adjustment for LA volumes and volume-derived indices of emptying function. In addition, LAε variables were independently associated with incident ischemic stroke risk after adjustment for incident AF and LV GLS, 2 factors that might have been suspected to account for the association.
Left atrial strain may be a better indicator than LA volumes of risk of cerebrovascular events. Left atrial strain assessment detects atrial myopathy earlier than LA enlargement, suggesting that the latter occurs in more advanced stages of LA dysfunction.6 Previously, impaired LAε was observed among patients with prior cryptogenic stroke.20 A report21 from the Cardiovascular Health Study also suggested an association between reduced LA reservoir strain and incident stroke, although technical aspects of the LAε measurement (use of analogic images digitized at a later time, with consequent low frame rate) and the race and ethnicity characteristics of the studied cohort (mainly White, with a small Black subgroup) significantly limited the applicability of the results to the general population. By comparison, the present study cohort is enriched with self-identified Hispanic and Black participants. Moreover, we also examined LAε rate, thus adding the analysis of LA conduit and pump functions to that of reservoir function alone. Left atrial strain in our study was also associated with incident ischemic stroke after adjustment for subclinical cerebrovascular disease, further reinforcing the independent association between atrial myopathy and new-onset stroke.
Nevertheless, the mechanism linking LA dysfunction with stroke is unclear. One explanation might be the presence of undiagnosed AF. Negative longitudinal LASR during LA contraction has been shown to be associated with the risk of new-onset AF and future embolic events with greater accuracy than markers of LA size.22 Similarly, reduced positive longitudinal LASR during ventricular systole was associated with development of AF after a cryptogenic stroke, suggesting a possible benefit associated with anticoagulation in patients with reduced LAε when AF has not yet been documented.20,23 Recently, positive longitudinal LAε was shown to be associated with new-onset AF among individuals younger than 65 years but not among older individuals.24 In the present study, the association between impaired LAε and ischemic stroke persisted after adjustment for new-onset AF, suggesting that LAε may be an important marker of atrial myopathy in the absence of AF. Recently, it has been argued that an LA source for cerebrovascular events might exist in the absence of AF,25 given the dissociation between the timing of AF episodes and stroke onset in individuals undergoing prolonged monitoring of cardiac rhythm.26,27 Therefore, AF might be a marker rather than causal risk factor in some patients. The ongoing ARCADIA (Atrial Cardiopathy and Antithrombotic Drugs in Prevention After Cryptogenic Stroke) study will provide further insights into atrial myopathy by assessing the effect of apixaban vs aspirin in secondary prevention after cryptogenic stroke.28
Left atrial dysfunction may also serve as a proxy for LV dysfunction, which may be involved in cardiac embolism.29 Reduced LAε and LV GLS have been associated with AF in patients with prior cryptogenic stroke.30 Negative longitudinal LASR during LA contraction was an independent factor associated with AF regardless of the size of the LA, whereas LV GLS was associated with AF only in patients with abnormal LA volumes. Previous studies by our group reported that LV GLS was an independent factor associated with incident AF31 and was associated with subclinical cerebrovascular disease.32 In the present study, we found that positive longitudinal LAε and negative longitudinal LASR during LA contraction remained significantly associated with incident ischemic stroke even after adjustment for LV GLS, which may exclude a significant role for subclinical LV dysfunction in explaining our results.
Atherosclerosis might also play a role in the association of LA functional impairment and risk of ischemic stroke. A shared subclinical atherosclerotic background might link LA dysfunction and ischemic stroke via endothelial dysfunction from an inflammatory mechanism, fibrotic remodeling of the atrial and arterial walls, and increased thrombotic risk.33 Consequently, LA dysfunction may be a marker of severity of coexisting cardiovascular disease, which might be involved in the excess stroke risk.34
The CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years [doubled], diabetes, stroke [doubled], vascular disease, age 65-74 years, and sex category [female]) score is the most frequently used clinical risk score for stroke risk assessment and anticoagulation therapy in patients with and without AF.35 Recently, the new P2-CHA2DS2-VASc (abnormal P wave axis [doubled], congestive heart failure, hypertension, age ≥75 years [doubled], diabetes, stroke [doubled], vascular disease, age 65-74 years, and sex category [female]) score incorporated the measurement of P-wave indices on electrocardiography, resulting in an improvement in the assessment of stroke risk in patients with AF.36 Left atrial strain also showed incremental diagnostic value vs the CHA2DS2-VASc score for the stratification of embolic risk in patients with AF and the assessment of poststroke mortality.37 In the present study, the association of LAε with incident ischemic stroke was present among participants with normal LA size, highlighting the importance of a functional, rather than structural, assessment of the LA to evaluate stroke risk. Left atrial strain may identify individuals who could benefit from closer or more frequent rhythm monitoring to detect subclinical AF episodes and thus inform earlier prevention. Intensive blood pressure lowering may also reduce adverse atrial remodeling and stroke risk, as hypertension is one of the main risk factors for LA dysfunction.
