Predictors of Mortality, Rehospitalization for Syncope and Cardiovascular Events in Patients With Cardiovascular Syncope

Transcription

1 Circ J 2017; 81: doi: /circj.CJ ORIGINAL ARTICLE Arrhythmia/Electrophysiology Predictors of Mortality, Rehospitalization for Syncope and Cardiovascular Events in Patients With Cardiovascular Syncope Tatsuya Onuki, PhD; Makoto Shoji, PhD; Yuya Nakamura, MD; Ko Ogawa, MD; Akinori Ochi, MD; Koichiro Inokuchi, MD; Shiro Kawasaki, PhD; Yoshimi Onishi, PhD; Yoshimasa Onuma, MD; Yumi Munetsugu, MD; Miwa Kikuchi, PhD; Hiroyuki Ito, PhD; Yoshino Minoura, PhD; Norikazu Watanabe, PhD; Taro Adachi, PhD; Mitsuharu Kawamura, PhD; Taku Asano, PhD; Kaoru Tanno, PhD; Youichi Kobayashi, PhD Background: Predictors of poor outcomes remain unknown for cardiovascular syncope patients after discharge. Methods and Results: We reviewed the medical records of consecutive patients admitted to hospital with cardiovascular syncope. We then performed Cox stepwise logistic regression analysis to identify significant independent factors for death, rehospitalization for syncope, and cardiovascular events. The study group was 206 patients with cardiovascular syncope. Of them, bradycardia was diagnosed in 50%, tachycardia in 27%, and structural disease in 23%. During a 1-year follow-up period, 18 (8%) and 45 (23%) patients, respectively, were rehospitalized for syncope or a cardiovascular event, and 10 (4%) died. Independent predictors of cardiovascular events were systolic blood pressure <100 mmhg (odds ratio [OR] 3.25; 95%confidence interval [CI] , P=0.006) and implantation of a pacemaker (OR 0.19; 95% CI , P=0.0005) (inverse association). Drug-induced syncope (OR 4.57; 95% CI , P=0.007) was an independent risk factor for rehospitalization. Finally, a history of congestive heart failure (OR 11.0; 95% CI , P=0.0006) and systolic blood pressure <100 mmhg (OR 5.40; 95% CI , P=0.02) were identified as significant independent prognostic factors for death. Conclusions: Drug-induced syncope, hypotension, no indication for a pacemaker, and a history of congestive heart failure are risk factors post-discharge for patients with cardiovascular syncope and careful follow-up of these patients for at least 1 year is recommended. Key Words: Cardiovascular syncope; Predictors; Outcomes Syncope is defined as a sudden loss of consciousness associated with loss of postural tone and spontaneous recovery. Syncope is common in Western countries, accounting for 1 6% of medical admissions and up to 3% of emergency department visits. 1 There are various causes of syncope and the prognosis differs according to cause. 1,2 Although prognosis has been shown to be similar between patients with reflex syncope and controls without syncope, it has also been shown to be worse among patients with cardiovascular syncope. 2 In addition, it is well known that cardiovascular syncope is associated with a high risk of recurrent syncope, 2 indicating that patients with cardiovascular syncope should receive careful follow-up after starting relevant therapy. However, although many factors have been identified for risk stratification of patients with syncope, 3 predictors of risk after discharge are poorly understood for this specific patient group. If these risks can be clarified, clinicians could provide more targeted follow-up. Thus, in this study, we aimed to identify the predictors of death, rehospitalization for syncope and cardiovascular events after discharge of patients with cardiovascular syncope. Methods We retrospectively reviewed the medical records of consecutive patients admitted to hospital for syncope between January 1, 2009 and December 31, Informed consent Received December 18, 2016; revised manuscript received April 11, 2017; accepted April 18, 2017; released online May 25, 2017 Time for primary review: 28 days Division of Cardiology, Department of Medicine, Showa University, Tokyo (T.O., M.S., Y.N., K.O., A.O., K.I., S.K., Y. Onishi, Y. Onuma, Y. Munetsugu, H.I., Y. Minoura, N.W., T. Adachi, M. Kawamura, Y.K.); Cardiovascular Center, Showa University Koto Toyosu Hospital, Tokyo (M. Kikuchi, K.T.); and Division of Cardiology, Department of Medicine, Showa