Prevalence and Clinical Correlation of Left Ventricular Systolic Dysfunction in African Americans with Ischemic Stroke Saravana Devulapalli, MD,* Shariff Dunlap, MD,* Nakia Wilson, MD,† Stephen Cockburn, MD,† Mohankumar Kurukumbi, MD,* Prafulla Mehrotra, MD,‡ and Annapurni Jayam-Trouth, MD*

Background: The goal of the present study was to determine the prevalence of left ventricular systolic dysfunction (LVSD) and associated clinical correlates in African Americans (AA) diagnosed with ischemic stroke (IS). Methods: Retrospective chart analysis was done on all diagnosed AA IS patients between January 2010 and March 2012. Patients with atrial fibrillation were excluded. A total of 147 patients were included in the study. Transthoracic 2-dimensional echocardiography was used to assess left ventricular systolic function, and study groups were categorized as normal, mild, moderate, and severely abnormal, based on the ejection fraction (EF). Available imaging studies were analyzed for data collection. Logistic regression and Pearson chi-square tests were performed. Results: Normal EF was present in 114 of 147 patients (78%). Mild abnormality was present in 9 of 147 (6%), moderate in 8 of 147 (5%), and severe in 16 of 147 (11%) patients. In patients with mildly reduced EF, smoking was the most common (RF). In patients with moderately and severely reduced EFs, hypertension was the most common RF. History of smoking was commonly found in systolic dysfunction group compared with normal group (P 5 .001). Logistic regression analysis revealed that smoking and advanced age were the significant predictors for LVSD. Large-vessel IS were more common in systolic dysfunction group than normal EF group (P 5.017). Conclusions: Prevalence of LVSD in AA with IS was 22% in our study. Smoking was a significant modifiable RF associated with systolic dysfunction. A history of smoking and higher age could predict the occurrence of LVSD. There were more large-vessel IS in patients with LVSD. Key Words: Left ventricular systolic dysfunction—African Americans— ischemic stroke—risk factors—clinical characteristics. Ó 2014 by National Stroke Association

Introduction

From the *Department of Neurology, Howard University Hospital, Washington, District of Columbia; †Howard University College of Medicine, Washington, District of Columbia; and ‡Department of Cardiology, Howard University Hospital, Washington, District of Columbia. Received December 15, 2013; accepted January 27, 2014. Grant support: None declared. Address correspondence to Mohankumar Kurukumbi, MD, Department of Neurology, Howard University Hospital, 2041 Georgia Ave, Washington, DC 20060. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.01.026

Left ventricular systolic dysfunction (LVSD) even without congestive heart failure (CHF) is a known risk factor (RF) for ischemic stroke (IS).1 Embolic strokes mostly originate from the heart with morbidity and mortality depending on the location of the stroke.2 Asymptomatic LVSD has an estimated prevalence of 3%-6% in the general population.3 CHF is associated with a 2- to 3-fold increase in the relative risk of stroke.1 Asymptomatic LVSD is at least twice as common as overt CHF and may be a precursor to symptomatic heart failure.1 African Americans (AA) are in the high-risk category for stroke, but there is a paucity of data regarding LVSD as an RF for IS in this population.4 Stroke mortality

Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2014: pp 1-4

1

S. DEVULAPALLI ET AL.

2

rates are approximately 50% higher in AA than whites, and at age 55, the risk of dying from stroke is 3 times greater for AA than whites.5

Methods Retrospective chart analysis was done on all diagnosed IS patients, admitted to Howard University Hospital, which takes care of a community of primarily AA, from January 2010 to March 2012. Only AA were included in the study. Study was approved by the institutional review board. Clinical RFs were obtained from documented records, and these were consistently reported. Patients with atrial fibrillation (AF) as reported by electrocardiogram were excluded. Transthoracic 2-dimensional echocardiography was used to assess ejection fraction (EF) and was done on all the patients with IS admitted to our institution over the period of the study. Groups were categorized as normal (EF . 50%), mildly abnormal (EF 5 41%-50%), moderately abnormal (EF 5 31%-40%), or severely abnormal (EF , 30%).1,6 IS locations were determined by computerized tomography scan and magnetic resonance (MR) imaging. MR angiography and ultrasound of carotids were used for assessing blood vessel pathology. CT scan of brain was done in all patients. In the vast majority, MR imaging of brain and MR angiography of neck were also available. Most of the imaging studies were done without contrast.

