Usefulness of N-Terminal ProeB-Type Natriuretic Peptide Increase as a Marker for Cardiac Arrhythmia in Patients With Syncope

Transcription

1 Usefulness of N-Terminal ProeB-Type Natriuretic Peptide Increase as a Marker for Cardiac Arrhythmia in Patients With Syncope Giorgio Costantino, MD a, *, Monica Solbiati, MD a, Giovanni Casazza, PhD b, Mattia Bonzi, MD a, Tarcisio Vago, MD c, Nicola Montano, MD, PhD a, Daniel McDermott, MD d, James Quinn, MD, MS e, and Raffaello Furlan, MD f B-type natriuretic peptides (BNPs) have been investigated as biomarkers for risk stratification of patients with syncope. Their concentration can be influenced by age and comorbidities. In the present study, we compared the change in N-terminal proeb-type natriuretic peptide (NT-proBNP) levels within 6 hours in patients with vasovagal and arrhythmic syncope to determine whether this change can predict arrhythmic syncope. Using a case-control design, 33 patients were enrolled. Of the 33 patients, 18 with arrhythmic syncope, as they underwent controlled ventricular tachycardia or ventricular fibrillation (VF) during device safety testing of an implantable cardioverter defibrillator implantation or battery replacement, were compared with 15 patients, who during a tilttable test were diagnosed with vasovagal syncope (VS). For each patient, a blood sample for NT-proBNP evaluation was collected at baseline and 6 hours after the episode of ventricular tachycardia, VF, or VS. We calculated the percentage of increase in the 6-hour NTproBNP concentration between the 2 groups using nonparametric techniques. We also calculated the area under a receiver operating characteristic curve with the 95% confidence intervals. The 6-hour change in the NT-proBNP concentrations between patients who had had an episode of ventricular tachycardia or VF and patients with VS was significantly different, with a median increase of 32% in the ventricular tachycardia or VF group versus 5% in the VS group (p <0.01). The area under a receiver operating characteristic curve to predict arrhythmic syncope was 0.8 (95% confidence interval 0.65 to 0.95). In conclusion, the results of the present study suggest that a 6-hour NT-proBNP increase might be able to predict arrhythmic syncope. Future work is needed to confirm these findings in undifferentiated emergency department patients who present with syncope. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;113:98e102) B-type natriuretic peptides (BNPs) have recently been investigated as biomarkers for risk stratification of patients with syncope. To date, only a single absolute plasma concentration of BNP and N-terminal proeb-type natriuretic peptide (NT-proBNP) has been considered. 1e4 However, the timing of this single sampling in relation to the event and the patient s age or co-morbidities have proved to be problematic. 5,6 Although a single absolute value appears limited, measuring a change in the levels over time could have some value. In a recent work, we reported a Dipartimento di Scienze Biomediche e Cliniche L. Sacco, Medicina ad Indirizzo Fisiopatologico, Ospedale L. Sacco, Università degli Studi di Milano, Milano, Italy; b Dipartimento di Scienze Biomediche e Cliniche L. Sacco, Università degli Studi di Milano, Milano, Italy; c Laboratorio di Endocrinologia, Ospedale L. Sacco, Milano, Italy; d Division of Emergency Medicine, University of California, San Francisco, California; e Division of Emergency Medicine, Stanford University School of Medicine, Stanford, California; and f Medicina Interna, Istituto Clinico Humanitas Rozzano, Università degli Studi di Milano, Milano, Italy. Manuscript received July 4, 2013; revised manuscript received and accepted August 19, See page 101 for disclosure information. *Corresponding author: Tel: (þ39) ; fax: (þ39) address: giorgic2@gmail.com (G. Costantino). a significant increase in the plasma concentration of both BNP and NT-proBNP 6 hours after a controlled episode of ventricular tachycardia or ventricular fibrillation (VF). 7 However, in that study, we did not measure the BNPs after a vasovagal event, the most common known cause of syncope. The goal of the present study was to compare the changes in the plasma concentrations of NT-proBNP during a 6-hour period in 2 different experimental models of syncope, vasovagal and arrhythmic syncope. Methods A total of 33 patients were considered, 18 were collected from a previously described 7 experimental model of the arrhythmic syncope as they underwent controlled ventricular tachycardia or VF during device safety testing of an implantable cardioverter defibrillator (ICD). These patients were compared with 15 patients who had had a syncopal or near-syncopal episode during a table-tilt test with pharmacologic challenge. All patients provided written informed consent, and the study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution s human research committee. Our assumption was that VF induced in the cardiac electrophysiology laboratory to test ICD functioning would /13/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved.

