International Journal of Cardiology

Transcription

1 International Journal of Cardiology 168 (2013) Contents lists available at ScienceDirect International Journal of Cardiology journal homepage: Ventriculoatrial conduction and related pacemaker-mediated arrhythmias in patients implanted for atrioventricular block: An old problem revisited Sergio Richter a,, Andreas Muessigbrodt a, Jozef Salmas a, Michael Doering a, Ulrike Wetzel a, Arash Arya a, Gerhard Hindricks a, Pedro Brugada b, Carsten W. Israel c a Department of Electrophysiology, Heart Center, University of Leipzig, Leipzig, Germany b Heart Rhythm Management Centre, Cardiovascular Centre, UZ Brussel (VUB), Brussels, Belgium c Department of Cardiology, Evangelical Hospital Bielefeld, Bielefeld, Germany article info abstract Article history: Received 6 August 2012 Received in revised form 9 February 2013 Accepted 6 April 2013 Available online 29 April 2013 Keywords: Atrioventricular block Implanted cardiac pacing device DDD mode Ventriculoatrial conduction Pacemaker-mediated tachyarrhythmias Introduction: Ventriculoatrial (VA) conduction and related pacemaker-mediated tachyarrhythmias (PMT) have not been systematically investigated in a large cohort of patients implanted for symptomatic atrioventricular (AV) block. Methods and results: Two hundred fifty consecutive patients (71 ± 14 years, 63% male) implanted for symptomatic second- or third-degree AV block were screened for retrograde VA conduction and related PMTs including endless loop tachycardia (ELT) and repetitive nonreentrant VA synchrony (RNRVAS). After a mean post-implantation period of 38 ± 12 months, AV block was persistent in 137 (55%) and variable in 113 (45%) patients. Retrograde 1:1 conduction was present in 76 patients (30%) with a mean VA conduction time of 258 ± 65 ms. The incidence of VA conduction varied considerably according to the presence and type of anterograde conduction block. Retrograde conduction was present in 24 of 137 patients (18%) with persistent AV block. Fifteen of the 76 patients (20%) with VA conduction had 1 documented PMT. The recorded arrhythmias were ELT in 11 and RNRVAS in 4 patients. VA conduction time was significantly longer in patients with than without PMT (297 ± 64 vs. 249 ± 62 ms, resp., P b 0.01). None of the patients without VA conduction had documented PMTs. Conclusions: Retrograde VA conduction and related PMTs are not uncommon in patients implanted for symptomatic AV block. Testing for retrograde conduction should therefore be performed in all patients with AV block in order to optimize device programming and prevent PMTs Elsevier Ireland Ltd. All rights reserved. 1. Introduction Retrograde ventriculoatrial (VA) conduction has been reported to be present in up to 50 70% of the cases [1 3]. In structurally and electrophysiologically normal hearts, retrograde VA conduction does usually not manifest clinically, and is merely of academic interest. However, in the presence of dual atrioventricular (AV) nodal pathways or an accessory AV pathway, the propensity to intrinsic retrograde VA conduction paves the way for a macroreentry utilizing the normal conduction system as retrograde limb of the circuit. Cardiac pacing devices operating in an atrial sensing mode provide an artificial second AV pathway and, similarly, may support reentry using the device for anterograde and the normal conduction system for retrograde conduction. This pacemakermediated reentrant tachycardia, so-called endless loop tachycardia (ELT), has been observed and understood early after introduction of dual-chamber pacing in subjects with intrinsic retrograde VA Corresponding author at: Department of Electrophysiology, Heart Center, University of Leipzig, Strümpellstr. 39, Leipzig, Germany. Tel.: ; fax: address: s.richter@med.uni-leipzig.de (S. Richter). conduction [4 6]. Understanding the reentrant circuit of this pacemaker-mediated tachyarrhythmia (PMT), it is tempting to assume that retrograde VA conduction and related PMTs may not be of clinical relevance in subjects with advanced anterograde AV block. However, the paradoxical phenomenon of retrograde VA conduction in complete heart block has long been recognized [7 9]. Small studies have demonstrated that this phenomenon is not infrequent at all: depending on the site of anterograde AV block, retrograde VA conduction was present in up to 25% in subjects with complete AV block [3,9]. However, data on retrograde VA conduction in patients with variable and persistent AV block is limited, and the incidence of related, clinically relevant PMTs in this selected patient population remains unknown. We therefore aimed to investigate anterograde and retrograde conduction characteristics and the susceptibility to PMTs in a large cohort of patients implanted for symptomatic AV block. 2. Methods 2.1. Study population Between October 2009 and September 2010, all consecutive patients undergoing routine follow-up of a dual-chamber pacing or cardiac resynchronization therapy (CRT) /$ see front matter 2013 Elsevier Ireland Ltd. All rights reserved.

