Effect of left ventricular ejection fraction Spectrum on 1‐Year mortality in patients with acute ischemic stroke or transient ischemic attack

Abstract Aims We aimed to investigate the association of the left ventricular ejection fraction (LVEF) spectrum with 1‐year clinical outcomes in patients with acute ischemic stroke (AIS) or transient ischemic attack (TIA). Methods In a prospective registry for the Third China National Stroke Registry (CNSR‐III), AIS or TIA patients with echocardiography records during hospitalization were recruited. All LVEFs were categorized into intervals of 5% in width. The lowest and highest intervals are ≤40% and >70%, respectively. The primary outcome was all‐cause death at 1 year. Cox proportional hazards regression analysis was performed to investigate the association between baseline LVEF and clinical outcomes. Results This analysis included a total of 14,053 patients. In total, 418 patients died during 1‐year follow‐up. Overall, LVEF ≤60% was associated with a higher risk of all‐cause death compared to LVEF >60%, independent of demographic and clinical characteristics (aHR 1.29 [95% CI 1.06–1.58]; p = 0.01). The cumulative incidence of all‐cause death was significantly different among the eight LVEF groups that survival declined successively with the decrease of LVEF (log‐rank p ≤ 0.0001). Conclusions Patients with AIS or TIA with decreased LVEF (≤60%) had a lower 1‐year survival rate after onset. LVEF 50%–60% even within the normal range, may still contribute to poor outcomes in AIS or TIA. Comprehensive evaluation of cardiac function after acute ischemic cerebrovascular disease should be strengthened.


| INTRODUC TI ON
Left ventricular ejection fraction (LVEF) is the most widely used indicator for quantifying left ventricular systolic function and is associated with cardiovascular outcomes. 1Recently, a large sample size cohort study based on community populations from a regional healthcare system demonstrated a U-shaped relationship between all-cause death and LVEF.Notably, deviation of LVEF from 60% to 65% was associated with poorer survival regardless of age, sex, or other relevant confounders, which arouse the attention to the people with supra-normal LVEF. 2 Although LVEF is routinely reported in patients with acute ischemic cerebrovascular disease during hospitalization, its impact on clinical outcomes has not been adequately investigated.4][5][6][7] However, none of these studies have investigated the association of broad LVEF spectrum with survival in patients with acute ischemic cerebrovascular disease in a large-scale clinical practice data.And whether supra-normal LVEF leads to increased risk remains unknown in these population.
We hypothesize that data from a large-scale clinical registry study will provide new insights into the impact of LVEF spectrum on clinical outcomes in patients with acute ischemic cerebrovascular disease.In this large, nationwide, multicenter, long-term follow-up prospective cohort registry for acute ischemic cerebrovascular disease, we aimed to investigate the relationship between the LVEF spectrum and clinical outcomes and to explore whether supranormal LVEF increased the risk of poor outcomes.

| Study design and participants
Participants were recruited from CNSR-III (Third China National Stroke Registry). 8The CNSR-III is a nationwide prospective registry that enrolled patients with acute ischemic stroke (AIS) or transient ischemic attack (TIA) performed in 201 hospitals across China between August 2015 and March 2018. 8Participants were consecutively enrolled if meeting the following criteria: (1) >18 years old; (2)   diagnosis of AIS or TIA within 7 days; (3) informed consent from participant or legally authorized representative.The study design and methods of the CNSR-III have been reported previously. 8Patients in the CNSR-III with echocardiography records during hospitalization were recruited to the current analysis.Those with missing data for LVEF were excluded from this analysis.

| Standard protocol approvals, registrations and patient consents
The study protocol was evaluated and approved by the ethics committee of Beijing Tiantan Hospital and each participating site.All patients or their legal representatives provided written informed consent to participate.

