In 1971, Francis Fontan published his first successful surgical treatment for tricuspid atresia which later was extended to other structural heart defects with single-ventricle physiology. Reference Fontan and Baudet1–Reference Gentles, Mayer and Gauvreau3 The goal of the Fontan procedure is to separate the systemic and the pulmonary venous return, reducing hypoxaemia and thromboembolic events and enabling patients with single-ventricle physiology a moderate survival and quality of life. After several modifications of the Fontan procedure, a staged Fontan completion with bidirectional cavopulmonary shunt and later total cavopulmonary connection is currently a standard procedure for patients with functionally univentricular hearts, which leads to drastically improved outcomes. Reference Downing, Allen and Glatz4–Reference Pundi, Johnson and Dearani9 Due to this improvement in life expectancy and functional status in patients with total cavopulmonary connection, the evaluation of long-term complications gains more importance. Brady-arrhythmia is one of the complications besides thromboembolism, protein-losing enteropathy, plastic bronchitis, and heart failure. Reference Dahlqvist, Wiklund and Karlsson10 Commonly observed in patients after Fontan procedure are brady-arrhythmic rhythm abnormalities such as sinus node dysfunction with a previous described incidence of 9–60% and atrioventricular dysfunction/atrioventricular block with a described incidence of 2–16 %. Reference Nakano, Kado and Tatewaki7,Reference Ono, Kasnar-Samprec and Hager8,Reference Dahlqvist, Wiklund and Karlsson10–Reference Okólska, Karkowski and Kuniewicz12 These rhythm disturbances could cause severe clinical problems and are the most common indication for pacemaker implantation in Fontan patients. Reference Downing, Allen and Glatz4–Reference Pundi, Johnson and Dearani9 Although tachy-arrhythmic rhythm abnormalities such as atrial tachycardia, atrial fibrillation, atrial flutter, and ventricular arrhythmia are commonly described in the mentioned group of patients often needing pharmacological or interventional catheter ablation treatment, brady-arrhythmia in Fontan patients are not fully studied. In this study, we aimed to determine the type and prevalence of brady-arrhythmic rhythm abnormalities in our large cohort of patients who underwent total cavopulmonary connection. We also evaluated the outcomes and complications after pacemaker implantation and estimated the influence of pacemaker implantation in Fontan circulation on survival, development of protein-losing enteropathy/ plastic bronchitis, and thromboembolic complications.
Methods
Ethical statement
Institutional Review Board of the Technical University of Munich approved this study on 27 June, 2022 (approved number 2022-303-S-KH), and the need for individual patient consent was waived due to the retrospective nature of the study.
Patients and data collection
A single-centre retrospective cohort study of 620 consecutive patients who underwent a total cavopulmonary connection at the German Heart Center Munich from May 1994 to December 2021 was performed. Medical records of each patient were depicted using electrical and paper documents, obtained from in-hospital and outpatient records. Survival and follow-up data were obtained from our institutional single-ventricle database, which is regularly tracked. In patients who underwent pacemaker implantation, pacemaker analysis was conducted regularly at 6-month intervals. Outpatient evaluations consisted of real-time telemetry of battery and lead measurements. To analyse the pacing voltages, we calculated the minimum energy threshold (MET) according to our previous study. Reference Stanner, Horndasch and Vitanova13
Operative techniques
The operative techniques for total cavopulmonary connection are described in previous reports. Reference Ono, Kasnar-Samprec and Hager8,Reference Schreiber, Hörer and Vogt14 Lateral tunnel total cavopulmonary connection was performed in 50 patients in the early era. In January 1999, extracardiac total cavopulmonary connection was introduced, and it became the standard procedure at our institution since May 2002. Reference Schreiber, Hörer and Vogt14 Fenestration was not performed routinely and was only used for high-risk patients. Reference Ono, Kasnar-Samprec and Hager8 As for the pacemaker implantation and revision technique, it was performed either through a midline sternotomy, sub-xiphoid approach, or lateral thoracotomy, based on previous cardiac operation and morphologic cardiac position in the thorax. Usually, bipolar steroid-eluting suture-on type epicardial leads were fixated with two 5/0 polypropylene sutures. When the midline sternotomy or sub-xiphoid approach was done, atrial leads were attached to the lateral wall of the right atrium, and ventricular lead was often fixed on the diaphragmatic ventricular surface. When the left lateral thoracotomy was chosen, the atrial lead was positioned on the left atrium, and the ventricular lead was positioned near the apex of the ventricle. During the latter study period, our current approach has generally been to place leads on the left atrial appendage through a lateral thoracotomy in patients with prior complex atrial operations. The surplus of leads is addressed by creating loops of the electrode within the pericardium and the pacemaker pocket, which was usually located in the abdomen.