University Fujigaoka Hospital, Yokohama (T. Asano), Japan Mailing address: Tatsuya Onuki, PhD, Division of Cardiology, Department of Medicine, Showa University, Hatanodai, Shinagawa-ku, Tokyo , Japan. onutatu@med.showa-u.ac.jp ISSN All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 1396 ONUKI T et al. Table 1. (A) Patients Characteristics, (B) Risk Prediction Score in 206 Patients With Cardiovascular Syncope, (C) Incidence of Cardiovascular Events, Rehospitalization for Syncope, and Death After 1 Year A Total patients n 206 Baseline characteristics Mean age (years) 69±15 Age >80 years 50 (24%) Men 136 (66%) Hypertension 115 (56%) Diabetes mellitus 45 (22%) Smoking 77 (37%) Malignancy 37 (17%) Stroke 13 (6%) sbp <100 mmhg 31 (15%) First syncope 114 (55%) EF <50% 33 (16%) History of CHF 38 (18%) Abnormal ECG 139 (67%) Coronary disease 59 (29%) Non-IHD 34 (17%) Pacemaker 67 (33%) Implantable cardioverter defibrillator 32 (16%) Catheter therapy 44 (21%) Drug-induced syncope 30 (15%) Injury 39 (19%) Circumstances of syncope During effort 62 (30%) While supine 24 (11%) After moving from supine to standing 21 (10%) position During urination or defecation 14 (7%) Drinking 8 (4%) While taking a bath 9 (4%) While driving 7 (3%) Characteristics of syncope Prodromal symptoms 119 (58%) Palpitation 23 (11%) Chest pain/chest uncomfortable 22 (11%) Vertigo 26 (13%) Asthenia 4 (2%) Diaphoresis 4 (2%) Blurred vision 18 (9%) Nausea 13 (6%) Dyspnea 6 (3%) Etiology of syncope Sick sinus syndrome 63 (31%) Atrioventricular block 34 (17%) Bradycardic atrial fibrillation 5 (2%) Supraventricular tachycardia 19 (9%) Ventricular tachycardia 37 (18%) Acute coronary syndrome 16 (8%) Vasospastic spastic angina 17 (8%) Valvular disease 5 (2%) Hypertrophic obstructive cardiomyopathy 6 (3%) Pulmonary thromboembolism 3 (1%) Other 1 (0.4%) (Table 1 continued the next column.) A Total patients Medications Antiplatelet drug 78 (38%) Anticoagulant 68 (33%) β-blocker 68 (33%) β-blocker low dose 43 (21%) β-blocker normal or high dose 33 (16%) α-blocker 14 (7%) ACEI/ARB 98 (48%) Diuretic 60 (29%) Loop diuretic 32 (16%) Potassium-sparing diuretic 19 (9%) Thiazide diuretic 6 (3%) Calcium-channel blocker 72 (35%) Coronary vasodilator 41 (20%) Antidepressant 39 (19%) Hypnotic 6 (3%) Statin 69 (33%) Antiarrhythmic drug 41 (20%) B OESIL score n 0 12 (6%) 1 42 (20%) 2 80 (39%) 3 52 (25%) 4 20 (10%) C Outcome n Cardiovascular event 45 (22%) Rehospitalization for syncope 18 (9%) Death 10 (5%) *Electrocardiogram (ECG) was considered abnormal if there were rhythm abnormalities, atrioventricular or bifascicular block or other intraventricular conduction disorders, left or right ventricular hypertrophy, Brugada-type ECG, long QT interval, old myocardial infarction, or ST-segment and T-wave abnormalities consistent with, or possibly related, to myocardial ischemia. ACEI, angiotensinconverting enzyme inhibitor; ARB, angiotensin receptor blocker; CHF, congestive heart failure; EF, ejection fraction; IHD, ischemic heart disease; OESIL, Osservatorio Epidemiologico sulla Sincope nel Lazio; sbp, systolic blood pressure. was given by the patients and the study was approved by the Ethics Committee of Showa University, Japan. We excluded patients with non-cardiovascular syncope caused by either reflex syncope or orthostatic hypotension and also those with unexplained syncope that could not be diagnosed during admission. We included all patients admitted for syncope as a result of cardiovascular disease. The medical records of participants were reviewed for the etiology of cardiac syncope, clinical history, including activity at the time of syncope, situation and characteristics of syncope, comorbidities (e.g., hypertension, diabetes mellitus, dyslipidemia, prior stroke/transient ischemic attack, congestive heart failure (CHF), ischemic heart disease (IHD), and non-ihd), syncope-related therapy (pacemaker, implantable cardioverter defibrillator, catheter intervention), druginduced syncope (cases of drugs causing cardiovascular syncope, e.g., antiarrhythmic drugs induce bradycardia and tachycardia arrhythmias, and diuretics or herbal medicine can cause electrolyte disturbances that induce arrhythmic syncope), other medications, cardiac and neurological exam-

