Journal of the Neurological Sciences 301 (2011) 8689

Contents lists available at ScienceDirect

Journal of the Neurological Sciences

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s

Brain natriuretic peptide is a marker associated with thrombus in stroke patients with atrial brillation
Yoko Okada , Kensaku Shibazaki, Kazumi Kimura, Noriko Matsumoto, Yasuyuki Iguchi, Junya Aoki, Kazuto Kobayashi, Kennichiro Sakai
Department of Stroke Medicine, Kawasaki Medical School, Okayama, Japan

a r t i c l e

i n f o

a b s t r a c t
Background: Patients with atrial brillation (AF) and atrial thrombus are at high risk of thromboembolic events. We investigated whether BNP levels can serve as a biological marker of thrombus. Methods: We prospectively enrolled patients with AF within 7 days of an ischemic stroke and transient ischemic attack (TIA). We measured BNP levels in all patients while they underwent transesophageal echocardiography (TEE) and then assigned them to groups based on the presence (positive group) or absence (negative group) of left atrial thrombus. Factors associated with atrial thrombus were investigated using multivariate logistic regression analysis. Results: Of the 67 (male, n = 40; mean age, 76.5 11.1 years) enrolled patients, 17 (25.4%) had left atrial thrombus. The incidence of hypertension was signicantly higher in the positive, than in the negative group (88.2% vs. 58.0%, p = 0.020). The BNP level was also signicantly higher in the positive, than in the negative group (median (interquartile range) 189.8 (141.4-473.2) vs. 117.9 (70.3-187.1) pg/ml, p = 0.012). The optimal cut-off value, sensitivity, and specicity of BNP levels to distinguish the positive, from the negative group were 140.0 pg/ml, 76.5%, and 62.0%, respectively. Multivariate logistic regression analysis demonstrated that a BNP concentration of N 140.0 pg/ml (odds ratio, 5.62; 95% CI, 1.3922.66, p = 0.015) was an independent factor associated with thrombus. Conclusion: Levels of BNP can serve as a marker of left atrial thrombus in acute ischemic stroke and TIA in patients with AF. 2010 Elsevier B.V. All rights reserved.

Article history: Received 3 February 2010 Received in revised form 14 September 2010 Accepted 19 October 2010 Available online 20 November 2010 Keywords: Brain natriuretic peptide Acute ischemic stroke Atrial brillation Transesopageal echocardiography Emboli detection Stroke biomarkers Anticoagulation Cardioembolic stroke

1. Introduction Atrial brillation (AF) is a common arrhythmia and a major risk factor for ischemic stroke. Thus, patients with ischemic stroke or transient ischemic attacks (TIA) and AF are at high risk of recurrent stroke [1]. Patients with left atrial thrombus are at particularly high risk for thromboembolic events, and anticoagulant agents reduce the likelihood of such events [2,3]. Therefore, to identify left atrial thrombus in acute ischemic stroke is important, and anticoagulant therapy can prevent further brain ischemia. Although transesophageal echocardiography (TEE) is a useful clinical tool for identifying actual thrombi and for visualizing spontaneous echo contrast (SEC) in patients with AF, its semiinvasive nature precludes its application to patients with acute stroke. Brain natriuretic peptide (BNP) is a 32-amino acid polypeptide with a 17-amino acid ring structure that was isolated from the porcine brain in 1988. It is a diuretic and vasodilatory factor that is released
Corresponding author. Department of Geriatric Medicine and Neurology, Ehime University, Shitsukawa, Toon City, Ehime 791-0295, Japan. Tel.: + 81 89 960 5851; fax: + 81 89 960 5852. E-mail address: yokokada@mac.com (Y. Okada). 0022-510X/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2010.10.017

mainly from the ventricular myocardium. A recent study has shown that BNP is a marker of congestive heart failure [4]. Plasma BNP levels are elevated in patients with acute ischemic stroke [5,6], particularly when accompanied by AF [6,7]. A recent report concluded that plasma BNP might be a useful marker of vulnerability to thromboembolism in patients with nonvalvular AF [8]. The present study investigates whether BNP levels could serve as a useful marker of left atrial thrombus during acute ischemic stroke and TIA in patients with AF. 2. Patients and methods Between November 2006 and June 2008, we prospectively enrolled patients with AF within 7 days of onset of acute ischemic stroke and TIA who underwent TEE. Patients were excluded if they did not agree with the TEE examination, or if TEE could not be performed because of severe conditions such as large brain infarction with herniation, respiratory or cardiac failure. Patients with old myocardial infarction (OMI), hypertrophic cardiomyopathy (HCM), and dialysisdependent chronic renal failure were also excluded from the present study because plasma BNP levels are increased under these circumstances [9]. This study followed the principles outlined in the