Diagnosis of brady-arrhythmia and indication for pacemaker implantation and revision
Patients' rhythms were regularly checked in-hospital and also in outpatient clinics. A standard 12-lead electrocardiogram was made during rest in the supine position. When brady-arrhythmia was suspected, patients underwent 24-hour Holter recording during normal daily activity, using commercially available Holter systems and the records were reviewed by experienced analysts. The predominant rhythm was defined as the rhythm that was present during > 50% of the time during the Holter recording. The pacemaker generator and leads were analysed on the first day after surgery, before discharge from hospital and consistently every 6 months during the follow-up. Lead parameters measured during follow-up were (1) sensed P- and R-wave amplitudes (in millivolts) if an underlying heart rhythm was present, (2) impedance of implanted leads (in ohms), and (3) minimal pacing threshold (in volts) and pulse width (in milliseconds) where consistent capture was still possible. To get comparable and reliable values for all patients, we calculated the MET according to following formula:
$$\matrix{{{\rm{MET}}(\mu {\rm{J}}) = [{\rm{pacing}}{\mkern 1mu} {\rm{voltage}}\;{{({\rm{V}})}^2} \times {\rm{pulse}}{\mkern 1mu} {\rm{width}}{\mkern 1mu} ({\rm{ms}})} \hfill \cr
{\quad \quad \quad \quad \quad \times {{10}^6}]/[{\rm{impedance}}\;(\Omega ) \times {{10}^3}{\mkern 1mu} {\rm{ms/s}}]} \hfill \cr } $$
Pacing and sensing lead thresholds for impedance and calculated MET were compared after 1, 3, 5, 10, 15, and 20 years after pacemaker implantation. Lead failure was defined as the need for replacement or abandonment on the basis of the following: (1) fracture or insulation break; (2) increasing pacing or sensing thresholds; and (3) phrenic or myopotential stimulation. Infections were classified separately into (1) superficial infection (with or without positive blood culture) or (2) deep infection necessitating generator removal. Lead data were censored for elective change, death, or orthotropic heart transplantation.
Statistical analysis
Exploratory data analysis was performed by using descriptive measures. Categorical variables are presented as absolute numbers and percentages. A Chi-square test was used for categorical data. Continuous variables are expressed as medians with interquartile ranges. An independent sample t-test was used to compare normally distributed variables. The Mann–Whitney U-test was used for variables that were not normally distributed. Levene’s test was used to differentiate between normally and non-normally distributions. Transplant-free survival following total cavopulmonary connection, freedom from pacemaker implantation following total cavopulmonary connection, and freedom from lead revision after initial pacemaker implantation were estimated using the Kaplan–Meier method. Risk factor analysis for the pacemaker implantation before total cavopulmonary connection was performed using linear regression model, and those analyses after total cavopulmonary connection were performed using Cox regression model. Data analysis was performed using SPSS version 28.0 for Windows (IBM, Ehningen, Germany) and R-statistical software (state package).