Our findings may have other possible implications. For example, in the univariable analysis, β-blocker use was associated with an excess risk of ischemic stroke. These agents have been associated with LA dysfunction in individuals with hypertension,38 and thus avoidance in some patients with reduced LAε might be considered, although prospective validation would be required.
Strengths and Limitations
The present study has several strengths. It used advanced Doppler echocardiographic techniques to examine the association of LAε and LASR with incident ischemic stroke in a multiethnic, community-based sample of older adults without a history of AF or stroke. Our analyses were adjusted for pertinent covariates, including several echocardiographic variables. We showed an independent association of LAε variables with ischemic stroke risk, adjusting for the presence of subclinical cerebrovascular disease at baseline, incident AF, and LV GLS.
However, our study also has important limitations. First, the study population predominantly included older adults; therefore, our results may not apply to younger individuals. However, stroke is more common among elderly individuals; thus, our cohort is an appropriate population in which to examine this topic. Second, our population is of prevalent Hispanic ethnicity, which may limit the generalizability of our findings to other ethnicities. Third, the small number of ischemic stroke events makes our results exploratory, with replication needed in larger cohorts with larger numbers of ischemic stroke events. Nevertheless, the results are directionally similar to those reported in the Cardiovascular Health Study.21 Fourth, although we adjusted for the occurrence of clinically detected, sustained AF during follow-up, we did not systematically perform prolonged monitoring of cardiac rhythm; therefore, the possibility that transient episodes of AF may account in part for the results cannot be excluded. However, the present study is not intended to establish reduced LAε as an ischemic stroke mechanism but rather as an independent strong and early marker of ischemic stroke risk, an observation that holds true regardless of the associated stroke mechanism.
Conclusions
The findings of this cohort study suggest that LAε and LASR are significantly associated with incident ischemic stroke in a community-based cohort of predominantly elderly individuals. This association was independent of stroke risk factors, measures of LA size, LV GLS, subclinical cerebrovascular disease, and incident AF. The independent association was present even among participants with normal LA size. The assessment of LAε and LASR by speckle-tracking echocardiography may improve stroke risk stratification in elderly individuals.
eFigure. Flow of Participants Included in the Study
eTable 1. Association of LAε and LASR With Incident Ischemic Stroke – All Covariates Are Shown
eTable 2. Association of LAε and LASR With Incident Ischemic Stroke Among Participants With LAVI<23.7 mL/m2
eTable 3. Association of LAε and LASR With Incident Ischemic Stroke Among Participants With LAVI≥23.7 mL/m2
Data Sharing Statement
References
- 1.Benjamin EJ, Virani SS, Callaway CW, et al. ; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee . Heart disease and stroke statistics—2018 update: a report from the American Heart Association. Circulation. 2018;137(12):e67-e492. doi: 10.1161/CIR.0000000000000558 [DOI] [PubMed] [Google Scholar]
- 2.Brown RD, Whisnant JP, Sicks JD, O’Fallon WM, Wiebers DO. Stroke incidence, prevalence, and survival: secular trends in Rochester, Minnesota, through 1989. Stroke. 1996;27(3):373-380. [PubMed] [Google Scholar]
- 3.O’Donnell MJ, Xavier D, Liu L, et al. ; INTERSTROKE investigators . Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet. 2010;376(9735):112-123. doi: 10.1016/S0140-6736(10)60834-3 [DOI] [PubMed] [Google Scholar]
- 4.Russo C, Jin Z, Homma S, et al. LA phasic volumes and reservoir function in the elderly by real-time 3D echocardiography: normal values, prognostic significance, and clinical correlates. JACC Cardiovasc Imaging. 2017;10(9):976-985. doi: 10.1016/j.jcmg.2016.07.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cameli M, Lisi M, Focardi M, et al. Left atrial deformation analysis by speckle tracking echocardiography for prediction of cardiovascular outcomes. Am J Cardiol. 2012;110(2):264-269. doi: 10.1016/j.amjcard.2012.03.022 [DOI] [PubMed] [Google Scholar]