3 Risk Predictors in CV Syncope Patients 1397 Figure 1. Etiology of cardiac syncope in study participants. ACS, acute coronary syndrome; af, atrial fibrillation; AVB, atrioventricular block; HOCM, hypertrophic obstructive cardiomyopathy; PTE, pulmonary thromboembolism; SSS, sick sinus syndrome; SVT, supraventricular tachycardia; VSA, vasospastic angina; VT, ventricular tachycardia. inations, blood pressure (BP) on consultation, body mass index, smoking, and ECG results. Patients were diagnosed by 2 specialists in cardiac medicine, using the 2009 guidelines for the diagnosis and management of syncope produced by the European Society of Cardiology. 3 Diagnoses were sorted into 3 categories: reflex syncope, orthostatic hypotension-related syncope, and cardiovascular syncope. Reflex syncope was defined as syncope triggered by pain, micturition or defecation, prolonged standing, stress, or as syncope induced by a head-up tilt test with a prodrome similar to clinical symptoms. Syncope caused by orthostatic hypotension included not only general initial, classic, and delayed orthostatic hypotension but also orthostatic hypotension with reflex syncope and postural orthostatic tachycardia syndrome. Orthostatic hypotension was diagnosed on the basis of an abnormal decrease in BP during an active standing test, or head-up tilt test. Cardiovascular syncope included that caused by vascular disease, arrhythmias, IHD, and mechanical cardiovascular causes, including aortic valve stenosis and hypertrophic obstructive cardiomyopathy (HOCM). Arrhythmic syncope was diagnosed on the basis of documented arrhythmic events or a positive electrophysiological study, while IHD was diagnosed by ECG and coronary angiography, including an acetylcholine stress test. Organic cardiac causes of syncope (i.e., aortic valve stenosis and HOCM) were confirmed by echocardiography. Vascular syncope was diagnosed by computed tomography, echocardiography, or pulmonary pressure obtained using a Swan-Ganz catheter. We reviewed the rates of mortality, rehospitalization for syncope and cardiovascular events after discharge among patients with cardiovascular syncope. Cardiovascular events were defined as those requiring hospital admission and were limited to CHF, ischemic attacks (including vasospastic angina), arrhythmias, hemorrhage, and cerebrovascular accidents. Follow-up data after discharge were obtained from outpatient visits or via telephone follow-up. We excluded patients who could not be followed up after 1 year as well as those who died during admission. For risk stratification, the Osservatorio Epidemiologico sulla Sincope nel Lazio (OESIL) risk score 4 was calculated for all patients. The OESIL score included the following 4 characteristics for adverse outcome: age >65years, clinical history of cardiovascular disease (CHF, peripheral arterial disease, or stroke or transient ischemic attack), syncope without prodromal symptoms, and abnormal ECG findings (atrioventricular or intraventricular conduction disorders, left or right ventricular hypertrophy, left-axis deviation, past myocardial infarction, or ST-segment and T-wave abnormalities consistent with or possibly related to myocardial ischemia). Statistical Analysis Data are reported as mean ± standard deviations. Continuous and categorical variables were compared using Mann- Whitney U-test, or chi-squared test, as appropriate. Cox stepwise logistic regression analysis was performed to identify significant independent prognostic factors for death, rehospitalization for syncope, and cardiovascular events, and the odds ratios (ORs) and 95% confidence intervals (CIs) are presented. The time from discharge was analyzed using Kaplan-Meier curves and compared using log-rank test, with P<0.05 considered statistically significant. JMP software version 13.0 was used for the analysis (SAS, Cary, NC, USA). Results We reviewed the medical records of 407 consecutive patients, of whom 186 were excluded because of diagnoses of possible non-cardiac or unexplained syncope. Of the 221 patients with cardiovascular syncope, 5 (2%) were excluded during follow-up and 10 (5%) were excluded because they died during admission. The remaining 206 patients (93%) were enrolled for further study and their clinical characteristics are shown in Table 1A. The mean age of participants was 69±15 years (range years) and 136 (66%) were men. Comorbidities included hypertension (n=115; 56%), diabetes mellitus (n=45; 22%), coronary disease (n=59; 29%), and history of CHF (n=38; 18%), and 99 patients (48%) had a pacing device (implantable cardioverter defibrillator or pacemaker). Major