Y. Okada et al. / Journal of the Neurological Sciences 301 (2011) 8689

87

Declaration of Helsinki and was approved by the Ethics Committee of Kawasaki Medical School Hospital. Atrial brillation was diagnosed from a history of AF, 12-lead electrocardiography (ECG) ndings upon admission, as well as ECG and 24-h Holter ECG ndings during hospitalization. We assigned the patients according to the presence (positive group) or absence (negative group) of left atrial thrombus and measured BNP levels at the time of TEE. All patients also underwent computed tomography or magnetic resonance imaging. We assessed age, gender, prior congestive heart failure, prior cerebral infarction, anticoagulants use before TEE, National Institutes of Health stroke scale (NIHSS) score upon admission [10], functional outcome at hospital discharge using the modied Rankin scale (mRS) [11], and cardiothoracic ratios (CTR) on chest X-rays. We also evaluated the following vascular risk factors: hypertension (dened as use of antihypertensive agents, systolic blood pressure 140 mm Hg or a diastolic blood pressure 90 mm Hg before, or 2 weeks after stroke onset); diabetes mellitus (dened as use of oral hypoglycemic agents or insulin, or fasting blood glucose 126 mg/dl, or glycosylated hemoglobin 6.4%); hyperlipidemia (dened as use of antihyperlipidemic agents or serum cholesterol 220 mg/dl) and smoking habit (dened as a history of cigarette smoking during the preceding 3 months). Blood samples were withdrawn upon admission from all patients to determine baseline values for the main hemostatic variables (leukocyte count, platelet count, high sensitive C-reactive protein (CRP), Prothrombin Time-International Normalized Ratio (PT-INR), D-dimer, thrombin-antithrombin III complex (TAT) and brinogen). 2.1. Echocardiography We performed TEE using an HDI 5000 (Philips Medical Systems, Bothell, WA, USA) with a 4- to 7-MHz wideband multiplane transducer. After local pharyngeal anesthesia with lidocaine jelly and spray, patients were placed in the left lateral position for transducer insertion. The left atrium and the left atrium appendage (LAA) were observed in longitudinal views to detect left atrial thrombus and SEC, the severity of which was semi-quantitatively graded (scored from 0 to 4) [12]. Velocity proles of the LAA were obtained by placing the pulsed Doppler sample volume at 12 cm into the orice of the LAA. The emptying ow velocity signals within each R-R interval were averaged over a minimum of ve cardiac cycles. The area of the LAA was measured in B-mode, short axis views with the aortic valve. The results were recorded on super VHS videotapes and reviewed. 2.2. BNP measurements Whole blood samples were collected from a peripheral vein at the time of TEE into tubes containing ethylenediamine tetraacetic acid and then BNP levels were measured using a uorescent immunochromatographic assay (SHIONOSPOT BNP, Shionogi & Co. Ltd., Osaka, Japan) within 15 min. The normal BNP value at our institution is 18.4 pg/ml and the assay detection limit is 5.9 pg/ml. 2.3. Statistical analysis We compared clinical characteristics including BNP levels between the two groups using the 2 and MannWhitney U tests. Factors associated with BNP levels were examined using the MannWhitney U test and linear regression analysis. We subanalyzed the association between BNP and TEE ndings including LAA ow velocity, SEC grade and LAA area. The optimal cut-off points of each continuous variable to discriminate the positive from the negative group were determined from receiver operating characteristics (ROC) curves. Finally, factors with a probability of b 0.1 on univariate analysis and the optimal level

of plasma BNP were entered into a multivariate analysis to determine adjusted odds ratios. Data were statistically analyzed using Stat View (version 5) and SPSS (version 11) software. Differences were considered statistically signicant at the level of p b 0.05.

