High free fatty acid level is associated with recurrent stroke in cardioembolic stroke patients Jeong-Yoon Choi, MD Ji-Sun Kim, MD Ji Hyun Kim, MD, PhD Kyungmi Oh, MD, PhD Seong-Beom Koh, MD, PhD Woo-Keun Seo, MD, PhD
Correspondence to Dr. Seo: [email protected]
ABSTRACT
Objective: To determine whether the plasma level of free fatty acid (FFA) could be associated with recurrent stroke in cardioembolic (CE) stroke patients.
Methods: We analyzed data from 669 acute ischemic stroke patients and examined the association between FFA concentration and recurrent stroke in CE stroke patients compared with non-CE stroke patients. Results: The baseline plasma FFA concentration (mEq/L) was approximately 1.5-fold higher in CE stroke patients (1.01 6 0.63) than in non-CE stroke patients (0.72 6 0.51). Multivariate logistic analysis showed that an increased level of FFA was significantly associated with CE stroke (hazard ratio [HR] 2.124, confidence interval [CI] 1.492–3.024). During the mean follow-up period of 25.4 months, a total of 56 (8.4%) patients experienced a stroke recurrence. The recurrence rate did not differ between patients with CE (10.5%) and non-CE (8.0%) stroke (p 5 0.396). In CE stroke patients, an elevated baseline FFA concentration was independently associated with stroke recurrence (HR 2.711, CI 1.056–6.959). However, there was no association between FFA and stroke recurrence in non-CE stroke patients.
Conclusion: In this retrospective registry-based observational study, CE stroke seemed to be associated with elevated plasma level of FFA. In addition, the present study suggested that an elevated FFA concentration could be a useful indicator for predicting recurrent stroke in CE stroke patients. Neurology® 2014;82:1142–1148 GLOSSARY AF 5 atrial fibrillation; CAD 5 coronary artery disease; CE 5 cardioembolic; CRP 5 C-reactive protein; FFA 5 free fatty acid; HR 5 hazard ratio; NIHSS 5 NIH Stroke Scale.
Free fatty acid (FFA) is the preferred energy source for human organs, including the heart, liver, and skeletal muscle.1 FFA enters the circulatory system from lipolysis of triglycerides stored in adipocytes, and the plasma FFA concentration is closely connected with lipid metabolism. Elevated FFA concentration is linked to several risk factors for atherosclerosis, including abdominal obesity,2 arterial hypertension,3 and insulin resistance.4 Therefore, an elevated FFA concentration could be associated with atherosclerotic cardiovascular accidents. In fact, some studies have suggested a linkage between FFA and coronary artery disease (CAD).5,6 Several studies have also demonstrated an association between FFA and arrhythmia.7–11 Moreover, a recent study has shown that elevated plasma FFA concentration is a risk factor for atrial fibrillation (AF).12 As FFA is linked to both atherosclerosis and arrhythmia, FFA may share a link to ischemic stroke attributed to atherosclerosis or arrhythmia. However, the effect of FFA on ischemic stroke is primarily unknown. It was recently reported that elevated FFA concentration was associated with cardioembolic (CE) stroke, but this association was not observed in non-CE stroke.13 AF could potentially act as a mediator between FFA and stroke caused by cardioembolism based on previous observations.12,13 However, a causal relationship between elevated FFA concentration and CE stroke remained to be discovered. To identify the effect of FFA in CE stroke, the prognostic value Editorial, page 1110 From the Department of Neurology (J.-Y.C.), Korea University College of Medicine, Korea University Ansan Hospital, Ansan; Department of Neurology (J.-S.K.), Korea University College of Medicine, Korea University Anam Hospital, Seoul; and Department of Neurology (J.H., K.O., S.-B.K., W.-K.S.), Korea University College of Medicine, Korea University Guro Hospital, Seoul. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article. 1142
© 2014 American Academy of Neurology
of FFA concentration on stroke recurrence based on the mechanisms underlying the stroke should be determined. Therefore, in the present study, we investigated the relationship between FFA and recurrent stroke in CE stroke in comparison with non-CE stroke. METHODS Study population. From January 2008 to May 2010, we recruited patients within 7 days of the onset of acute ischemic stroke using the prospectively collected hospital-based stroke registry (Korea University Stroke Registry–Guro arm). The details of this registry are discussed elsewhere.13,14 Patients were initially evaluated for demographic and clinical stroke risk factors and were treated according to established guidelines for the management of acute stroke and prevention of secondary stroke.15 The exclusion criteria for this study were defined as follows: (1) patients who were diagnosed with a TIA without a relevant lesion; (2) patients who died during initial admission; (3) patients without measurement of baseline FFA levels; and (4) patients lost to follow-up within 30 days of discharge. During the study period, a total of 864 patients with acute ischemic stroke and appropriate radiologic findings were recruited. Of these patients, 20 died at initial admission, 79 had no measurement of their baseline FFA level, and 96 were lost to follow-up. Finally, 669 patients were analyzed (figure). Clinical and laboratory parameters including hypertension, diabetes, lipid profiles, smoking, and body mass index were assessed as in previous studies.13 The FFA plasma concentration was determined with a colorimetric assay (ADIVA 1650; Siemens, Erlangen, Germany)
Figure
Flow diagram for inclusion of the patients
using fasting blood obtained on date or the subsequent day. All stroke subtypes according to the For the analysis, patients were non-CE stroke groups.
either the patient’s admission patients were classified into 5 TOAST stroke classification.16 divided into CE stroke and
Assessment of recurrent stroke. The primary outcome in this study was a recurrent stroke, defined by symptoms correlating with lesions visible upon brain imaging. Recurrent stroke included both ischemic and hemorrhagic stroke. Other major vascular events including CAD and peripheral artery embolisms were not included. The data regarding patient outcome were obtained via regular outpatient visits with the subjects or from family members. In addition, a trained interviewer conducted telephone interviews with patients who were unable to visit the outpatient clinics. Standard protocol approval, registrations, and patient consents. The study protocol was approved and supervised by the Institutional Review Board of Korea University Guro Hospital (KUGH12221). Since this study was a retrospective registry-based observational study, the board permitted the study to be conducted in the absence of patient consent.
Statistical analyses. Data are presented as mean with SD or median with interquartile range for continuous variables and numbers and percentages for categorical variables. For the comparison of baseline characteristics between CE stroke and nonCE stroke patients, a x2 test or Fisher exact test was used for categorical variables, whereas Student t test or Mann-Whitney U test was applied for continuous variables. To determine the baseline characteristics associated with current stroke mechanism (CE or non-CE stroke), logistic regression analysis was applied. Univariate and multivariate analyses by Cox proportional hazard models were performed to determine predictors of stroke recurrence in CE and non-CE stroke patients (criteria for inclusion: p , 0.05; criteria for exclusion: p . 0.10). Results are presented as hazard ratio (HR) and 95% confidence interval. In the Cox proportional hazard regression models, both interactions and main effect analyses were included. Specifically, an equality of magnitude of HR between CE and non-CE stroke was used to analyze main effect. An equality of HR directionalities between stroke groups was tested through interaction analysis. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Inc., Cary, NC), and p values less than 0.05 were regarded as statistically significant. RESULTS Baseline characteristics. The
FFA 5 free fatty acid.