2 S. Richter et al. / International Journal of Cardiology 168 (2013) device were screened for documented second- or third-degree AV block. Subjects were included in the present study if they met the following entry criteria: (1) documented second-orthird-degreeavblockattimeof device implantation, (2) implanted dual chamber or CRT cardiac pacing device with a functioning right atrial lead in place, (3) programmed to DDD(R) mode for the last 12 months, (4) absence of atrial tachyarrhythmias at time of evaluation, and (5) absence of overt ventricular preexcitation or known retrogradely conducting accessory pathway. Moreover, features for automatic atrial and ventricular threshold measurements (if available) needed to be disabled in the devices used in this study. All included subjects gave informed consent to participate in the registry. AV block was defined variable if a different type of AV conduction pattern was documented at time of testing for VA conduction. Subjects fulfilling the inclusion criteria were assessed for the presence and timing of VA conduction and related, clinically relevant PMTs including the classic type of ELT and repetitive nonreentrant VA synchrony (RNRVAS) [6,10,11]. Relevant clinical data including data on age, gender, underlying cardiovascular disease, implanted cardiac pacing device, preexisting conduction abnormalities (if recorded), and AV-node active antiarrhythmic drugs were gathered at time of evaluation. Subjects were classified according to the type of documented AV block at time of implantation, anterograde and retrograde conduction at evaluation, and documented PMTs Assessment of anterograde AV conduction Testing for anterograde AV conduction was performed in the VVI mode at the lowest programmable intervention rate (i.e. 30 beats per minute (bpm)) in devices from the manufactures St. Jude Medical, Biotronik, Medtronic, and Boston Scientific or in the inhibited mode in Medtronic devices (Fig. 1A). A 12-lead electrocardiogram (ECG) was recorded simultaneously to assess for AV conduction, intrinsic PR interval, and ventricular conduction pattern. In the presence of a Wenckebach-type second-degree AV block, the shortest intrinsic PR interval was taken for analysis. An escape rhythm was considered to be present if a regular, intrinsic ventricular activity of >30 bpm occurred dissociated from sinus rhythm during rhythm recording for at least 1 monitor length. In case of severe sinus bradycardia or atrial standstill, AV conduction was assessed in the AAI mode at 60 bpm Assessment of retrograde VA conduction Retrograde VA conduction was assessed in the VVI or VDI mode starting at a pacing rate of 10 bpm above the intrinsic sinus rate. To detect and analyze retrograde conduction, the devices were programmed temporarily to display the atrial electrogram during ventricular pacing. In subjects with retrograde conduction, the VA conduction time was measured from the ventricular pacing stimulus to the intrinsic atrial deflection in the atrial electrogram at the lowest ventricular pacing rate that produced 1:1 VA conduction (Fig. 1B). Incremental ventricular pacing to assess for VA Wenckebach conduction was performed as long as clinically tolerated up to a rate of 180 bpm Pacemaker-mediated arrhythmias All subjects were screened for documented PMTs that occurred spontaneously during follow-up or were inducible during atrial pacing threshold testing and/or ventricular pacing (Fig. 2). All spontaneous episodes with a stored electrogram classified as PMT were analyzed. All automatic PMT detections without a stored electrogram were discarded with respect to the analysis. To test for inducible, clinically relevant PMT, atrial pacing threshold testing was performed in the DDD mode using the programmed parameters that had previously been arranged according to the patient's clinical needs. A PMT was defined as a pacemaker-mediated arrhythmia depending on retrograde VA conduction upon a paced ventricular activation (VA synchrony). Two different forms of PMTs related to VA synchrony have been distinguished: (1) ELT or repetitive reentrant VA synchrony, the classic type of a pacemaker-mediated reentrant tachycardia, and (2) RNRVAS, the repetitive nonreentrant variant of VA synchrony (or so-called AV desynchronization arrhythmia). Electrophysiological features of ELT and RNRVAS have been described in detail elsewhere [5,6,10,11]. In addition, the mode of spontaneous arrhythmia initiation and termination by automatic pacemaker algorithms was analyzed and considered for rhythm diagnosis. The VA conduction time during PMT was measured from the ventricular pacing stimulus in the ventricular electrogram to the intrinsic atrial deflection in the atrial electrogram and compared with the VA conduction time during ventricular pacing at a similar rate (Fig. 1) Statistical analysis Data are presented as means ± SD. The chi-square test and the Fisher's exact test were used to compare categorical variables. Continuous variables between two groups were analyzed by the Student t test or the Mann Whitney test as appropriate. The Woolf test and exact Cochran Mantel Haenszel test have been applied to test for interaction between complete AV block, retrograde VA conduction, and type of implanted device. A probability value b0.05 was considered statistically significant. Statistical analysis was performed using SPSS 14.0 software package (SPSS Inc., Chicago, Illinois, USA). 