- 6.Leong DP, Joyce E, Debonnaire P, et al. Left atrial dysfunction in the pathogenesis of cryptogenic stroke: novel insights from speckle-tracking echocardiography. J Am Soc Echocardiogr. 2017;30(1):71-79. doi: 10.1016/j.echo.2016.09.013 [DOI] [PubMed] [Google Scholar]
- 7.Sacco RL, Khatri M, Rundek T, et al. Improving global vascular risk prediction with behavioral and anthropometric factors: the multiethnic NOMAS (Northern Manhattan Cohort Study). J Am Coll Cardiol. 2009;54(24):2303-2311. doi: 10.1016/j.jacc.2009.07.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 9.Lang RM, Bierig M, Devereux RB, et al. ; Chamber Quantification Writing Group; American Society of Echocardiography’s Guidelines and Standards Committee; European Association of Echocardiography . Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005;18(12):1440-1463. doi: 10.1016/j.echo.2005.10.005 [DOI] [PubMed] [Google Scholar]
- 10.Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation. 1977;55(4):613-618. doi: 10.1161/01.CIR.55.4.613 [DOI] [PubMed] [Google Scholar]
- 11.Mannina C, Jin Z, Russo C, et al. Effect of hypertension and diabetes on subclinical left ventricular systolic dysfunction in a predominantly elderly population-based cohort. Eur J Prev Cardiol. 2020;27(19):2173-2175. doi: 10.1177/2047487319872571 [DOI] [PubMed] [Google Scholar]
- 12.Russo C, Jin Z, Elkind MS, et al. Prevalence and prognostic value of subclinical left ventricular systolic dysfunction by global longitudinal strain in a community-based cohort. Eur J Heart Fail. 2014;16(12):1301-1309. doi: 10.1002/ejhf.154 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Russo C, Jin Z, Liu R, et al. LA volumes and reservoir function are associated with subclinical cerebrovascular disease: the CABL (Cardiovascular Abnormalities and Brain Lesions) study. JACC Cardiovasc Imaging. 2013;6(3):313-323. doi: 10.1016/j.jcmg.2012.10.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Willey JZ, Moon YP, Paik MC, et al. Lower prevalence of silent brain infarcts in the physically active: the Northern Manhattan Study. Neurology. 2011;76(24):2112-2118. doi: 10.1212/WNL.0b013e31821f4472 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Adams HP Jr, Bendixen BH, Kappelle LJ, et al. Classification of subtype of acute ischemic stroke: definitions for use in a multicenter clinical trial: TOAST: Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993;24(1):35-41. doi: 10.1161/01.STR.24.1.35 [DOI] [PubMed] [Google Scholar]
- 16.Voigt JU, Pedrizzetti G, Lysyansky P, et al. Definitions for a common standard for 2D speckle tracking echocardiography: consensus document of the EACVI/ASE/Industry Task Force to Standardize Deformation Imaging. J Am Soc Echocardiogr. 2015;28(2):183-193. doi: 10.1016/j.echo.2014.11.003 [DOI] [PubMed] [Google Scholar]
- 17.Kamel H, Bartz TM, Longstreth WT Jr, et al. Association between left atrial abnormality on ECG and vascular brain injury on MRI in the Cardiovascular Health Study. Stroke. 2015;46(3):711-716. doi: 10.1161/STROKEAHA.114.007762 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yaghi S, Moon YP, Mora-McLaughlin C, et al. Left atrial enlargement and stroke recurrence: the Northern Manhattan Stroke Study. Stroke. 2015;46(6):1488-1493. doi: 10.1161/STROKEAHA.115.008711 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Barnes ME, Miyasaka Y, Seward JB, et al. Left atrial volume in the prediction of first ischemic stroke in an elderly cohort without atrial fibrillation. Mayo Clin Proc. 2004;79(8):1008-1014. doi: 10.4065/79.8.1008 [DOI] [PubMed] [Google Scholar]
- 20.Pagola J, González-Alujas T, Flores A, et al. Left atria strain is a surrogate marker for detection of atrial fibrillation in cryptogenic strokes. Stroke. 2014;45(8):e164-e166. doi: 10.1161/STROKEAHA.114.005540 [DOI] [PubMed] [Google Scholar]
- 21.Kamel H, Bartz TM, Longstreth WT Jr, et al. Cardiac mechanics and incident ischemic stroke: the Cardiovascular Health Study. Sci Rep. 2021;11(1):17358. doi: 10.1038/s41598-021-96702-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hirose T, Kawasaki M, Tanaka R, et al. Left atrial function assessed by speckle tracking echocardiography as a predictor of new-onset non-valvular atrial fibrillation: results from a prospective study in 580 adults. Eur Heart J Cardiovasc Imaging. 2012;13(3):243-250. doi: 10.1093/ejechocard/jer251 [DOI] [PubMed] [Google Scholar]