2 Arrhythmias and Conduction Disturbances/NT-ProBNP and Syncope 99 Table 1 Demographic and clinical features of the study population Variable Case Group (Arrhythmic) (n ¼ 18) Control Group (Vasovagal) (n ¼ 15) Age (yrs) Gender Male 14 6 Female* 4 9 ICD implantation 9 Battery replacement 9 Implantation indication Ventricular arrhythmias 14 Cardiac arrest 4 Heart failure 6 Coronary artery disease 10 Dilated cardiomyopathy 14 Ischemic 9 Idiopathic 4 Toxic (chemotherapy) 1 Mean ejection fraction (%) Mean VF duration (s) Range of VF (s) 6e30 Co-morbidities Arterial hypertension 16 (89) 12 (80) Diabetes mellitus type 2 5 (28) 7 (47) Atrial fibrillation 5 (28) 4 (27) Coronary artery disease* 16 (89) 4 (27) Parkinson s disease 0 3 (20) Cancer 3 (18) 2 (13) Heart failure* 12 (67) 2 (13) None 1 (6) 2 (13) Syncope 5 (33) Presyncope 10 (67) Data are presented as mean SD, n, or n (%). *p<0.05. be a unique clinical model mimicking the hemodynamic changes occurring during arrhythmic syncope. The ICD is composed of a subcutaneously placed generator with 1 electrodes seated in the cardiac chambers. It is able to sense heart electric activity and, when required, to deliver electric pacing (antibradycardia therapy) or an electric shock (antitachycardia therapy). Therefore, the ICD is generally used to prevent sudden cardiac death. Some investigators have suggested performing tests of pacing, sensing, and cardioversion threshold after implantation to evaluate the functionality of the ICD. Hence, with the patient under sedation, an episode of VF is induced, and the ICD must recognize the arrhythmia and deliver an effective electric shock. 8 This procedure can be performed during both initial implantation and subsequent battery replacement. The tilt-table test is used to diagnose neurally mediated syncope. 9 It is performed using an electrically driven tilt table, which enables maintenance of the patient in a head-up position without lower limb movements. The experimental protocol we adopted had 3 components. At the beginning, the patient was maintained in a clinostatic position for 5 minutes. Then, the table was tilted to 75. After 15 minutes, if significant hypotension or syncope had not occurred, 300 to 400 mg of sublingual nitroglycerin were administered, and the test was continued until the development of syncope or near-syncope or for 10 minutes. During the test, continuous electrocardiographic Figure 1. The NT-proBNP percentage of variation in each subject of the case (arrhythmic) and control (vasovagal) groups. p <0.05 for cases versus controls. Note, NT-proNP increases >60% of baseline values were observed only in those who had undergone controlled ventricular arrhythmias and a >25% increase provided the best sensitivity and specificity. and noninvasive beat-to-beat arterial pressure monitoring were performed. The test was performed in the morning after a period 48 hours from the episode of syncope. The patient had fasted for 8 hours. The examination findings were considered negative if the patient had not developed syncope or nearsyncope. The tilt-table test findings were considered positive whenever the signs and symptoms (e.g., diaphoresis, lightheadedness, nausea, tunnel vision, dizziness, weakness, pallor) of syncope or near-syncope occurred in association with significant hypotension (absence of a peripheral pulse or a pressure reduction of 20 mm Hg and a systolic pressure of 90 mm Hg). For each patient, a blood sample for NT-proBNP evaluation was collected at baseline (i.e., before ICD implantation or the tilt-table test) and 6 hours after the episode of VF or vasovagal syncope (VS). The concentration of NT-proBNP for the arrhythmic model was assessed using the Electro Chemo Luminescence kit (Roche Diagnostics, Indianapolis, Indiana). The concentration of NT-proBNP for the vasovagal model was assessed using the Immulite 2000 kit (Siemens Healthcare Diagnostics, Tarrytown, New York). The NTproBNP variation must be compared with its minimum concentration change that can be considered relevant. This has been defined as the reference change value, which is determined using the analytic variability of the laboratory method and the intraindividual variability of the marker. The reference change value for NT-proBNP amounts to 11%. 5 An internal analysis showed that both the Roche Elecsys and the Siemens Immulite kits have the same analytic variability (4.2%). For every patient, we calculated the relative variation of the 6-hour NT-proBNP concentration compared with baseline. Variations <1% and those with a negative value were considered variations of 0%. Categorical variables are reported as frequencies and percentages. We used nonparametric techniques to compare continuous variables, reported as the median and range, as appropriate. In particular, the Wilcoxon 2-sample test was used to assess the differences in the plasmatic concentration variation of NT-proBNP between the cases (arrhythmic) and controls