3. Results 3.1. Study population and clinical characteristics The study population consisted of 250 consecutive patients (157 male, 63%) with a mean age of 71 ± 14 years (median 75 years, range 10 98). All subjects had symptomatic second- or third-degree AV block documented at time of device implantation. A total of 199 patients (80%) presented initially with complete AV block and 51 patients (20%) with (n + 1):1 AV block. Retrograde VA conduction and the course of AV conduction block was assessed after a mean post-implantation period of 38 ± 12 months. AV block was persistent in 137 (55%) and variable in 113 (45%) subjects. Fig. 3 depicts the course of AV conduction block in patients with baseline complete AV block and baseline AV block II with (n + 1):1 AV conduction. Retrograde conduction was present in 76 patients (30%) with a mean VA conduction time of 258 ± 65 ms (range ms). In 61 of the 76 patients with retrograde conduction, VA conduction time was also assessed at faster ventricular pacing rate and/or at the retrograde Wenckebach point. There was no significant difference in VA conduction time (248 ± 57 vs. 267 ± 67 ms, P = 0.10) measured at the 2 different mean pacing cycle lengths (669 ± 121 and 539 ± 113 ms, respectively). Table 1 summarizes the clinical characteristics of the study population and two subgroups according to the presence of retrograde VA conduction. Subjects with retrograde conduction were more frequently implanted with a pacemaker (88 vs. 66%, respectively, P b 0.001) and less frequently with a CRT system (6 vs. 27%, respectively, P b 0.001) than those without retrograde conduction. There was no significant heterogeneity between the 3 device groups (pacemaker, ICD, and CRT) regarding the type of anterograde AV block (P = 0.73). Left ventricular ejection fraction was greater in subjects with than without retrograde conduction (54 ± 15 vs. 49 ± 16%, respectively, P = 0.04). Importantly, there was no significant difference in antiarrhythmic drug therapy between the two groups Electrophysiological characteristics of patients with and without VA conduction Table 2 compares the electrophysiological characteristics of patients with and without 1:1 retrograde VA conduction. The presence and type of AV block differed significantly between the two groups. Both at time of implantation and assessment for retrograde conduction, complete AV block was more often present in patients without retrograde VA conduction compared to those with retrograde conduction. After testing for interaction, the frequency of VA conduction block was significantly higher for patients with than without complete AV block (P b ). Patients with retrograde conduction had a significantly higher incidence of variable AV conduction block (68 vs. 35%, respectively, P b 0.001) and presented more often with 1:1 AV conduction with or without PR prolongation at time of evaluation compared with patients without retrograde conduction. In line with these findings, the mean PR interval of conducted beats was significantly shorter in patients with than without retrograde VA conduction (238 ± 57 vs. 285 ± 86 ms, respectively, P b 0.001). However, there were no significant differences in the presence and type of preexisting conduction defects and QRS duration between the two groups. During a mean device follow-up of 72 ± 59 months, the incidence of new-onset atrial fibrillation (AF) developing after implantation did not differ significantly between patients with and without VA conduction (26 vs. 37%, respectively, P = 0.11) nor between patients with and without related PMTs (27 vs. 34%, respectively, P =0.59).Thedevice follow-up duration was, however, significantly shorter in patients with than without retrograde VA conduction (50 ± 34 vs. 81 ± 66 months, respectively, P b 0.001). Table 3 depicts the electrophysiological characteristics of all patients with persistent AV block and 1:1 retrograde VA conduction.

3 3302 S. Richter et al. / International Journal of Cardiology 168 (2013) Fig. 1. Example of a 76-year-old male implanted with a dual-chamber pacemaker (St. Jude Medical, Identity DR) who presented with persistent complete AV block, 1:1 retrograde VA conduction, and documented ELT (patient #5; Table 3). Shown are surface ECG leads III and I (10 mm/mv), the intracardiac bipolar atrial electrogram (2.6 mm/mv) and the marker annotation. A, sinus rhythm with complete AV block during VVI pacing at 30 bpm. Note intermittent retrograde conduction that becomes evident by an advanced, distinct bipolar atrial electrogram (asterisk) and negative P wave on surface ECG lead III. B, 1:1 retrograde conduction with a stable VA interval (VA stim ) of 260 ms is present during VVI pacing at 100 bpm. C, atrial pacing threshold testing: after loss of atrial capture at 0.5 V, an ELT initiates with the first retrograde atrial event sensed outside the programmed PVARP of 250 ms (arrow). The automatic PMT detection algorithm starts after termination of the threshold testing and return to the programmed parameters ( Programmiert ). After 8 VP-AS intervals the AV delay is shortened by 50 ms (to 219 ms). Because of the subsequent stable VP-AS interval PMT is diagnosed and terminated by non-tracking of the retrograde atrial event. Note that the VA interval during ELT (VA ELT ) is similar to that measured during VVI pacing at 100 bpm. Retrograde conduction was demonstrated