| Data collection
All data were collected by trained investigators and monitored by an independent contract research organization throughout the study period.Baseline data included patient demographics, risk factors (ischemic stroke, hypertension, diabetes mellitus, dyslipidemia, and chronic kidney disease), smoking status, medical history of heart disease (coronary heart disease, myocardial infarction, atrial fibrillation, valvular heart defect, dilated cardiomyopathy, and heart failure), National Institutes of Health Stroke Scale (NIHSS) score at admission, laboratory test, etiological classification, echocardiography parameters, and medication at discharge.The etiological classification was conducted by the TOAST (Trial of Org 10,172 in Acute Stroke Treatment) criteria. 9EF estimation was extracted from transthoracic echocardiography results performed during hospitalization.All LVEFs were categorized into intervals 5% in width and inclusive of the higher threshold.The lowest and highest intervals were ≤40% and >70%, respectively.Therefore, we classified patients into eight categories according to the degree of LVEF.A common LVEF of 61%-65% was established as the reference group for all mortality comparisons. 2her echocardiography parameters including local ventricular wall motion abnormalities, septal thickness, posterior wall thickness, and left ventricular end diastolic diameter were also obtained.

| Outcomes measurement
Each patient was followed up according to the study protocol by trained research coordinators who were blinded to the baseline clinical information.The primary outcome was all-cause death at 1 year defined as death from any cause and confirmed by a death certification from the attended hospital or the local civil registry.And the secondary outcomes included stroke recurrence and functional dependence at 1 year.Stroke recurrence was defined as new ischemic stroke and recurrent hemorrhagic stroke (intracerebral hemorrhage and subarachnoid hemorrhage).Functional dependence was defined as modified Rankin Scale (mRS) of 3-5.

| Statistical analysis
Continuous variables were presented as mean ± SD or medians (interquartile range).Categorical variables were presented as numbers (percentages).Student's t tests or Mann-Whitney U tests were used for the comparison of continuous variables.χ 2 tests or Fisher's exact tests were used for the comparison of categorical variables.Cox proportional hazards regression was performed to model time-todeath and time-to-stoke recurrence according to LVEF levels, and the results are reported as hazard ratios (HR) with 95% confidence intervals (CI).Multivariate logistic regression was used to analyze the independent associations between LVEF intervals and the functional dependence, and the results are reported as odds ratios (OR) with their 95% CI.In the multivariable analyses, variables including age, sex, height, body mass index, smoking, history of hypertension, history of diabetes, using statins at discharge, using antiplatelets at discharge, using anticoagulants at discharge, using antihypertensives at discharge were adjusted in model 1.In model 2, baseline NIHSS score, white blood cells were further added, to adjust for its potential impact on outcomes.In addition, survival analysis was performed by the Kaplan-Meier method.Survival curves were compared between each LVEF level using log-rank analysis.All statistical analyses were performed using SAS software v 9.4 (SAS Institute Inc).A two-tailed value of p < 0.05 was considered statistically significant.

| Baseline characteristics
Among the 15,166 patients in the CNSR-III, a total of 14,451 AIS or TIA patients were recruited in this study by excluding 516 patients with missing clinical data and 199 patients with no transthoracic echocardiography results.After exclusion of 398 patients with missing 1-year follow-up, a total of 14,053 patients were finally included in this analysis (Figure 1).The mean age of the study population was 62.2 ± 11.2 years, and 68.2% of all the included patients were male.
There was a total of 4202 (29.9%) patients with LVEF≤60%.The demographics and clinical characteristics of included patients are shown among LVEF groups divided by 5% intervals, with the lowest and highest intervals being ≤40% and >70%, respectively (Table 1).
Compared with higher values of LVEF groups, patients with lower LVEF were older and more likely to be male; were more likely to have higher height, hematocrit, WBC, C-reactive protein, blood urea nitrogen, and creatinine level.Lower LVEF patients were more likely to have chronic kidney disease, coronary heart disease, myocardial infarction, atrial fibrillation, valvular heart defect, dilated cardiomyopathy, heart failure, and less likely to have a history of hypertension.Furthermore, patients with lower LVEF had a higher proportion of cardiogenic embolism in TOAST classification, a more severe neurologic deficit on admission, and a higher percentage of receiving anticoagulant agents at discharge, but a lower percentage of receiving antiplatelet agents and statins agents.

