Author + information
- Received September 13, 2018
- Revision received January 29, 2019
- Accepted January 31, 2019
- Published online May 20, 2019.
- Jayaprakash Shenthar, MD, DMa,∗ (, )
- Mukund Aravind Prabhu, MD, DMa,
- Bharatraj Banavalikar, MD, DMa,
- David G. Benditt, MDb and
- Deepak Padmanabhan, MD, DMa
- aElectrophysiology Unit, Department of Cardiology, Sri Jayadeva Institute of Cardiovascular Sciences and Research, Bangalore, Karnataka, India
- bUniversity of Minnesota Medical School, Minneapolis, Minnesota
- ↵∗Address for correspondence:
Prof. Jayaprakash Shenthar, Electrophysiology Unit, Department of Cardiology, Sri Jayadeva Institute of Cardiovascular Sciences and Research, 9th Block Jayanagar, Bannerghatta Road, Bangalore 560069, India.
Objectives This study sought to determine the cause of recurrent syncope and clinical outcomes by using the head-up tilt test (HUTT) and an insertable loop recorder (ILR) in patients with structural heart disease (SHD) and negative electrophysiology study (EPS) results.
Background Patients with syncope and SHD with negative EPS findings have a low risk of sudden cardiac arrest. Nevertheless, the cause of recurrent syncope and the outcomes in these patients are not well characterized.
Methods This prospective study evaluated syncope patients with SHD and negative EPS results by using HUTT (with sublingual nitroglycerine [NTG] provocation as needed) and ILR. A total of 41 SHD patients (27 patients [66%] had coronary arterial disease, and 14 patients [34.15%] had dilated cardiomyopathy with mean EF of 42 ± 4.8% [range 30% to 49%]) were included.
Results HUTT findings were positive in 25 patients (61%) in group A and negative in 16 patients (39%) in group B. An ILR was implanted in 21 of 25 group A patients (84%) and in 12 of 16 group B patients (75%), and they were followed for 15 ± 8 months. During follow-up, 17 of 21 patients (81%) in group A and 5 of 12 patients (41.7%) in group B had ILR evidence consistent with reflex syncope. One group B patient had documented atrioventricular block and underwent pacemaker implantation. There were no malignant ventricular arrhythmias or deaths on follow-up.
Conclusions Reflex syncope is the most common cause of syncope and accounts for approximately 60% of cases in patients with SHD, negative EPS results, left ventricular systolic dysfunction with left ventricular EF >30%, and not in heart failure.
Syncope/collapse is a common clinical condition and is among the most frequent diagnoses leading to emergency department visits and hospital admissions (1–3). Furthermore, in patients who had syncope in the setting of a structurally normal heart, a vasovagal reflex mechanism is the most common diagnosis (4,5).
As a rule, mortality risk is low in reflex syncope patients, especially those without structural heart disease (SHD). Their major concerns are risk of injury and impact on quality of life (6). In patients with SHD, syncope per se does not necessarily herald an adverse prognosis if there is no documented ventricular arrhythmia (5). In fact, it is estimated that among SHD patients with syncope, approximately one-third, including those with inducible arrhythmia episodes, may have reflex syncope (4); however, substantial uncertainty remains.
This study used an insertable loop recorder (ILR) to determine the cause of recurrent syncope and assess subsequent clinical outcomes in syncope patients with SHD, no previously documented ventricular arrhythmia episodes, negative electrophysiology study (EPS) findings, and a positive head-up tilt test (HUTT) score.
The present study was undertaken prospectively in SHD patients with a history of recurrent syncope, evaluated at the syncope clinic of a tertiary care cardiac center in south India. The institutional review board approved the study, and all subjects consented to the study.
Study subjects consisted of consecutive patients with a history of syncope, with moderate left ventricular (LV) dysfunction due to coronary arterial disease or DCM, no documented spontaneous ventricular tachyarrhythmia episodes on 24-h Holter recording, and negative EPS findings.
All study patients underwent a thorough clinical history from the patient or from eyewitnesses if present, or both if possible. A comprehensive physical examination included postural blood pressure recordings. Cardiovascular system assessment also included 12-lead electrocardiography (ECG) and an echocardiogram (i.e., chamber dimensions, valvular function, and LV ejection fraction [LVEF]).