in 24 of the 137 subjects (18%) with persistent (n + 1):1 or complete AV block at time of evaluation. Eleven patients had complete (patients #1 to 11) and 13 (patients #12 to 24) had 2:1 AV conduction block. Of important note, the majority of these patients (63%) had documentation of complete bundle-branch block or bifascicular block (most commonly right bundle-branch block plus left anterior fascicular block) prior to device implantation Retrograde VA conduction according to presence and type of AV block at testing At the time of testing for retrograde VA conduction, complete AV block was present in 126 patients (50%), AV block with (n + 1):1 AV conduction ratio in 23 patients (9%), Wenckebach-type second-degree AV block in 12 patients (5%), and 1:1 AV conduction with prolonged PR interval in 71 patients (28%). Eighteen patients (7%) had 1:1 AV conduction with normal PR interval. The incidence of retrograde VA conduction varied considerably according to the presence and type of anterograde conduction block (Table 4). No more than 9% of the patients with complete AV block had 1:1 retrograde VA conduction. In contrast, 13 of the 23 patients (57%) with (n + 1):1 AV block had 1:1 retrograde VA conduction. As shown in Table 3, all 13 patients had 2:1 AV conduction block (patients #12 to 24) and the majority of them (9 patients) had preexisting complete bundle branch block or bifascicular block. Notably, the mean intrinsic PR interval of conducted beats during 2:1 block was rather short in these patients but did not differ significantly from patients with 2:1 AV block and no retrograde conduction (232 ± 59 vs. 288 ± 97 ms, P = ns). Retrograde VA conduction was present in 25%, 52%, and 67% of subjects with Wenckebach-type second-degree AV block, 1:1 AV conduction with prolonged PR interval, and 1:1 AV conduction with normal PR interval, respectively. The VA

4 S. Richter et al. / International Journal of Cardiology 168 (2013) Fig. 2. Induction of ELT by ventricular pacing and atrial pacing threshold testing in a 69-year-old female with a dual-chamber pacemaker (Medtronic, Kappa DR) implanted for complete AV block (patient #2; Table 3). Shown are surface ECG lead II (0.2 mv/mm), the marker annotation, and the intracardiac bipolar atrial electrogram (0.5 mv/mm). A, complete AV block without intrinsic ventricular activity b30 bpm. Note that retrograde VA conduction is seen following the last ventricular paced beat (asterisk). B, VVI pacing at 100 bpm demonstrates 1:1 retrograde conduction with a long VA interval (340 ms). Upon cessation of ventricular pacing, an ELT initiates with the first sensed retrograde atrial event (arrow). In this case, ELT is automatically detected and terminated by PVARP prolongation rendering the following retrograde atrial event refractory (AR). C, induction of ELT by atrial pacing threshold testing. The first sensed atrial event after loss of capture at 0.25 V is a sinus beat (arrow) that is tracked to the ventricle and initiates ELT. The automatic PMT detection algorithm terminates ELT by PVARP prolongation. The programmed PVARP was 310 ms. Paper speed: 25 mm/s. interval was significantly shorter in patients with 1:1 AV conduction and normal PR interval compared to all other patients with any type of anterograde AV block (P 0.03) Pacemaker-mediated arrhythmias Fifteen patients (6%) had 1 documented PMT. The recorded arrhythmias were diagnosed as ELT in 11 and RNRVAS in 4 patients. All patients with documented PMT had retrograde VA conduction at testing irrespective of the presence and type of anterograde conduction block. Conversely, no PMT occurred spontaneously or during atrial pacing threshold testing in any patient without retrograde VA conduction. Among patients with VA conduction, VA conduction time was significantly longer in patients with than without PMT (297 ± 64 vs. 249 ± 62 ms, respectively, P b 0.01). The VA conduction time during documented PMT was similar to that measured during ventricular pacing at a similar rate (307 ± 69 vs. 296 ± 67 ms, respectively, P =0.68; mean Δ 9 ± 14 ms). Interestingly, among the 24 patients with persistent second- or third-degree AV block and 1:1 retrograde VA conduction, 8 (33%) had a documented PMT, that was diagnosed as ELT in

5 3304 S. Richter et al. / International Journal of Cardiology 168 (2013) Table 2 Electrophysiological characteristics of the patients according to retrograde VA conduction. VAC present n = 76 (30%) VAC absent n = 174 (70%) P value Baseline Type of AVB AVB III, n (%) 52 (68) 147 (84) AVB II (n + 1):1, n (%) 24 (32) 27 (16) Persistent, n (%) 24 (32) 113 (65) b Variable, n (%) 52 (68) 61 (35) b Preexisting conduction defect, n (%) # LBBB 11/68 (16) 36/146 (25) 0.21 RBBB 20/68 (29) 43/146 (29) 1.00 Bifascicular block (RBBB + LFB) 13/68 (19) 21/146 (14) 0.42 None 32/68 (47) 48/146 (33) 0.07 Intrinsic QRS duration, ms 118 ± ± Testing Sinus cycle length 824 ± ± AV conduction AVB III, n (%) 11 (14) 115 (66) b AVB II (n + 1):1, n (%) 13 (17) 10 (6) AVB II Wenckebach, n (%) 3 (4) 9 (5) 0.76 AVB I, n (%) 37 (49) 34 (20) b No AVB, n (%) 12 (16) 6 (3) Intrinsic PR interval, ms 238 ± ± 86 b0.001 Pacing cycle length, ms 677 ± ± VA interval, ms 258 ± PMT, n (%) 15 (20) 0 (0) b Fig. 3. Distribution of the different types of AV conduction block recorded at follow-up in (A) patients with baseline AVB III (n = 199; blue bar) and (B) patients with baseline AV block II with (n + 1):1 AV conduction (n = 51; blue bar). Presented are the percentages of patients with different AV conduction block patterns at time of testing for VA conduction (red bars). AVB, atrioventricular block; AVB II W, Wenckebach-type second-degree AV block. Plus-minus values are means ± SD. AV, atrioventricular; AVB, atrioventricular block; VA, ventriculoatrial; VAC, ventriculoatrial conduction; LBBB, left bundle branch block; RBBB, right bundle branch block; LFB, left fascicular block; PMT, pacemaker-mediated tachyarrhythmia. # Pre-implantation 12-lead ECG recording sinus rhythm and intrinsic AV conduction was not available in all subjects. 