Several patients had more than 1 IS in different locations. Combination strokes were categorized as IS extending in any 2 adjacent locations, for example, frontoparietal, temporo-occipital, and strokes involving the basal ganglia extending to the internal capsule. Whole middle cerebral artery, anterior cerebral artery, posterior cerebral artery territory, and basilar artery territory strokes were categorized as complete vessel strokes. Combination and complete vessel strokes were considered as large-vessel ischemic strokes. We analyzed the strokes with respect to the location, namely right, left, or bilateral. The volume of infarctions was not studied. Statistical analysis was performed to test the significance of our data. Significant predictors for LVSD were determined by doing logistic regression analysis. Pearson chi-square test was performed to obtain P value by using SPSS software.

Results A total of 147 IS patients were identified. Normal EF was present in 114 patients (78%) and abnormal EF in 33 (22%). Of the 33 patients in LVSD group, 22 were men. Mild abnormality was present in 9 (6%), moderate in 8 (5%), and severe in 16 (11%) patients. In the 9 patients with mildly reduced EF, smoking was the most common RF. In addition, stroke was mostly on the right side of the brain, and the most frequent location was the cerebellum. In the 8 patients with moderately reduced EF and the 16 patients with

Table 1. Degree of LVSD and clinical correlates Variables

Mild (n 5 9)

Mod (n 5 8)

Severe (n 5 16)

Significance (P)

Male Female Age (mean): Hypertension Diabetes Smoking Low-density lipoprotein, .70 mg/dL Intracranial stenosis Carotid atherosclerosis On antiplatelets On statins Left side only Right side only Both sides Frontal Parietal Temporal Occipital Thalamus Basal ganglia Combination and complete Cerebellum Brain stem Internal capsule

7 2 61.1 5 (55.6) 3 (33.3) 8 (88.9) 2 (22.2) 1 (11.1) 1 (11.1) 6 (66.7) 6 (66.7) 3 (33.3) 4 (44.4) 2 (22.2) 2 (22.2) 1 (11.1) 1 (11.1) 1 (11.1) 2 (22.2) 3 (33.3) 2 (22.2) 4 (44.4) 1 (11.1) 3 (37.3)

5 3 66 5 (62.5) 2 (25) 4 (50) 2 (25) None None 1 (12.5) 1 (12.5) 2 (25) 2 (25) 4 (50) 2 (25) 1 (12.5) None 1 (12.5) 1 (12.5) 1 (12.5) 4 (50) 2 (25) 1 (12.5) None

10 6 70 13 (81.3) 4 (25) 9 (56.3) 7 (43.8) 1 (6.3) 6 (37.5) 3 (18.8) 4 (25) 6 (37.5) 3 (18.8) 7 (43.8) 5 (31.3) 4 (25) None 1 (6.3) 1 (6.3) 7 (43.8) 7 (43.8) 4 (25) 2 (12.5) 1 (6.3)

.407 .607 .003 .029 .621 .080 .032 .091 .747 .561 .471 .945 .432 .549 .912 .095 .370 .050 .165 .455 .043

PREVALENCE AND CLINICAL CORRELATION OF LEFT VENTRICULAR SYSTOLIC DYSFUNCTION

Table 2. Comparison of non-LVSD and LVSD groups

Variables Male Female Age (mean) Hypertension Diabetes Smoking Low-density lipoprotein, .70 mg/dL Intracranial stenosis Carotid atherosclerosis On antiplatelets On statins Left side only Right side only Both sides Frontal Parietal Temporal Occipital Thalamus Basal ganglia Combination and complete Cerebellum Brain stem Internal capsule

Table 3. Logistic regression analysis of factors contributing to LVSD

Non-LVSD LVSD Significance (n 5 114) (n 5 33) (P) 55 59 62 87 (76.3) 45 (39.5) 37 (32.5) 69 (60.5)