A staff electrophysiologist evaluated the 12-lead ECG and the 24-h Holter recording obtained from all study subjects. Coronary angiography and, when necessary, cardiac magnetic resonance, were used to evaluate the presence and severity of coronary arterial disease, DCM, or other SHDs. Inducible ischemia was excluded in patients with ischemic heart disease by treadmill test, dobutamine stress test, or stress thallium scan.
EPS were undertaken in all patients, consisting of assessment of sinus node function and atrioventricular (AV) conduction during the EPS. Programmed stimulation was performed at 2 cycle lengths (600 and 450 ms), with up to 3 extra stimuli from the right ventricular apex and right ventricular outflow tract to assess for inducibility of tachyarrhythmia episodes at baseline and with an infusion of isoprenaline at 1 to 4 mg/min to achieve a heart rate of 110 to 120 beats/min. Patients with negative EPS findings underwent HUTT, using nitroglycerin (NTG) provocation if needed (i.e., the Italian protocol) (7).
Exclusion criteria included documented or inducible ventricular tachyarrhythmia episodes, inducible ischemia, LVEF <30%, presence of heart failure, significant bradyarrhythmias such as sinus node dysfunction, or AV block documented either on ECG or 24-h Holter and patients with implanted devices such as pacemakers, implantable cardioverter-defibrillators, and cardiac resynchronization therapy devices, or if the patient did not give consent for the study. Patients with bundle branch block or bifascicular or trifascicular block with a prolonged His to ventricle interval >70 ms, abnormal sinus node function (i.e., corrected sinus node recovery time >550 ms) on EPS were excluded. Patients treated with nitrates were also excluded because nitrate tolerance potentially affecting the response to NTG provocation during HUTT.
The modified VASIS (Vasovagal Syncope International Study) classification was used to classify the syncopal events during the HUTT (8). Briefly, the responses associated with a positive HUTT were:
Type 1 (mixed): heart rate drops at the time of syncope with the ventricular rate not <40 beats/min or is <40 beats/min for <10 s with or without asystole of <3 s.
Type 2A: cardio-inhibitory without asystole; heart rate falls to a ventricular rate <40 beats/min for >10 s, but asystole of >3 s does not occur.
Type 2B: cardio-inhibitory with asystole; asystole occurs >3 s, and heart rate fall coincides with or precedes blood pressure fall.
Type 3: vasodepressor; heart rate does not drop more than 10% from its peak, but at the time of syncope blood pressure falls precipitously.
Classification of arrhythmic events during HUTT
Development of new arrhythmia episodes that were not present during the 30-min observation and baseline tilt were documented. An arrhythmia event was defined as tachyarrhythmia if the heart rate was >100 beats/min after excluding sinus tachycardia. The criterion for bradyarrhythmia was sinus rate of <40 beats/min, development of junctional rhythm, pause >3 s, or any degree of AV block. The criterion for premature beats was a newly developed atrial, junctional, or ventricular ectopic heart beats. A new onset arrhythmia episode such as atrial fibrillation, atrial tachycardia, ventricular tachycardia, or ventricular fibrillation that occurred during the test was considered a significant arrhythmic event.
Implantation of an insertable loop recorder
Implantation of an ILR was offered to all patients with negative EPS findings with or without a positive HUTT result. In those who agreed to the implantation, a Reveal XT insertable cardiac monitor (Medtronic Inc., Minneapolis, Minnesota) was implanted subcutaneously in the left parasternal area. The ILR was programmed to save up to 3 manual activations of 7.5 min. Automatic activations were programmed to: 1) rapid ventricular tachycardia (R-R interval: 260 ms in at least 30 of 40 consecutive beats); 2) ventricular tachycardia (R-R interval: 261 to 340 ms in 16 consecutive beats); 3) a pause of >3 s; or 4) bradycardia (heart rate: 30 beats/min in 4 consecutive beats).
ILR recordings were retrieved as soon as was feasible in case of syncope and assessed for any ventricular or atrial tachyarrhythmia, and the heart rate response during the episode. The ISSUE (International Study on Syncope of Uncertain Etiology) classification was used to document ECG syncope on ILR (9). Briefly, the classifications were as follows:
Type 1: asystole with R-R pause >3 s (type 1A sinus arrest, type 1B sinus bradycardia plus AV block, type 1C AV block).