6 and RNRVAS in 2 of them (Table 3). As shown in Table 3, the programmed postventricular atrial refractory period (PVARP) was shorter than the VA interval in all 6 patients with ELT and longer Table 1 Clinical characteristics of the patients according to retrograde VA conduction. All patients n = 250 VAC present n=76 (30%) VAC absent n = 174 (70%) P value Age, y 71 ± ± ± Male, n (%) 157 (63) 44 (58) 113 (65) 0.32 Cardiac pacing device, n (%) Pacemaker 181 (72) 67 (88) 114 (66) b0.001 ICD 16 (6) 3 (4) 13 (7) 0.40 CRT 53 (21) 6 (8) 47 (27) b0.001 Cardiovascular disease, n (%) Hypertension 194 (78) 59 (78) 135 (78) 1.00 Coronary artery disease 119 (48) 29 (38) 90 (52) 0.05 Valvular disease 61 (24) 14 (18) 47 (27) 0.20 Cardiomyopathy 92 (37) 21 (28) 71 (41) 0.06 None 55 (22) 24 (32) 31 (18) 0.02 Sick sinus syndrome, n (%) 27 (11) 6 (8) 21 (12) 0.38 Atrial fibrillation, n (%) 68 (27) 21 (28) 47 (27) 1.00 Echocardiographic parameters LVEF, % 51 ± ± ± IVS, mm 12.9 ± ± ± AV-node active drugs, n (%) Beta-blockers 160 (64) 42 (55) 118 (68) 0.06 Calcium antagonists 4 (2) 1 (1) 3 (2) 1.00 Digitalis 29 (12) 5 (6.5) 24 (14) 0.13 Sotalol 2 (1) 1 (1) 1 (0.5) 1.00 Amiodarone 9 (4) 2 (3) 7 (4) 0.73 Plus-minus values are means ± SD. AV, atrioventricular; VA, ventriculoatrial; VAC, ventriculoatrial conduction; ICD, implantable cardioverter-defibrillator; CRT, cardiac resynchronization therapy; LVEF, left ventricular ejection fraction; IVS, interventricular septum. in the 2 patients with RNRVAS. Figs. 1 and 2 show representative examples of 2 distinct patients with persistent complete AV block, 1:1 retrograde VA conduction, and documented ELT (patient #5 and patient #2, respectively; Table 3). Notably, intermittent retrograde conduction could already be detected during VVI pacing at 30 bpm depending on the preceding A-VP interval (Figs. 1A and 2A, asterisks). An episode of induced RNRVAS in a patient with persistent 2:1 AV block and retrograde conduction (patient #20; Table 3) is demonstrated in Fig Discussion 4.1. Main study findings This study systematically analyzed retrograde conduction and related PMTs in a large cohort of device recipients with different types of anterograde AV block. It reveals several clinically important findings: (1) second- and third-degree AV block was variable in 45% of the cases during a mean post-implantation period of >3 years; (2) retrograde VA conduction was present in 30% of the whole study population and in 18% of patients with persistent (n + 1):1 or complete AV block; (3) PMT was documented in 6% of all study patients and in 33% of patients with persistent (n + 1):1 or complete AV block and 1:1 retrograde VA conduction. In contrast, no PMT occurred in any patient without retrograde conduction. These data should call attention to careful assessment of anterograde and retrograde conduction and appropriate device programming at time of device implantation and during follow-up Retrograde VA conduction and anterograde AV block The presence of retrograde conduction in complete heart block has been considered a paradoxical and rare phenomenon [7,8]. However, previous small studies found retrograde VA conduction in up to

6 S. Richter et al. / International Journal of Cardiology 168 (2013) Table 3 Electrophysiological characteristics of patients with persistent AV block and retrograde VA conduction (n = 24). No. Age (y) Sex BBB LFB QRS AV block Sinus CL PR interval Pacing CL VA ELT CL ELT VA ELT RNRVAS VA RNRVAS 1 72 M NA NA NA III F RBBB LAFB 140 III M LBBB III M NA NA NA III M III M III F LBBB III F III - # M RBBB III F RBBB III F RBBB LAFB 140 III (VA criterium) M : M RBBB LAFB 140 2: M RBBB LAFB 160 2: M RBBB LAFB 140 2: F : M : F RBBB LAFB 160 2: M : M RBBB : M RBBB LAFB 150 2: M RBBB LAFB 150 2: M LBBB : (VA criterium) F RBBB LAFB 150 2: BBB, bundle branch block; RBBB, right bundle branch block; LBBB, left bundle branch block; LFB, left fascicular block; LAFB, left anterior fascicular block; AV, atrioventricular; VA, ventriculoatrial; CL, cycle length; ELT, endless loop tachycardia; RNRVAS, repetitive nonreentrant VA synchrony; PVARP, post-ventricular atrial refractory period; NA, no data available. # no intrinsic atrial activation (sinus arrest). PVARP 25% in patients with complete AV block [3,9]. Using intracardiac His bundle recording and ventricular pacing in 57 patients with complete AV block, Touboul et al. [9] elegantly demonstrated that the site of anterograde conduction block was located at an infra-hisian level in almost all patients with retrograde conduction. Similar observations were reported by Josephson [12]. Retrograde conduction was rarely observed in patients with AV nodal block in the absence of a retrogradely conducting accessory pathway [2,9,12,13]. It has therefore been