baseline characteristics of the 669 subjects (429 male, age 5 65.13 6 12.53 years) are presented in table 1. Among the study participants, large-artery atherosclerosis (206, 30.8%) was the most common cause of ischemic stroke, followed by undetermined etiology (170, 25.4%), small-vessel occlusion (168, 25.1%), cardioembolism (105, 15.7%), and other-determined etiology (20, 3.0%). Consequently, 564 of the patients were grouped into non-CE stroke. Of the CE stroke patients, all exhibited at least one of the established risk factors for cardioembolism; of these patients, 79 had AF, 13 had patent foramen ovale or atrial septal defects, 10 had valvular heart disease, 9 had acute or recent myocardial infarction, and 1 had congestive heart failure with low ejection fraction. Six patients Neurology 82
April 1, 2014
1143
Table 1
Demographic, laboratory, and clinical characteristics of 669 stroke patients
Sex, male
p Valuea
All (n 5 669)
CE (n 5 105)
Non-CE (n 5 564)
429 (64.1)
62 (59.0)
367 (65.1)
0.237
Age, y
65.13 6 12.53
67.50 6 12.04
64.69 12.58
0.021
Hypertension
410 (62.3)
57 (55.3)
353 (63.6)
0.112
Diabetes
174 (27.3)
19 (18.6)
155 (28.9)
0.032
Smoking
240 (37.3)
25 (24.5)
215 (39.7)
0.004
History of CAD
19 (2.8)
7 (6.7)
12 (2.1)
0.010
Recurrent stroke
56 (8.4)
11 (10.5)
45 (8.0)
0.396
Previous stroke
87 (13.0)
13 (12.4)
74 (13.1)
0.836
0.76 6 0.54
1.01 6 0.63
0.72 6 0.51
(0.66, 0.45–0.92)
(0.84, 0.60–1.43)
(0.64, 0.43–0.88)
Fasting glucose, mg/dL
142.68 6 63.96 (123.0, 103.0–156.0)
140.98 6 50.61 (128.0, 107.5–158.0)
142.99 6 66.18 (122.0, 102.0–155.0)
Total cholesterol, mg/dL
177.36 6 41.14 (174.0, 150.0–201.0)
167.82 6 40.13 (169.0, 141.0–86.0)
179.12 6 41.11 (174.0, 152.0–203.0)
0.011
LDL cholesterol, mg/dL
114.13 6 34.41
106.75 6 35.52
115.51 6 34.06
0.011b
HDL cholesterol, mg/dL
44.30 6 24.15 (42.0, 35.0–50.0)
44.93 6 12.10 (44.0, 36.0–53.5)
44.18 6 25.80 (42.0, 35.0–49.0)
0.101
Free fatty acid, mEq/L
,0.001
0.221
Triglycerides, mg/dL
135.11 6 96.17 (112.0, 76.00–162.0)
101.70 6 61.65 (80.0, 62.5–123.0)
141.33 6 100.12 (119.0, 82.0–170.0)
,0.001
CRP, mg/L
3.90 6 4.08 (1.43, 0.59–4.45)
12.81 6 28.40 (2.99, 0.85–12.05)
6.10 6 19.56 (1.29, 0.57–3.70)
,0.001
Initial NIHSS
3.90 6 4.08 (3.0, 1.0–5.0)
6.47 6 6.25 (4.0, 1.0–10.5)
3.42 6 3.32 (2.0, 1.0–4.0)
,0.001
Warfarin use
110 (16.4)
66 (62.9)
39 (7.0)
,0.001
Antidyslipidemic drugs
464 (69.4)
56 (53.3)
408 (72.3)
,0.001
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale. Values are n (%) or mean 6 SD (median, interquartile range). a p Values were calculated with Mann-Whitney U test for continuous variables or x2 test for categorical variables. b p Value by Student t test.
presented with more than 2 potential risk factors for cardioembolism. The initial NIH Stroke Scale (NIHSS) score was higher in CE stroke patients, and vitamin K antagonists were more frequently prescribed for CE patients than non-CE patients at discharge. The average FFA concentration (mEq/L) for each stroke etiology was 0.73 6 0.46 for large-vessel atherosclerosis, 1.01 6 0.63 for cardioembolism, 0.60 6 0.38 for small-vessel occlusion, 0.84 6 0.69 for other determined etiology, and 0.81 6 0.61 for undetermined etiology. The average FFA concentration was higher in CE stroke patients than in non-CE stroke patients (1.01 6 0.63, 0.84 [0.60–1.43] vs 0.72 6 0.51, 0.64 [0.43– 0.88], p , 0.001). In addition, patients in the CE stroke group were older and frequently had a history of prior CAD and an increased level of C-reactive protein (CRP). In contrast, a higher prevalence of diabetes and a more frequent history of smoking were found in patients with non-CE stroke. Levels of fasting glucose, total cholesterol, low-density lipoprotein cholesterol, and triglycerides were increased in the nonCE stroke group. Furthermore, antidyslipidemic drugs were more frequently prescribed to patients 1144
Neurology 82
April 1, 2014
in the non-CE stroke group. No differences were found in sex, prevalence of hypertension and history of previous stroke, or the level of high-density lipoprotein cholesterol. Baseline FFA level and CE stroke. Logistic regression analysis using the baseline characteristics demonstrated that an increased FFA concentration, the presence of prior CAD, and a higher initial NIHSS score were independently associated with CE stroke. However, only increased levels of triglycerides were associated with non-CE stroke (table 2). Predictors of recurrent stroke in CE stroke. The mean follow-up duration was 25.4 months (median 26.6 months, interquartile range 14.6–35.4 months), during which 56 (8.4%) of the 669 patients experienced a recurrence of stroke, 11 CE stroke (8 ischemic and 3 hemorrhagic) and 45 non-CE stroke (35 ischemic and 10 hemorrhagic). Stroke recurrence was confirmed by appropriate brain imaging scans in all 56 patients. The annual stroke recurrence rate was 4.0%. At the end of the study, the overall rate of recurrence for CE stroke patients did not differ from that of non-CE stroke patients (10.5% vs 8.5%, p 5 0.396). Univariate and multivariate logistic regression
Table 2
Univariate and multivariate logistic regression to prediction of CE infarction All patients (n 5 669) Univariate OR (95% CI)
p
Multivariate OR (95% CI)
Sex, male
1.292 (0.844–1.978)
0.238
Age, y
1.019 (1.001–1.037)
0.036
Hypertension
1.420 (0.933–2.161)
0.102
Diabetes
1.715 (1.010–2.915)
Smoking History of CAD
p
1.000 (0.982–1.019)
0.978
0.046
0.582 (0.326–1.039)
0.067
1.986 (1.229–3.210)
0.005
0.612 (0.359–1.044)
0.071
0.330 (0.129–0.849)
0.021
3.322 (1.121–9.842)
0.030
Previous stroke
0.921 (0.491–1.729)
0.799
Free fatty acid, mEq/L
2.293 (1.644–3.197)
,0.001
2.124 (1.492–3.024)
,0.001
Fasting glucose, mg/dL
0.999 (0.996–1.003)
0.769
Total cholesterol, mg/dL
0.993 (0.987–0.998)
0.010
1.002 (0.989–1.015)
0.757
LDL cholesterol, mg/dL
0.992 (0.86–0.999)
0.017
0.992 (0.977–1.007)
0.280
HDL cholesterol, mg/dL
1.001 (0.994–1.009)
0.771
Triglycerides, mg/dL
0.992 (0.988–0.995)
,0.001
0.995 (0.990–0.999)
0.011
Initial NIHSS
1.156 (1.106–1.208)
,0.001
1.120 (1.068–1.174)
,0.001
CRP, mg/L
1.010 (1.002–1.019)
0.012
1.007 (0.999–1.016)
0.090
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CI 5 confidence interval; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale; OR 5 odds ratio.
analyses showed that an increased FFA concentration and lower total cholesterol were independently associated with stroke recurrences in CE stroke patients (tables 3 and 4). In non-CE stroke patients, no variables were associated with stroke recurrences. In the subgroup analyses with recurrent ischemic stroke patients, FFA concentration showed a trend of association with recurrent ischemic stroke in CE stroke (HR 2.249 [0.808–6.260]) without statistical significance. Different effects of risk factor on stroke recurrence according to the stroke groups. Main effect analysis re-
vealed no variables that showed a difference in HR magnitude based on the stroke mechanisms. In interaction analyses, the plasma FFA concentration showed a different HR directionality between CE and non-CE stroke (p 5 0.022) in univariate analysis, but this was not present in multivariate analysis. DISCUSSION In this study, plasma FFA concentration was associated with stroke caused by CE. The baseline level of FFA in CE stroke patients was approximately 1.5-fold higher than that of non-CE stroke patients. Furthermore, we showed that plasma FFA concentration was independently associated with recurrent stroke only in the CE stroke group. Moreover, we found that FFA may potentially play different roles in stroke recurrence depending on the underlying cause of the stroke.
