Several surgical techniques can be used to repair transposition of the great arteries with ventricular septal defect and pulmonary stenosis. The traditional surgical technique is the Rastelli operation.Reference Rastelli, Wallace and Ongley 1 Long-term problems include sub-aortic obstruction caused by the long intra-cardiac tunnel and the need for re-operations due to homograft or xenograft replacements. These late complications and other individual-specific anatomic features, such as small restrictive ventricular septal defects, precluding the use of a Rastelli procedure, resulted in the search for alternative potentially better surgical approaches for a bi-ventricular repair in this group of patients. This led to the introduction of operations like Réparation à l’étage ventriculaire,Reference Lecompte, Neveux and Leca 2 aortic translocationReference Nikaidoh 3 , or the Ross–Konno switch operation.Reference Haas 4
More recently, the en bloc rotation of the arterial trunkReference Yamagishi, Shuntoh and Matsushita 5 has been introduced as a new option for anatomical repair. The aortic and the pulmonary arterial roots are cut out in one block and rotated by 180 degrees. This avoids the placement of a conduit from the right ventricle to the pulmonary artery, and may even preserve the pulmonary valve in the majority of cases. Furthermore, this may enable bi-ventricular repair in complex anatomical situations, when Rastelli-repair and other surgical techniques might fail as mentioned above; however, at present, there are no studies that can clearly demonstrate the superiority of one surgical procedure over others.
Arrhythmia, sudden cardiac death, and dyssynchrony are important factors for long-term morbidity and mortality in congenital heart surgery.Reference Lan, Lee and Wetzel 6 Although the impact of the Rastelli procedure, aortic translocation, and the Ross–Konno switch operation on arrhythmias has already been reported, there is no information available regarding the en bloc rotation of the arterial trunk and arrhythmias. Therefore, the aim of this study was to evaluate the effects of en bloc rotation of the arterial trunk on the conduction system.
Operative technique
The aorta is detached just above the coronary ostia. Similar to the arterial switch procedure, these ostia are cut out and mobilised. The pulmonary artery is separated at the same height as the aorta (Fig 1). After inspection of the pulmonary valve and sizing with Hegar dilators, a transverse incision is made into the sub-aortic conus (Fig 1a). It is necessary to excise the arterial trunk including the sub-arterial part of the conus septum totally en bloc. Then the remnant of the conus septum is transected (Fig 1b). It is important to spare the papillary muscle of the conus, papillary muscle of Lancisi. Subsequently, a U-shaped pericardial patch is used to close the ventricular septal defect by enlarging the left ventricular outflow tract at the same time. The aortic root is brought over the left outflow tract and the pulmonary root over the right ventricular outflow tract by rotating the truncal block by 180°. Then the truncal block is re-implanted (Fig 1c), and the coronaries are re-implanted to the aorta followed by a Lecompte manoeuvre.Reference Mair, Sames-Dolzer, Vondrys, Lechner and Tulzer 7 , Reference Mair 8
( a ) En bloc excision of the truncus arteriosus (dotted line). ( b ) Incision of the conus septum. ( c ) Re-implantation of the truncus arteriosus after rotation by 180°.
Methods
Since 2003, 16 consecutive patients with transposition of the great arteries, ventricular septal defect, and pulmonary stenosis were treated in our institution by en bloc rotation of the arterial trunk. Their median age at surgery was 103 days (inter-quartile range 369–16 days), the median operative weight was 4.5 kg (inter-quartile range 8.3–3.47 kg), and the median height was 55 cm (inter-quartile range 72–52 cm); ten patients had transposition of the great arteries with ventricular septal defect and pulmonary stenosis, and 6 patients had a double-outlet right ventricle with transposition of the great arteries and pulmonary stenosis. Among all, two patients needed a Rashkind procedure before the en bloc rotation; three other patients had had previous operations; and two patients had undergone a Glenn procedure in other centres, because bi-ventricular repair was not thought to be possible because of the complex anatomic situation and position of the VSD. Furthermore, one patient had undergone a modified Blalock–Taussig-shunt. In these three patients, the Glenn and Blalock–Taussig-shunt takedown was performed at the same time with the en bloc rotation. A trans-annular patch had to be placed in three patients; one patient had an 8p-syndrome, a rare chromosomal disorder with deletion of the short arm (p) of chromosome 8, resulting in various abnormalities including skeletal malformations, mental retardation, and congenital heart disease. Pre-operative, post-operative, and follow-up electrocardiograms were reviewed retrospectively for QRS duration, QRS pattern, re-polarisation abnormalities, and QTc. The charts of the patients were reviewed for relevant post-operative arrhythmias. The first electrocardiogram was obtained at median 3.5 (1–30) days before surgery, the first electrocardiograms after surgery at median 12 (3–27) days later. The median follow-up time was 585.5 (11–2572) days.