postulated that the presence of retrograde conduction points to the His-Purkinje system (HPS) as site of anterograde conduction block [9,12]. In line with these observations is our finding that the majority of patients with persistent 2:1 or complete AV block and 1:1 retrograde VA conduction had complete bundle branch block or bifascicular block documented prior to the development of anterograde AV block (Table 3). Interestingly, 57% of our patients with (n + 1):1 AV block had 1:1 retrograde VA conduction. All these patients had 2:1 AV block with a slightly prolonged PR interval (232 ± 59 ms). Typically, in most of these patients right bundle branch block with left anterior Table 4 Retrograde VA conduction according to the presence and type of AV block at testing. Patients, n VA conduction, n (%) VA interval, ms AVB III (9) 268 ± 60 AVB II (n + 1): (57) 275 ± 78 AVB II Wenckebach 12 3 (25) 293 ± 93 AVB I (52) 259 ± 64 Normal PR interval (67) 215 ± 37 # Plus-minus values are means ± SD. AV, atrioventricular; AVB, atrioventricular block; VA, ventriculoatrial. # P 0.03 compared to VA interval in any type of AV block. fascicular block was present with intrinsic AV conduction. These findings strongly support a subnodal site of the 2:1 AV block in patients with retrograde conduction. However, there was no significant difference in preexisting ventricular conduction abnormalities and QRS duration between subjects with and without 1:1 retrograde conduction, a finding that is well in accordance with previously reported data [12,14]. Given that the presence of anterograde conduction abnormalities located in the HPS does not directly influence retrograde conduction through the AV node itself, it is tempting to speculate that AV nodal conduction constitutes the major determinate for retrograde VA conduction. In fact, we found that patients with retrograde conduction had significantly more often variable AV conduction block (68 vs. 35%) and, accordingly, presented more often with 1:1 AV conduction (65 vs. 23%) and normal or slightly prolonged PR interval (238 ± 57 vs. 285 ± 86 ms) at time of evaluation compared to those without retrograde conduction. In keeping with this, Westveer et al. [1] showed that AV nodal effective refractory period and AV nodal conduction times (PR interval and AH interval) but not conduction abnormalities in the HPS (QRS duration and HV interval) were associated with retrograde conduction in patients with 1:1 AV conduction. In particular, a PR interval >250 ms predicted retrograde VA conduction block [1]. Further support for a link between AV nodal conduction properties and retrograde conduction is provided by our observation that the VA interval was significantly shorter in patients with 1:1 AV conduction and normal PR interval (b200 ms) compared to all other patients with any type of anterograde AV block (Table 4). In addition, the presence of retrograde VA conduction block correlated with the degree of anterograde AV block (in particular AV block III ) with the exception of 2:1 (infra-his) block. Our data support the notion that delay in retrograde conduction preferentially occurs at the level of the AV node rather than in the HPS [13,15].

7 3306 S. Richter et al. / International Journal of Cardiology 168 (2013) Fig. 4. Initiation of RNRVAS by atrial pacing threshold testing in a 71-year-old male with a dual-chamber pacemaker (Medtronic, Relia DR) implanted for symptomatic 2:1 AV block (patient #20; Table 3). Shown are surface ECG lead I (0.2 mv/mm), the marker annotation, and the intracardiac bipolar atrial electrogram (0.5 mv/mm). A, testing for anterograde AV conduction in the inhibited mode reveals 2:1 AV block with a short PR interval of 160 ms in the presence of complete RBBB (suggesting a subnodal site of block). Note the pronounced ventriculophasic phenomenon. B, VVI pacing at 90 bpm demonstrates 1:1 retrograde conduction with a short VA interval of 200 ms. C, initiation of RNRVAS by atrial pacing threshold testing. Loss of atrial capture at 0.5 V (asterisk) enables the subsequent ventricular paced beat to conduct retrogradely. Detection of the retrograde atrial event is precluded by the PVARP of 250 ms (AR). The atrial escape interval times out to release an atrial stimulus during atrial refractoriness resulting in functional noncapture. Paper speed: 25 mm/s. However, the mechanism by which 1:1 retrograde VA conduction may occur with anterograde AV block is not completely understood. In the absence of an accessory AV pathway, unidirectional block in the damaged specialized conduction system along with intact retrograde nodal conduction seems to be operative Pacemaker-mediated arrhythmias The present study investigated the incidence of pacemaker-mediated arrhythmias related to VA synchrony in a large cohort of patients implanted for symptomatic AV block. In our series, 6% of the whole study population and 20% of patients with retrograde conduction had at least 1 documented PMT. All patients with documented PMT exhibited retrograde VA conduction irrespective of the presence and type of anterograde AV block. Interestingly, one third of patients with persistent 2:1 or complete AV block and evidence of retrograde conduction had clinically relevant PMTs (Table 3). Vice versa, more than 50% of all patients with documented PMTs had persistent (n + 1):1 or complete AV block. These unexpected findings suggest an underestimation of this clinically relevant problem leading to inappropriate device programming in this particular patient population. In fact, VA conduction time was significantly longer in patients with than without PMT (297 ± 64 vs. 249 ± 62 ms). Considering a nominal value of 250 ms for the PVARP in most contemporary dual-chamber pacemakers, it is not surprising that patients with long retrograde conduction times (>250 ms) suffered from ELT if no testing for retrograde

