Summary
Background
Few studies have compared conventional testing with prolonged monitoring using an implantable loop recorder (ILR) following the first syncope episode in patients with bundle branch block (BBB) and negative workup.
Objectives
To compare two syncope evaluation strategies–primary use of an ILR (Group 1) versus conventional testing (Group 2)–and to estimate the prevalence of significant arrhythmias in the ILR patient subset.
Methods
From January 2005 to December 2010, 78 patients admitted after one syncope episode were randomized to ILR ( n = 41) or conventional follow-up ( n = 37). Mean follow-up was 27 ± 12 months.
Results
Mean age was 76 ± 8 years and 30 patients were women (38.5%); 18 presented cardiomyopathy (23%) and 12 had a history of atrial fibrillation (15.4%). Mean left ventricular ejection fraction was 56.5 ± 11% and mean His-to-ventricle interval was 55 ± 6 ms based on negative electrophysiological study (EPS). Electrocardiogram abnormalities involved: 34 left bundle branch blocks (BBBs); 11 right BBBs; and 33 bifascicular blocks. Overall, 21 patients (27%) developed significant arrhythmic events: ventricular tachycardia ( n = 1; 1.3%); sudden death ( n = 2; 2.6%); third-degree atrioventricular (AV) block ( n = 14; 18%); sick sinus syndrome ( n = 4; 5.1%). In 19 (24.4%) patients, relevant arrhythmias were detected, with a significant difference between the ILR group ( n = 15/41; 36.6%) and the conventional follow-up group ( n = 4/37; 10.8%) ( P = 0.02). Eighteen patients were implanted with pacemakers; one received an implantable defibrillator. No predictors of AV block were identified in the ILR group.
Conclusions
In this randomized prospective study, the ILR strategy proved largely superior to conventional follow-up in detecting recurrent events, with a potential impact on therapeutic management. This observation highlights the usefulness of early monitoring in patients with BBB and negative EPS even after the first syncope episode but an empiric pacemaker strategy remains to be validated in this selected population.
Résumé
Contexte
Les données publiées sur le mécanisme des syncopes isolées chez les patients avec bloc de branche complet (BBC) et bilan électrophysiologique négatif sont peu nombreuses.
Objectifs
De comparer pour les patients avec première syncope et BBC une stratégie avec mise en place d’un moniteur ECG implantable (MEI) (Group I) versus un suivi conventionnel clinique et holter (Group II).
Méthodes
De janvier 2005 à décembre 2010, 78 patients (76 ± 8 ans) ont été randomisés, 41 patients dans le groupe MEI et 37 patients dans le groupe suivi conventionnel. Le suivi moyen était de 27 ± 12 mois.
Résultats
Les troubles de conduction étaient les suivants : 34 BBC gauche, 11 BBC droit et 33 blocs bifasciculaires. Dans l’ensemble de la population, 21 patients (27 %) ont développé un évènement rythmique : une TV dans un cas (1,3 %), une mort subite dans deux cas (2,6 %), un BAV du 3 e degré chez 14 patients (18 %) et une dysfonction sinusale dans quatre cas (5,1 %). Pour 19 patients (24,4 %), la détection d’un évènement rythmique a été possible, avec une différence significative entre les deux groupes de suivi : MEI ( n = 15/41 ; 36,6 %) versus suivi conventionnel ( n = 4/37 ; 10,8 %) ( p = 0,02). Par conséquent, 18 patients ont bénéficié de l’implantation d’un pacemaker et un patient de la mise en place d’un défibrillateur.
Conclusions
Cette étude prospective randomisée multicentrique chez des patients avec un épisode de syncope associé à un BBC et une exploration électrophysiologique négative montre que la stratégie du MEI est très largement supérieure au suivi traditionnel. Dans le groupe avec MEI, la prévalence des évènements rythmiques était de 36,6 % à 2,5 ans (14,6 % par an) versus 10,8 % (4,3 % par an) dans le groupe de suivi conventionnel.
Background
Cardiac syncope doubles the risk of all-cause death, while increasing the risk of fatal and non-fatal cardiovascular events . Soteriades et al. demonstrated that subjects with syncope of unknown aetiology comprise a heterogeneous patient group with an increased risk of death . The current approach to investigating patients with unexplained syncope involves short-term electrocardiographic monitoring or tests such as head-up tilt testing and electrophysiological study (EPS) . Recent advances in long-term monitoring with an implantable loop recorder (ILR) have enabled clinicians to obtain a correlation between symptoms and rhythm in the majority of patients . Despite the number of published reports on ILRs, only a few randomized studies are available, with their major limitations being small sample size and short-term follow-up . Furthermore, most published trials included patients with recurrent syncope, with three to seven syncope episodes .