Results
During the follow-up period, one early death, but no late deaths, was reported; two patients developed grade I aortic regurgitation and five patients a residual supra-valvar or valvar pulmonary stenosis without the need for intervention during the study period; and three patients required re-operations. In addition, one patient had a re-operation 9 months after surgery because of a significant residual ventricular septal defect, one patient needed a re-operation 4 years later because of a pseudo-aneurysm, and one patient received a permanent pacemaker after 2 months. This patient with double-outlet right ventricle, transposition of the great arteries, and sub-valvar and valvar pulmonary stenosis had transient complete atrio-ventricular block post-operatively, which resolved within the first 10 days after surgery; however, Mobitz type II atrio-ventricular block developed during the follow-up needing subsequent pacemaker implantation. This patient also had supra-ventricular tachycardia post-operatively, which was treated with digoxin. After three months, the medication was stopped with no further recurrence. The patient who died had complete atrio-ventricular block along with double-outlet right ventricle, transposition of the great arteries, sub-valvar and valvar pulmonary stenosis, and straddling of mitral valve. Another patient with transposition of the great arteries, ventricular septal defect, sub-valvar and valvar pulmonary stenosis developed junctional ectopic tachycardia, and one patient, with transposition of the great arteries, ventricular septal defect, and valvar pulmonary stenosis, had short periods of ventricular tachycardia. In both the patients, short-term antiarrhythmic treatment was effective to terminate the arrhythmia with no further recurrence. All the patients except for the one patient on digoxin were discharged without antiarrhythmic treatment.
A transient atrial rhythm immediately after surgery was observed in one patient. None of the patients had a typical right bundle branch block pattern before surgery, but it was present in 8/16 patients (50%) after surgery and persisted during follow-up. All the patients, without right bundle branch block pattern, had a median QRS duration of 65 ms (54–112 ms) before surgery, 62 ms (54–122 ms) after surgery, and 84 ms (66–128 ms) at the last follow-up visit. In the right bundle branch block group, the median QRS duration was 100 ms (86–116 ms) immediately after surgery and 100 ms (92–128 ms) at the last follow-up. In 12 of 16 patients, there were non-specific ST changes and negative T-waves, which persisted during the follow-up with unknown significance for the future.
Discussion
It has been shown that the en bloc rotation of the arterial trunk can be performed safely and with promising results even in small children.Reference Mair, Sames-Dolzer, Vondrys, Lechner and Tulzer 7 This provides a better post-operative physiological result than other surgical techniques with the native aortic valve in the systemic circulation and a right ventricular outflow tract with either the native pulmonary valve or a trans-annular patch. Significant damage to the conduction system caused by en bloc rotation may place these advantages at risk. The data of this retrospective study demonstrate that, despite the very complex nature of cardiac surgery, the negative effects of this procedure on the conduction system were low and comparable with other surgical techniques.
Arrhythmia, sudden cardiac death, and dyssynchrony are important factors for early and long-term morbidity and mortality in congenital heart surgery. The risk factors for early post-operative arrhythmias reported in the literature include longer cardiopulmonary bypass time and higher surgical complexity. The step-wise approach of incision of the sub-aortic conus septum, excision of the complete arterial trunk, and transection of the conus septum during the en bloc rotation of the arterial trunk may cause damage to the conduction system.