Type 2: bradycardia with decrease in heart rate >30% (type 2A) or <40 beats/min for >10 s (type 2B).
Type 3: no or slight rhythm variation.
Type 4: tachycardia. Increase in heart rate >30% or >120 beats/min.
After patients received ILR implantation, they were followed in the outpatient clinic every 3 months until a diagnosis was reached or until the end of service of the device. Once the device reached its end of service, it was explanted with no further replacement. and patients were followed in the outpatient clinic. In those cases in which the patient refused ILR implantation, they were followed in the outpatient clinic. Endpoints were either syncope recurrence or mortality.
Data are presented as mean ± SD. Patient data was compared using Fisher exact test. Continuous variables were compared using Student’s t-test in the case of normal distribution of values and the nonparametric Mann-Whitney U test in the case of an asymmetric distribution. A p value <0.05 was considered significant.
From January 2010 to June 2015, 70 patients who had syncope with SHDs were evaluated. Seven patients were excluded from the study because they had documented episodes of symptomatic ventricular arrhythmia. Of the 63 patients who underwent EPS, 22 patients with inducible ventricular arrhythmia (n = 17) and 5 patients with prolonged HV intervals >70 ms (2 with left bundle branch block, and 1 with bifascicular and 1 with trifascicular block) were excluded from the study. The final study population consisted of 41 SHD patients with a history of syncope and no inducible ventricular arrhythmia (Figure 1).
The mean age of the patients was 59 ± 16 years (range 16 to 85 years), and the majority were males (73.2%). There were 27 patients (65.9%) with ischemic heart disease and 14 patients (34.1%) with DCM. The mean LVEF was 42 ± 4.8% (range 30% to 49%). The mean number of syncopal episodes was 4.4 per year (range 3 to 7). Baseline parameters of the study population are summarized in Online Table 1.
Outcome of the HUTT results
HUTT results were positive in 25 of the 41 patients (61%); test results were positive without pharmacological provocation in 6 of 25 patients (24%) and with NTG provocation in 19 or 25 patients (76%). The most common response was the modified VASIS type III noted in 13 patients (52%), type I in 6 patients (24%), type 2B in 5 patients (20%), and type 2A in 1 patient (4%). There was a good correlation between the clinical symptoms and HUTT-induced symptoms in these patients.
Episodes of arrhythmia during the HUTT
Arrhythmias occurred during the HUTT in 29 of 41 patients (71%) (Online Table 2). Bradyarrhythmias occurred as a part of the positive response in 12 patients (29.3%). Sinus bradycardia (with no pauses >3 s) was the most common arrhythmia seen in 7 patients (17.1%). Sinus pause of >3 s was documented in 5 patients (12.2%) with a mean pause duration of 3.6 ± 0.53 s.
The most common arrhythmia during the HUTT was ectopic beats, which occurred in 14 patients (34.1%). Ventricular ectopic beats were observed in 7 patients (17.07%), atrial ectopic beats in 3 patients (7.32%), and junctional ectopic beats in 3 patients (7.32%). Three patients (7.32%) had tachyarrhythmias with atrial tachycardia seen in 2 patients (4.9%) during the HUTT. The tachycardia rate varied between 123 and 176 beats/min with a mean heart rate of 153 beats/min. The tachycardia lasted for 32 s in 1 patient and 50 s in another and was asymptomatic and terminated spontaneously with no requirement for cardioversion. Transient atrial fibrillation lasting for 3 min developed in 1 patient (2.4%) with a mean heart rate of 164 beats/min and normal blood pressure. Except for palpitations, there were no other specific complaints from the patient. There were no instances of ventricular tachyarrhythmias noted in any subjects during the HUTT.
Group A consisted of 21 of 25 patients (84%) with positive HUTT results who underwent ILR implantation (Table 1). Group B consisted of 12 of 16 patients (75%) with negative HUTT results who underwent ILR implantation. Four patients in each group declined ILR placement.