LVEF ≤40%
(n = 106) Figure 2).No significant associations were identified in multivariate logistic regression analyses of secondary outcomes (Table 2, model 2).The distribution of mRS at 1 year according to LVEF levels is shown in Figure 3.A significant shift was observed toward poor functional outcomes in patients with LVEF ≤60% (Figure 3).
The baseline characteristics and outcomes according to LVEF above and below 60% were further compared (Table S1).In a propensity-matched analysis, LVEF ≤60% and >60% were matched on age, sex, height, BMI, history of ischemic stroke, baseline NIHSS score, toast classification.A total of 4153 matched pairs were identified.In multivariate logistic regression analyses, LVEF ≤60 was associated with higher risk of all-cause death compared to LVEF >60% (aHR 1.29 [95% CI 1.06-1.58];p = 0.01) after adjusting for all confounders considered in the study (Table S3).

| Subgroup analyses
Results of subgroup analyses of the primary outcome are shown in Table 3.There was no heterogeneity in the effects of LVEF levels on the primary outcome between subgroups classified by age, sex, stroke cause, previous history of ischemic stroke, and baseline NIHSS score.Of note, patients with LVEF ≤60% were significantly associated with the primary outcome in certain subgroups of special interests, including male, patients with non-cardiogenic stroke, baseline NIHSS score ≥4, and those without previous ischemic stroke, but there were no significant interactions.

