Atrioventricular nodal reentrant tachycardia (AVNRT) is a supraventricular tachycardia (SVT) that arises from a reentry circuit within the region of the atrioventricular node (AVN) (Fig. 1). It should be distinguished from atrioventricular reentrant tachycardias, Wolf-Parkinson-White syndrome for example, which are caused from accessory atrioventricular connections away from the direct region of the AVN, but also causing reentrant circuits.1 Antegrade conduction in AVNRT occurs by slow pathways posteroinferior to the AVN near the base of the triangle of Koch. Retrograde conduction is via fast pathways anterosuperior to the AVN around the apex of the triangle of Koch (Fig. 1). Controversy remains whether reentrant pathways in AVNRT are functional or anatomic.2
Medical management is the mainstay of treatment for AVNRT. In the past, treatment for AVNRT included surgical division of the bundle of His, which resulted in permanent third-degree heart block and often required implantation of a permanent pacemaker. In 1982, Scheinman et al3 demonstrated percutaneous ablation of the bundle of His. In the mid-1980s, Dr. Cox developed a surgical AVN modification whereby cryosurgery was performed on the peri-atrioventricular (AV) nodal tissue to disrupt the AV nodal input pathways. The development of this surgical procedure was to avoid third-degree heart block that often followed percutaneous ablation in its early days.4,5 Currently, this complication is rare with the percutaneous approach, which now, along with medical therapy, almost entirely defines the management of this disease process. Nevertheless, percutaneous ablation occasionally fails. We describe the robotic approach to such a case in a patient with concomitant atrial fibrillation and moderate mitral insufficiency.
A 65-year-old man presented with a several-year history of palpitations and AVNRT. Two years before his surgical procedure, the patient underwent electrophysiological study and attempted ablation for his paroxysmal SVT. During that study, he was found to have an easily inducible typical slow-fast AV nodal as well as slightly prolonged atrial His bundle (AH) interval. Catheter cryoablation of the slow pathway AV nodal complex was partially successful because of the fact that the entire slow pathway could not be ablated secondary to mild AH prolongation during cryoablation. At this point, it was decided to treat the patient with beta-blocker therapy, and in case that medical therapy failed, further ablative procedures would be considered; however, because of the proximity of his slow pathway tissue to his AV node, further catheter ablation would be unlikely. Despite treatment with beta-blockers, the patient had at least weekly episodes of symptomatic SVT. In addition, the patient was being followed up for moderate-severe mitral regurgitation and atrial fibrillation and now exhibited New York Heart Association class II heart failure symptoms. The patient was otherwise healthy aside from previous appendectomy and tonsillectomy. Transesophageal echocardiography demonstrated an anteriorly directed regurgitant mitral jet resulting from posterior leaflet prolapse and normal ventricular function. Coronary angiography revealed normal coronary arteries.
Under general anesthesia with a double-lumen endotracheal tube, the patient was positioned in the left lateral decubitus position. Percutaneous right internal jugular venous cannulation and right femoral venous and arterial cannulation via a 2-cm incision were attained. A 4-cm right anterolateral thoracotomy was made, and retraction was provided by a soft tissue retractor. The pericardium was opened anterior to the phrenic nerve and retracted, and the right hemithorax was flooded with carbon dioxide. Ports for the robotic arms were positioned in the third, fourth, and sixth intercostal spaces. A transthoracic cross-clamp was positioned, and after the establishment of cardiopulmonary bypass, a cardioplegia cannula was secured in the ascending aorta.
Temporary epicardial pacing wires were positioned directly on the right atrium and ventricle to monitor AV conduction. Upon entering the operating theater, the patient maintained a regular SVT but developed atrial fibrillation during cannulation. Because the atrial fibrillation was refractory to electrical cardioversion, we decided to proceed with concomitant maze procedure. The robot was deployed, a right atriotomy was made, and the da Vinci left atrial retractor was used to expose the triangle of Koch. Cardiotomy suckers were placed in the coronary sinus and right atrium, as the heart was not arrested for the AVN modification to continually assess the electrocardiogram. A series of cryolesions were made along the tendon of Todaro beginning at the coronary sinus and moving to the apex of the triangle of Koch, as previously described by Cox et al5 and depicted in the inset of Figure 1 (see Video http://links.lww.com/INNOV/A7, which demonstrates the surgical procedure). Heart block was not attained while making these lesions, so we proceeded with the second lesion along the septal leaflet of the tricuspid valve, again starting from the coronary sinus. While working up toward the apex of the triangle of Koch, heart block developed. A couple of lesions were then made along the base of the triangle of Koch at the coronary sinus.