Several possible explanations exist for the association found between plasma FFA concentration and CE stroke in this study.17 First, FFA may generate arrhythmia. Previously, we demonstrated that the association between FFA and CE stroke might result from the association between FFA and AF.13 Numerous experimental studies have also suggested the arrhythmogenicity of FFA. Studies using rats showed that if the molar ratio of FFA to albumin was sufficiently high, FFA could be directly arrhythmogenic in the absence of ischemia.18 In an ischemic condition, impaired FFA metabolism in heart mitochondria leads to an increase of cytosolic Ca21 influx and Ca21-dependent reperfusion arrhythmias.19 An FFA-induced detergent effect could also explain the generation of arrhythmia.20 Several clinical observations and epidemiologic studies indirectly support this assumption.7–11 The arrhythmogenicity of FFA may have been confirmed recently in a population-based prospective cohort study using 4,175 men, in which an increased risk of AF in subjects with higher FFA concentration was shown at a 10-year follow-up.12 Second, based on the association between FFA and stroke recurrence in CE stroke patients, elevated FFA might have thrombogenicity in patients with CE stroke.17 In support of this idea, animal studies have demonstrated systemic thromboembolism after FFA infusion,21 with activation of factor XII by stearic acids postulated as a possible mechanism.22 Although FFA is mostly bound to albumin under normal conditions, a Neurology 82
April 1, 2014
1145
Table 3
Univariate Cox regression analysis to predict stroke recurrence CE (n 5 105), univariate HR (95% CI)
Non-CE (n 5 564), univariate HR (95% CI)
p Valuea (main effect)
p Valueb (interaction term)
Sex, male
1.839 (0.560–6.036)
0.726 (0.374–1.406)
0.9332
0.1797
Age, y
1.053 (0.989–1.120)
1.008 (0.984–1.032)
0.2589
0.1965
Hypertension
2.266 (0.600–8.552)
0.979 (0.536–1.791)
0.7592
0.2596
Diabetes
1.755 (0.466–6.619)
1.192 (0.625–2.273)
0.5024
0.6070
Smoking
0.587 (0.127–2.721)
1.180 (0.653–2.134)
0.1718
0.4055
History of CAD
1.212 (0.155–9.479)
3.081 (0.746–12.727)
0.3101
0.4643
History of stroke
2.333 (0.617–8.819)
1.523 (0.707–3.279)
0.5477
0.5855
a
Fasting glucose, mg/dL
1.001 (0.997–1.005)
1.008 (1.001–1.016)
0.7654
0.0657
Free fatty acid, mEq/L
3.295 (1.483–7.325)a
1.071 (0.625–1.834)
0.1275
0.0221
a
0.0800
0.1091
a
Total cholesterol, mg/dL
a
0.974 (0.957–0.992)
0.990 (0.982–0.998)
LDL cholesterol, mg/dL
0.983 (0.963–1.003)
0.988 (0.978–0.998)
0.5082
0.6294
HDL cholesterol, mg/dL
0.960 (0.915–1.008)
0.979 (0.951–1.008)
0.3345
0.4979
Triglycerides, mg/dL
0.995 (0.984–1.007)
1.001 (0.998–1.004)
0.1748
0.3786
CRP, mg/dL
1.008 (0.996–1.019)
0.998 (0.978–1.018)
0.6376
0.4004
NIHSS
0.956 (0.852–1.074)
1.055 (0.972–1.145)
0.0987
0.1740
Warfarin use
1.357 (0.359–5.125)
1.604 (0.633–4.069)
0.7739
0.8400
Antidyslipidemic drugs
0.775 (0.412–1.458)
0.957 (0.291–3.140)
0.9416
0.7590
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CI 5 confidence interval; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; HR 5 hazard ratio; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale. a p Values are for cardioembolic stroke group main effects, which test an equality of magnitude of HR between groups. b p Values are for an interaction term between each of the variables and the cardioembolic stroke group, which tests an equality of HR directionality between groups.
small portion of FFA could freely exist in serum.23 Moreover, the concentration of unbound FFA is positively correlated with the total concentration of FFA; thus, systemic thromboembolism may occur more easily when the molar ratio of FFA to albumin is high (FFA/albumin . 2).24 However, we could not exclude the possibility of a confounding effect of AF on FFA and recurring CE stroke, as FFA and AF were found to be significantly associated in a previous study13 and AF was the most common CE risk factor in the present study. Finally, an increase of plasma FFA in CE stroke patients in our studies may reflect a metabolic state of CE stroke attributed to AF. Noticeably, the
Table 4
triglyceride level was negatively associated with CE stroke both in the present study and in previous studies.13 It has been suggested that elevated triglyceride level is a marker for metabolic syndrome in patients with stroke or cardiovascular diseases.25 Increased FFA influx to the liver results in secretion of very low-density lipoprotein, which consequently results in the production of atherogenic triglyceride-rich lipoprotein.26 Therefore, an elevated plasma FFA concentration could be associated with hypertriglyceridemia, obesity, insulin resistance, and cardiovascular disease.27 However, this pathophysiologic mechanism is contrary to our results. In fact, the plasma FFA concentration has an opposite effect on stroke recurrence, based on
Multivariate Cox proportional hazard regression analysis with interactions to predict stroke recurrence
CE (n 5 105), HR (95% CI)
Non-CE (n 5 564), HR (95% CI)
p Valuea (interaction term)
Free fatty acid, mEq/L
2.711 (1.056–6.959)
1.119 (0.666–1.879)
0.1071
Total cholesterol, mg/dL
0.972 (0.948–0.995)
0.996 (0.978–1.015)
0.1103
LDL cholesterol, mg/dL
1.009 (0.983–1.036)
0.992 (0.971–1.014)a
0.3326
Fasting glucose, mg/dL
1.001 (0.993–1.010)
1.001 (0.997–1.005)a
0.8974
Abbreviations: CE 5 cardioembolic; CI 5 confidence interval; HR 5 hazard ratio; LDL 5 low-density lipoprotein. a p Values are for interaction terms between each of the variables and the cardioembolic stroke group, which tests an equality of HR directionality between groups. 1146
Neurology 82
April 1, 2014
the stroke mechanisms, in our univariate analysis (0.0221, interaction p value). To explain our results, it may be useful to place them in the context of AF-induced metabolic changes. AF is known to cause a hypermetabolic condition.28 Animal studies have demonstrated that AF induces an increase in atrial oxygen consumption and coronary flow up to threefold as compared to sinus rhythm.29 In addition, histologic changes in the atria including cellular hypertrophy, myolysis, distortion of mitochondrial shape, and perinuclear glycogen accumulation may reflect the high energy demands of cardiac myocyte under the condition of chronic AF.30 Because FFA, derived from triglyceride through lipolysis, is the primary energy source of cardiac myocyte,31 increased levels of plasma FFA and decreased triglyceride levels may reflect the metabolic consequences of energy demands on AF. The present study has several limitations. First, the study population was relatively small and was only composed of a single ethnic population. To better generalize our data, a long-term prospective cohort study with a larger population may be necessary. Second, serial FFA measurements were not performed. The baseline FFA measurement could have been influenced by several factors, such as cerebral infarction or myocardial ischemia. In addition, several infectious conditions that are commonly associated with stroke could have resulted in elevated levels of plasma FFA through catecholamine-induced responses. Consequently, the exact effect of FFA on CE stroke and stroke recurrence in CE stroke may be obscured. However, we attempted to minimize the effects of other confounding factors by performing group comparisons and multivariate analysis including stroke severity (NIHSS score), ischemic heart condition (prior history of CAD), and markers for inflammation (CRP levels). Finally, we discussed the role of FFA in relation to recurrent ischemic stroke in CE stroke, since there were no laboratory or clinical studies investigating a causal relationship between FFA and intracranial hemorrhage until now. However, the effect of FFA on recurrent ischemic stroke did not reach statistical significance. This might be partially attributed to small number of recurrent strokes. Further studies are needed for clarification of the association. The present study suggests that an elevated FFA concentration could serve as a marker for stroke caused by cardioembolism. In addition, FFA concentration may be predictive for stroke recurrence following a CE stroke. Although FFA could be involved in CE stroke by several mechanisms, each mechanism may play some part in common pathophysiologic cascades. Further experimental studies should focus on determining the pathophysiologic mechanisms of FFA on CE stroke. In addition, therapeutic trials should be
performed to test whether reduction of plasma FFA results in reduced recurrence of CE stroke. AUTHOR CONTRIBUTIONS Jeong-Yoon Choi designed the research and drafted the manuscript. Ji-Sun Kim acquired the data, analyzed the data, and revised the manuscript. Ji Hyun Kim acquired the data, analyzed the data, and made critical revision of the manuscript. Kyungmi Oh acquired the data, interpreted the data, and made critical revision of the manuscript. Seong-Beom Koh acquired the data and made critical revision of the manuscript. Woo-Keun Seo supervised the project, designed the research, performed statistical analysis, and drafted the manuscript.
