OUP user menu

Electrocardiographic characteristics of atrioventricular block induced by tilt testing

Dorota Zyśko, Jacek Gajek, Edward Koźluk, Walentyna Mazurek
DOI: http://dx.doi.org/10.1093/europace/eun299 225-230 First published online: 5 November 2008

Abstract

Aims The electrocardiographic (ECG) characteristics of atrioventricular (AV) block during reflex syncope may be unique due to the presence of hypervagotonia. The aim of the present study was to define the ECG characteristics of the AV block induced by neurocardiogenic reflex provoked by tilt testing (TT).

Methods and results A series of 31 patients with presumed vasovagal syncope and AV block provoked by TT was studied. The duration of PP and PR interval, AV block grade and type, concomitant arrhythmias, and timing of the AV block occurrence were assessed. The AV block occurred at TT termination in 26 patients, in the recovery in 4 patients, and in both periods in 1 patient. Atrioventricular block was preceded by sinus slowing, and sinus rhythm during AV block was slow and instable. Mobitz I, 2:1 second-degree AV block, and advanced second-degree AV block were recognized in 35.5, 48.4, and 67.8% of patients, respectively. Third-degree AV block was diagnosed in 41.9% of patients. Twenty-one patients had at least two AV block forms. The most prevalent concomitant arrhythmia was junctional escape rhythm (61.3%).

Conclusion (i) The occurrence of the AV block during neurocardiogenic reaction induced by TT is always preceded by sinus rhythm slowing and usually by PR interval prolongation. (ii) The AV block provoked by TT usually occurs at TT termination, but may occur even in the recovery period in a supine position. Sometimes the AV block may be present both at TT termination and during the recovery period.

Keywords
  • Tilt testing
  • Vasovagal syncope
  • Atrioventricular block

Introduction

Syncope is a common medical problem and, if related to the atrioventricular (AV) conduction disturbances, may be an indication for permanent pacemaker implantation.1 Transient AV block is sometimes recorded during prolonged electrocardiographic (ECG) recordings. When it occurs, distinguishing between AV block that is due to degenerative disease of the conduction system and functional, vagally induced AV block due to neurocardiogenic reaction has important prognostic and therapeutic implications. Atrioventricular block during tilt testing (TT) is a rare finding.24 The majority of reports in the literature present only a single case of ECG characteristics during spontaneous or provoked reflex syncope.57 It is therefore important to further describe the ECG characteristics of the vagally induced AV block.

The aim of the study was to present the ECG characteristics of the AV block induced by TT in patients with vasovagal syncope.

Methods

Population

A series of 31 consecutive patients with AV block provoked by TT out of 786 subjects were referred to our laboratory for TT in the period March 1999 to March 2008. The clinical variables of patients and ECG characteristics of the developed AV block during TT were analysed.

Tilt testing protocol and measurements

The following equipment was used: motorized tilt table (with mattress and footrest), Oxford 4500 ECG Holter recorder, and ECG cardiomonitor. Tilt testing was performed in the morning, according to the Italian protocol (passive phase lasting 20 min and active phase after pharmacological provocation lasting 15 min) with sublingual administration of 400 µg nitroglycerin, if syncope did not develop during the initial passive phase.8 The phase of TT when the patient fainted was noted. Electrocardiographic Holter recordings were automatically processed on the Medilog Optima Jet system from Oxford. This procedure was followed by a careful manual review to limit any potential artefacts.

Each patient was lying supine at least 30 min before the onset of TT and at least 10 min after TT termination. The ECG Holter monitoring was started at least 15 min before the onset of TT and continued at least 60 min after the TT was completed.

The occurrence of AV block during TT or in the recovery period was noted.

Definitions

The AV conduction disturbances were classified as second- and third-degree AV blocks.9,10 Forms of the second-degree AV block:

  1. Mobitz I second-degree AV block.

  2. 2:1 AV block—one conducted P-wave for each P-wave blocked.

  3. Mobitz II second-degree AV block—occurrence of the single non-conducted P-wave.

  4. Advanced AV block—two or more consecutive P-waves are blocked.

The instability of the sinus rhythm during parasympathetic activation in the setting of neurocardiogenic reaction made the classical definitions of second- and third-degree AV block of no use. The short sequences of AV conduction disturbances did not help differentiate between advanced second-degree and third-degree AV blocks. The third-degree AV block can most probably be diagnosed if there are non-conducted P-waves as well as junctional or ventricular escape rhythms without the signs of retrograde conduction.

Electrocardiographic analysis

The term asystole referred to no QRS complexes and no P-waves for at least 3 s. If P-waves were observed during this period, it was called ventricular asystole with non-conducted P-waves. When sinus arrest with no escape rhythm was present for a period of at least 3 s after a blocked P-wave, it was called asystole after blocked P-wave.