In patients with both syncope and bundle branch block (BBB), syncope is suspected to be attributed to atrioventricular (AV) block, with EPS being able to predict the development of AV block in 87% of patients . In patients with BBB and negative EPS, the risk of developing a stable AV block was shown to be close to 20% after 4 years, with the risk of syncope recurrence being close to 40% at 3 years . Given this clinical setting, Brignole et al. found that syncope recurrences were mainly accounted for by paroxysmal AV block, with the risk estimated at 34% at 15 months (27% incidence/year), suggesting that some EPS results were, in fact, false negatives . Accordingly, an ILR was shown to be able to establish a symptom-rhythm correlation in most patients with recurrent syncope . To date, there are little data available regarding patients with a first syncope episode, BBB and negative workup, including EPS. Moreover, in this subset of patients with initial negative EPS results, no randomized studies have compared a conventional monitoring strategy (clinical, electrocardiogram and standard Holter monitoring) with ILR following a first syncope episode, despite the European Society of Cardiology guidelines of 2009 . In these guidelines, patients with left BBB are considered at high risk and syncope in these patients should lead to an ILR (class I, level B) . The recommendation for patients with syncope and left BBB could lead to pacemaker implantation but the level of proof is considered as class IIa, level C . A recent paper underlined the role of ILRs in a subset of patients despite negative EPS but patients could have had more than one syncope episode .
The aim of this multicentre prospective study was to compare two syncope evaluation strategies–namely, the use of an ILR (Group 1) versus conventional follow-up (Group 2)–in a population of patients with BBB and negative EPS following a first syncope episode, in addition to assessing the prevalence of significant arrhythmic events in the ILR patient subset.
Methods
Study population
The study protocol was approved by the Institutional Research Board of the Saint-Étienne Hospital and the Ethics Committee in October 2004. The study was supported by the Ministère Français de la Santé (Projet Hospitalier de Recherche Clinique, 2003) and Saint-Étienne University Hospital.
Inclusion criteria
Patients admitted after one syncope episode were invited to participate in this prospective randomized trial comparing two diagnostic approaches to syncope. Consecutive patients were included in the trial if they met the following inclusion criteria: single syncopal episode associated with any type of BBB with QRS greater or equal to 120 ms; no evidence of second- or third-degree AV block; and negative workup including EPS. Prior to enrolment, patients underwent clinical assessment, involving postural blood pressure testing, baseline ambulatory monitoring or inpatient telemetry for at least 24 hours and transthoracic echocardiogram. Baseline Holter assessment was considered negative if patients did not experience syncope or presyncope reminiscent of their referral symptoms during the recording and if there was no evidence of the following: asymptomatic second- or third-degree AV block; pauses of at least 3 seconds; sustained supraventricular tachycardia or greater or equal to 10 beats of wide QRS complex tachycardia likely to represent ventricular tachycardia. Before enrolment, additional neurological or cardiovascular testing was performed by the referring physician, although this was not mandatory according to the protocol. Patients were excluded if one of the following conditions was found: left ventricular ejection fraction (LVEF) less or equal to 35%; unlikelihood of surviving 1 year; or inability to attend follow-up or give informed consent. Patients with LVEF less than 60% were considered as having cardiomyopathy and thus were included. The difference between ischaemic cardiomyopathy and dilated cardiomyopathy was based on the coronary angiography values. Patients with a typical presentation of neurally mediated syncope at baseline were diagnosed as such and excluded from participating in the study. The clinical symptoms of neurally mediated syncope were induced by upright posture, with a prodromal phase, including the feeling of warmth and excessive sweating, followed by postepisode complaints of fatigue. The EPS included the measurement of the sinus node recovery time in addition to the measurement of the His to ventricle (HV) interval at baseline and under stress during incremental atrial pacing, although if the baseline assessment was inconclusive, the EPS was continued with pharmacological provocation using a slow infusion of ajmaline (1 mg/kg intravenously). Furthermore, the EPS involved the assessment of the inducibility of ventricular arrhythmia by programmed ventricular stimulation and supraventricular arrhythmia by any atrial stimulation protocol.
Exclusion criteria
In line with the published literature, the EPS was considered diagnostic and resulted in patients being excluded from the study if one of the following criteria was met: sinus bradycardia and abnormal sinus node recovery time; baseline HV interval greater or equal to 70 ms, second- or third-degree His-Purkinje block shown during incremental atrial pacing or high-degree His-Purkinje block provoked by intravenous administration of ajmaline; induction of sustained monomorphic ventricular tachycardia; induction of rapid supraventricular arrhythmia associated with hypotensive or spontaneous symptoms; carotid sinus hypersensitivity; symptomatic orthostatic hypotension diagnosed by standing blood pressure measurement; subclavian steal syndrome.