STUDY FUNDING Supported by a grant from the Korean Stroke Society young investigator’s award (KSS-2009-003) and Korea University Grant (K1032861).
DISCLOSURE The authors report no disclosures relevant to the manuscript. Go to Neurology.org for full disclosures.
Received July 13, 2013. Accepted in final form November 18, 2013. REFERENCES 1. Ebbert JO, Jensen MD. Fat depots, free fatty acids, and dyslipidemia. Nutrients 2013;5:498–508. 2. Poirier P, Giles TD, Bray GA, et al. Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association scientific statement on obesity and heart disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation 2006;113: 898–918. 3. Fagot-Campagna A, Balkau B, Simon D, et al. High free fatty acid concentration: an independent risk factor for hypertension in the Paris prospective study. Int J Epidemiol 1998;27:808–813. 4. Bays H, Mandarino L, DeFronzo RA. Role of the adipocyte, free fatty acids, and ectopic fat in pathogenesis of type 2 diabetes mellitus: peroxisomal proliferator-activated receptor agonists provide a rational therapeutic approach. J Clin Endocrinol Metab 2004;89:463–478. 5. Jouven X, Charles MA, Desnos M, Ducimetière P. Circulating nonesterified fatty acid level as a predictive risk factor for sudden death in the population. Circulation 2001;104:756–761. 6. O’Donoghue M, de Lemos JA, Morrow DA, et al. Prognostic utility of heart-type fatty acid binding protein in patients with acute coronary syndromes. Circulation 2006;114:550–557. 7. Cocco G, Chu D. Rimonabant may induce atrial fibrillation. BMJ 2009;338:b1061. 8. Paolisso G, Gualdiero P, Manzella D, et al. Association of fasting plasma free fatty acid concentration and frequency of ventricular premature complexes in nonischemic noninsulin-dependent diabetic patients. Am J Cardiol 1997; 80:932–937. 9. Pilz S, Scharnagl H, Tiran B, et al. Free fatty acids are independently associated with all-cause and cardiovascular mortality in subjects with coronary artery disease. J Clin Endocrinol Metab 2006;91:2542–2547. 10. Pilz S, Scharnagl H, Tiran B, et al. Elevated plasma free fatty acids predict sudden cardiac death: a 6.85-year follow-up of 3315 patients after coronary angiography. Eur Heart J 2007;28:2763–2769. Neurology 82
April 1, 2014
1147
11.
12.
13.
14.
15.
16.
17. 18.
19.
Tansey MJ, Opie LH. Relation between plasma free fatty acids and arrhythmias within the first twelve hours of acute myocardial infarction. Lancet 1983;2:419–422. Khawaja O, Bartz TM, Ix JH, et al. Plasma free fatty acids and risk of atrial fibrillation (from the Cardiovascular Health Study). Am J Cardiol 2012;110:212–216. Seo WK, Kim J, Kim YH, et al. Elevated free fatty acid is associated with cardioembolic stroke subtype. Can J Neurol Sci 2011;38:874–879. Eun MY, Seo WK, Lee J, et al. Age-dependent predictors for recurrent stroke: the paradoxical role of triglycerides. Eur Neurol 2013;69:171–178. Sacco RL, Adams R, Albers G, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: a statement for healthcare professionals from the American Heart Association/American Stroke Association Council on Stroke: co-sponsored by the Council on Cardiovascular Radiology and Intervention: the American Academy of Neurology affirms the value of this guideline. Stroke 2006;37:577–617. 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;24:35–41. Spence JD. Free fatty acids and stroke from atrial fibrillation. Can J Neurol Sci 2011;38:803. Makiguchi M, Kawaguchi H, Tamura M, Yasuda H. Effect of palmitic acid and fatty acid binding protein on ventricular fibrillation threshold in the perfused rat heart. Cardiovasc Drugs Ther 1991;5:753–761. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343: 155–158.
20. 21.
22.
23. 24. 25.
26.
27. 28.
29.
30.
31.
Oliver MF. Sudden cardiac death: the lost fatty acid hypothesis. QJM 2006;99:701–709. Connor WE, Hoak JC, Warner ED. Massive thrombosis produced by fatty acid infusion. J Clin Invest 1963;42: 860–866. Didisheim P, Mibashan RS. Activation of Hageman factor (Factor Xii) by long-chain saturated fatty acids. Thromb Diath Haemorrh 1963;143:346–353. Richieri GV, Kleinfeld AM. Unbound free fatty acid levels in human serum. J Lipid Res 1995;36:229–240. Hoak JC. Stearic acid, clotting, and thrombosis. Am J Clin Nutr 1994;60:1050S–1053S. Miller M, Stone NJ, Ballantyne C, et al. Triglycerides and cardiovascular disease: a scientific statement from the American heart association. Circulation 2011;123:2292–2333. Ip S, Lichtenstein AH, Chung M, Lau J, Balk EM. Systematic review: association of low-density lipoprotein subfractions with cardiovascular outcomes. Ann Intern Med 2009;150:474–484. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2010;375:181–183. Barth AS, Merk S, Arnoldi E, et al. Reprogramming of the human atrial transcriptome in permanent atrial fibrillation: expression of a ventricular-like genomic signature. Circ Res 2005;96:1022–1029. White CW, Kerber RE, Weiss HR, Marcus ML. The effects of atrial fibrillation on atrial pressure-volume and flow relationships. Circ Res 1982;51:205–215. Thijssen VL, Ausma J, Borgers M. Structural remodelling during chronic atrial fibrillation: act of programmed cell survival. Cardiovasc Res 2001;52:14–24. Pilz S, Marz W. Free fatty acids as a cardiovascular risk factor. Clin Chem Lab Med 2008;46:429–434.
Does Your Salary, Profitability Match Your Peers? The 2013 Neurology Compensation and Productivity Report from the American Academy of Neurology is now available. Compare your salary and practice profitability with your peers in your community and nationally. Based on 2012 data from hundreds of neurologists and neurology practice managers, this is the most recent and reliable information for the neurology profession. Use the Neurology Compensation and Productivity Report to:
• Compare your salary, productivity, and practice characteristics to peers • Evaluate physician and non-physician provider performance compared to your peers • Discover fair-market value based on your subspecialty, region, and practice type • Analyze whether it makes sense to expand your practice • Identify variances in key metrics for use in practice improvements Learn more at www.aan.com/view/report.
1148
Neurology 82
April 1, 2014
High free fatty acid level is associated with recurrent stroke in cardioembolic stroke patients Jeong-Yoon Choi, Ji-Sun Kim, Ji Hyun Kim, et al. Neurology 2014;82;1142-1148 Published Online before print February 28, 2014 DOI 10.1212/WNL.0000000000000264 This information is current as of February 28, 2014 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/82/13/1142.full.html
Supplementary Material
Supplementary material can be found at: http://www.neurology.org/content/suppl/2014/02/28/WNL.0000000000 000264.DC1.html
References
This article cites 31 articles, 15 of which you can access for free at: http://www.neurology.org/content/82/13/1142.full.html##ref-list-1
Citations
This article has been cited by 1 HighWire-hosted articles: http://www.neurology.org/content/82/13/1142.full.html##otherarticles
Subspecialty Collections
This article, along with others on similar topics, appears in the following collection(s): All Cerebrovascular disease/Stroke http://www.neurology.org//cgi/collection/all_cerebrovascular_disease_ stroke Cardiac http://www.neurology.org//cgi/collection/cardiac Embolism http://www.neurology.org//cgi/collection/embolism
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 2014 American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
Correspondence to Dr. Seo: [email protected]
ABSTRACT
Objective: To determine whether the plasma level of free fatty acid (FFA) could be associated with recurrent stroke in cardioembolic (CE) stroke patients.