22 11 66.6 23 (69.7) 9 (27.3) 21 (63.6) 11 (33.3)

.440 .200 .001 .006

15 (13.2)

2 (6.1)

.262

42 (36.8)

7 (21.2)

.093

50 (43.9) 40 (35.1) 29 (25.4) 29 (25.4) 56 (49.1) 28 (24.6) 12 (10.5) 8 (7) 8 (7) 38 (33.3) 49 (43.1) 22 (19.3)

10 (30.3) 11 (33.3) 11 (33.3) 9 (27.3) 13 (39.4) 9 (27.3) 6 (18.2) 1 (3) 3 (9.1) 4 (12.1) 11 (33.3) 13 (39.4)

.163 .852 .370 .832 .324 .752 .237 .400 .690 .018 .321 .017

18 (15.8) 29 (25.4) 37 (32.5)

10 (30.3) 4 (12.1) 4 (12.1)

.062 .106 .022

Variables

severely reduced EF, hypertension (HTN) was the most common RF. The stroke location was mostly bilateral in the brain for both groups. The most frequent location was frontoparietal for the moderate group and basal ganglia for the severe group. Carotid atherosclerosis and largevessel IS were more common in the severe LVSD group. The 2 most common RFs for all the 3 groups were HTN and smoking, with smoking being statistically significant. Complete internal carotid artery occlusion was present in 1 patient, and intracranial stenosis was present in 2 patients. Details of patient characteristics in the abnormal EF group are listed in Table 1. Comparison of patient populations with and without LVSD is listed in Table 2. Logistic regression analysis done on the data showed that higher age and a history of smoking are significant clinical associations with LVSD, and the details are listed in Table 3.

Discussion The prevalence of LVSD in AA with IS is not well studied. Increased risk of stroke with decreased left ventricular function has been reported previously.1 An exclusive AA population is unique to our study. This study is larger than the

3

Age Sex Hypertension Diabetes Smoking Low-density lipoprotein, .70 mg/dL

Significance (P)

Odds ratio

95% Confidence interval of odds ratio

.017 .160 .893 .280 .001 .109

1.054 1.997 1.078 .574 5.516 .403

1.009-1.101 .760-5.244 .358-3.244 .210-1.572 1.944-15.650 .133-1.224

Northern Manhattan study, which included 87 AA with IS.1 Furthermore, we are describing the location of stroke and other clinical variables in our population with LVSD. The prevalence of LVSD was 22% (33 of 147 patients) in our population compared with 24% (65 of 270 patients) in the Northern Manhattan study, which included a multiethnic population consisting of 32% AA.1 Although the prevalence of LVSD in the 2 studies appears to be similar, a breakdown showing the actual prevalence in the AA population was not done in the Northern Manhattan study. It is interesting that although lacunar infarctions from small-vessel disease have been reported to be more common in the AA,7 large-vessel ischemic strokes (combination and complete) were more frequent in our study population with LVSD. Interestingly, the temporal and occipital lobes were the least susceptible to IS in all patients. HTN was the most common RF in both the groups; however, smoking was statistically significant in the group with systolic dysfunction. Smoking is a known RF for the development of LVSD in clinical and experimental studies.8 Cigarette smoke includes more than 4000 chemical substances, including nicotine, polycyclic aromatic hydrocarbons, and oxidative gasses, most of which are cardiotoxic.8 Nicotine promotes cardiomyocyte apoptosis by inducing oxidative stress and disrupting apoptosis-related gene expression.8 Even though our data show differences in stroke locations and other clinical variables listed in Tables 1-3 in the comparison groups, most of the results are not statistically significant. This could be secondary to a low sample number, and larger numbers could give more significant results. Modification of RFs, particularly smoking cessation and control of HTN, are important to prevent occurrence of LVSD. LVSD could be secondary to multiple etiologies like coronary artery disease, HTN, diabetes mellitus, alcohol, and cardiomyopathy in our patient population. The likelihood of coronary artery disease in patients with