3. Results We enrolled 72 patients with AF who underwent TEE for ischemic stroke and TIA. Those with OMI (n = 1), HCM (n = 2) and dialysisdependent chronic renal failure (n = 2) were excluded. Therefore, the study included 67 patients (17 with TIA; 27 females, age, (mean SD) 76.5 11.1 years). The mean SD of the NIHSS score on admission and of the mRS at discharge were 7.7 8.0 and 2.3 1.8, respectively. Seventeen (25.4%) patients had left atrial thrombus. Table 1 shows the baseline characteristics of the two groups. The proportion of hypertension (negative vs. positive: 58.0% vs. 88.2%; p = 0.020) was signicantly higher in the positive, than in the negative group. On the other hand, platelet count (21.2 8.4 vs. 16.3 5.6 10 4 / l, p = 0.022) was signicantly lower in the positive than in the negative group. None of the other variables signicantly differed. 3.1. TEE ndings The interval from stroke onset to TEE did not differ between the two groups (7.9 5.7 vs. 10.8 8.2 days, p = 0.239). Flow velocity in the LAA (28.1 15.7 vs. 14.5 3.2 cm/s, p b 0.001) was signicantly lower in the positive, than in the negative group. On the other hand, the SEC grade (1.9 1.1 vs. 3.3 0.6, p b 0.001) was signicantly higher in the positive, than in the negative group whereas the LAA

Table 1 Clinical characteristics. Negative group n = 50 Age (years) Female gender Hypertension Diabetes mellitus Hyperlipidemia Smoking status Prior congestive heart failure Prior ischemic stroke Anticoagulants use before TEE TIA Cardiothoracic ratio NIHSS on admission mRS at discharge 01 23 45 Death WBC (/l) PLT (10,000/l) CRP (mg/dl) PT-INR 2.0 PT-INR before TEE 2.0 D-dimer (g/ml) TAT (ng/ml) Fibrinogen (mg/dl) BNP (pg/ml) Median (IQR) Mean SD 76.4 11.7 17 (34.0) 29 (58.0) 10 (20.0) 7 (14.0) 26 (52.0) 5 (10.0) 15 (30.0) 36 (72.0) 8 (16.0) 60.2 7.0 7.0 7.8 2.6 (05) 22 (44.0) 12 (24.0) 15 (30.0) 0 (0.0) 6936 2772 21.2 8.4 1.36 3.18 1.21 0.29 2 (4.0) 1.48 0.53 5 (27.8) 2.4 3.0 9.3 11.3 318 117 117.9 (70.3187.1) 146.5 119.0 Positive group n = 17 76.5 9.5 10 (58.8) 15 (88.2) 6 (35.3) 3 (17.6) 7 (41.2) 3 (17.6) 4 (23.5) 15 (83.3) 0 (0.0) 62.3 5.2 9.8 8.3 2.2 (06) 7 (41.2) 4 (23.5) 6 (35.3) 1 (5.9) 6669 2108 16.3 5.6 1.58 5.43 1.36 0.38 2 (11.8) 1.69 0.61 7 (14.0) 3.2 3.9 11.5 11.8 329 70 189.8 (141.4473.2) 307.3 270.6 0.729 0.065 0.020 0.171 0.492 0.313 0.326 0.430 0.124 0.082 0.293 0.235 0.301 0.839 0.969 0.684 0.560 0.960 0.022 0.828 0.116 0.243 0.258 0.369 0.747 0.197 0.221 0.012 p

Data are shown as mean SD or number (%). TIA, transient ischemic attack; WBC, white blood cell; PLT, platelets; CRP, C-reactive protein; TAT, thrombin-antithrombin III complex; BNP, brain natriuretic peptide; NIHSS, National Institutes of Health stroke scale; mRS, modied Rankin scale; IQR, interquartile range.

88 Table 2 TEE ndings. Negative group LA SEC SEC grade LAA out ow (cm/s) LAA area (cm2) 42 (84.0) 1.9 1.1 28.1 15.7 6.17 1.92

Y. Okada et al. / Journal of the Neurological Sciences 301 (2011) 8689 Table 3 Association between BNP and clinical characteristics. Positive group 17 (100.0) 3.3 0.6 14.5 3.2 6.81 2.63 p 0.082 b 0.001 b 0.001 0.384 Age NIHSS on admission mRS at discharge LAA ow velocity (cm/s) p 0.023 0.031 0.005 0.037 r 0.277 0.263 0.340 0.258

Data are shown as means SD or number (%). LA SEC, left atrial spontaneous echo contrast; LAA, left atrium appendage.