Methods: We analyzed data from 669 acute ischemic stroke patients and examined the association between FFA concentration and recurrent stroke in CE stroke patients compared with non-CE stroke patients. Results: The baseline plasma FFA concentration (mEq/L) was approximately 1.5-fold higher in CE stroke patients (1.01 6 0.63) than in non-CE stroke patients (0.72 6 0.51). Multivariate logistic analysis showed that an increased level of FFA was significantly associated with CE stroke (hazard ratio [HR] 2.124, confidence interval [CI] 1.492–3.024). During the mean follow-up period of 25.4 months, a total of 56 (8.4%) patients experienced a stroke recurrence. The recurrence rate did not differ between patients with CE (10.5%) and non-CE (8.0%) stroke (p 5 0.396). In CE stroke patients, an elevated baseline FFA concentration was independently associated with stroke recurrence (HR 2.711, CI 1.056–6.959). However, there was no association between FFA and stroke recurrence in non-CE stroke patients.
Conclusion: In this retrospective registry-based observational study, CE stroke seemed to be associated with elevated plasma level of FFA. In addition, the present study suggested that an elevated FFA concentration could be a useful indicator for predicting recurrent stroke in CE stroke patients. Neurology® 2014;82:1142–1148 GLOSSARY AF 5 atrial fibrillation; CAD 5 coronary artery disease; CE 5 cardioembolic; CRP 5 C-reactive protein; FFA 5 free fatty acid; HR 5 hazard ratio; NIHSS 5 NIH Stroke Scale.
Free fatty acid (FFA) is the preferred energy source for human organs, including the heart, liver, and skeletal muscle.1 FFA enters the circulatory system from lipolysis of triglycerides stored in adipocytes, and the plasma FFA concentration is closely connected with lipid metabolism. Elevated FFA concentration is linked to several risk factors for atherosclerosis, including abdominal obesity,2 arterial hypertension,3 and insulin resistance.4 Therefore, an elevated FFA concentration could be associated with atherosclerotic cardiovascular accidents. In fact, some studies have suggested a linkage between FFA and coronary artery disease (CAD).5,6 Several studies have also demonstrated an association between FFA and arrhythmia.7–11 Moreover, a recent study has shown that elevated plasma FFA concentration is a risk factor for atrial fibrillation (AF).12 As FFA is linked to both atherosclerosis and arrhythmia, FFA may share a link to ischemic stroke attributed to atherosclerosis or arrhythmia. However, the effect of FFA on ischemic stroke is primarily unknown. It was recently reported that elevated FFA concentration was associated with cardioembolic (CE) stroke, but this association was not observed in non-CE stroke.13 AF could potentially act as a mediator between FFA and stroke caused by cardioembolism based on previous observations.12,13 However, a causal relationship between elevated FFA concentration and CE stroke remained to be discovered. To identify the effect of FFA in CE stroke, the prognostic value Editorial, page 1110 From the Department of Neurology (J.-Y.C.), Korea University College of Medicine, Korea University Ansan Hospital, Ansan; Department of Neurology (J.-S.K.), Korea University College of Medicine, Korea University Anam Hospital, Seoul; and Department of Neurology (J.H., K.O., S.-B.K., W.-K.S.), Korea University College of Medicine, Korea University Guro Hospital, Seoul. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article. 1142
© 2014 American Academy of Neurology
of FFA concentration on stroke recurrence based on the mechanisms underlying the stroke should be determined. Therefore, in the present study, we investigated the relationship between FFA and recurrent stroke in CE stroke in comparison with non-CE stroke. METHODS Study population. From January 2008 to May 2010, we recruited patients within 7 days of the onset of acute ischemic stroke using the prospectively collected hospital-based stroke registry (Korea University Stroke Registry–Guro arm). The details of this registry are discussed elsewhere.13,14 Patients were initially evaluated for demographic and clinical stroke risk factors and were treated according to established guidelines for the management of acute stroke and prevention of secondary stroke.15 The exclusion criteria for this study were defined as follows: (1) patients who were diagnosed with a TIA without a relevant lesion; (2) patients who died during initial admission; (3) patients without measurement of baseline FFA levels; and (4) patients lost to follow-up within 30 days of discharge. During the study period, a total of 864 patients with acute ischemic stroke and appropriate radiologic findings were recruited. Of these patients, 20 died at initial admission, 79 had no measurement of their baseline FFA level, and 96 were lost to follow-up. Finally, 669 patients were analyzed (figure). Clinical and laboratory parameters including hypertension, diabetes, lipid profiles, smoking, and body mass index were assessed as in previous studies.13 The FFA plasma concentration was determined with a colorimetric assay (ADIVA 1650; Siemens, Erlangen, Germany)
Figure
Flow diagram for inclusion of the patients
using fasting blood obtained on date or the subsequent day. All stroke subtypes according to the For the analysis, patients were non-CE stroke groups.
either the patient’s admission patients were classified into 5 TOAST stroke classification.16 divided into CE stroke and
Assessment of recurrent stroke. The primary outcome in this study was a recurrent stroke, defined by symptoms correlating with lesions visible upon brain imaging. Recurrent stroke included both ischemic and hemorrhagic stroke. Other major vascular events including CAD and peripheral artery embolisms were not included. The data regarding patient outcome were obtained via regular outpatient visits with the subjects or from family members. In addition, a trained interviewer conducted telephone interviews with patients who were unable to visit the outpatient clinics. Standard protocol approval, registrations, and patient consents. The study protocol was approved and supervised by the Institutional Review Board of Korea University Guro Hospital (KUGH12221). Since this study was a retrospective registry-based observational study, the board permitted the study to be conducted in the absence of patient consent.
Statistical analyses. Data are presented as mean with SD or median with interquartile range for continuous variables and numbers and percentages for categorical variables. For the comparison of baseline characteristics between CE stroke and nonCE stroke patients, a x2 test or Fisher exact test was used for categorical variables, whereas Student t test or Mann-Whitney U test was applied for continuous variables. To determine the baseline characteristics associated with current stroke mechanism (CE or non-CE stroke), logistic regression analysis was applied. Univariate and multivariate analyses by Cox proportional hazard models were performed to determine predictors of stroke recurrence in CE and non-CE stroke patients (criteria for inclusion: p , 0.05; criteria for exclusion: p . 0.10). Results are presented as hazard ratio (HR) and 95% confidence interval. In the Cox proportional hazard regression models, both interactions and main effect analyses were included. Specifically, an equality of magnitude of HR between CE and non-CE stroke was used to analyze main effect. An equality of HR directionalities between stroke groups was tested through interaction analysis. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Inc., Cary, NC), and p values less than 0.05 were regarded as statistically significant. RESULTS Baseline characteristics. The
FFA 5 free fatty acid.