ILR group
Patients randomly assigned to the prolonged monitoring strategy were implanted with the Reveal ILR (Medtronic model 9526 Reveal Plus, replaced by model 9528 Reveal DX after 2008 and, more recently, by model XT Reveal; Minneapolis, MN, USA) in the left upper chest region under local anaesthesia following intravenous administration of 1 g of cefazolin. The ILR was a continuous electrocardiogram monitor capable of providing spontaneous automatic single-lead electrocardiogram recordings for up to 42 minutes. If patients experienced spontaneous symptoms, they were invited to press the button in order to ‘freeze’ the prior electrocardiogram recording, which was downloaded using a standard pacemaker programmer (Medtronic 9290C). After implantation, the patients, along with their family members or friends, were instructed in how to use the activator. The recommended programme mode involved one manual event and 13 automatic events for 42 minutes of storage. The resulting memory configuration meant that the automated backup of manual activations was able to detect bradycardia or pauses in addition to prespecified extreme rates or pauses (typically < 30 beats/min, > 160 beats/min and pauses > 3 s). Patients were told to activate the device after each syncope episode. In the absence of recurrent symptoms, the device detected any asymptomatic heart rate changes that were likely to provide clinical insights into the potential causes of the syncope . After ILR implantation, patients had follow-up visits every 3 months until the first symptomatic or asymptomatic episode documented by electrocardiogram or until 36 months. The mechanism of syncope was designated by the endpoints as defined by the committee members, who analysed the full set of all episodes. In the case of battery depletion prior to the study end, a second ILR was implanted.
Conventional strategy group
Patients randomly assigned to conventional follow-up were seen in the outpatient department at 3, 6, 12, 15, 18, 21, 24, 27, 30 and 33 months after randomization and at the study end (36 months). At each visit, arrhythmic or cardiovascular events were recorded and a 12-lead electrocardiogram was obtained, with follow-up continued in order to record any additional endpoints other than the initial endpoint. At each visit, a Holter monitor was used for 7 days, with analyses performed using the R.Test Evolution (RTE) event recorder (Novacor, Rueil Malmaison, France) and two electrodes placed on the patient’s body . The RTE event recorder ensured continuous electrocardiogram analysis, with any abnormal events being automatically stored in a 20-minute solid-state memory, which was autonomous for up to 7 days. Additionally, the patient was able to trigger the Holter manually . The RTE was programmed to recognize 10 types of arrhythmic events and one category of ischaemic event . The patients were instructed to report any clinical abnormality that occurred during the recording, by providing a detailed description and precise frequency of clinical symptoms. All recordings were analysed by two independent observers, with a third being used in case of discrepancies.
Endpoints
Clinically significant symptomatic and asymptomatic arrhythmias were defined as follows: pause more than 5 seconds; third-degree AV block; heart rate less than 30 beats/min for more than 10 seconds while awake; more than 10 beats of wide complex tachycardia consistent with ventricular tachycardia; and more than 30 beats of narrow complex tachycardia more than 165 beats/min. Borderline asymptomatic arrhythmias were defined as follows: less than 10 seconds of second- or third-degree AV block (Mobitz II); or heart rate less than 30 beats/min for less than 10 seconds while asleep. These endpoints were based on the guidelines from the European Society of Cardiology, the American College of Cardiology, the American Heart Association and the North American Society of Pacing and Electrophysiology regarding the implantation of cardiac pacemakers and antiarrhythmia devices . The primary study endpoint was reached when a symptom-rhythm correlation was obtained by manual activation of the device after spontaneous symptoms or when a prespecified significant asymptomatic arrhythmia was observed. Patients with borderline arrhythmias continued to be monitored without intervention. The primary endpoint was the time to the occurrence of significant symptomatic or asymptomatic events such as those defined above, thus requiring the implantation of a pacing or antiarrhythmic device based on the aforementioned guidelines .
Statistical analysis
Baseline patient characteristics were compared between the ILR group and the conventional group using Fisher’s exact test for categorical variables and the two-sample t -test for continuous variables, as appropriate. Summary values were reported as proportions and means ± standard deviations. For all time-to-event analyses, rates were estimated using the Kaplan-Meier method and compared by the log-rank test. Cox regression was used to calculate the hazard ratio and 95% confidence interval of risk of events in a first model and risk of AV block III in a second model between the ILR group and conventional group. Crude and adjusted hazard ratios were presented. Patient data were censored at the time of last follow-up, withdrawal from the study or non-rhythmic death. Symptomatic and asymptomatic events episodes were recorded, with a safety monitoring board reviewing the recordings. All reported levels of significance were two-sided. A probability value of P < 0.05 was considered statistically significant. It was estimated that 80 patients would need to be enrolled to detect a 20% event-difference at 2 years in favour of the ILR group, with 80% power, a two-sided 0.05 α level and a two-sided 0.20 β level, assuming that 5% of patients would be lost to follow-up in this elderly population. Statistical procedures were performed using SPSS for Windows (version 15.0) and Statview. All authors had full access to the data, take responsibility for its integrity and have read and approved the final manuscript.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
