OUP user menu

Implantable loop recorder for recurrent syncope: influence of cardiac conduction abnormalities showing up on resting electrocardiogram and of underlying cardiac disease on follow-up developments

Bertrand Pierre, Laurent Fauchier, Guillaume Breard, Olivier Marie, Philippe Poret, Dominique Babuty
DOI: http://dx.doi.org/10.1093/europace/eun039 477-481 First published online: 5 March 2008


Aims The implantable loop recorder is a useful diagnostic tool in dealing with recurrent syncope in patients. We tested to determine the influence of cardiac conduction abnormalities that turn up on resting electrocardiogram (ECG) and the impact of underlying cardiac disease on developments during follow-up.

Methods and results Ninety-five consecutive patients received an implantable loop recorder to monitor recurrent syncope (n = 4.9±3.8) of unknown aetiology after cardiac investigations, including an electrophysiological study. Resting ECG was abnormal, suggesting an arrhythmic syncope, in 29 (30.5%) patients, while 21 (22.1%) patients had an underlying cardiac disease. During an average follow-up period of 10.2±5.2 months, 43 (45.2%) patients developed a new syncope associated in 27 of them (62.8%) with an arrhythmic event. Syncope was no more frequent in the subgroup of patients with cardiac conduction abnormalities on resting ECG, while the frequency of arrhythmic events was similar whether or not the ECG was normal. In the subgroup of patients with cardiac disease with normal left ventricular ejection fraction, the occurrence of syncope was less frequent, and the number of arrhythmic events was no greater in these patients.

Conclusion Implantable loop recorder is a useful diagnostic tool for recurrent syncope of unknown aetiology.

  • Syncope
  • Implantable loop recorder
  • Bundle branch block
  • Cardiac disease
  • Atrioventricular block
  • Sinus arrest

Syncope is frequent in the general population,1 and the identification of cardiac aetiologies is mandatory in view of their poor prognosis.1 Despite evaluation that includes tilt-table testing and electrophysiological studies, at least 30% of recurrent syncopes remain unexplained.2,3 It has therefore been proposed to implant loop recorders to monitor cardiac rhythm in these patients for ∼14 months.4,5 The relevance of this new diagnostic tool has been clearly demonstrated, because at least 50% of patients suffer a new syncope, corresponding to ∼50% of the time to an arrhythmic event.2,3,68 However, some authors have suggested that the arrhythmic events can be underestimated in patients with bundle branch block on resting electrocardiogram (ECG).9 In the present study, we tested the impact of cardiac conduction abnormalities appearing on resting ECG and also of the presence of underlying cardiac disease on the recurrence and origin of syncope.


From July 1999 to January 2006, 95 consecutive patients complaining of recurrent syncope (at least three episodes) received an implantable loop recorder (Reveal Plus, Medtronic, Minneapolis, MN, USA) after cardiac and neurological evaluation. The initial evaluation included normal clinical examination, verification of the absence of orthostatic hypotension, normal 24-h ambulatory ECG recording, negative tilt-table testing, and a standard neurological appraisal. The criteria for ECG abnormalities suggesting an arrhythmic syncope, as set out in the guidelines of the European Society of Cardiology,10 were used to assess abnormal resting ECG. The strategy was to push diagnostic workup until the syncope mechanism was clearly documented. Every patient was subjected to an electrophysiological study to confirm normal infra-Hisian conduction (HV interval < 70 ms), a Wenckebach point during incremental atrial pacing >150 bpm, normal corrected sinus node recovery time (<550 ms), normal response to sinus carotid massage, absence of inducible ventricular and supraventricular arrhythmia. The presence of cardiac disease was determined through clinical examination and ECG. Patients with a severely depressed left ventricular ejection fraction (LVEF) (<30–35%) and candidates for a prophylactic implantable-cardioverter defibrillator (ICD) were excluded. The loop recorder was implanted subcutaneously, and the device was programmed to record an ECG for 6 min before and 1 min after activation. The device could be activated manually by the patient after syncope or automatically on the occurrence of bradycardia <30 min, ventricular arrest >3 s, or tachycardia >180 bpm during 32 beats. The total memory capacity of the device was set at 42 min. Patients were checked every 3 months until a new syncope occurred, marking the endpoint of the study. The device was explanted after the primary endpoint was reached or at the end of the life of the device (14 months as recommended by the maker of the implantable loop recorder).

Statistical analyses

Comparisons between groups were made using the Student's t-test for continuous variables. ANOVA and chi-square or Fisher exact tests were performed when appropriate to test for statistical differences. A P-value < 0.05 was considered statistically significant. When applicable, data were expressed as mean ± SD. The time to occurrence of clinical events was analysed using the Kaplan–Meier method. Statview 4.5 (Abacus Concepts, Berkeley, CA, USA) was used for statistical analyses.