baseline characteristics of the 669 subjects (429 male, age 5 65.13 6 12.53 years) are presented in table 1. Among the study participants, large-artery atherosclerosis (206, 30.8%) was the most common cause of ischemic stroke, followed by undetermined etiology (170, 25.4%), small-vessel occlusion (168, 25.1%), cardioembolism (105, 15.7%), and other-determined etiology (20, 3.0%). Consequently, 564 of the patients were grouped into non-CE stroke. Of the CE stroke patients, all exhibited at least one of the established risk factors for cardioembolism; of these patients, 79 had AF, 13 had patent foramen ovale or atrial septal defects, 10 had valvular heart disease, 9 had acute or recent myocardial infarction, and 1 had congestive heart failure with low ejection fraction. Six patients Neurology 82
April 1, 2014
1143
Table 1
Demographic, laboratory, and clinical characteristics of 669 stroke patients
Sex, male
p Valuea
All (n 5 669)
CE (n 5 105)
Non-CE (n 5 564)
429 (64.1)
62 (59.0)
367 (65.1)
0.237
Age, y
65.13 6 12.53
67.50 6 12.04
64.69 12.58
0.021
Hypertension
410 (62.3)
57 (55.3)
353 (63.6)
0.112
Diabetes
174 (27.3)
19 (18.6)
155 (28.9)
0.032
Smoking
240 (37.3)
25 (24.5)
215 (39.7)
0.004
History of CAD
19 (2.8)
7 (6.7)
12 (2.1)
0.010
Recurrent stroke
56 (8.4)
11 (10.5)
45 (8.0)
0.396
Previous stroke
87 (13.0)
13 (12.4)
74 (13.1)
0.836
0.76 6 0.54
1.01 6 0.63
0.72 6 0.51
(0.66, 0.45–0.92)
(0.84, 0.60–1.43)
(0.64, 0.43–0.88)
Fasting glucose, mg/dL
142.68 6 63.96 (123.0, 103.0–156.0)
140.98 6 50.61 (128.0, 107.5–158.0)
142.99 6 66.18 (122.0, 102.0–155.0)
Total cholesterol, mg/dL
177.36 6 41.14 (174.0, 150.0–201.0)
167.82 6 40.13 (169.0, 141.0–86.0)
179.12 6 41.11 (174.0, 152.0–203.0)
0.011
LDL cholesterol, mg/dL
114.13 6 34.41
106.75 6 35.52
115.51 6 34.06
0.011b
HDL cholesterol, mg/dL
44.30 6 24.15 (42.0, 35.0–50.0)
44.93 6 12.10 (44.0, 36.0–53.5)
44.18 6 25.80 (42.0, 35.0–49.0)
0.101
Free fatty acid, mEq/L
,0.001
0.221
Triglycerides, mg/dL
135.11 6 96.17 (112.0, 76.00–162.0)
101.70 6 61.65 (80.0, 62.5–123.0)
141.33 6 100.12 (119.0, 82.0–170.0)
,0.001
CRP, mg/L
3.90 6 4.08 (1.43, 0.59–4.45)
12.81 6 28.40 (2.99, 0.85–12.05)
6.10 6 19.56 (1.29, 0.57–3.70)
,0.001
Initial NIHSS
3.90 6 4.08 (3.0, 1.0–5.0)
6.47 6 6.25 (4.0, 1.0–10.5)
3.42 6 3.32 (2.0, 1.0–4.0)
,0.001
Warfarin use
110 (16.4)
66 (62.9)
39 (7.0)
,0.001
Antidyslipidemic drugs
464 (69.4)
56 (53.3)
408 (72.3)
,0.001
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale. Values are n (%) or mean 6 SD (median, interquartile range). a p Values were calculated with Mann-Whitney U test for continuous variables or x2 test for categorical variables. b p Value by Student t test.
presented with more than 2 potential risk factors for cardioembolism. The initial NIH Stroke Scale (NIHSS) score was higher in CE stroke patients, and vitamin K antagonists were more frequently prescribed for CE patients than non-CE patients at discharge. The average FFA concentration (mEq/L) for each stroke etiology was 0.73 6 0.46 for large-vessel atherosclerosis, 1.01 6 0.63 for cardioembolism, 0.60 6 0.38 for small-vessel occlusion, 0.84 6 0.69 for other determined etiology, and 0.81 6 0.61 for undetermined etiology. The average FFA concentration was higher in CE stroke patients than in non-CE stroke patients (1.01 6 0.63, 0.84 [0.60–1.43] vs 0.72 6 0.51, 0.64 [0.43– 0.88], p , 0.001). In addition, patients in the CE stroke group were older and frequently had a history of prior CAD and an increased level of C-reactive protein (CRP). In contrast, a higher prevalence of diabetes and a more frequent history of smoking were found in patients with non-CE stroke. Levels of fasting glucose, total cholesterol, low-density lipoprotein cholesterol, and triglycerides were increased in the nonCE stroke group. Furthermore, antidyslipidemic drugs were more frequently prescribed to patients 1144
Neurology 82
April 1, 2014
in the non-CE stroke group. No differences were found in sex, prevalence of hypertension and history of previous stroke, or the level of high-density lipoprotein cholesterol. Baseline FFA level and CE stroke. Logistic regression analysis using the baseline characteristics demonstrated that an increased FFA concentration, the presence of prior CAD, and a higher initial NIHSS score were independently associated with CE stroke. However, only increased levels of triglycerides were associated with non-CE stroke (table 2). Predictors of recurrent stroke in CE stroke. The mean follow-up duration was 25.4 months (median 26.6 months, interquartile range 14.6–35.4 months), during which 56 (8.4%) of the 669 patients experienced a recurrence of stroke, 11 CE stroke (8 ischemic and 3 hemorrhagic) and 45 non-CE stroke (35 ischemic and 10 hemorrhagic). Stroke recurrence was confirmed by appropriate brain imaging scans in all 56 patients. The annual stroke recurrence rate was 4.0%. At the end of the study, the overall rate of recurrence for CE stroke patients did not differ from that of non-CE stroke patients (10.5% vs 8.5%, p 5 0.396). Univariate and multivariate logistic regression
Table 2
Univariate and multivariate logistic regression to prediction of CE infarction All patients (n 5 669) Univariate OR (95% CI)
p
Multivariate OR (95% CI)
Sex, male
1.292 (0.844–1.978)
0.238
Age, y
1.019 (1.001–1.037)
0.036
Hypertension
1.420 (0.933–2.161)
0.102
Diabetes
1.715 (1.010–2.915)
Smoking History of CAD
p
1.000 (0.982–1.019)
0.978
0.046
0.582 (0.326–1.039)
0.067
1.986 (1.229–3.210)
0.005
0.612 (0.359–1.044)
0.071
0.330 (0.129–0.849)
0.021
3.322 (1.121–9.842)
0.030
Previous stroke
0.921 (0.491–1.729)
0.799
Free fatty acid, mEq/L
2.293 (1.644–3.197)
,0.001
2.124 (1.492–3.024)
,0.001
Fasting glucose, mg/dL
0.999 (0.996–1.003)
0.769
Total cholesterol, mg/dL
0.993 (0.987–0.998)
0.010
1.002 (0.989–1.015)
0.757
LDL cholesterol, mg/dL
0.992 (0.86–0.999)
0.017
0.992 (0.977–1.007)
0.280
HDL cholesterol, mg/dL
1.001 (0.994–1.009)
0.771
Triglycerides, mg/dL
0.992 (0.988–0.995)
,0.001
0.995 (0.990–0.999)
0.011
Initial NIHSS
1.156 (1.106–1.208)
,0.001
1.120 (1.068–1.174)
,0.001
CRP, mg/L
1.010 (1.002–1.019)
0.012
1.007 (0.999–1.016)
0.090
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CI 5 confidence interval; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale; OR 5 odds ratio.
analyses showed that an increased FFA concentration and lower total cholesterol were independently associated with stroke recurrences in CE stroke patients (tables 3 and 4). In non-CE stroke patients, no variables were associated with stroke recurrences. In the subgroup analyses with recurrent ischemic stroke patients, FFA concentration showed a trend of association with recurrent ischemic stroke in CE stroke (HR 2.249 [0.808–6.260]) without statistical significance. Different effects of risk factor on stroke recurrence according to the stroke groups. Main effect analysis re-
vealed no variables that showed a difference in HR magnitude based on the stroke mechanisms. In interaction analyses, the plasma FFA concentration showed a different HR directionality between CE and non-CE stroke (p 5 0.022) in univariate analysis, but this was not present in multivariate analysis. DISCUSSION In this study, plasma FFA concentration was associated with stroke caused by CE. The baseline level of FFA in CE stroke patients was approximately 1.5-fold higher than that of non-CE stroke patients. Furthermore, we showed that plasma FFA concentration was independently associated with recurrent stroke only in the CE stroke group. Moreover, we found that FFA may potentially play different roles in stroke recurrence depending on the underlying cause of the stroke.