The following ECG variables were measured:

  1. PP and PR intervals at the following time points:

    • at the baseline,

    • at the maximal heart rate during TT before heart rate slowing,

    • before AV block (the last conducted P-wave before AV block),

    • after AV block (the first conducted P-wave).

  2. The shortest and the longest PP intervals during AV block not interrupted by any QRS complex.

  3. The longest difference between two consecutive PP intervals during AV block.

The following features were assessed to describe the ECG characteristics of the AV block induced by TT:

  • the type of AV block,

  • the presence of an asystole, the presence of ventricular asystole with non-conducted P-waves, and the presence of an asystole after blocked P-waves,

  • the presence of non-conducted P-waves during escape rhythm,

  • number of non-conducted P-waves,

  • the presence of junctional or ventricular escape rhythms,

  • the duration of heart rate below 40 per min before escape rhythm emerged.

Statistical analysis

Continuous variables were presented as mean ± standard deviation or median with inter-quartile range as appropriate; discrete variables were presented as percentages.

For normally distributed data, analysis of variance (ANOVA) was used to test for significant differences among variables. Post hoc multiple t-test comparison was then performed to determine the variables that were significantly different and a Bonferroni correction of the P-value significance was applied to adjust to the problem of inflating the type-I error. For non-normally distributed data, the Kruskal–Wallis ANOVA was used to test for statistical difference among variables. For pair-by-pair comparison, Mann–Whitney U or Wilcoxon tests were used as appropriate.

Discrete variables were compared with χ2 test with Yates correction if necessary. A P-value of <0.05 was considered significant.

Results

The study group with the AV block constituted 3.9% of the whole examined population and 5.2% of the patients with positive TT. In this group, 17 patients (2.2 and 2.9%, respectively) had only the AV block without sinus arrest. The asystolic response without the AV block was present in 17.7% of all patients and in 23.3% of the patients with positive TT.

Patient characteristics

The demographics, clinical, and TT data of the study group are presented in Table 1.

View this table:
Table 1

Demographics, clinical, and tilt testing data in the studied population with tilt testing-induced atrioventricular block

Number of patients31
Age (years)44.5 ± 19.9
Female gender (%)74.2
Number of syncopal spells (median, inter-quartile range)3 (1–12)
Traumatic injuries related to syncope (%)38.7
Treatment with β-blocker (%)3.2
Arterial hypertension (%)32.3
Diabetes mellitus (%)3.2
Previous myocardial infarction (%)3.2
First-degree atrioventricular block at baseline (%)19.4
Nitroglycerine provocation (%)77.4
Syncope at tilt testing termination (%)96.8
Reproduction of the spontaneous symptoms (%)87.1

Timing and duration of atrioventricular block provoked by tilt testing

The time of the AV block occurrence was TT termination in 26 patients, TT termination and recovery period in 1 patient, and the recovery period only in 4 patients. When the AV block occurred at TT termination, it was accompanied by syncope in each patient. In patients with the AV block during recovery, presyncope only was present in four of five patients.

Electrocardiographic characteristics

The ECG characteristics during AV block are listed in Table 2.

View this table:
Table 2

The electrocardiographic characteristics during atrioventricular block

AV block at TT termination (%)87.1
AV block duration (s)47.6 ± 77.2 (6–360)
Mobitz I AV block (%)35.5
2:1 second-degree AV block (%)48.4
Advanced AV block (%)67.8
Third-degree AV block (%)41.9
Total number of non-conducted P-waves (n)10.4 ± 7.4 (1–30)
Presence of asystole (%)45.2
Presence of asystole after blocked P-wave (%)19.4
Presence of ventricular asystole with non-conducted P-waves (%)58.1
Junctional escape rhythm (%)61.3
Ventricular escape rhythm (%)3.2
The duration of heart rate below 40 per min before escape rhythm (s)11.9 ± 11.6
The longest PP interval during AV block (ms) not interrupted by any QRS complexes (n = 28)1559 ± 531
The shortest PP interval during AV block (ms) not interrupted by any QRS complexes (n = 23)1006 ± 267
The longest difference between the duration of two consecutive PP intervals during AV block (ms)326 ± 278
  • AV, atrioventricular and TT, tilt testing.

In our study group, the most prevalent ECG pattern during neurocardiogenic reaction was sinus rhythm slowing and PR interval prolongation until Mobitz I AV block occurred. Then advanced AV block followed, presenting as ventricular asystole with non-conducted P-waves, which often terminated in asystole. The other common feature of an AV block during TT was 1:1 conduction separated by various forms of second-degree AV blocks. In 10 patients, one form of AV block was observed and so were two forms in 14 patients. Three and four forms of AV blocks were present in six and one patient, respectively.