4 1398 ONUKI T et al. Table 2. (A) Baseline Characteristics and Comparison of Patients With and Without Cardiovascular Events, Rehospitalization for Syncope and Death, (B) Circumstances, Characteristics and Etiology of Syncope and Comparison of Patients With and Without Cardiovascular Events, Rehospitalization for Syncope and Death, (C) Medications and Comparison of Patients With and Without Cardiovascular Events, Rehospitalization for Syncope and Death A Total Cardiovascular event Rehospitalization for syncope Death patients ( ) (+) P value ( ) (+) P value ( ) (+) P value n Baseline characteristics Age >80 years 50 (24%) 42 (26%) 8 (18%) NS 48 (26%) 2 (11%) NS 46 (23%) 4 (40%) NS Male 136 (66%) 109 (68%) 27 (60%) NS 127 (68%) 9 (50%) NS 130 (66%) 6 (60%) NS Hypertension 115 (56%) 93 (58%) 22 (49%) NS 107 (57%) 8 (44%) NS 112 (57%) 3 (30%) NS Diabetes mellitus 45 (22%) 35 (22%) 10 (22%) NS 41 (22%) 4 (22%) NS 43 (22%) 2 (20%) NS Smoking 77 (37%) 61 (38%) 16 (36%) NS 71 (38%) 6 (33%) NS 75 (38%) 2 (20%) NS Malignancy 37 (17%) 32 (20%) 5 (11%) NS 34 (18%) 3 (17%) NS 34 (17%) 3 (30%) NS Stroke 13 (6%) 8 (5%) 5 (11%) NS 12 (6%) 1 (6%) NS 12 (7%) 1 (10%) NS sbp <100 mmhg 31 (15%) 16 (10%) 15 (33%) (14%) 4 (22%) NS 26 (13%) 5 (50%) First syncope 114 (55%) 90 (56%) 24 (53%) NS 107 (58%) 7 (39%) NS 108 (55%) 6 (60%) NS EF <50% 33 (16%) 23 (14%) 10 (22%) NS 29 (15%) 4 (22%) NS 29 (15%) 4 (40%) 0.03 History of CHF 38 (18%) 27 (17%) 11 (24%) NS 36 (19%) 2 (11%) NS 31 (16%) 7 (70%) < Abnormal ECG 139 (67%) 105 (65%) 34 (76%) NS 128 (68%) 11 (61%) NS 132 (67%) 7 (70%) NS Coronary disease 59 (29%) 42 (26%) 17 (38%) NS 55 (29%) 4 (22%) NS 56 (29%) 3 (30%) NS Non-IHD 34 (17%) 22 (14%) 12 (27%) (16%) 3 (17%) NS 30 (15%) 4 (40%) 0.04 Pacemaker 67 (33%) 63 (39%) 4 (9%) (35%) 1 (6%) (33%) 2 (20%) NS Implantable 32 (16%) 24 (15%) 8 (18%) NS 29 (15%) 3 (17%) NS 29 (15%) 3 (30%) NS cardioverter defibrillator Catheter therapy 44 (21%) 31 (19%) 13 (29%) NS 40 (21%) 4 (22%) NS 42 (21%) 2 (20%) NS Drug-induced syncope 30 (15%) 19 (12%) 11 (24%) (12%) 7 (39%) (14%) 3 (30%) NS B Total Cardiovascular event Rehospitalization for syncope Death patients ( ) (+) P value ( ) (+) P value ( ) (+) P value n Circumstances of syncope Injury 39 (19%) 31 (19%) 8 (18%) NS 38 (20%) 1 (6%) NS 36 (18%) 3 (30%) NS During effort 62 (30%) 50 (31%) 12 (27%) NS 60 (32%) 2 (11%) NS 58 (30%) 4 (40%) NS While supine 24 (11%) 18 (11%) 6 (13%) NS 21 (11%) 3 (17%) NS 21 (11%) 3 (30%) NS After moving from supine to standing position 21 (10%) 16 (10%) 5 (11%) NS 20 (11%) 1 (6%) NS 21 (11%) 0 (0%) NS During urination or 15 (7%) 8 (5%) 6 (13%) NS 11 (6%) 3 (17%) NS 14 (7%) 0 (0%) NS defecation Drinking 8 (4%) 1 (2%) 7 (4%) NS 8 (4%) 0 (0%) NS 8 (4%) 0 (0%) NS While taking bath 9 (4%) 7 (4%) 2 (4%) NS 9 (5%) 0 (0%) NS 8 (4%) 0 (0%) NS While driving 7 (3%) 7 (4%) 0 (0%) NS 7 (4%) 0 (0%) NS 7 (4%) 0 (0%) NS Characteristics of syncope Prodromal 119 (58%) 98 (61%) 21 (47%) NS 109 (58%) 10 (56%) NS 114 (58%) 5 (50%) NS symptoms Palpitation 23 (11%) 19 (12%) 4 (9%) NS 21 (11%) 2 (11%) NS 22 (11%) 1 (10%) NS Chest pain/ 22 (11%) 17 (11%) 5 (11%) NS 20 (11%) 2 (11%) NS 22 (11%) 0 (0%) NS discomfort Vertigo 26 (13%) 23 (14%) 3 (7%) NS 23 (12%) 3 (13%) NS 26 (13%) 0 (0%) NS Asthenia 4 (2%) 4 (2%) 0 (0%) NS 4 (2%) 0 (0%) NS 4 (2%) 0 (0%) NS Diaphoresis 4 (2%) 4 (2%) 0 (0%) NS 4 (2%) 0 (0%) NS 4 (2%) 0 (0%) NS Blurred vision 18 (9%) 16 (10%) 2 (4%) NS 17 (9%) 1 (5%) NS 16 (8%) 2 (20%) NS Nausea 13 (6%) 10 (6%) 3 (7%) NS 11 (6%) 2 (11%) NS 11 (5%) 2 (20%) NS Dyspnea 6 (3%) 4 (2%) 2 (4%) NS 4 (2%) 2 (11%) (3%) 1 (10%) NS Stagger 15 (7%) 12 (7%) 3 (7%) NS 14 (7%) 1 (6%) NS 15 (8%) 0 (0%) NS (Table 2 continued the next page.)