Of the 95 patients included, 29 (30.5%) had cardiac conduction abnormalities suggesting an arrhythmic syncope on resting ECG, and 21 (22.1%) presented cardiac disease. LVEF was relatively well preserved (>35%) in all patients. The characteristics of the patients are summarized in Table 1. Forty-three patients (45.2%) developed a new syncope during an average follow-up period of 10.2 ± 5.2 months. The interval between the implantation of the loop recorder and the recurrence of syncope was 5.4 ± 4.6 months. The actuarial estimate is 78.9, 72.6, 66.3, 54.7% at 3, 6, 9, and 14 months (Figure 1A). An arrhythmic event was observed in 27 of the 43 patients with recurrent syncope (62.8%): complete atrioventricular block (n = 5), ventricular fibrillation (n = 1), sustained ventricular tachycardia (n = 2), rapid non-sustained ventricular tachycardia (224 bpm) (n = 1), atrial fibrillation with fast ventricular response (n = 1), supraventricular tachycardia (n = 1), sinus arrest (n = 16). Twenty patients received a permanent pacemaker for complete atrioventricular block or sinus arrest; one sinus arrest was successfully treated with theophylline in a young patient, and three patients received an ICD. Two patients were treated with a class I antiarrhythmic agent or beta-blockers for supraventricular arrhythmia. In 16 patients (37.2%), the event was considered non-arrhythmic because no bradycardia or tachyarrhythmia was recorded by the implantable loop recorder. Three of them, considered neuromediated syncope because of sinus tachycardia recorded just before syncope, were treated successfully with lifestyle changes, tilt training, or beta-blockers. No patient complained of a new syncope during a further follow-up period of 8.5 ± 4.9 months (minimum 1, maximum 18 months).

Figure 1

Recurrence of syncope. Event-free curves: (A) All patients; (B) comparison of patients according to cardiac conduction abnormalities on resting electrocardiogram; (C) comparison of patients according to underlying cardiac disease.

View this table:
Table 1

Clinical characteristics of patients (n = 95)

Age (mean ± SD, in years)64.3 ± 17.3
Male, n (%)57 (60)
Syncope episodes (mean ± SD)4.9 ± 3.8
Electrocardiogram abnormalities suggesting an arrhythmic syncope, n (%)29 (30.5)
Bifascicular block (left bundle branch block or right bundle branch block combined with left anterior fascicular block)12
Right bundle branch block with QRS duration >0.12 s17
Underlying cardiac disease, n (%)21 (22.1)
Coronary artery disease with heart failure10
Coronary artery disease with myocardial infarctus7
Aortic stenosis3
Congenital cardiac disease1

Fifty-four patients (56.8%) showed no cardiac disease or cardiac conduction abnormalities. Twenty-seven of these patients (50%) complained of recurrent syncope, and 18 of them (33.3%) suffered an observed arrhythmic event. The most frequent arrhythmic event was sinus arrest (n = 13).

Nine patients (9.4%) showed cardiac disease and cardiac conduction abnormalities. Two of these patients (22.2%) complained of recurrent syncope, and one of them suffered a complete atrioventricular block.

Surprisingly, the recurrence of syncope tended to be less frequent in the subgroup of patients with cardiac conduction abnormalities on resting ECG (Figure 1B), and the rate of arrhythmic events (notably complete atrioventricular block, occurring in only 13.7% of patients with ECG abnormalities, Table 2) was not significantly different in this subgroup of patients (Figure 2B). The rate of sinus arrest was higher in the subgroup of patients without cardiac conduction abnormalities but the difference is not significant (13 vs. 3, P = 0.13).

Figure 2

Recurrence of arrhythmic events. Event-free curves: (A) all patients; (B) comparison of patients according to cardiac conduction abnormalities on resting electrocardiogram; (C) comparison of patients according to underlying cardiac disease.

View this table:
Table 2

Events stored by implantable loop recorder (n = 95)

PatientsTotalResting electrocardiogramCardiac disease
n (%)9529 (30.5)66 (69.4)21 (22.1)74 (77.8)
Age (mean ± SD)64.3 ± 17.374.1 ± 1360.1 ± 1770.1 ± 1662.7 ± 17
Syncope recurrence, n (%)43 (45.2)12 (41.3)31 (46.9)ns6 (28.5)37 (50)ns
Arrhythmic event, n (%)27 (28.4)8 (27.5)19 (28.7)ns2 (9.5)25 (33.7)P = 0.02
Complete atrioventricular block54114
Ventricular fibrillation10101
Sustained ventricular tachycardia21102
Non-sustained ventricular tachycardia10110
Atrial fibrillation10101
Supraventricular tachycardia10101
Sinus arrest16313016

Syncope recurred less frequently in the subgroup of patients with cardiac disease than in the subgroup of patients without cardiac disease (Figure 1C). The rate of arrhythmic events was significantly lower in the subgroup of patients with cardiac disease (Figure 2C, Table 2). No sinus arrest was documented in this subgroup of patients. One 76-year-old patient with a normal ECG, normal LVEF, and no spontaneous or inducible arrhythmia died of a ventricular fibrillation recorded by the device.