8 S. Richter et al. / International Journal of Cardiology 168 (2013) conduction had previously been performed. Appropriate programming of the PVARP has been demonstrated to effectively prevent ELT in virtually all patients implanted with a dual-chamber pacemaker [16].Ingeneral, the PVARP should be programmed at 50 to 75 ms in excess of the determined VA conduction time [17]. In a previous study including 50 patients with advanced AV block, PVARP adjustment effectively prevented induction of ELT in all patients with retrograde conduction; an arbitrarily programmed PVARP of 300 ms offered protection against inducible ELT in all patients without retrograde conduction [17,18]. Importantly, during a 19-month follow-up, retrograde conduction remained absent in 94% of the patients with VA conduction block at time of implantation [18]. Inlinewiththis,wefoundno stored PMT in any patient without retrograde conduction, irrespective of initial device programming. These findings suggest that VA conduction block predicts absence of retrograde conduction and spontaneous or inducible PMTs during follow-up in patients implanted for second- or third-degree AV block. Nevertheless, it has been recommended as safety precaution to program a PVARP of 300 ms even in patients with advanced AV block and no retrograde conduction [17,18]. Programming a long PVARP can cause 2:1 block in AV block with fast sinus rates and predisposes to RNRVAS with unfavorable hemodynamic effects. In our series, only 5% of patients with retrograde conduction had documented RNRVAS with VA conduction times (mean 240 ± 37 ms) shorter than the programmed PVARP. The low incidence of RNRVAS in our study population can be explained by the fact that RNRVAS depends also on pacing parameters other than PVARP, such as AV interval, lower rate interval, and sensor parameters [10,11]. In patients with AV block and long intrinsic AV conduction times, a short AV delay in concert with a relatively long lower rate interval (i.e ms) and passive sensor should be programmed to optimize AV synchrony and minimize atrial pacing, rendering RNRVAS unlikely. Another reason of the low incidence of RNRVAS in our patient population may have been that some device algorithms do not detect them since they do not store high atrial rates with paced atrial cycles. It should be emphasized, that all patients were programmed to the DDD pacing mode during follow-up and that programming was not modified to facilitate PMT induction during atrial pacing threshold testing. With this approach we intended to provide a realistic clinical condition and highlighted an underestimated clinically relevant problem encountered in patients implanted for second- or third-degree AV block Retrograde VA conduction and new-onset AF Retrogradely conducting spontaneous or paced ventricular beats can occasionally induce AF [19,20]. It is therefore tempting to speculate that the presence of retrograde VA conduction in patients with an implanted cardiac pacing device might be associated with AF. In line with a recent study by Baranchuk et al. [21], the incidence of new-onset AF after device implantation, however, did not differ significantly between patients with and without retrograde VA conduction. The observed tendency towards a higher incidence of new-onset AF in patients without retrograde VA conduction (37 vs. 26% in patients with retrograde VA conduction) was most likely related to the substantially longer follow-up period in these patients. Moreover, the occurrence of PMTs was not associated with an increased incidence of new-onset AF in our patient population. These data suggest that the presence of retrograde VA conduction and related PMTs has no significant impact on the incidence of new-onset AF during long-term follow-up in AV block patients implanted with a dual-chamber pacing or CRT device Limitations Study limitations include the absence of retrograde conduction testing at time of implantation. Therefore, the present study cannot provide information about the long-term course of VA conduction and the potential influence of permanent pacing on VA conduction in this particular patient population. However, only 5% of patients with VA conduction block at time of implantation develop retrograde conduction during follow-up [1,17,18]. Moreover, in the present study, retrograde conduction was assessed by ventricular pacing. Previous studies suggested that introduction of ventricular extrastimuli during AV sequential pacing may be more accurate to assess retrograde conduction compared to ventricular pacing alone [22 24]. Of interest, this AV sequential pacing method resulted in significantly longer VA conduction times than did ventricular pacing [22], a finding that could explain why ELT may occur despite PVARP adjustment according to a VA interval assessed only by ventricular pacing. Therefore, the incidence of retrograde conduction may have been underestimated in our patient