5 Risk Predictors in CV Syncope Patients 1399 B Total patients Cardiovascular event Rehospitalization for syncope Death ( ) (+) P value ( ) (+) P value ( ) (+) P value Etiology of syncope Sick sinus syndrome 63 (31%) 53 (33%) 10 (22%) NS 57 (30%) 6 (33%) NS 59 (30%) 4 (40%) NS Atrioventricular block 34 (17%) 30 (19%) 4 (9%) NS 33 (18%) 1 (6%) NS 32 (16%) 2 (20%) NS Bradycardic atrial 5 (2%) 5 (3%) 1 (2%) NS 4 (2%) 1 (2%) NS 5 (3%) 0 (0%) NS fibrillation Supraventricular tachycardia 19 (9%) 16 (10%) 3 (7%) NS 18 (10%) 1 (6%) NS 19 (10%) 0 (0%) NS Ventricular tachycardia 37 (18%) 25 (16%) 12 (27%) NS 30 (16%) 7 (39%) (17%) 4 (40%) NS Acute coronary 16 (8%) 11 (7%) 5 (11%) NS 16 (9%) 0 (0%) NS 16 (9%) 0 (0%) NS syndrome Vasospastic spastic 17 (8%) 13 (8%) 4 (9%) NS 15 (8%) 2 (11%) NS 17 (9%) 0 (0%) NS angina Valvular disease 5 (2%) 3 (2%) 2 (4%) NS 4 (2%) 1 (6%) NS 5 (2%) 0 (0%) NS Hypertrophic 6 (3%) 5 (3%) 1 (2%) NS 6 (3%) 0 (0%) NS 6 (3%) 0 (0%) NS obstructive cardiomyopathy Pulmonary 3 (1%) 1 (1%) 2 (4%) NS 3 (1%) 0 (0%) NS 3 (0%) 0 (0%) NS thromboembolism Other 1 (0.4%) 0 (0%) 1 (2%) NS 1 (1%) 0 (0%) NS 1 (1%) 0 (0%) NS C Total Cardiovascular event Rehospitalization for syncope Death patients ( ) (+) P value ( ) (+) P value ( ) (+) P value n Medications Antiplatelet drug 78 (38%) 58 (36%) 20 (44%) NS 71 (38%) 7 (39%) NS 74 (38%) 4 (40%) NS Anticoagulant 68 (33%) 49 (30%) 19 (42%) NS 61 (32%) 7 (39%) NS 63 (32%) 5 (50%) NS β-blocker 68 (33%) 52 (32%) 16 (36%) NS 61 (32%) 7 (39%) NS 64 (33%) 4 (40%) NS β-blocker low dose 43 (21%) 35 (21%) 8 (18%) NS 39 (21%) 4 (22%) NS 40 (20%) 3 (30%) NS β-blocker normal or high dose 33 (16%) 16 (10%) 7 (16%) NS 21 (11%) 2 (11%) NS 22 (11%) 1 (10%) NS α-blocker 14 (7%) 11 (7%) 3 (7%) NS 13 (7%) 1 (6%) NS 14 (7%) 0 (0%) NS ACEI/ARB 98 (48%) 79 (49%) 19 (42%) NS 92 (49%) 6 (33%) NS 94 (48%) 4 (40%) NS Loop diuretic 32 (16%) 24 (15%) 8 (18%) NS 30 (16%) 2 (11%) NS 27 (14%) 5 (50%) Potassium-sparing 19 (9%) 11 (7%) 8 (18%) (10%) 1 (6%) NS 17 (9%) 2 (20%) NS diuretic Thiazide diuretic 6 (3%) 6 (4%) 0 (0%) NS 6 (4%) 0 (0%) NS 6 (4%) 0 (0%) NS Calcium-channel blocker 72 (35%) 59 (37%) 13 (29%) NS 66 (35%) 6 (33%) NS 71 (36%) 1 (10%) NS Coronary vasodilator 41 (20%) 30 (19%) 11 (24%) NS 37 (20%) 4 (22%) NS 41 (21%) 0 (0%) NS Antidepressants 39 (19%) 33 (20%) 6 (13%) NS 35 (19%) 4 (22%) NS 38 (19%) 1 (10%) NS Hypnotic 6 (3%) 5 (3%) 1 (2%) NS 5 (3%) 1 (6%) NS 5 (3%) 1 (10%) NS Statin 69 (33%) 50 (31%) 19 (42%) NS 63 (34%) 6 (33%) NS 66 (34%) 3 (30%) NS Antiarrhythmic drug 41 (20%) 32 (20%) 9 (20%) NS 37 (20%) 4 (22%) NS 40 (20%) 1 (10%) NS Abbreviations as in Table 1. or minor injury was caused to 39 patients (19%) during their syncopal event. As for the circumstances of the syncope, 62 (30%) and 24 (11%) patients, respectively, experienced syncope during effort or while supine. Prodromal symptoms occurred in 119 patients (58%) and included palpitation (n=23; 11%), chest pain or discomfort (n=22; 11%), vertigo (n=26; 13%), and blurred vision (n=18; 9%). Figure 1 shows the various diagnoses during admission. At the end of the diagnostic evaluation, sick sinus syndrome was established in 63 patients (31%), atrioventricular block or bradycardic atrial fibrillation in 39 patients (19%), supraventricular tachycardia in 19 patients (9%), ventricular tachycardia in 37 patients (18%), acute coronary syndrome or vasospastic angina in 33 patients (16%), valvular disease or HOCM in 11 patients (5%), pulmonary thromboembolism in 3 patients (1%) and other in 1 patient (0.4%). Table 1B shows the distribution of OESIL scores in the patients: 12 (6%), 42 (20%), 80 (39%), 52 (25%), and 20 (10%) patients. At 1 year after discharge, 45 and 18 patients, respectively, were re-hospitalized for a cardiovascular event or syncope (Table 1C). Implantation of a pacemaker was less frequent in patients with a cardiovascular event (9%) than in those with a non-cardiovascular event (39%) (P=0.0001). In contrast, systolic BP (sbp) <100 mmhg (33% vs. 10%, P=0.0001) and non-ihd (27% vs. 14%, P=0.04) were significantly more frequent in patients with cardiovascular events (Table 2A). Stepwise logistic regression indicated that sbp <100 mmhg