Several possible explanations exist for the association found between plasma FFA concentration and CE stroke in this study.17 First, FFA may generate arrhythmia. Previously, we demonstrated that the association between FFA and CE stroke might result from the association between FFA and AF.13 Numerous experimental studies have also suggested the arrhythmogenicity of FFA. Studies using rats showed that if the molar ratio of FFA to albumin was sufficiently high, FFA could be directly arrhythmogenic in the absence of ischemia.18 In an ischemic condition, impaired FFA metabolism in heart mitochondria leads to an increase of cytosolic Ca21 influx and Ca21-dependent reperfusion arrhythmias.19 An FFA-induced detergent effect could also explain the generation of arrhythmia.20 Several clinical observations and epidemiologic studies indirectly support this assumption.7–11 The arrhythmogenicity of FFA may have been confirmed recently in a population-based prospective cohort study using 4,175 men, in which an increased risk of AF in subjects with higher FFA concentration was shown at a 10-year follow-up.12 Second, based on the association between FFA and stroke recurrence in CE stroke patients, elevated FFA might have thrombogenicity in patients with CE stroke.17 In support of this idea, animal studies have demonstrated systemic thromboembolism after FFA infusion,21 with activation of factor XII by stearic acids postulated as a possible mechanism.22 Although FFA is mostly bound to albumin under normal conditions, a Neurology 82
April 1, 2014
1145
Table 3
Univariate Cox regression analysis to predict stroke recurrence CE (n 5 105), univariate HR (95% CI)
Non-CE (n 5 564), univariate HR (95% CI)
p Valuea (main effect)
p Valueb (interaction term)
Sex, male
1.839 (0.560–6.036)
0.726 (0.374–1.406)
0.9332
0.1797
Age, y
1.053 (0.989–1.120)
1.008 (0.984–1.032)
0.2589
0.1965
Hypertension
2.266 (0.600–8.552)
0.979 (0.536–1.791)
0.7592
0.2596
Diabetes
1.755 (0.466–6.619)
1.192 (0.625–2.273)
0.5024
0.6070
Smoking
0.587 (0.127–2.721)
1.180 (0.653–2.134)
0.1718
0.4055
History of CAD
1.212 (0.155–9.479)
3.081 (0.746–12.727)
0.3101
0.4643
History of stroke
2.333 (0.617–8.819)
1.523 (0.707–3.279)
0.5477
0.5855
a
Fasting glucose, mg/dL
1.001 (0.997–1.005)
1.008 (1.001–1.016)
0.7654
0.0657
Free fatty acid, mEq/L
3.295 (1.483–7.325)a
1.071 (0.625–1.834)
0.1275
0.0221
a
0.0800
0.1091
a
Total cholesterol, mg/dL
a
0.974 (0.957–0.992)
0.990 (0.982–0.998)
LDL cholesterol, mg/dL
0.983 (0.963–1.003)
0.988 (0.978–0.998)
0.5082
0.6294
HDL cholesterol, mg/dL
0.960 (0.915–1.008)
0.979 (0.951–1.008)
0.3345
0.4979
Triglycerides, mg/dL
0.995 (0.984–1.007)
1.001 (0.998–1.004)
0.1748
0.3786
CRP, mg/dL
1.008 (0.996–1.019)
0.998 (0.978–1.018)
0.6376
0.4004
NIHSS
0.956 (0.852–1.074)
1.055 (0.972–1.145)
0.0987
0.1740
Warfarin use
1.357 (0.359–5.125)
1.604 (0.633–4.069)
0.7739
0.8400
Antidyslipidemic drugs
0.775 (0.412–1.458)
0.957 (0.291–3.140)
0.9416
0.7590
Abbreviations: CAD 5 coronary artery disease; CE 5 cardioembolic; CI 5 confidence interval; CRP 5 C-reactive protein; HDL 5 high-density lipoprotein; HR 5 hazard ratio; LDL 5 low-density lipoprotein; NIHSS 5 NIH Stroke Scale. a p Values are for cardioembolic stroke group main effects, which test an equality of magnitude of HR between groups. b p Values are for an interaction term between each of the variables and the cardioembolic stroke group, which tests an equality of HR directionality between groups.
small portion of FFA could freely exist in serum.23 Moreover, the concentration of unbound FFA is positively correlated with the total concentration of FFA; thus, systemic thromboembolism may occur more easily when the molar ratio of FFA to albumin is high (FFA/albumin . 2).24 However, we could not exclude the possibility of a confounding effect of AF on FFA and recurring CE stroke, as FFA and AF were found to be significantly associated in a previous study13 and AF was the most common CE risk factor in the present study. Finally, an increase of plasma FFA in CE stroke patients in our studies may reflect a metabolic state of CE stroke attributed to AF. Noticeably, the
Table 4
triglyceride level was negatively associated with CE stroke both in the present study and in previous studies.13 It has been suggested that elevated triglyceride level is a marker for metabolic syndrome in patients with stroke or cardiovascular diseases.25 Increased FFA influx to the liver results in secretion of very low-density lipoprotein, which consequently results in the production of atherogenic triglyceride-rich lipoprotein.26 Therefore, an elevated plasma FFA concentration could be associated with hypertriglyceridemia, obesity, insulin resistance, and cardiovascular disease.27 However, this pathophysiologic mechanism is contrary to our results. In fact, the plasma FFA concentration has an opposite effect on stroke recurrence, based on
Multivariate Cox proportional hazard regression analysis with interactions to predict stroke recurrence
CE (n 5 105), HR (95% CI)
Non-CE (n 5 564), HR (95% CI)
p Valuea (interaction term)
Free fatty acid, mEq/L
2.711 (1.056–6.959)
1.119 (0.666–1.879)
0.1071
Total cholesterol, mg/dL
0.972 (0.948–0.995)
0.996 (0.978–1.015)
0.1103
LDL cholesterol, mg/dL
1.009 (0.983–1.036)
0.992 (0.971–1.014)a
0.3326
Fasting glucose, mg/dL
1.001 (0.993–1.010)
1.001 (0.997–1.005)a
0.8974
Abbreviations: CE 5 cardioembolic; CI 5 confidence interval; HR 5 hazard ratio; LDL 5 low-density lipoprotein. a p Values are for interaction terms between each of the variables and the cardioembolic stroke group, which tests an equality of HR directionality between groups. 1146
Neurology 82
April 1, 2014
the stroke mechanisms, in our univariate analysis (0.0221, interaction p value). To explain our results, it may be useful to place them in the context of AF-induced metabolic changes. AF is known to cause a hypermetabolic condition.28 Animal studies have demonstrated that AF induces an increase in atrial oxygen consumption and coronary flow up to threefold as compared to sinus rhythm.29 In addition, histologic changes in the atria including cellular hypertrophy, myolysis, distortion of mitochondrial shape, and perinuclear glycogen accumulation may reflect the high energy demands of cardiac myocyte under the condition of chronic AF.30 Because FFA, derived from triglyceride through lipolysis, is the primary energy source of cardiac myocyte,31 increased levels of plasma FFA and decreased triglyceride levels may reflect the metabolic consequences of energy demands on AF. The present study has several limitations. First, the study population was relatively small and was only composed of a single ethnic population. To better generalize our data, a long-term prospective cohort study with a larger population may be necessary. Second, serial FFA measurements were not performed. The baseline FFA measurement could have been influenced by several factors, such as cerebral infarction or myocardial ischemia. In addition, several infectious conditions that are commonly associated with stroke could have resulted in elevated levels of plasma FFA through catecholamine-induced responses. Consequently, the exact effect of FFA on CE stroke and stroke recurrence in CE stroke may be obscured. However, we attempted to minimize the effects of other confounding factors by performing group comparisons and multivariate analysis including stroke severity (NIHSS score), ischemic heart condition (prior history of CAD), and markers for inflammation (CRP levels). Finally, we discussed the role of FFA in relation to recurrent ischemic stroke in CE stroke, since there were no laboratory or clinical studies investigating a causal relationship between FFA and intracranial hemorrhage until now. However, the effect of FFA on recurrent ischemic stroke did not reach statistical significance. This might be partially attributed to small number of recurrent strokes. Further studies are needed for clarification of the association. The present study suggests that an elevated FFA concentration could serve as a marker for stroke caused by cardioembolism. In addition, FFA concentration may be predictive for stroke recurrence following a CE stroke. Although FFA could be involved in CE stroke by several mechanisms, each mechanism may play some part in common pathophysiologic cascades. Further experimental studies should focus on determining the pathophysiologic mechanisms of FFA on CE stroke. In addition, therapeutic trials should be
performed to test whether reduction of plasma FFA results in reduced recurrence of CE stroke. AUTHOR CONTRIBUTIONS Jeong-Yoon Choi designed the research and drafted the manuscript. Ji-Sun Kim acquired the data, analyzed the data, and revised the manuscript. Ji Hyun Kim acquired the data, analyzed the data, and made critical revision of the manuscript. Kyungmi Oh acquired the data, interpreted the data, and made critical revision of the manuscript. Seong-Beom Koh acquired the data and made critical revision of the manuscript. Woo-Keun Seo supervised the project, designed the research, performed statistical analysis, and drafted the manuscript.