The prevalence of arrhythmias early in the post-operative period in congenital heart disease may be up to 70–80%, including minor rhythm disturbances such as premature ectopic beats. Grosse–Wortmann et al reported relevant rhythm disorders in 29.6% of neonates and in 38.9% of older children by Holter monitoring in a comprehensive evaluation of post-operative arrhythmias following surgery for congenital heart disease.Reference Grosse-Wortmann, Kreitz and Grabitz 9 In our study, there was a similar prevalence of 25% (four patients) with relevant rhythm disorders. Grosse–Wortmann et al reported 7% supra-ventricular tachycardia and 3% ventricular tachycardia, which compares well with our data, where we had a prevalence of 6% (one supraventricular tachycardia and one ventricular tachycardia).
Junctional ectopic tachycardia is a serious tachyarrhythmia with haemodynamic impact. It has a reported mortality of up to 14%.Reference Mildh, Hiippala, Rautiainen, Pettilä, Sairanen and Happonen 10 Longer cardiopulmonary bypass times and higher levels of troponin-T, as a marker for significant myocardial injury after open-heart surgery in infants, are reported as possible risk factors for developing junctional ectopic tachycardia post-operatively. The anatomy and the surgical procedure may contribute to the overall risk. Therefore, it was a major concern that the extensive surgery of en bloc rotation of the arterial trunk may lead to a higher prevalence of junctional ectopic tachycardia. In our small series of patients, we have found a prevalence of 6%, which compares well with the 2–15% reported in earlier studies.Reference Mildh, Hiippala, Rautiainen, Pettilä, Sairanen and Happonen 10 , Reference Zampi, Hirsch and Gurney 11 Zampi et al reported prevalence of 25.6% in a sub-group of patients with transposition of the great arteries and ventricular septal defect.Reference Zampi, Hirsch and Gurney 11
Complete atrio-ventricular block is of major concern after complex cardiac surgery involving ventricular septal defect closure. The reported range in the literature varies between 4 and 14%. The need for pacemaker implantation due to complete atrio-ventricular block after the Rastelli operation was reported to be 4% in one study.Reference Hazekamp, Gomez and Koolbergen 12 In another comparable Rastelli cohort with a mean follow-up of 8.6±5.6 years, Brown et al reported a prevalence of 5% requiring permanent pacemaker implantation.Reference Brown, Ruzmetov, Huynh, Rodefeld, Turrentine and Fiore 13 Di Carlo et al reported a prevalence of 8% in their long-term results of the Réparation-à-l’Étage-ventriculaire procedure.Reference Di Carlo, Tomasco, Cohen, Vouhé and Lecompte 14 Höhrer et al had five patients (12.8%) who developed complete atrio-ventricular block post-operatively and 8% who required permanent pacemaker implantation.Reference Hörer, Schreiber and Dworak 15 Honjo et al reported that two out of 14 (14%) patients who underwent a Nikaidoh or Rastelli operation needed permanent pacemaker implantation during a median follow-up period of 65 (4–153) months.Reference Honjo, Kotani and Bharucha 16 In our study, we had two cases of atrio-ventricular blocks, representing 12.5% of our patient cohorts. This relatively high prevalence may be due to the fact that in one of these two patients the atrio-ventricular block was most likely due to additional extensive cardiac surgery with a Glenn takedown, reconstruction of the pulmonary arteries, and repair of a straddling mitral valve.
Right bundle branch block may be an electrocardiographic marker for the risk of late sudden cardiac death and arrhythmia. Prolonged QRS duration may lead to adverse left ventricular mechanics with bi-ventricular electromechanical dyssynchrony, as reported in adults with repaired tetralogy of Fallot.Reference Berul, Hill and Geggel 17 , Reference Tzemos, Harris and Carasso 18 In a comparable Rastelli cohort, with a mean follow-up of 8.6±5.6 years, Brown et al reported a prevalence of 77% for post-operative right bundle branch block, and Höhrer et al a prevalence 64% in their cohort of nine patients. The right bundle branch block prevalence of 50% in our study was even lower and compares well with the reported data.
Limitations
A limiting factor is the relatively short period of observation. It is known that ventricular tachycardia, sudden cardiac death, and complex forms of re-entry tachycardia causing significant morbidity can develop later in life. Another limitation is the low number of patients. More patients and longer follow-up periods are needed to confirm the results of this study.
Conclusion
Although en bloc rotation of the arterial trunk is a complex surgical procedure with extensive placement of sutures and long bypass time, negative effects on the conduction system were limited and comparable with other surgical procedures.
Acknowledgements
None.
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of Interest
None.