Classification of atrioventricular block

The sinus rhythm slowed down before an AV block occurred in each patient and it was instable during the AV block. This is shown in Tables 2 and 3. The lack of a stable sinus rhythm indicated a vagally mediated AV block. The PR interval increased before the AV block occurrence in 30 of 31 patients. In one patient, it did not differ from its value just before the onset of the heart rate slowing. The frequency of AV block forms is presented in Table 2. The representative ECG Holter recordings of observed AV conduction disturbances are shown in Figures 13.

Figure 1

Electrocardiographic Holter recordings at 12.5 mm/s, 3 consecutive strips. The electrocardiographic tracing begins with sinus rhythm slowing, PR prolongation and then 2:1 (48.4% of patients) atrioventricular block occurs, ending with Mobitz I (35.5% of patients) second-degree atrioventricular block.

Figure 2

Electrocardiographic Holter recordings at 12.5 mm/s, 3 consecutive strips. The electrocardiographic tracing begins with sinus rhythm slowing, PR prolongation and then advanced (67.8% of patients) atrioventricular block occurs, terminating with 2:1 second-degree atrioventricular block.

Figure 3

Electrocardiographic Holter recordings at 12.5 mm/s, 3 consecutive strips. The electrocardiographic tracing begins with sinus rhythm slowing, PR prolongation and then Mobitz I atrioventricular block occurs, terminating with third-degree (41.9%) atrioventricular block with ventricular escape rhythm.

View this table:
Table 3

Electrocardiographic parameters in patients with atrioventricular block induced by tilt testing

PP interval (ms)PR interval (ms)
Baseline909 ± 132178 ± 40
Maximal heart rate before slowing605 ± 121*147 ± 26*
Before last conducted P-wave1105 ± 443*,**239 ± 76*,**
After atrioventricular block n = 301130 ± 296*,**186 ± 53**,***
  • Baseline, maximal, before atrioventricular block, and after atrioventricular block, respectively, the PP and PR interval at the baseline conditions, at the maximal heart rate before slowing onset, just before atrioventricular block, and the first conducted P-wave after atrioventricular block.

  • * P < 0.001 vs. baseline.

  • **P < 0.001 vs. maximal heart rate before slowing.

  • ***P < 0.001 vs. before AV block.

Electrocardiographic parameters

The assessed ECG parameters are shown in Table 3.

Discussion

The AV block during the neurocardiogenic reflex is a rare phenomenon. In daily life circumstances, it was documented during vasovagal and situational syncope.6,7,11 The AV block induced by TT was reported both in vasovagal patients and in healthy volunteers.4,7,11,12

Incidence of atrioventricular block provoked by tilt testing

The AV block provoked by TT represents the cardioinhibitory component of the neurocardiogenic reflex that had lower incidence than sinus arrest, sinoatrial block, or sinus bradycardia. Although vagally mediated reflex exerts an effect on the sinoatrial and AV nodes, the effect of vagal stimulation on sinoatrial node usually predominates, thus masking the effect of vagal stimulation on the AV node that makes the paroxysmal AV block uncommon. The sinus arrest especially does not allow AV conduction disturbances to be recognized because of the lack of atrial electrical activity. In our study, the incidence of the AV block in patients with positive TT was as low as in Vanerio et al.5 and Brignole et al.13 studies.

The timing of atrioventricular block provoked by tilt testing

We observed the events of AV block provoked by TT both at TT termination and during the recovery phase with the higher incidence in the former period. During the recovery after vasovagal syncope, the predisposition to syncope recurrence is increased during daily life circumstances. Our findings regarding the AV block occurrence during the recovery period indicate the presence of the vagal predominance after reflex syncope, which could disturb the adaptation to an upright position. The presence of the AV block exclusively in the recovery phase is probably very rare and we did not find similar reports in the literature.

Syncope or presyncope during atrioventricular block induced by tilt testing

In our study, the loss of consciousness occurred in each case of AV block present at TT termination. The AV block occurring in the recovery phase was manifested much more frequently as presyncope. After the analysis of implantable loop recorder memory recordings, Lombardi et al.14 established that during presyncopal events one-third of patients had an advanced AV block. More frequent prevalence of presyncope as a clinical manifestation of the AV block in the study of Lombardi et al. may indicate the predominance of another aetiology of syncope than neurocardiogenic reflex in their population with AV blocks. The most important difference between these populations was the mean age of the groups—their patients were older than ours.