6 1400 ONUKI T et al. Table 3. Independent Predictors of Cardiovascular Events, Rehospitalization for Syncope and All-Cause Death Cardiovascular event 45 (22%) Independent predictors sbp <100 mmhg (OR 3.25; 95% CI , P=0.006) Implantation of a pacemaker (inverse association) (OR 0.19; 95% CI , P=0.0005) Rehospitalization for syncope 18 (9%) Independent predictors Drug-induced syncope (OR 4.57; 95% CI , P=0.007) All-cause death 10 (5%) Independent predictors History of CHF (OR 11.0; 95% CI , P=0.0006) sbp <100 mmhg (OR 5.40; 95% CI , P=0.02) CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1. Figure 3. Comparison of syncope-free curves for rehospitalization rates of patients with and without drug-induced syncope. Figure 2. (A) Comparison of cardiovascular event-free survival curves for patients with and without a pacemaker. (B) Comparison of the cardiovascular event-free survival curves of patients with systolic blood pressure (sbp) <100 mmhg and those with sbp 100 mmhg. (OR 3.25; 95% CI , P=0.006) and implantation of a pacemaker (OR 0.19; 95% CI , P=0.0005) (inverse association) were independent risk factors for rehospitalization with a cardiovascular event (Table 3). Patients with sbp <100 mmhg or without an indication for a pacemaker for syncope consistently developed cardiovascular events within 1 year of discharge (Figure 2A,B). Pacemaker was less common in patients rehospitalized for syncope (6%) than in those who were not (35%) (P=0.01). Also, drug-induced syncope was more frequent in patients rehospitalized for syncope (39%) than in those who were not (12%) (P=0.002) (Table 2A). Syncope caused by ventricular tachyarrhythmia (39% vs. 16%, P=0.02) was more frequent in patients rehospitalized for syncope (Table 2B). Drug-induced syncope was a significant independent risk factor for rehospitalization for syncope (OR 4.57; 95% CI , P=0.007) (Table 3). Indeed, rehospitalization for syncope was much more common from 6 months after discharge among patients with drug-induced syncope (P=0.001) (Figure 3; Table S1). A total of 10 patients had died at 1 year after discharge (Table 1C). Compared with those who survived, sbp <100 mmhg (50% vs. 13%, P=0.002), ejection fraction <50% (40% vs. 15%, P=0.03), use of loop diuretic (50% vs. 14%, P=0.002) and the presence of non-ihd (40% vs. 15%, P=0.03) were more frequent in the patients who died (Table 2A,C). Specifically, patient deaths were higher among those with a history of CHF (70%) than those without (16%) (P<0.0001) (Table 2A). Significant independent risk factors for death were a history of CHF (OR 11.0; 95% CI , P=0.0006) and sbp <100 mmhg (OR 5.40; 95% CI , P=0.02) (Table 3). Figure 4A,B shows the mortality-free survival curves for patients with sbp <100 mmhg and a history of CHF after discharge. In the present study, the OESIL score did not predict death or rehospitalization for syncope and cardiovascular events. Discussion It is well known that the prognosis of reflex syncope is comparable to that of healthy subjects without syncope 2 and that cardiovascular syncope is not only associated with a worse prognosis but is also a strong predictor of death and sudden death. 2,5 Indeed, most patients with cardiovascular syncope have comorbidities and the treatment of these may itself cause syncope. There is a need for physicians to prevent critical cardiovascular events and recurrent syncope

7 Risk Predictors in CV Syncope Patients 1401 after discharging patients admitted with cardiovascular syncope. Thus, to improve follow-up, we identified critical predictors of risk in the year after initial treatment for cardiovascular syncope. There are no previous reports about cardiovascular events or rehospitalization for syncope after initial diagnosis and discharge of patients with cardiovascular syncope. In this study, we showed that the significant predictors of cardiovascular event were sbp <100 mmhg and the absence of a pacemaker. Most cardiac diseases are likely to induce syncope, 3 with arrhythmias being particularly prevalent in the present study and considered causative in approximately 65% of cases. These patients have an elevated risk of sudden death or arrhythmia-induced syncope, and may benefit from implantation of a cardiac pacing device or discontinuation of a culprit drug. Patients with sbp <100 mmhg (P=0.006) and patients without a pacemaker (P=0.0005) consistently developed cardiovascular events within the year after discharge. Patients with a pacemaker may, therefore, be relatively more protected against rehospitalization for cardiovascular events, and we should focus our follow-up efforts on patients with no indication for a pacemaker. The most significant independent risk factor for rehospitalization because of syncope was drug-induced syncope (P=0.007). Most drugs responsible for cardiovascular syncope were antiarrhythmics, or diuretics or cathartics that caused electrolyte disturbances. When considering all patients with syncope, including those with non-cardiac diagnoses, drug-induced syncope was not a major risk factor for recurrence. 