Results
Patient characteristics and perioperative data
Patient characteristics of 620 patients who underwent total cavopulmonary connection in this study are shown in Table 1. Most frequent primary diagnosis was hypoplastic left heart syndrome (28%), and most frequent associated cardiac anomaly was transposition of the great arteries (34%). Median age at total cavopulmonary connection was 2.3 (interquartile ranges: 1.8–3.4) years. Extracardiac total cavopulmonary connection was performed in 570 (92%) of the patients, and fenestration was placed in only 46 (7%) patients.
Table 1. Patient characteristics
Follow-up and survival
Median follow-up period was 5.8 (1.0–13.0) years. There were 15 late deaths after hospital discharge and 4 heart transplantations. No patient underwent Fontan takedown. Transplant-free survival at 5, 10, 15, and 20 years after total cavopulmonary connection was 96.8, 94.6, 94.2, and 91.3%, respectively.
Incidence of brady-arrhythmia and timing of pacemaker implantation
During the study period, a total of 52 (8.4 %) patients presented with brady-arrhythmia and needed pacemaker implantation during or after the staged Fontan palliation. As an indication for pacing, atrioventricular block third degree in 25 patients, atrioventricular block second degree in 4 patients, sick sinus syndrome (sinus node dysfunction) in 16 patients, and junctional escape rhythm in 7 patients were described (Table 2). Pacemaker implantation was performed before total cavopulmonary connection in 16 patients, concomitant with total cavopulmonary connection in 8 patients, and after total cavopulmonary connection in 28 patients (median 1.8 years after total cavopulmonary connection). As for the pacing lead, 49 patients underwent surgical epicardial pacing lead implantation and three patients underwent transcatheter endocardial pacing lead implantation.
Table 2. Timing and indication for pacemaker implantation
TCPC = total cavopulmonary connection.
Initial pacemaker implantation
Among 52 patients who needed pacemaker implantation, a surgical epicardial pacemaker implantation was performed in 49 patients and transcatheter intravenous pacemaker implantation was performed in 3 patients. Among three patients with transcatheter pacemaker implantation, the first patient developed sinus node dysfunction 9 years post-operatively after lateral tunnel total cavopulmonary connection, and we implanted transvenous atrial pacing and sensing (AAI) pacemaker leads to the lateral wall of the total cavopulmonary connection tunnel. The second patient developed sinus node dysfunction 2 years after extracardiac total cavopulmonary connection, and we implanted transvenous AAI pacemaker leads to the right atrium through fenestration. The last patient after extracardiac total cavopulmonary connection was readmitted because of protein-losing enteropathy and Fontan failure. The patient demonstrated sinus node dysfunction and a transvenous AAI pacemaker was implanted. The leads were placed through catheter intervention into the right atrium through the superior vena cava, left pulmonary artery, and left atrium. Among 49 patients who underwent surgical pacemaker implantation, most of the patients (n = 45) received DDD (both atrium and ventricle sensed and paced) pacemaker. In five patients, who developed brady-arrhythmia in newborns or infants (around initial palliation period), ventricular pacing and sensing (VVI) pacemaker was implanted initially and was upgraded to DDD system at the time of second-stage palliation. Operative techniques included 46 median sternotomies and 3 subxiphoid incisions. Steroid-eluting epicardial leads were used in all patients. Cardiac re-synchronisation therapy with multisite pacing was applied in one patient. The patient with double-inlet left ventricle, transposition of the great arteries, and pulmonary atresia underwent a central shunt at 1-month-old and bidirectional cavopulmonary shunt at 7 months old. Non-fenestrated extracardiac total cavopulmonary connection was performed using 18-mm Goretex conduit at 3 years old. Five years later, he was readmitted because of highly reduced systemic ventricular function with severe aortic regurgitation and mild mitral regurgitation. Aortic valve replacement (SJM Regent 23 mm) and mitral valve repair were performed and a complete atrioventricular block developed post-operatively. Nineteen days post-operatively, biventricular pacemaker system was implanted (Medtronic Insync III 8042). One ventricular lead was placed on the apex of the ventricle, and another ventricular lead was placed on the frontal surface of the ventricle just below the sternum. In spite of good biventricular pacing, his systemic ventricular function did not improve and he died of heart failure at 9 years old.