Electrocardiographic characteristics of atrioventricular block provoked by tilt testing

The first feature of AV blocks related to neurocardiogenic reflex provoked by TT was the substantial slowing of the sinus rhythm and the presence of huge differences among PP intervals. This observation is in line with the observation of Vanerio et al.5 who described a similar pattern of ECG characteristic of the AV conduction disturbances in their case series. Sinus bradycardia plus AV block was also present during syncopal episode documented by implantable loop recorder but the sudden onset of the AV block was shown almost at the same sinus rate.7 Our results clearly indicate the functional, vagal origin of TT-induced AV blocks. The practical implementation of the described ECG characteristics enables one to distinguish the vagally induced conduction disturbances from the organic disease.

In the literature, AV block during TT-induced syncope is often called the second- or the third-degree AV block although the most important criterion—the stable sinus rhythm—is not met. Sinus rhythm slowing and irregular sinus rhythm observed during the AV block due to neurocardiogenic reflex indicate its vagal origin although several consecutive non-conducted P-waves suggest advanced second-degree or third-degree AV block. This interpretation is concordant with the interpretation of other investigators such as Brignole et al. who considered that the association of sinus bradycardia and AV block at the same time strongly suggests vagal reflex be the mechanism responsible for the event.15 However, the level where the conduction disturbance occurs: atrial-His or His-ventricle, is not known. Carotid sinus massage—a manoeuvre that increases the vagal activity—could produce the prolongation of conduction or even its disruption on the atrial-His and His-ventricle level.16

The second feature of the AV conduction disturbances during neurocardiogenic reflex was their rather gradual than abrupt emergence. They were usually preceded by prolongation of the PR interval. Advanced second-degree AV block with at least 2 consecutive non-conducted P-waves was in 40% preceded by an event of Mobitz I or 2:1 second-degree AV block. Vanerio et al.5 observed the Wenckebach block and 2:1 AV block before the third-degree AV block in three of their six patients undergoing TT. The patients with other than neurocardiogenic reflex aetiology of the AV block do not have this feature. Deharo et al. described the intermittent third-degree AV block recorded by implantable loop recorder in three patients with negative TT and non-vasovagal medical history. The onset of AV conduction disturbances was sudden, with no change in the PP interval. In one case, an origin other than neurocardiogenic was confirmed by its maintenance for several hours, which is unusual for reflex syncope. The authors reported no Mobitz I block or 2:1 AV block before the third-degree AV block in those patients.11 The coexistence of the supranodal and infranodal AV conduction disturbances is extremely rarely found. In this case, the presence of the Mobitz I second-degree AV block can exclude the infranodal conduction disturbances.9

Our ECG characteristics of the TT-induced AV block resemble the spontaneous events observed by Castellanos et al. They described the asymptomatic events of two to six non-conducted P-waves concomitant with heart rate slowing during ECG Holter monitoring in vasovagal patients with positive TT. The increase in sinus rate or lack of decrease during the episode of the AV block argues against underlying neurocardiogenic reflex.17

The onset of the AV block may occur at the relatively high sinus rate before significant sinus slowing. Such a case was reported by Sra et al.18 Their patient had a sinus cycle of 480 ms at the AV block onset. In our case series, the shortest sinus cycle before the AV block was 554 ms, which was only 74 ms longer than described above. Nevertheless, the sinus cycle decreased before the AV block occurrence.

The third feature of the AV block observed in our study was a frequent coexistence of sinus arrest. Sinus arrest could follow or precede the AV block event that is the consequence of the coexistence of vagal influences on the sinoatrial as well as AV node.

The last feature of the AV block due to neurocardiogenic reflex, not commonly seen, is its manifestation not only as a single episode of the AV block but also sometimes as recurrent episodes lasting for several minutes even if the patient remains in a supine position after TT.

Limitations

The vasovagal syncope may overlap the sick sinus syndrome and paroxysmal AV conduction disturbances in up to 30% of patients. In our patients, with the AV block provoked by TT, electrophysiological AV conduction properties were not extensively studied. In the baseline conditions, 5 patients had the first-degree AV block with PR interval from 207 to 270 ms, which persevered to the maximal heart rate increase before sinus rhythm slowing during neurocardiogenic reflex in 2 patients. The 24 h ECG Holter monitoring in these patients did not reveal any episodes of the second or third-degree AV blocks. The transoesophageal atrial pacing was proposed to nine patients, and one of them refused. In two patients, minor abnormalities were found: in one person Mobitz I periodicity occurred at 120 per min and in another corrected sinus node recovery time was 650 ms, in the others we did not find any abnormalities.

Conclusions

  1. The occurrence of the AV block during neurocardiogenic reaction induced by TT is always preceded by sinus rhythm slowing and usually by PR interval prolongation.

  2. Atrioventricular block provoked by TT usually occurs at TT termination, but may emerge even in the recovery period in a supine position. Sometimes the AV block may be present both at TT termination and during the recovery period.

  3. Subsidiary pacemaker activity is present in more than 50% of the patients.

Conflict of interest: none declared.

References

View Abstract