S. DEVULAPALLI ET AL.

4 9

symptomatic intracranial arterial stenosis is around 50%. We do not know the coronary artery disease burden in our patient population. AF, which is commonly coexistent with CHF, is an RF for stroke.10 Even though we excluded patients presenting with AF in our study, paroxysmal AF cannot be completely ruled out. The mechanism underlying the association between LVSD and IS is not clear although cardioembolism is a well-recognized factor. Ischemic strokes varied from lacunar to large-vessel strokes in our study population. Small-vessel disease secondary to HTN and diabetes could be the most likely etiology for all lacunar strokes in our study including the LVSD population. Largevessel ischemic strokes in the LVSD group could be secondary to embolism and arterial atherosclerosis. Morbidity and mortality from ischemic strokes depend on the location of the stroke. Our study is unique in trying to identify a common location of stroke in LVSD patients. Large-vessel strokes were more common in the LVSD group, and ischemic strokes from small-vessel disease in structures like brainstem, internal capsule, thalamus, and basal ganglia were more common in the non-LVSD group. In both LVSD and non-LVSD groups, there was no predilection for unilateral, single-site stroke occurrence in our study. Age seemed to play a role in that the LVSD occurred in older patients, with an average age of 66, compared with 62 years. The long-term prognosis of patients with IS may be improved with early diagnosis of LVSD and modification of RFs.

Conclusion This is the largest study to observe the prevalence and correlation between the extent of LVSD and clinical characteristics of strokes in AA. The prevalence of LVSD in AA with IS was 22% in our study. HTN was the most common RF for both normal and abnormal EF groups. However, smoking was a statistically significant modifiable RF in patients with LVSD. Patients with LVSD were older than non-LVSD group.

LVSD was more common in men than women in our study. There were more large-vessel ischemic strokes with LVSD. There was no significant correlation between the laterality of stroke and LVSD. There was an association between the location of IS and severity of LVSD, but the sample was small. Further large prospective studies could provide more insight on the prevalence of LVSD in AA and associated factors, which could be predictive of LVSD.

References 1. Hays AG, Sacco RL, Rundek T, et al. Left ventricular systolic dysfunction and the risk of ischemic stroke in a multiethnic population. Stroke 2006;37:1715-1719. 2. Pullicino PM, Halperin JL, Thompson JL. Stroke in patients with heart failure and reduced left ventricular ejection fraction. Neurology 2000;54:288-294. 3. Wang TJ, Levy D, Benjamin EJ, et al. The epidemiology of ‘‘asymptomatic’’ left ventricular systolic dysfunction: implications for screening. Ann Intern Med 2003;138: 907-916. 4. Rosamond WD, Folsom AR, Chambless LE, et al. Stroke incidence and survival among middle-aged adults: 9year follow- up of the Atherosclerosis Risk in Communities (ARIC) cohort. Stroke 1999;30:736-743. 5. Howard G, Labarthe DR, Hu J, et al. Regional differences in African Americans’ high risk for stroke: the remarkable burden of stroke for southern African Americans. Ann Epidemiol 2005;17. 6. Wang TJ, Evans JC, Benjamin EJ, et al. Natural history of asymptomatic left ventricular systolic dysfunction in the community. Circulation 2003;108:977-982. 7. Gorelick PB, Leurgans S, Richardson D, et al, for the AAASPS Investigators. African American Antiplatelet Stroke Prevention Study: clinical trial design. J Stroke Cerebrovasc Dis 1998;7:426-434. 8. Zhou X, Li C, Xu W, et al. Protective effects of valsartan against cigarette smoke induced left ventricular systolic dysfunction in rats. Int J Cardiol 2013;167:677-680. 9. Arenillas JF, Candell-Riera J, Romero-Farina G, et al. Silent myocardial ischemia in patients with symptomatic intracranial atherosclerosis: associated factors. Stroke 2005;36:1201-1206. 10. Lubitz SA, Benjamin EJ, Ellinor PT. Atrial fibrillation in congestive heart failure. Heart Fail Clin 2010;6:187-200.

Prevalence and clinical correlation of left ventricular systolic dysfunction in african americans with ischemic stroke.

The goal of the present study was to determine the prevalence of left ventricular systolic dysfunction (LVSD) and associated clinical correlates in Af...
118KB Sizes 0 Downloads 3 Views