The mean duration of follow-up in ILR patients was 15 ± 8 months. In group A patients, 17 of the 21 patients (81%) had syncope during follow-up. Of the 17 patients in group A, ILR recordings were suggestive of reflex syncope, with type 1A (Figure 2) in 1 of 17 patients (5.9%), type 1B in 1 of 17 patients (5.9%), type 2A in 4 of 17 patients (23.5%), type 2B in 5 of 17 patients (29.4%), and type 3 in 6 of 17 patients (35.3%), according to the ISSUE classification. In group B, 5 of the 12 patients (41.7%) had ILR recordings suggestive of reflex syncope, with type 1A in 1 of 5 patients (20%), type 2A in 2 of 5 patients (40%), and type 3 in 2 of 5 patients (40%). One patient with coronary arterial disease in group B had intermittent complete AV block with no increase in atrial rate, suggesting an AV conduction disease (Figure 3), and underwent permanent pacemaker implantation. The patient who developed complete AV block had baseline right bundle branch block, normal axis, with a normal HV interval of 50 ms, and an AV Wenckebach 450-ms drive cycle length during EPS.
A comparison of the patients with positive HUTT results who had a syncopal episode during ILR (Figure 4) showed that, of the 4 patients with prolonged asystolic pause (VASIS type 2B) during HUTT, 1 patient had bradycardia with a decrease in heart rate >30% (ISSUE type 2A), and 2 patients had heart rate of <40 beats/min for >10 s (ISSUE type 2B) during ILR.
During 15 ± 8 months of follow-up, there were no significant ventricular or atrial tachyarrhythmias or deaths in either of the 2 ILR groups. Eight patients, 4 in each group, refused ILR but continued to follow-up in the syncope clinic. In the patients who refused ILR, syncope recurred in 2 of 4 patients (50%) in group A and 1 of 4 patients (25%) in group B, with no deaths.
Three group A subjects and 2 group B patients were lost to follow-up. Telephone conversations revealed that they had no recurrences and were unwilling to come to the hospital.
The present study assessed causes and outcomes of syncope in patients with SHD and negative EPS. There were 4 principal observations. First, approximately 60% of patients in this study had positive HUTT results (Central Illustration). Second, in patients with positive HUTT results, 81% had findings suggestive of reflex syncope documented on ILR. Third, in patients with negative HUTT results, 50% had ECG observations suggestive of reflex syncope on ILR. Finally, there were no deaths or clinically significant arrhythmia events detected during follow-up, suggesting that the medium-term prognosis is favorable in SHD patients and negative EPS findings.
In syncope/collapse patients with SHD, a thorough evaluation is essential to exclude cardiac causes. In this regard, the presence of SHD does not necessarily imply that the syncopal event has a cardiac/arrhythmic cause with malignant connotation (5,10), but the latter is certainly a high priority to exclude, given the potential mortality risk. Nevertheless, patients with underlying SHD can have nonarrhythmic causes of syncope, such as reflex or drug-induced syncope/collapse (10,11).
The HUTT is safe, inexpensive, and above all, noninvasive. In addition, it has a relatively high diagnostic yield in the evaluation of syncope (3,12). However, most studies of HUTT (with or without pharmacological provocation) have been conducted in patients with structurally normal hearts (13–17). As seen in the present study, however, even in SHD patients, the Italian protocol procedure is safe, unlike HUTT conducted with isoprenaline provocation, where significant episodes of arrhythmia can be precipitated with isoprenaline infusion (4,18,19).
Previous work suggests that approximately one-third of SHD patients with syncope/collapse have a reflex basis for their symptoms, and consequently, the HUTT may be useful as part of their overall evaluation (4). Furthermore, it is suggested that the HUTT may reasonably be used for evaluation of syncope in SHD patients with negative EPS findings (20,21).
Prolonged ILR ECG monitoring has become an invaluable tool in the evaluation of patients with syncope of unknown origin. Garcia-Civera et al. (21) noted that, in SHD patients with syncope of unknown cause with negative EPS, HUTT results were positive in 57% of the cohort and that selective use of EPS, HUTT, and ILR resulted in a diagnosis in 78% of patients. In the present study, 56% of patients had ILR findings most consistent with reflex syncope (see criteria summarized above), with a higher incidence in those with a positive HUTT score (81%) compared to those with a negative HUTT score (50%). Results of the present study agree with prior observations regarding the incidence of reflex syncope, even in SHD patients deemed to be at high risk (22). Furthermore, the medium-term mortality appears to be low in SHD patients with negative EPS findings, which is also consistent with the most common cause of syncope, that is, the reflex syncope in origin (5,22,23).