cohort. Furthermore, pharmacological stimulation with atropine and/or isoprenaline or exercise testing could have unmasked concealed nodal VA conduction in some patients without retrograde conduction at baseline condition [25,26]. The latter two limitations, however, seemed not to be of clinical importance as none of our patients with VA conduction block experienced spontaneous PMT episodes during a mean follow-up of >3 years. Finally, it cannot be entirely ruled out that retrograde conduction was by way of a concealed accessory AV pathway in individual patients with fast VA conduction. However, no patient with retrograde conduction had a history of fast and regular palpitations or documented supraventricular tachycardia suggestive of an orthodromic AV reentrant tachycardia Conclusions Retrograde conduction occurs in 30% of patients implanted for second- or third-degree AV block and can cause PMTs. Careful examination of anterograde and retrograde conduction should therefore be performed at least once at time of discharge after device implantation in all patients with documented AV block in order to optimize device programming and prevent PMTs. Acknowledgment We are greatly indebted to Kerstin Nowotnik for her skilled technical assistance. References [1] Westveer DC, Stewart JR, Goodfleish R, Gordon S, Timmis GC. Prevalence and significance of ventriculoatrial conduction. Pacing Clin Electrophysiol 1984;7(5): [2] Akhtar M. Retrograde conduction in man. Pacing Clin Electrophysiol 1981;4(5): [3] Hayes DL, Furman S. Atrio-ventricular and ventriculo-atrial conduction times in patients undergoing pacemaker implant. Pacing Clin Electrophysiol 1983;6(1 Pt 1): [4] Furman S, Fisher JD. Endless loop tachycardia in an AV universal [DDD] pacemaker. Pacing Clin Electrophysiol 1982;5(4): [5] Den Dulk K, Lindemans FW, Bar FW, Wellens HJ. Pacemaker related tachycardias. Pacing Clin Electrophysiol 1982;5(4): [6] Bathen J, Gundersen T, Forfang K. Tachycardias related to atrial synchronous ventricular pacing. Pacing Clin Electrophysiol 1982;5(4): [7] Adams CW. Retrograde atrial conduction with complete heart block following implantation of an internal ventricular pacemaker. Dis Chest 1963;43: [8] Castillo C, Samet P. Retrograde conduction in complete heart block. Br Heart J 1967;29(4): [9] Touboul P, Huerta F, Delahaye JP. Retrograde conduction in complete atrioventricular block. Study using His bundle recordings. Br Heart J 1976;38(7): [10] Barold SS. Repetitive reentrant and non-reentrant ventriculoatrial synchrony in dual chamber pacing. Clin Cardiol 1991;14(9): [11] Barold SS, Levine PA. Pacemaker repetitive nonreentrant ventriculoatrial synchronous rhythm. A review. J Interv Card Electrophysiol 2001;5(1): [12] Josephson ME. Clinical cardiac electrophysiology: techniques and interpretation. 4th ed. Philadelphia: Lippincott Williams and Wilkins; p [13] Schuilenburg RM, editor. Patterns of VA conduction in the human heart in the presence of normal and abnormal AV conduction. Philadelphia: Lea Febiger; 1976.

9 3308 S. Richter et al. / International Journal of Cardiology 168 (2013) [14] Akhtar M, Damato AN, Batsford WP, Ruskin JN, Ogunkelu JB. A comparative analysis of antegrade and retrograde conduction patterns in man. Circulation 1975;52(5): [15] Narula OS. Retrograde pre-excitation. Comparison of antegrade and retrograde conduction intervals in man. Circulation 1974;50(6): [16] Hayes DL, Holmes Jr DR, Vlietstra RE, Osborn MJ. Changing experience with dual chamber (DDD) pacemakers. J Am Coll Cardiol 1984;4(3): [17] Barold SS, Falkoff MD, Ong LS, Heinle RA. Electrocardiography of contemporary DDD pacemakers. Basic concepts, upper rate response, retrograde ventriculoatrial conduction and differential diagnosis of pacemaker tachycardias. Philadelphia: W.B. Saunders; [18] van Mechelen R, Ruiter J, Vanderkerckhove Y, de Boer H, Hagemeijer F. Prevalence of retrograde conduction in heart block after DDD pacemaker implantation. Am J Cardiol 1986;57(10): [19] Shen EN, Sung RJ. Initiation of atrial fibrillation by spontaneous ventricular premature beats in concealed Wolff-Parkinson-White syndrome. Am Heart J 1982;103(5): [20] Israel CW. The role of pacing mode in the development of atrial fibrillation. Europace 2006;8(2): [21] Baranchuk A, Greiss I, Simpson C, et al. The presence of ventriculo-atrial conduction is not associated with new atrial fibrillation in patients with dual chamber pacemakers. Heart Rhythm 2011;8(5):S11. [22] Webb CR, Spielman SR, Greenspan AM, Yacone LA, Horowitz LN. Improved method for evaluating ventriculoatrial conduction before implantation of atrial-sensing dual chamber pacemakers. J Am Coll Cardiol 1985;5(6): [23] Mahmud R, Denker S, Lehmann MH, Akhtar M. Effect of atrioventricular sequential pacing in patients with no ventriculoatrial conduction. J Am Coll Cardiol 1984;4(2): [24] Mahmud R, Lehmann M, Denker S, Gilbert CJ, Akhtar M. Atrioventricular sequential pacing: differential effect on retrograde conduction related to level of impulse collision. Circulation 1983;68(1): [25] Hariman RJ, Pasquariello JL, Gomes JA, Holtzman R, el-sherif N. Autonomic dependence of ventriculoatrial conduction. Am J Cardiol 1985;56(4): [26] Cazeau S, Daubert C, Mabo P, et al. Dynamic electrophysiology of ventriculoatrial conduction: implications for DDD and DDDR pacing. Pacing Clin Electrophysiol 1990;13(12 Pt 1):

10 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具