2 However, the prognosis and recurrence of syncope is unclear in the subgroup of patients with drug-induced syncope. There are reports that despite discontinuing culprit drugs, some patients with drug-induced sick sinus syndrome and atrioventricular block later require a pacemaker. 6,7 Additionally, discontinuation of an antiarrhythmic drug may itself induce tachyarrhythmias. We consider that druginduced factors are, therefore, plausible risk factors for rehospitalization because of cardiovascular syncope. Considering that recurrent syncope occurred within 6 months in most cases (P=0.001), drug-induced syncope should be monitored more closely during this period for the persistence of critical recurrent syncope that requires hospitalization. Significant independent prognostic factors for death include a history of CHF and sbp of <100 mmhg (P= and P=0.02, respectively). In patients with HF, the sbp on admission has been shown to be an important prognostic factor for in-hospital and post-discharge death. 8,9 Indeed, in a cohort of Japanese patients with HF, low sbp was predictive of in-hospital death. 10 In addition, hypotension has been associated with short-term serious outcomes in patients presenting to emergency departments with syncope or near syncope. 11 We identified that patients in this study with cardiovascular syncope and low BP were at increased risk of death after 1 year. The main causes of cardiovascular syncope are structural heart disease and arrhythmias, which result in decreased cardiac output, hemodynamic impairment, and a subsequent decrease in cerebral blood flow. The pulmonary congestion and systemic edema of CHF are induced by insufficient cardiac output. Although the general symptoms of HF and cardiovascular syncope are different, the pathology of HF resembles cardiovascular syncope. It may be reasonable to conclude that low BP is a risk factor for death in both HF and cardiovascular syncope for these reasons. By the same token, it is understandable that a his- Figure 4. (A) Comparison of mortality-free survival curves for patients with systolic blood pressure (sbp) <100 mmhg and those with sbp 100 mmhg. (B) Comparison of mortality-free survival curves for patients with and without a history of congestive heart failure (CHF). tory of CHF was a significant independent prognostic factor for death. In a previous study, Khera et al 12 reported the independent risk factors for death and rehospitalization for syncope in elderly patients with syncope. They showed that diabetes mellitus, a history of smoking, non-use of statins, CHF, and acute coronary syndrome were independent risk factors for both outcomes. In another study, Sule et al added that for hospitalized patients with syncope, a history of diabetes, coronary artery bypass graft surgery, malignancy, narcotic use, smoking, atrial fibrillation, and volume depletion were significant independent risk factors for death, whereas diabetes, atrial fibrillation, and smoking were risk factors for rehospitalization. 13 However, these results differ from ours, because patients with non-cardiovascular syncope and unexplained syncope were included, or because observation times were different. The OESIL score did not predict poor outcome in present study, even though it is generally useful for risk stratification of patients with syncope in the short-term. 4,14 Studies of syncope differ in the of clinical setting, patients ages, and diagnostic definitions, making comparisons between them difficult. 3 In addition, cardiovascular syncope has the risk of sudden death, which is the most adverse event. It is well known that the etiology of sudden death is different between Western countries and Asia. 15,16 Indeed, IHD associated with arteriosclerosis causes sudden death more frequently in Western countries than in Asia. Thus, it seems to be important to establish the risk factors of syncope in Asia.