STUDY FUNDING Supported by a grant from the Korean Stroke Society young investigator’s award (KSS-2009-003) and Korea University Grant (K1032861).
DISCLOSURE The authors report no disclosures relevant to the manuscript. Go to Neurology.org for full disclosures.
Received July 13, 2013. Accepted in final form November 18, 2013. REFERENCES 1. Ebbert JO, Jensen MD. Fat depots, free fatty acids, and dyslipidemia. Nutrients 2013;5:498–508. 2. Poirier P, Giles TD, Bray GA, et al. Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association scientific statement on obesity and heart disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation 2006;113: 898–918. 3. Fagot-Campagna A, Balkau B, Simon D, et al. High free fatty acid concentration: an independent risk factor for hypertension in the Paris prospective study. Int J Epidemiol 1998;27:808–813. 4. Bays H, Mandarino L, DeFronzo RA. Role of the adipocyte, free fatty acids, and ectopic fat in pathogenesis of type 2 diabetes mellitus: peroxisomal proliferator-activated receptor agonists provide a rational therapeutic approach. J Clin Endocrinol Metab 2004;89:463–478. 5. Jouven X, Charles MA, Desnos M, Ducimetière P. Circulating nonesterified fatty acid level as a predictive risk factor for sudden death in the population. Circulation 2001;104:756–761. 6. O’Donoghue M, de Lemos JA, Morrow DA, et al. Prognostic utility of heart-type fatty acid binding protein in patients with acute coronary syndromes. Circulation 2006;114:550–557. 7. Cocco G, Chu D. Rimonabant may induce atrial fibrillation. BMJ 2009;338:b1061. 8. Paolisso G, Gualdiero P, Manzella D, et al. Association of fasting plasma free fatty acid concentration and frequency of ventricular premature complexes in nonischemic noninsulin-dependent diabetic patients. Am J Cardiol 1997; 80:932–937. 9. Pilz S, Scharnagl H, Tiran B, et al. Free fatty acids are independently associated with all-cause and cardiovascular mortality in subjects with coronary artery disease. J Clin Endocrinol Metab 2006;91:2542–2547. 10. Pilz S, Scharnagl H, Tiran B, et al. Elevated plasma free fatty acids predict sudden cardiac death: a 6.85-year follow-up of 3315 patients after coronary angiography. Eur Heart J 2007;28:2763–2769. Neurology 82
April 1, 2014
1147
11.
12.
13.
14.
15.
16.
17. 18.
19.
Tansey MJ, Opie LH. Relation between plasma free fatty acids and arrhythmias within the first twelve hours of acute myocardial infarction. Lancet 1983;2:419–422. Khawaja O, Bartz TM, Ix JH, et al. Plasma free fatty acids and risk of atrial fibrillation (from the Cardiovascular Health Study). Am J Cardiol 2012;110:212–216. Seo WK, Kim J, Kim YH, et al. Elevated free fatty acid is associated with cardioembolic stroke subtype. Can J Neurol Sci 2011;38:874–879. Eun MY, Seo WK, Lee J, et al. Age-dependent predictors for recurrent stroke: the paradoxical role of triglycerides. Eur Neurol 2013;69:171–178. Sacco RL, Adams R, Albers G, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: a statement for healthcare professionals from the American Heart Association/American Stroke Association Council on Stroke: co-sponsored by the Council on Cardiovascular Radiology and Intervention: the American Academy of Neurology affirms the value of this guideline. Stroke 2006;37:577–617. 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;24:35–41. Spence JD. Free fatty acids and stroke from atrial fibrillation. Can J Neurol Sci 2011;38:803. Makiguchi M, Kawaguchi H, Tamura M, Yasuda H. Effect of palmitic acid and fatty acid binding protein on ventricular fibrillation threshold in the perfused rat heart. Cardiovasc Drugs Ther 1991;5:753–761. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343: 155–158.
20. 21.
22.
23. 24. 25.
26.
27. 28.
29.
30.
31.
Oliver MF. Sudden cardiac death: the lost fatty acid hypothesis. QJM 2006;99:701–709. Connor WE, Hoak JC, Warner ED. Massive thrombosis produced by fatty acid infusion. J Clin Invest 1963;42: 860–866. Didisheim P, Mibashan RS. Activation of Hageman factor (Factor Xii) by long-chain saturated fatty acids. Thromb Diath Haemorrh 1963;143:346–353. Richieri GV, Kleinfeld AM. Unbound free fatty acid levels in human serum. J Lipid Res 1995;36:229–240. Hoak JC. Stearic acid, clotting, and thrombosis. Am J Clin Nutr 1994;60:1050S–1053S. Miller M, Stone NJ, Ballantyne C, et al. Triglycerides and cardiovascular disease: a scientific statement from the American heart association. Circulation 2011;123:2292–2333. Ip S, Lichtenstein AH, Chung M, Lau J, Balk EM. Systematic review: association of low-density lipoprotein subfractions with cardiovascular outcomes. Ann Intern Med 2009;150:474–484. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2010;375:181–183. Barth AS, Merk S, Arnoldi E, et al. Reprogramming of the human atrial transcriptome in permanent atrial fibrillation: expression of a ventricular-like genomic signature. Circ Res 2005;96:1022–1029. White CW, Kerber RE, Weiss HR, Marcus ML. The effects of atrial fibrillation on atrial pressure-volume and flow relationships. Circ Res 1982;51:205–215. Thijssen VL, Ausma J, Borgers M. Structural remodelling during chronic atrial fibrillation: act of programmed cell survival. Cardiovasc Res 2001;52:14–24. Pilz S, Marz W. Free fatty acids as a cardiovascular risk factor. Clin Chem Lab Med 2008;46:429–434.
Does Your Salary, Profitability Match Your Peers? The 2013 Neurology Compensation and Productivity Report from the American Academy of Neurology is now available. Compare your salary and practice profitability with your peers in your community and nationally. Based on 2012 data from hundreds of neurologists and neurology practice managers, this is the most recent and reliable information for the neurology profession. Use the Neurology Compensation and Productivity Report to:
• Compare your salary, productivity, and practice characteristics to peers • Evaluate physician and non-physician provider performance compared to your peers • Discover fair-market value based on your subspecialty, region, and practice type • Analyze whether it makes sense to expand your practice • Identify variances in key metrics for use in practice improvements Learn more at www.aan.com/view/report.
1148
Neurology 82
April 1, 2014
High free fatty acid level is associated with recurrent stroke in cardioembolic stroke patients Jeong-Yoon Choi, Ji-Sun Kim, Ji Hyun Kim, et al. Neurology 2014;82;1142-1148 Published Online before print February 28, 2014 DOI 10.1212/WNL.0000000000000264 This information is current as of February 28, 2014 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/82/13/1142.full.html
Supplementary Material
Supplementary material can be found at: http://www.neurology.org/content/suppl/2014/02/28/WNL.0000000000 000264.DC1.html
References
This article cites 31 articles, 15 of which you can access for free at: http://www.neurology.org/content/82/13/1142.full.html##ref-list-1
Citations
This article has been cited by 1 HighWire-hosted articles: http://www.neurology.org/content/82/13/1142.full.html##otherarticles
Subspecialty Collections
This article, along with others on similar topics, appears in the following collection(s): All Cerebrovascular disease/Stroke http://www.neurology.org//cgi/collection/all_cerebrovascular_disease_ stroke Cardiac http://www.neurology.org//cgi/collection/cardiac Embolism http://www.neurology.org//cgi/collection/embolism
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 2014 American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