Perioperative complications included one superficial pacemaker pocket infection, which was conservatively treated. Phrenic pacing was not reported. Freedom from pacemaker implantation following total cavopulmonary connection at 5, 10, and 15 years was 94, 92, and 91%, respectively (Fig. 1).
Figure 1. Freedom from pacemaker implantation after TCPC. TCPC, total cavopulmonary connection.
Outcomes after pacemaker implantation
After initial pacemaker implantation, 12 patients had pacemaker lead events. The causes of event included 9 pacemaker lead dysfunctions (sensing failure, lead fracture, and increasing pacing thresholds), 3 pacemaker infections (2 primary and 1 secondary to lead dysfunction), and 1 pacemaker dislocation (Table 3). Eight of nine patients with pacemaker lead dysfunction had revision of pacemaker lead (implantation of new leads). Two patients underwent left thoracotomy, and the remaining seven underwent median sternotomy. One patient who was admitted due to severe Fontan failure had atrial lead dysfunction during cardiopulmonary resuscitation. Because of the patient`s bad condition, the pacing mode was switched to VVI mode and a pacemaker lead revision was not performed. Three patients developed infections. The first patient had pacemaker pocket infection 9 years after total cavopulmonary connection but recovered with conservative therapy. The second patient developed pacemaker leads and battery infection, and the pacemaker system was surgically removed. A new pacemaker was implanted 2 months later. The third patient with congenitally corrected transposition of the great arteries and congenital atrioventricular block III had a modified Blalock–Taussig shunt and VVI pacemaker implantation 13 days after birth. The patient developed mediastinitis 1 month post-operatively. The pacemaker system was surgically removed, and a DDD pacemaker was implanted concomitantly with bidirectional cavopulmonary shunt at 6 months. One patient with VVI pacemaker implantation in infancy developed dislocation of pulse generator battery and was therefore revised. One patient underwent transcatheter pacemaker implantation after DDD pacemaker implantation. There were three patients who were successfully converted to catheter-based transvenous pacemaker implantation through pulmonary artery to right atrium. Freedom from pacemaker lead revision at 5 and 10 years after initial pacemaker implantation was 91 and 78%, respectively (Fig. 2). Freedom from pacemaker battery exchange at 5 and 10 years after initial pacemaker implantation was 81 and 16%, respectively (Supplementary figure S1). Atrial and ventricular lead impedances over time are shown in Figure 3. Atrial impedance did not significantly change over time until 20 years (Fig. 3a), and ventricular impedance was stable until 20 years (Fig. 3b). MET of atrial lead and of ventricular lead over time are shown in Figure 4. MET of atrial leads tended to decrease from 1 year to at 3 years (1.1 ± 1.8 µJ–0.5 ± 0.5 µJ, p = 0.13), then the value was stable until 20 years. MET of ventricular leads was stable until 15 years (2.2 ± 3.6 µJ) but then increased to 4.8 ± 6.2 µJ at 20 years (p = 0.007).
Figure 2. Freedom from pacemaker lead revision after initial pacemaker implantation.
Figure 3. Box-and-whiskers dot plots showing change of atrial (a) and ventricular (b) pacemaker lead thresholds for impedance after initial pacemaker implantation. The top and bottom whiskers mark the minimum and maximum values, the upper and lower borders of the box represent the upper and lower quartiles, and the middle horizontal line represents the median.
Figure 4. Box-and-whiskers dot plots showing change of atrial (a) and ventricular (b) MET after initial pacemaker implantation. The top and bottom whiskers mark the minimum and maximum values, the upper and lower borders of the box represent the upper and lower quartiles, and the middle horizontal line represents the median.
Table 3. Events after pacemaker implantation
TCPC, total cavopulmonary connection, VVI, ventricular pacing and sensing.
* This patient had secondary infection.