8 1402 ONUKI T et al. Study Limitations This study has the limitations of an observational, retrospective, single-center study and will need to be repeated in a larger trial to confirm our results. Indeed, our results may overestimate identified risk factors because only a few patients experienced adverse events. Moreover, we could not account for the factors that influenced the tests chosen by the attending physician, which may have affected diagnosis, although in a previous report 17 we could determine the diagnosis of 81.7% of all syncope patients. Conclusions We identified the significant independent risk factors for 3 poor outcomes in patients with cardiovascular syncope. First, drug-induced syncope was the main factor associated with rehospitalization for syncope. Second, hypotension and the presence of a pacemaker device (inverse association) were significant independent risk factors for cardiovascular events. Finally, a history of CHF and hypotension were independent risk factors for death. Drug-induced syncope, hypotension, the presence of a pacemaker device (inverse association), and a history of CHF were risk factors for poor outcomes in patients with cardiovascular syncope. We recommend careful follow-up of patients with these risk factors for at least 1 year after discharge. Conflict of Interest There are no conflicts of interest or financial disclosures to declare regarding this study. Acknowledgments We thank the medical clerk and cardiology nursing staff at Showa University Hospital, Japan. None. Grants References 1. Goldschlager N, Epstein AE, Grubb BP, Olshansky B, Prystowsky E, Roberts WC, et al. Etiologic considerations in the patient with syncope and an apparently normal heart. Arch Intern Med 2003; 163: Soteriades ES, Evans JC, Larson MG, Chen MH, Chen L, Benjamin EJ, et al. Incidence and prognosis of syncope. N Engl J Med 2002; 347: Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB, et al. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 2009; 30: Colivicchi F, Ammirati F, Melina D, Guido V, Imperoli G, Santini M, et al. Development and prospective validation of a risk stratification system for patients with syncope in the emergency department: The OESIL risk score. Eur Heart J 2003; 24: Alboni P, Brignole M, Menozzi C, Raviele A, Del Rosso A, Dinelli M, et al. Diagnostic value of history in patients with syncope with or without heart disease. J Am Coll Cardiol 2001; 37: Inada K, Yamane T, Tokutake K, Yokoyama K, Mishima T, Hioki M, et al. The role of successful catheter ablation in patients with paroxysmal atrial fibrillation and prolonged sinus pauses: Outcome during a 5-year follow-up. Europace 2014; 16: Osmonov D, Erdinler I, Ozcan KS, Altay S, Turkkan C, Yildirim E, et al. Management of patients with drug-induced atrioventricular block. Pacing Clin Electrophysiol 2012; 35: Choi DJ, Han S, Jeon ES, Cho MC, Kim JJ, Yoo BS, et al. Characteristics, outcomes and predictors of long-term mortality for patients hospitalized for acute heart failure: A report from the korean heart failure registry. Korean Circ J 2011; 41: Lee SE, Cho HJ, Lee HY, Yang HM, Choi JO, Jeon ES, et al. A multicentre cohort study of acute heart failure syndromes in Korea: Rationale, design, and interim observations of the Korean Acute Heart Failure (KORAHF) registry. Eur J Heart Fail 2014; 16: Tsutsui H, Tsuchihashi-Makaya M, Kinugawa S, Goto D, Takeshita A; JCARE-CARD Investigators. Clinical characteristics and outcome of hospitalized patients with heart failure in Japan. Circ J 2006; 70: Quinn JV, Stiell IG, McDermott DA, Sellers KL, Kohn MA, Wells GA. Derivation of the San Francisco syncope rule to predict patients with short-term serious outcomes. Ann Emerg Med 2004; 43: Khera S, Palaniswamy C, Aronow WS, Sule S, Doshi JV, Adapa S, et al. Predictors of mortality, rehospitalization for syncope, and cardiac syncope in 352 consecutive elderly patients with syncope. J Am Med Dir Assoc 2013; 14: Sule S, Palaniswamy C, Aronow WS, Ahn C, Peterson SJ, Adapa S, et al. Etiology of syncope in patients hospitalized with syncope and predictors of mortality and rehospitalization for syncope at 27-month follow-up. Clin Cardiol 2011; 34: Costantino G, Furlan R. Syncope risk stratification in the emergency department. Cardiol Clin 2013; 31: Murakoshi N, Aonuma K. Epidemiology of arrhythmias and sudden cardiac death in Asia. Circ J 2013; 77: Tanno K, Miyoshi F, Watanabe N, Minoura Y, Kawamura M, Ryu S, et al. Are the MADIT II criteria for ICD implantation appropriate for Japanese patients? Circ J 2005; 69: Onuki T, Ito H, Ochi A, Chiba Y, Kawasaki S, Onishi Y, et al. Single center experience in Japanese patients with syncope. J Cardiol 2015; 66: Supplementary Files Supplementary File 1 Table S1. Comparison of patients with and without drug-induced syncope Please find supplementary file(s);