NIHSS, National Institutes of Health stroke scale; mRS, modied Rankin scale; LAA, left atrium appendage.

area did not differ between the two groups (6.17 1.92 vs. 6.81 2.63 cm2, p = 0.384; Table 2). 3.2. Levels of BNP The median (interquartile range, IQR) BNP level was 129.4 (76.0 221.4) pg/ml. The value was signicantly higher in the positive, than in the negative group (median (IQR) 189.8 (141.4473.2) vs. 117.9 (70.3187.1) pg/ml, p = 0.012; Fig. 1). The BNP level was positively related to clinical variables such as age (r = 0.277, p = 0.023), NIHSS on admission (r = 0.263, p = 0.031), and mRS at discharge (r = 0.340, p = 0.005). LAA ow velocity (r = 0.258, p = 0.037) was negatively associated with BNP levels. None of the other variables signicantly differed (Table 3). Female, hypertension, platelet count and BNP level were chosen as possible factors associated with left atrial thrombus. The optimal cutoff value to distinguish the positive from the negative group was assessed by analyzing ROC curves. The area under the curve using BNP to predict left atrial thrombus was 0.685 (p = 0.024). At a BNP level of 140.0 pg/ml, sensitivity and specicity were 76.5% and 64.0%, respectively. The cut-off platelet count that identied the positive group with the highest sensitivity and specicity was 15.510,000/l (58.8% and 71.9%, respectively). The ndings of the multivariate logistic regression analysis revealed that a BNP value of N 140.0 pg/ml (odds ratio, 5.62; 95%CI, 1.3922.66, p = 0.015) was an independent factor associated with left atrial thrombus. (Table 4). 4. Discussion We demonstrated that the mean BNP level in the AF patients with left atrial thrombus was signicantly higher than in those without thrombus. A BNP value of N 140 pg/ml was an independent factor of

BNP level (pg/ml)

1000

p = 0.012

800 600 400

left atrial thrombus in our patients with AF accompanied by acute ischemic stroke and TIA. Possible explanations for these ndings are as follows. Firstly, left atrial dysfunction such as low LAA ow velocity and high SEC grade are powerful predictors of left atrial thrombus formation [13,14]. In the present study, BNP levels signicantly and negatively correlated with LAA peak velocity, which is compatible with previous ndings [15,16]. Therefore, BNP might be increased in patients with AF and LAA dysfunction. Secondly, congestive heart failure is also an independent predictor of left atrial thrombus [17] and it frequently complicates ischemic stroke with AF. Therefore, the BNP level (as a marker of congestive heart failure) could also be elevated in patients with ischemic stroke and AF. Shimizu et al. reported that plasma BNP levels are higher in nonvalvular AF patients with, than without left atrial thrombus [6]. Therefore, we believe that high BNP levels in patients with AF accompanied by acute ischemic stroke and TIA are at high risk for left atrial thrombus. Although cardiac thrombus can be identied and spontaneous echo contrast can be visualized by TEE, this procedure is contraindicated for some patients with acute stroke because it is semiinvasive. A simple BNP assay has recently been developed that can determine BNP levels at bedside within about 15 minutes. Our study showed that the BNP level was associated with left atrial thrombus and left atrial dysfunction. Therefore, if a BNP value is N 140 pg/ml, left atrial thrombus should be considered, and such patients should be treated with anticoagulants such as heparin or warfarin as soon as possible to prevent further ischemic events. We demonstrated that BNP was associated with left atrial thrombus in the present study. However, BNP is not likely to be predictive marker of left atrial thrombus because of cut-off levels of BNP to distinguish the positive group from the negative group had low sensitivity and specicity. Not only heart dysfunction but also hypercoagulopathy and hyperviscosity may be associated with left thrombus formation. This study has several limitations. Firstly, not all patients with acute ischemic stroke and transient ischemic attack who are admitted to our hospital undergo TEE, which might have introduced a selection bias. Secondly, we did not assess cardiac function such as ejection fraction and left ventricular end diastolic pressure. Thirdly, the measurement time points differed between BNP and other hemostatic markers.

Table 4 Multivariate logistic regression analysis.