3 100 The American Journal of Cardiology ( Table 2 Diagnostic accuracy of N-terminal proeb-type natriuretic peptide (NT-proBNP) variation for different cutoffs to maximize negative and positive likelihood ratios and Youden index Variation* (%) Sensitivity (%) Specificity (%) LRþ LR 8 83 (66e100) 60 (35e85) 2.08 (1.08e4.01) 0.28 (0.09e0.85) (45e88) 87 (69e100) 5 (1.32e18.92) 0.38 (0.19e0.76) (21e67) 93 (81e100) 6.67 (0.94e47.46) 0.60 (0.39e0.92) Data in parentheses are 95% confidence intervals. LR ¼ likelihood ratio. * Percentage of variation in NT-proBNP from baseline (time 0) to 6 hours. Figure 2. ROC curve for NT-proBNP increase in predicting arrhythmic versus vasovagal syncope. (vasovagal). Univariate and multivariate logistic regression analyses were performed to assess the ability of the variation in NT-proBNP to discriminate between cases and controls, adjusting for gender and age. A receiver operating characteristic (ROC) analysis was performed to evaluate the ability of the relative variation in NT-proBNP to detect arrhythmic syncope. The sensitivity and specificity and negative and positive likelihood ratios of the NT-proBNP relative variation, with the 95% confidence intervals, were calculated for some thresholds to maximize the negative and positive likelihood ratios and Youden s index (sensitivity þ specificity 1). The area under the ROC curve with the 95% confidence interval was calculated. A p-value <0.05, 2-tailed, was considered statistically significant. Statistical analyses were performed using the Statistical Analysis Systems statistical software, version 9.2 (SAS Institute, Cary, North Carolina). Results The 2 groups were similar in terms of age, although significantly more women were in the VS group (60% vs 22%, p ¼ 0.03). Both groups had many co-morbidities (Table 1). In the VF group, the baseline values of NT-proBNP ranged from 96 pg/ml to a maximum of pg/ml. At 6 hours after VF, the mean NT-proBNP concentration was pg/ml. The median NT-proBNP concentration at baseline was pg/ ml; at 6 hours, it was pg/ml, and the median percentage of variation was 32%. We did not find any correlation between the duration of the episode of ventricular fibrillation and the NT-proBNP increase. In 4 patients with VF, no increase in NT-proBNP occurred (percentage of variation <11%). No characteristics were found that distinguished these patients from other patients with VF. In the VS group, the NT-proBNP baseline values ranged from <20 to pg/ml. The mean NT-proBNP value at 6 hours was pg/ml. The median value at baseline was 358 pg/ml and at 6 hours was 419 pg/ml, and the median percentage of variation was 5.1%. The 6-hour variation in the NT-proBNP concentration between the patients with an episode of VF and those with vasovagal syncope or presyncope was significantly (p <0.01) different (Figure 1). The logistic regression model confirmed the significant ability of NT-proBNP to discriminate between cases and controls, independent of age and gender (univariate p ¼ 0.025, multivariate p ¼ 0.028). The sensitivity and specificity of the different cutpoints of NT-proBNP change to discriminate vasovagal and arrhythmic syncope are listed in Table 2. The area under the ROC curve for the NT-proBNP variations was 0.8 (95% confidence interval 0.65 to 0.95; Figure 2). Discussion In the present study, we observed a greater increase in the 6-hour NT-proBNP plasma levels in patients with arrhythmic syncope than in those with a vasovagal event induced by gravitational stimulus. In addition, the area under the ROC curve of the variation in NT-proBNP to discriminate between vasovagal and arrhythmic syncope was promising and suggests the increase in NT-proBNP could be used to discriminate arrhythmic from vasovagal syncope. As shown in Figure 1, all the patients with an NTproBNP increase >60% were arrhythmic. This result suggests the possible use of NT-proBNP to recognize patients at high risk of adverse events who might merit hospital admission or strict monitoring. BNPs are released after cardiac myocite stretch and are currently used to discriminate between cardiac and pulmonary causes of acute shortness of breath. 10 Moreover, BNP plasma levels enhance in response to an increase in