In the present study, of the 4 patients with prolonged asystolic pause during the HUTT (VASIS type 2B), 1 patient had bradycardia with a decrease in heart rate >30% (ISSUE type 2A), and 2 patients had a heart rate of <40 beats/min for >10 s (ISSUE type 2B) during ILR (Figure 4). The findings from the present study differ from those reported in another study where 86% of patients with asystolic pauses during the HUTT (VASIS type 2B) had asystolic pauses (ISSUE type 1) on ILR (24). In the present study, there was no correlation between the HUTT response and ILR-documented syncope, and this is similar to what has been reported previously (25).
A number of limitations impact the interpretation of the findings reported here. First, the observations are derived from a small number of patients at a single center. Furthermore, the study examined a highly select group of SHD patients with syncope and no documented or inducible arrhythmias. However, these are patients who would be otherwise classified as having syncope of an unknown cause based on clinical guidelines and would have qualified for implantation of a defibrillator. Second, the ambulatory monitoring to exclude spontaneous episodes of arrhythmias prior to inclusion in the study was of a relatively short duration (24 h). A longer baseline monitoring period might have resulted in exclusion of additional patients. Third, in the present study, most of the patients had underlying coronary arterial disease, and further studies in a larger number of patients with other SHDs are needed. Fourth, the authors did not perform provocative testing with procainamide or flecainide as these were patients with SHD. Fifth, the results of the study do not apply to patients with documented arrhythmias, positive EPS, demonstrable ischemia or EF <30%, and/or heart failure. Finally, the duration of ILR monitoring was only 15 ± 8 months, and whether a more extended period of monitoring would have revealed a higher incidence of arrhythmic causes of syncope or greater mortality is not known.
Reflex syncope is the most common cause of transient loss of consciousness in patients with SHD and a history of syncope/collapse, a negative EPS, and LV dysfunction. Selective use of the HUTT and an ILR when needed, with clinical follow-up, appears to be a useful diagnostic strategy and can be undertaken safely in this SHD patient population. However, these results may not apply for patients with LVEF <30% and patients with heart failure.
COMPETENCY IN MEDICAL KNOWLEDGE: It is widely accepted that syncope/collapse in patients with SHD portends a worrisome prognosis. Potentially, this prognostic concern may lead physicians to more readily advise costly therapies, such as ablation, pacemakers, and implantable cardioverter-defibrillators. However, observations from the present study remind clinicians that syncope in SHD patients may not be directly related to the SHD itself but often originates from the lower risk reflex origin, comparable to the case in non-SHD fainters. Consequently, prior to accepting SHD as being directly responsible, all such patients merit careful assessment of the multiple potential causes of syncope/collapse before clinicians embark on a treatment strategy.
TRANSLATIONAL OUTLOOK: Current studies and practice guidelines appropriately emphasize the mortality risk associated with syncope in SHD patients. Without diminishing the value of such advice, the importance of substantiating a causal diagnosis must also be emphasized. To this end, the future of patient care should focus on the development and widespread use of novel remote cardiac monitoring technologies in order to establish with greater certainty the basis of syncope/collapse in individual patients.
Dr. Benditt is supported in part by a grant from the Dr. Earl E. Bakken Family for Heart-Brain research; and is a consultant for and holds equity in Medtronic Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- dilated cardiomyopathy
- ejection fraction
- electrophysiological study
- head-up tilt test
- insertable loop recorder
- left ventricle
- structural heart disease
- Received September 13, 2018.
- Revision received January 29, 2019.
- Accepted January 31, 2019.
- 2019 American College of Cardiology Foundation
- Pezawas T.,
- Stix G.,
- Kastner J.,
- Schneider B.,
- Wolzt M.,
- Schmidinger H.
- Benditt D.G.,
- Ferguson D.W.,
- Grubb B.P.,
- et al.
- Flammang D.,
- Erickson M.,
- McCarville S.,
- Church T.,
- Hamani D.,
- Donal E.
- Prabhu M.A.,
- Pillai V.,
- Shenthar J.
- Garcia-Civera R.,
- Ruiz-Granell R.,
- Morell-Cabedo S.,
- et al.
- Menozzi C.,
- Brignole M.,
- Garcia-Civera R.,
- et al.
- Brignole M.,
- Menozzi C.,
- Moya A.,
- et al.