| DISCUSS ION
In this large, long-term follow-up, prospective registry across China, we investigated the association between routinely echocardiographic assessed LVEF spectrum with 5% width intervals and clinical outcomes in patients with acute ischemic cerebrovascular disease.We found that LVEF ≤60% was associated with a higher risk of all-cause death compared to LVEF >60%, and the risk of all-cause death increased gradually with progressive LVEF reduction.Our study added new data to this research area by accessing this association in the broad LVEF spectrum.These results highlight that LVEF 50-60%, although within the normal range, may still contribute to poor outcomes in acute ischemic cerebrovascular disease.Assessment and monitoring of cardiac function are also important in these populations.
Our current study showed that patients with acute ischemic cerebrovascular disease with lower LVEF were generally older, more likely to have a history of heart diseases, higher NHISS on admission, a higher rate of cardiogenic embolism in TOAST types, a higher percentage of receiving anticoagulant agents at discharge accordingly, which was in line with previous findings. 3,4,10e reported prevalence of decreased LVEF in stroke patients ranges from 5% to 20%, [3][4][5][6] while our registry studies have demonstrated that the number of patients with acute ischemic There are conflicting results regarding the association of cardiac function (with or without LVEF data) and outcomes, including functional disability and stroke recurrence in stroke patients.[3][4][5][6][18][19][20] These inconsistent results may be partly due to the different definitions of heart failure and category of LVEF level.Our study focused on LVEF levels, which could be routinely obtained in clinical practice in patients with acute ischemic cerebrovascular disease.
LVEF grouping methods employed in previous studies have only targeted specific ranges or used a crude dichotomization. 4,21Our study assessed the association of a broad LVEF spectrum with clinical outcomes in a large, long-term follow-up, prospective registry.
We found that in patients with acute ischemic cerebrovascular disease with LVEF ≤60%, LVEF levels independently predicted the 1year risk of all-cause death, and the risk gradually increased with successive decreases in LVEF levels.Our study demonstrates that left ventricular systolic function is more closely related to mortality in patients with acute ischemic cerebrovascular disease, and even an LVEF of 50%-60%, which is generally considered to be normal, still has an increased risk of mortality.Therefore, comprehensive evaluation of cardiac function after AIS or TIA should be strengthened.
Our study added new data to this research area by accessing this association in the broad LVEF spectrum.
The reasons underlying the associations between lower LVEF and poor outcomes in patients with AIS or TIA were unclear and may be multifactorial.The etiology of AIS patients with lower LVEF was more likely to be cardioembolism in the TOAST classification, which may account for the poorer prognosis in this population.
Furthermore, given the close relationship between heart disease and LVEF levels, heart disease may be one of the contributing factors to this association.Patients with lower LVEF were more likely to receive anticoagulation treatment, which was associated with a higher risk of hemorrhagic complications leading to worse outcomes. 22In addition, identical conclusions that AIS patients with lower LVEF have poorer neurological function at admission were obtained in studies, which may be explained by autonomic dysfunction after vascular brain injury. 23Notably, an essential finding of this study was that patients with lower LVEF had higher levels of WBC and C-reactive protein, which were commonly regarded as sensitive indicators of inflammation.It has been found that patients with lower LVEF suffered more severe inflammation burden, 24,25 but this has not been verified in stroke patients.Our findings suggest that ischemic stroke patients with decreased LVEF may experience a greater inflammatory burden than patients with normal LVEF.Growing evidence suggests that the immune F I G U R E 3 Distribution of the modified Rankin Scale scores at 1 year according to LVEF levels.LVEF, left ventricular ejection function.
system plays a complex role in the pathophysiological changes following cerebral ischemia injury. 26Subsequent neuroinflammatory and systemic inflammatory processes following ischemia induce neuronal dysfunction, leading to larger infarct volumes and worse clinical outcomes. 16,26 patients with LVEF ≤60% (LVEF ≤40%: adjusted HR 4.27 [95% CI F I G U R E 1 Flowchart of study population selection.AIS, acute ischemic stroke; CNSR-III, Third China National Stroke Registry; TIA, transient ischemic attack; TTE, Transthoracic echocardiography.| 3521 WEI et al. Kaplan-Meier curves of LVEF level on all-cause death.LVEF, left ventricular ejection function.cerebrovascular disease with reduced LVEF (<50%) is less common (2.45%).Patients with LVEF <50% were reported to represent nearly 20% of the community population but included older patients and a higher proportion of patients with a history of heart disease.2Recent research has shown that several factors can influence the prognosis of stroke including demographic characteristics, 11 time to treatment, 12,13 comorbidities,14,15 blood markers,16 imaging findings,17 and so on.Identifying the factors that influence stroke prognosis helps clinicians predict outcomes, guide treatment decisions, and inform the development of new interventions.Our study provides new insight to this area by finding the association between readily available clinical indicator LVEF and survival in AIS or TIA patients.
which may affect clinical prognosis due to different efficacy and safety.Fourth, the reasons underlying the associations between lower LVEF and poor outcomes in patients with AIS or TIA were unclear.Further studies are needed to clarify this issue.Subgroup analysis of primary outcome according to LVEF levels.a higher risk of all-cause death compared to LVEF >60%, and the risk of all-cause death increased gradually with decreasing LVEF levels.LVEF values of 50-60%, although within the normal range, may still contribute to poor outcomes in acute ischemic cerebrovascular disease.Future research needs to explore its underlying mechanism.
Limitations of this study warrant discussion.First, although this largest sample size of the study revealed an association between LVEF values and stroke outcomes, the number of patients with reduced LVEF was relatively small.This may explain why mortality in the LVEF 56%-60% group did not reach statistical significance.5|CONCLUSIONOur study demonstrated that lower LVEF was associated with older age, cardiac comorbidities, more severe neurological deficits, and higher levels of inflammatory markers.LVEF ≤60% was associatedTA B L E 3Abbreviation: NIHSS, National Institutes of Health Stroke Scale.a Adjusted for age; sex; height; body mass index; smoking; history of hypertension, history of diabetes; using statins at discharge; using antiplatelets at discharge; using anticoagulants at discharge; using antihypertensives at discharge; baseline NIHSS score; white blood cells.with