200
Thrombus p Odds ratio 3.09 3.65 1.60 5.62 95%CI 0.8311.44 0.6520.39 0.426.13 1.3922.66

0 Negative Positive
Fig. 1. BNP levels of both groups. The BNP level was signicantly higher in the positive than in the negative group (median (IQR) 189.8 (141.4473.2) vs. 117.9 (70.3187.1) pg/ml, p = 0.012).

Female gender Hypertension PLT b 15.5 10,000 l BNP N 140 pg/ml

0.091 0.140 0.495 0.015

PLT, platelet count; BNP, brain natriuretic peptide.

Y. Okada et al. / Journal of the Neurological Sciences 301 (2011) 8689

89

5. Conclusion Levels of BNP could serve as a useful marker of left atrial thrombus in patients with AF accompanied by acute ischemic stroke and TIA. References
[1] Cardiogenic brain embolism. Cerebral embolism task force. Arch Neurol 1986;43: 7184. [2] Hart RG, Pearce LA, Koudstaal PJ. Transient ischemic attacks in patients with atrial brillation: Implications for secondary prevention: The European atrial brillation trial and Stroke Prevention in Atrial Fibrillation III trial. Stroke 2004;35:94851. [3] Lin HJ, Wolf PA, Kelly-Hayes M, Beiser AS, Kase CS, Benjamin EJ, et al. Stroke severity in atrial brillation. The Framingham study. Stroke 1996;27:17604. [4] Tsutamoto T, Wada A, Maeda K, Hisanaga T, Maeda Y, Fukai D, et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: Prognostic role of plasma brain natriuretic peptide concentration in patients with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:50916. [5] Montaner J, Perea-Gainza M, Delgado P, Ribo M, Chacon P, Rosell A, et al. Etiologic diagnosis of ischemic stroke subtypes with plasma biomarkers. Stroke 2008;39: 22807. [6] Shibazaki K, Kimura K, Iguchi Y, Okada Y, Inoue T. Plasma brain natriuretic peptide can be a biological marker to distinguish cardioembolic stroke from other stroke types in acute ischemic stroke. Intern Med 2009;48:25964. [7] Nakagawa K, Yamaguchi T, Seida M, Yamada S, Imae S, Tanaka Y, et al. Plasma concentrations of brain natriuretic peptide in patients with acute ischemic stroke. Cerebrovasc Dis (Basel) 2005;19:15764.

[8] Shimizu H, Murakami Y, Inoue S, Ohta Y, Nakamura K, Katoh H, et al. High plasma brain natriuretic polypeptide level as a marker of risk for thromboembolism in patients with nonvalvular atrial brillation. Stroke 2002;33:100510. [9] Buckley MG, Sethi D, Markandu ND, Sagnella GA, Singer DR, MacGregor GA. Plasma concentrations and comparisons of brain natriuretic peptide and atrial natriuretic peptide in normal subjects, cardiac transplant recipients and patients with dialysis-independent or dialysis-dependent chronic renal failure. Clin Sci Lond 1992;83:43744. [10] Lyden P, Brott T, Tilley B, Welch KM, Mascha EJ, Levine S, et al. Improved reliability of the NIH stroke scale using video training. NINDS tPA stroke study group. Stroke 1994;25:22206. [11] van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988;19: 6047. [12] Fatkin D, Loupas T, Jacobs N, Feneley MP. Quantication of blood echogenicity: Evaluation of a semiquantitative method of grading spontaneous echo contrast. Ultrasound Med Biol 1995;21:11918. [13] Hart RG, Halperin JL. Atrial brillation and stroke: Concepts and controversies. Stroke 2001;32:8038. [14] Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood ow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994;23:9619. [15] Igarashi Y, Kashimura K, Makiyama Y, Sato T, Ojima K, Aizawa Y. Left atrial appendage dysfunction in chronic nonvalvular atrial brillation is signicantly associated with an elevated level of brain natriuretic peptide and a prothrombotic state. Jpn Circ J 2001;65:78892. [16] Inoue S, Murakami Y, Sano K, Katoh H, Shimada T. Atrium as a source of brain natriuretic polypeptide in patients with atrial brillation. J Card Fail 2000;6:926. [17] Tsai LM, Lin LJ, Teng JK, Chen JH. Prevalence and clinical signicance of left atrial thrombus in nonrheumatic atrial brillation. Int J Cardiol 1997;58:1639.