- 23.Pathan F, Sivaraj E, Negishi K, et al. Use of atrial strain to predict atrial fibrillation after cerebral ischemia. JACC Cardiovasc Imaging. 2018;11(11):1557-1565. doi: 10.1016/j.jcmg.2017.07.027 [DOI] [PubMed] [Google Scholar]
- 24.Alhakak AS, Biering-Sørensen SR, Møgelvang R, et al. Usefulness of left atrial strain for predicting incident atrial fibrillation and ischaemic stroke in the general population. Eur Heart J Cardiovasc Imaging. 2022;23(3):363-371. doi: 10.1093/ehjci/jeaa287 [DOI] [PubMed] [Google Scholar]
- 25.Sajeev JK, Kalman JM, Dewey H, Cooke JC, Teh AW. The atrium and embolic stroke: myopathy not atrial fibrillation as the requisite determinant? JACC Clin Electrophysiol. 2020;6(3):251-261. doi: 10.1016/j.jacep.2019.12.013 [DOI] [PubMed] [Google Scholar]
- 26.Brambatti M, Connolly SJ, Gold MR, et al. ; ASSERT Investigators . Temporal relationship between subclinical atrial fibrillation and embolic events. Circulation. 2014;129(21):2094-2099. doi: 10.1161/CIRCULATIONAHA.113.007825 [DOI] [PubMed] [Google Scholar]
- 27.Healey JS, Connolly SJ, Gold MR, et al. ; ASSERT Investigators . Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 2012;366(2):120-129. doi: 10.1056/NEJMoa1105575 [DOI] [PubMed] [Google Scholar]
- 28.Kamel H, Longstreth WT Jr, Tirschwell DL, et al. The AtRial Cardiopathy and Antithrombotic Drugs In prevention After cryptogenic stroke randomized trial: rationale and methods. Int J Stroke. 2019;14(2):207-214. doi: 10.1177/1747493018799981 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Fuster V, Halperin JL. Left ventricular thrombi and cerebral embolism. N Engl J Med. 1989;320(6):392-394. doi: 10.1056/NEJM198902093200610 [DOI] [PubMed] [Google Scholar]
- 30.Kawakami H, Ramkumar S, Pathan F, Wright L, Marwick TH. Use of echocardiography to stratify the risk of atrial fibrillation: comparison of left atrial and ventricular strain. Eur Heart J Cardiovasc Imaging. 2020;21(4):399-407. [DOI] [PubMed] [Google Scholar]
- 31.Russo C, Jin Z, Sera F, et al. Left ventricular systolic dysfunction by longitudinal strain is an independent predictor of incident atrial fibrillation: a community-based cohort study. Circ Cardiovasc Imaging. 2015;8(8):e003520. doi: 10.1161/CIRCIMAGING.115.003520 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Russo C, Jin Z, Homma S, et al. Subclinical left ventricular dysfunction and silent cerebrovascular disease: the Cardiovascular Abnormalities and Brain Lesions (CABL) study. Circulation. 2013;128(10):1105-1111. doi: 10.1161/CIRCULATIONAHA.113.001984 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Larsen BS, Kumarathurai P, Falkenberg J, Nielsen OW, Sajadieh A. Excessive atrial ectopy and short atrial runs increase the risk of stroke beyond incident atrial fibrillation. J Am Coll Cardiol. 2015;66(3):232-241. doi: 10.1016/j.jacc.2015.05.018 [DOI] [PubMed] [Google Scholar]
- 34.Katayama T, Fujiwara N, Tsuruya Y. Factors contributing to left atrial enlargement in adults with normal left ventricular systolic function. J Cardiol. 2010;55(2):196-204. doi: 10.1016/j.jjcc.2009.10.008 [DOI] [PubMed] [Google Scholar]
- 35.Shen MJ, Arora R, Jalife J. Atrial myopathy. JACC Basic Transl Sci. 2019;4(5):640-654. doi: 10.1016/j.jacbts.2019.05.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Maheshwari A, Norby FL, Roetker NS, et al. Refining prediction of atrial fibrillation-related stroke using the P2-CHA2DS2-VASc Score. Circulation. 2019;139(2):180-191. doi: 10.1161/CIRCULATIONAHA.118.035411 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Obokata M, Negishi K, Kurosawa K, et al. Left atrial strain provides incremental value for embolism risk stratification over CHA2DS2-VASc score and indicates prognostic impact in patients with atrial fibrillation. J Am Soc Echocardiogr. 2014;27(7):709-716. doi: 10.1016/j.echo.2014.03.010 [DOI] [PubMed] [Google Scholar]
- 38.Sardana M, Syed AA, Hashmath Z, et al. Beta-blocker use is associated with impaired left atrial function in hypertension. J Am Heart Assoc. 2017;6(2):e005163. doi: 10.1161/JAHA.116.005163 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure. Flow of Participants Included in the Study
eTable 1. Association of LAε and LASR With Incident Ischemic Stroke – All Covariates Are Shown
eTable 2. Association of LAε and LASR With Incident Ischemic Stroke Among Participants With LAVI<23.7 mL/m2
eTable 3. Association of LAε and LASR With Incident Ischemic Stroke Among Participants With LAVI≥23.7 mL/m2
Data Sharing Statement