This study confirms the usefulness of the implantable loop recorder in monitoring recurrent syncope in the particular setting of patients having undergone a complete cardiological evaluation, including an electrophysiological study and who are considered at low risk for cardiac events. Forty-five percent of the patients developed a new syncope, and 28.4% an arrhythmic syncope. These results are comparable to those previously reported.3,6 Interestingly, we found that cardiac events in the subgroup of patients with cardiac conduction abnormalities on resting ECG were no more frequent than in patients showing no cardiac conduction abnormalities. The fact that these results contradict those obtained by Brignole et al.,9 who reported a high percentage (77%) of complete atrioventricular block, points to an empirical therapeutic strategy with pacing for this subgroup of patients. These discrepancies can be explained through a difference in how the population was selected. In our study, the patients had suffered at least three episodes of syncope without documented arrhythmia, and infra-Hisian conduction time (HV interval) was always normal. In Brignole's study, 42% of the patients had bifascicular block consisting of right bundle branch block (RBBB) associated with right axis deviation, considered severe atrioventricular conduction abnormalities exposing patients to complete atrioventricular block.9 In our study, the number of patients with bifascicular block was comparable (n = 12; 41.3%) and characterized by an RBBB and left axis deviation but never by right axis deviation. At the present time, patients with recurrent syncope and RBBB associated with right axis deviation are usually implanted with a permanent pacemaker. This may explain the very high percentage of complete atrioventricular block in Brignole’s study, a percentage which is not reported by any authors using the implantable loop recorder as a diagnostic tool for recurrent syncope.3,6 The most frequent bradycardia mechanism reported by others is sinus arrest or sinus bradycardia, as is also the case in our study (76.1%). This type of bradycardia could be due to sick sinus syndrome or neuromediated vagal syncope. In our study, the probability of a neuromediated syncope is low because syncope history evoked no vasovagal syncope. All patients underwent negative tilt-table testing, and the occurrence of sinus arrest was abrupt, unpreceded by progressively slowing heart rate on the implantable loop recorder recording. Moreover, these patients with sick sinus syndrome received a permanent pacemaker that proved to be beneficial in lowering the recurrence of syncope, whereas there is general agreement that implantation of a cardiac pacemaker is ineffective on vasovagal syncope. In our study, the implantable loop recorder appears to be a useful tool for diagnosing sick sinus syndrome, especially, if there are no cardiac conduction abnormalities on resting ECG and no cardiac disease. The electrophysiological study and 24-h ambulatory ECG recording are inadequate for the diagnosis of sick sinus syndrome in this subgroup of patients.

In patients with cardiac disease, syncope can be related to complete atrioventricular block or severe arrhythmia such as ventricular tachycardia.1 Therefore, ICD implantation may be proposed for patients suffering from recurrent syncope and showing altered LVEF.11 The work of other authors8,12 suggests that there are two main reasons for our failure to find frequent arrhythmic events combined with underlying cardiac disease: LVEF was unaltered in most of our patients, and programmed ventricular stimulation did not induce ventricular tachycardia. The exclusion of patients with a high risk of ventricular arrhythmia may be explained by the relatively high propensity of programmed ventricular stimulation to discern the arrhythmic risk13 and by the prophylactic implantation of an ICD in these patients, especially in those whose LVEF was altered.11


The implantable loop recorder is a useful diagnostic tool for recurrent syncope of unknown aetiology in patients with or without cardiac conduction abnormalities or cardiac disease. The absence of arrhythmic events was frequently reported in all patient subgroups. This argues against an empirical pacing strategy in patients with cardiac conduction abnormalities on resting ECG who are suffering from recurrent syncope, but whose electrophysiological study tests are normal. Therefore, the implantation of a loop recorder could be extended to patients with recurrent syncope and cardiac disease when the LVEF is relatively well preserved (>35%) and when right programmed ventricular stimulation is negative.


Funding to pay the Open Access publication charges for this article was provided by Medtronic.


The authors are indebted to Mr Steve Randel, native English assistant, for re-reading this manuscript.

Conflict of interest: none declared.


View Abstract