Risk factor analysis for pacemaker implantation
Factors associated with pacemaker implantation are shown in Table 4. Diagnosis of congenitally corrected transposition of the great arteries (odds ratio, 4.9; confidence interval, 1.4–17.4, p = 0.01) was identified as an associated factor for pacemaker implantation before total cavopulmonary connection. No variable was associated with pacemaker implantation after total cavopulmonary connection.
Table 4. Factors associated with pacemaker implantation
ccTGA = congenitally corrected transposition of the great arteries; DKS = Damus–Kay–Stansel; HLHS = hypoplastic left heart syndrome; PAB = pulmonary artery banding; PM = pacemaker; RV = right ventricle; TCPC = total cavopulmonary connection.
Impact of pacemaker on outcomes after total cavopulmonary connection
Impact of pacemaker rhythm on outcomes after the Fontan procedure was analysed using Cox regression models. Pacemaker implantation before or at Fontan procedure was not a risk factor for survival (p = 0.636, hazard ratio: 1.416, 95% CI: 0.335–5.979), protein-losing enteropathy/plastic bronchitis (p = 0.746, hazard ratio: 0.790, 95% CI: 0.190–3.291) or thromboembolic complications (p = 0.727, HR: 0.046, 95% CI: 0.000–1424127.3).
Comment
The present study evaluated the prevalence, therapy, and risk factors of brady-arrhythmic rhythm disturbances in patients after the total cavopulmonary connection at our centre over 25 years of follow-up. Among 620 patients who underwent total cavopulmonary connection, a total of 52 patients needed permanent pacemaker implantation. Pacemaker was needed before total cavopulmonary connection in 16 patients, whose main indication was atrioventricular block. Primary diagnosis of congenitally corrected transposition of the great arteries was a significant risk factor for pacemaker implantation before total cavopulmonary connection. Concomitantly with or following total cavopulmonary connection, 36 patients needed pacemaker, the main cause being sinus node dysfunction. There were 12 patients who needed pacemaker revisions, and all revision procedures were successful. Living with pacemaker in Fontan circulation was not a risk for survival, protein-losing enteropathy/plastic bronchitis, or thromboembolic complications after total cavopulmonary connection.
Incidence of brady arrhythmia and pacemaker implantation
After the Fontan procedure, 3–13% of patients may require pacemaker within the first 5 years after the operation. Reference Nakano, Kado and Tatewaki7,Reference Ono, Kasnar-Samprec and Hager8,Reference Dahlqvist, Wiklund and Karlsson10–Reference Okólska, Karkowski and Kuniewicz12 In our cohort, 8.4% of patients developed the need for pacemaker throughout the staged Fontan palliation. Patients with congenitally corrected transposition of the great arteries had a trend to acquire complete atrioventricular block that needed pacemaker implantation before the Fontan procedure. Compared to previous literature, Reference Nakano, Kado and Tatewaki7,Reference Dahlqvist, Wiklund and Karlsson10–Reference Okólska, Karkowski and Kuniewicz12 the most common indication for pacemaker implantation was not sinus node dysfunction but atrioventricular block third degree. This is because our cohort included relative high amount of patients with congenitally corrected transposition of the great arteries (5%). Following total cavopulmonary connection, sinus node dysfunction was the most frequent diagnosis for pacemaker implantation. Although endocardial pacing is generally preferred to epicardial pacing because of lower current values and pacing thresholds, there are certain circumstances in which the use of transvenous leads is impractical or contraindicated. Disadvantages include the limited vascular access in small children, risk of venous obstruction, and the need for growth accommodation. Generally, it may be impossible to gain venous access to the ventricle in patients undergoing the Fontan operation who require ventricular pacing. Furthermore, current modifications of the extracardiac conduit may exclude venous access to the atrium. However, with improvement in lead technology, such as steroid-eluting leads, the pacing thresholds are very similar to conventional endocardial leads.