4 Arrhythmias and Conduction Disturbances/NT-ProBNP and Syncope 101 ventricular filling pressure; thus, their use has gained interest as prognostic markers of other disorders such as pulmonary embolism, acute coronary syndrome, atrial fibrillation, and ventricular tachyarrhythmias. 11e14 In addition, BNPs were recently proposed as prognostic markers to predict adverse outcomes after syncope. This seems biologically important, because a history of congestive heart failure has consistently been found to be a risk factor for adverse outcomes. 1,2 Pfister et al reported that a single NT-proBNP value greater than the internal 156 pg/ml cutoff was characterized by a sensitivity as great as 90% for cardiac syncope, a clinical condition characterized by a poor prognosis. 2 BNP was also included in a risk scale aimed at identifying high-risk patients referred to the emergency department because of syncope. 1 However, these studies considered only the absolute NT-proBNP concentration. They also did not consider the timing of the blood sampling relative to the event and other factors such as age, gender, and comorbidities. This has led to concerns regarding its value as a 1-time test in the emergency department setting. 6 In our study, we considered the variation of serial measures of NT-proBNP plasma concentration to minimize the confounding role potentially played by diseases already present in the patient s clinical history. The results of our study have suggested that patients with VF, 6 hours after the arrhythmic event showed a greater increase in NT-proBNP plasma levels than those with orthostaticinduced, neurally mediated syncope. This suggests that the pathophysiologic mechanisms resulting in NT-proBNP plasma changes are different in the 2 clinical conditions. The exact pathophysiologic mechanism underlying the rapid increase in BNPs after arrhythmic syncope remains elusive, 15 and myocardial hypoxemia 16 and/or the stretch in ventricular myocites produced by the arrhythmic episode have been hypothesized. Because we previously found no correlation between the VF duration and the increase in BNPs, 7 the most relevant stimulus leading to the NT-proBNP increase is likely to be the myocardial stretch after a period of ventricular stunning by VF. This is in contrast to the minimal or no increase in NT-proNP in the VS group. An important finding of the present study was the high specificity characterizing the NT-proBNP 6-hour increase, which makes the 6-hour change of this biomarker a potential confirmatory index of syncope induced by ventricular arrhythmias. In addition, the results of the area under a ROC curve pointed to the suitability of the 6-hour NT-proBNP plasma increase in predicting arrhythmic syncope. If confirmed in future studies, our work suggests that the 6-hour NT-proBNP plasma change could be used as a valuable quantitative marker to predict previous ventricular arrhythmias in syncope patients and, thus, better stratifying the risk of these subjects. Our 2 populations were heterogeneous, but this was not surprising because the characteristics of patients with VF and VS differ greatly. This led to a wide range in the baseline NT-proBNP values; however, we did not emphasize the baseline levels but, rather, the percentage of change. Furthermore, we did a logistic regression model of gender and age and found the percentage of BNP increase was an independent predictor. In the present study, NT-proBNP was measured at 0 and 6 hours after the index event. However, in clinical practice, a delay often occurs before the patient presents with syncope. One could argue that our baseline NT-proBNP value at time 0 was artificial, making the 6-hour increase of the biomarker impossible to accurately calculate. However, in a previous study, 7 we found that the NT-proBNP increase started at 1 hour and had increased further at 3 hours but had reached a peak with a statistically