We used the ATS cryoprobe (CryoMaze Probe; ATS Medical, Plymouth, MN USA) with just the tip of the probe exposed. All lesions were made for 1 minute or until heart block developed. A 5-mm laparoscopic suction-irrigator using room-temperature saline is efficient at thawing the frozen myocardium while minimizing the amount of saline returned to the cardiopulmonary bypass machine. Visualization of the triangle of Koch was excellent using the robot. The da Vinci probe graspers were used to manipulate the cryoprobe. It is important to be sure that the cryolesions overlap.
The operation was completed with a Cox-Maze III lesion set using cryothermy, as described previously,6 and a mitral valve repair consisting of a triangular resection and implantation of a 37-mm annuloplasty band.
Postoperatively, the patient did well aside from urinary retention that delayed his discharge until postoperative day 7. At 3 weeks postoperatively, an electrocardiogram demonstrated first-degree heart block, and at 14 months, a stress test demonstrated normal sinus rhythm; up to this point, the patient remained entirely asymptomatic without rhythm modulating medication or anticoagulation. Echocardiography at 10 months found only trace mitral regurgitation, and the left ventricular ejection fraction was 60%. More rigorous follow-up was not possible as the patient would not consent for continuous monitoring.
Although surgical division of the AVN is effective and permanent, cryosurgical AVN modification acts as a “reversible knife,” permitting the surgeon to “creep up” on the AVN and bundle of His along a predefined anatomic lesion set without needing to directly identify the AVN. Following the described lesion set, the surgeon can ablate the reentrant pathways without mapping them. When cryoablation continues to the point of heart block as described above, the reentrant pathways should be contained within the ablation line. In addition, any potential future pathways should also be contained within the same anatomic lesion set. Consequently, we did not attempt to perform a directed ablation of the AVNR pathway. This procedure has been shown to be reproducible and safe both clinically and in animal models, so long as the lesions are made along the borders of the triangle of Koch as indicated and the probe is thawed upon the development of AV block. In a series of eight patients with AVNRT, this procedure performed via sternotomy effected a long-term cure in all patients without creating AV block. Even without medication, no patient experienced a recurrence of AVNRT in up to 5 years of follow-up.
The development of atrial fibrillation in this patient made it difficult to measure the AV interval. In this case, we could still successfully assess the effect of cryoablation by monitoring the ventricular electrocardiogram and watching for bradycardia and prolongation of the QRS complex. The onset of these changes in the electrocardiogram implies that origin of the conduction impulse had moved further down the conduction system. This finding further implies that the lesion extends to the normal conduction system and, therefore, that all additional inputs to the AVN have been treated.
Robotic assistance provided a number of important benefits. Excellent visualization and improved control of the cryoprobe over thoracoscopic instrumentation afforded by the robotic system helped to ensure that all cryolesions for the AVN modification and Cox-Maze procedure are overlapping. Any gaps between lesions would place the patient at high risk of recurrence. In addition, this procedure was an important step in the progression toward a completely endoscopic combined procedure.
Unfortunately, the patient did not consent for continuous monitoring after the procedure. However, the patient experienced significant symptoms during AVNRT episodes preoperatively, but was completely asymptomatic up to 14 months postoperatively while remaining off medication. All postoperative electrocardiograms demonstrated either first-degree heart block (at 3 weeks) or normal sinus rhythm.
Although this patient could have been offered a repeat ablation followed by a Cox-Maze procedure and mitral valve repair, we instead offered the patient a “one-stop-shop” by performing all three procedures concomitantly. We felt that this minimally invasive robotic approach was a reasonable given the safety and efficacy of each individual procedure. Although effective, it is not likely that robotic AVN modification will become a mainstream treatment option for AVNRT because of the success of the catheter-based approach. Robotic AVN modification may be an effective, minimally invasive treatment of AVNRT when medical and catheter-based treatment failed and in patients requiring concomitant robotic procedures.
The authors thank Dawne Colwell for her assistance in preparing the figures.
1.Lee P, Chen S, Hwang B. Atrioventricular node anatomy and physiology: implications for ablation of atrioventricular nodal reentrant tachycardia. Curr Opin Cardiol.
2.Kwaku KF, Josephson ME. Typical AVNRT—an update on mechanisms and therapy. Card Electrophysiol Rev.
3.Scheinman MM, Morady F, Hess DS, Gonzalez R. Catheter-induced ablation of the atrioventricular junction to control refractory supraventricular arrhythmias. JAMA.
4.Holman WL, Ikeshita M, Lease JG, et al. Alteration of antegrade atrioventricular conduction by cryoablation of peri-atrioventricular nodal tissue. Implications for the surgical treatment of atrioventricular nodal reentry tachycardia. J Thorac Cardiovasc Surg.
5.Cox JL, Holman WL, Cain ME. Cryosurgical treatment of atrioventricular node reentrant tachycardia. Circulation.
6.Rodriguez E, Cook RC, Chu MW, et al. Minimally invasive bi-atrial CryoMaze operation for atrial fibrillation. Operat Tech Thorac Cardiovasc Surg.