Results after pacemaker implantation
The outcomes after pacemaker implantation were excellent. There was no pacemaker-related death. We experienced three infections, but all of them were treated conservatively. In this study, nine patients developed pacemaker lead dysfunction and freedom from pacemaker lead revision at 10 years after total cavopulmonary connection was 78%. Cohen et al. reported that probability of freedom from lead failure at 10 years after pacemaker implantation was 60%, and it was comparable to those with biventricular patients. Reference Cohen, Vetter and Wernovsky15 Huntley et al. demonstrated that epicardial leads had higher rates of lead failure but similar generator longevity in comparison to endocardial leads in adult Fontan patients. Reference Huntley, Deshmukh and Warnes16 Impedance and MET value of both atrial and ventricular leads were stable in the long term. These results are consistence with the reports from Stanner et al. Reference Stanner, Horndasch and Vitanova13 and Huntley et al. Reference Huntley, Deshmukh and Warnes16 As for the influence of pacemaker rhythm in Fontan circulation, our results showed no influence on survival, incidence of thromboembolic complications, or incidence of protein-losing enteropathy/plastic bronchitis. The impact of pacemaker on survival is controversial. Fishberger et al. reported that the survival with pacemaker in patients after Fontan procedure was comparable to those without pacemaker, Reference Fishberger, Wernovsky and Gentles17 whereas Poh et al. demonstrated that having a permanent pacemaker and needing ventricular pacing is likely associated with an increased risk of death and transplantation in patients with a Fontan circulation. Reference Poh, Celermajer and Grigg18
Future prospective
In patients after extracardiac total cavopulmonary connection, conventional epicardial pacemaker implantation is highly invasive and a high-risk procedure due to repeat thoracotomy and instable haemodynamics of the patients, especially when patients are old. Repeated device hardware replacement, frequently required due to high rate of lead failure, is a challenging and significant problem for these patients. Recently, transcatheter transvenous pacemaker implantation in these patients is an emerging and promising technique, which has been established in selected institutes. Reference Hoyt, Moore and Shannon19–Reference Akazawa, Higaki and Nagai21 This techniques is less invasive, offers stable pacing electrical parameters, and solves the pacemaker lead dysfunction late after total cavopulmonary connection. Hoyt et al. showed the three patients who successfully applied this technique resulted in excellent midterm results without any detected complications. Reference Hoyt, Moore and Shannon19 Tunks et al. showed usefulness of this technique with the guide of a three-dimensional imaging model. Reference Tunks, Myers and Cohen20 Segar et al. demonstrated that transvenous pacemakers can be utilised with equal efficacy compared to epicardial pacemakers with trends towards longer lead longevity in transvenous pacemaker systems. Reference Segar, Maldonado, Brown and Law22 Future improvement of this technique might offer a novel solution and minimise the risk of pacemaker dysfunction for patients after extracardiac Fontan procedure.
Limitation
This study was limited by its retrospective, non-randomised, and single-centre design. Surgical and medical management may have changed during the study period, probably influencing the long-term outcomes. Because the follow-up periods were limited, impact of pacemaker rhythm on late outcomes such as systemic ventricular failure or Fontan-associated liver disease was not evaluated. Not all patients with a single ventricle have achieved Fontan completion. Given the incidence of inter-stage death, there might be a selection bias for this cohort.
Conclusions
In our large cohort of 620 patients following total cavopulmonary connection, freedom from permanent pacemaker implantation at 10, 15, and 20 years was 92, 91, and 88%, respectively. Epicardial leads can be safely placed surgically in patients during and after the staged Fontan palliation, and the use of epicardial leads is a reliable method with applicability in every child. Results after pacemaker implantation were excellent. Patients with congenitally corrected transposition of the great arteries are at increased risk for pacemaker implantation before total cavopulmonary connection. Having a pacemaker in the Fontan circulation had no adverse effect on survival, protein-losing enteropathy/plastic bronchitis, or thromboembolic complication.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951123002500
Acknowledgments
The authors thank Mrs. Martina Strbad for her excellent data management and secretarial support at the study center for congenital heart surgery, Department of Congenital and Pediatric Heart Surgery at the German Heart Center Munich.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Competing interests
None.