significant difference only at 6 to 9 hours after the ventricular arrhythmias. Therefore, provided that the patient is seen in the emergency department within 3 hours from syncope, an increase in NTproBNP within the next 6 hours would likely still be meaningful. However, this will require additional testing in the emergency department setting. Finally, our model of arrhythmic syncope was artificially induced under controlled conditions, and the sinus rhythm was restored by an appropriate direct current shock. The latter might be considered a factor potentially confounding the study because of the presence of the direct current shock. However, in patients with atrial fibrillation who underwent electrical cardioversion, the NT-proBNP values were not affected by a direct current shock. 13 Acknowledgment: We acknowledge Sachin Y. Paranjape, BS, for his thoughtful criticisms and help in reviewing the report. We acknowledge Siemens Diagnostics for the kit for NT-proBNP analysis in the control population. Disclosures The authors have no conflicts of interest to disclose. 1. Reed MJ, Newby DE, Coull AJ, Prescott RJ, Jacques KG, Gray AJ. The ROSE (risk stratification of syncope in the emergency department) study. J Am Coll Cardiol 2010;55:713e Pfister R, Diedrichs H, Larbig R, Erdmann E, Schneider CA. NT-pro- BNP for differential diagnosis in patients with syncope. Int J Cardiol 2009;133:51e Reed MJ, Newby DE, Coull AJ, Jacques KG, Prescott RJ, Gray AJ. Role of brain natriuretic peptide (BNP) in risk stratification of adult syncope. Emerg Med J 2007;24:769e Tanimoto K, Yukiiri K, Mizushige K, Takagi Y, Masugata H, Shinomiya K, Hosomi N, Takahashi T, Ohmori K, Kohno M. Usefulness of brain natriuretic peptide as a marker for separating cardiac and noncardiac causes of syncope. Am J Cardiol 2004;93:228e Wu AHB. Serial testing of B-type natriuretic peptide and NTpro-BNP for monitoring therapy of heart failure: the role of biologic variation in the interpretation of results. Am Heart J 2006;152:828e Costantino G, Solbiati M, Pisano G, Furlan R. NT-pro-BNP for differential diagnosis in patients with syncope. Int J Cardiol 2009;137: 298e Costantino G, Solbiati M, Sagone A, Vago T, Pisano G, Barbic F, Dipaola F, Casazza G, Viecca M, Furlan R. Time course of B-type natriuretic peptides changes after ventricular fibrillation: relationships with cardiac syncope. Int J Card 2011;153:333e Di Marco JP. Implantable cardioverter-defibrillators. N Engl J Med 2003;349: Task Force for the Diagnosis and Management of Syncope, European Society of Cardiology (ESC), European Heart Rhythm Association (EHRA), Heart Failure Association (HFA), Heart Rhythm Society (HRS), Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB, Deharo JC, Gajek J, Gjesdal K, Krahn A, Massin M, Pepi M, Pezawas T, Ruiz Granell R, Sarasin F, Ungar A, van Dijk JG, Walma EP, Wieling W. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 2009;30:2631e2671.

5 102 The American Journal of Cardiology ( 10. Maisel AS, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc P, Omland T, Storrow AB, Abraham WT, Wu AH, Clopton P, Steg PG, Westheim A, Knudsen CW, Perez A, Kazanegra R, Herrmann HC, McCullough PA; Breathing Not Properly Multinational Study Investigators. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med 2002;347: 161e Lega JC, Lacasse Y, Lakhal L, Provencher S. Natriuretic peptides and troponins in pulmonary embolism: a meta-analysis. Thorax 2009;64: 869e Heeschen C, Hamm CW, Mitrovic V, Lantelme NH, White HD; Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) Investigators. N-terminal pro-b-type natriuretic peptide levels for dynamic risk stratification of patients with acute coronary syndromes. Circulation 2004;110:3006e Wozakowska-Kapton B. Effect of sinus rhythm restoration on plasma brain natriuretic peptide in patients with atrial fibrillation. Am J Cardiol 2004;93:1555e Sutuvosky I, Katoh T, Ohno T, Honma H, Takayama H, Takano T. Relationship between brain natriuretic peptide, myocardial wall stress, and ventricular arrhythmia severity. Jpn Heart J 2004;45:771e Costantino G, Barbic F, Solbiati M, Dipaola F, Furlan R. The elusive mechanisms of B-type natriuretic peptides rise after ventricular fibrillation. Int J Cardiol 2012;156:247e Arjamaa O, Nikinmaa M. Does hypoxia directly regulate the natriuretic peptide system? Int J Cardiol 2012;154:372.