Case report
A 6-year-old male patient, after total cavopulmonary connection Fontan surgery for atrial situs solitus and atrioventricular concordance {S, D, D}, an upstair–downstair heart, i.e., a concordant criss-cross heart, a double outlet right ventricle with a small right ventricle and a non-committed ventricular septal defect and pulmonary stenosis, a persistent left superior caval vein was referred for supraventricular tachycardia. His electrocardiography showed a long RP’ narrow superior axis QRS regular tachycardia with a heart rate of 205 beats per min (Figure 1( a )). The first session of the electrophysiological study was performed using a 3D mapping system (CARTO®, Biosense Webster, Diamond Bar, CA, USA) under deep sedation with mechanical ventilation by tracheal intubation. We were forced to take a retroaortic approach because the conduit puncture approach was difficult due to the presence of a coronary vein between the atrium and the conduit located on the left side of the heart, on account of funnel chest, in addition to the dextrocardia (Figure 1( b )). We detected a His-bundle electrogram only at the posterior side of the atrial septum during sinus rhythm and atrial pacing from multiple sites. Ventricular pacing unmasked a ventriculoatrial pathway with decremental conduction. Ventricular extrastimulus pacing induced the clinical supraventricular tachycardia with a tachycardia cycle length of 280 ms following the ventriculoatrial jump (Figure 1( c )). Although the A-A interval was sometimes irregular, the tachycardia cycle length was determined by the preceding H-H interval, and the administration of adenosine terminated the tachycardia with ventriculoatrial block. Therefore, we diagnosed the tachycardia as fast-slow atrioventricular nodal reentrant tachycardia with an upper common pathway. As mapping during tachycardia continued, the tachycardia shifted to ventricular fibrillation due to drug-resistant hypotension. We had no choice but to defibrillate and complete the procedure.
( a )The 12-lead electrocardiograms of the long RP’ narrow QRS tachycardia with a heart rate of 205 beats per min. Red arrow: P’ wave preceding the QRS. ( b ) The schematic drawing of the patient’s heart morphology after Fontan surgery. The coronary sinus was enlarged by perfusion of the left superior caval vein into the right atrium in the period up to Glenn surgery. Ao = aorta; CSos = coronary sinus ostium; ICV = inferior caval vein; LA = left atrium; LSCV = left superior caval vein; LV = left ventricle; PA = pulmonary artery; RA = right atrium; RSCV = right superior caval vein; RV = right ventricle. ( c ) Surface and intracardiac electrocardiograms showing the induction of the tachycardia. His-bundle electrogram (blue arrow) preceded the ventricular potential during the tachycardia following the ventriculoatrial jump by ventricular extrastimulus pacing. d = distal; His = His potential (the catheter placed at the postero-septal wall of the right. atrium via aorta); p = proximal; RV = right ventricle (the catheter placed at the right ventricle via aorta).

The second session was performed under shallow sedation with the oral airway. To increase the information on atrial activation sequence, we used the oesophageal catheter and two ring-shaped catheters for the upper and lower positions of the conduit (Figure 2( a )). Verapamil (1 mg, 0.06 mg/kg) was administered in order to prolong the tachycardia cycle length. After recovery from a transient loss of ventriculoatrial conduction, clinical supraventricular tachycardia with a cycle length of 480 ms and constant A-A intervals was induced in a similar property by programmed stimulation. The tachycardia cycle length was prolonged, with an increase in the atrioventricular interval and no change in the ventriculoatrial interval (Figures 2( b ), 2( c )). Prior to verapamil administration, blood pressure was difficult to maintain during tachycardia, but it was well maintained after administration. We detected the earliest atrial potential with a quasi-slow pathway potential during the tachycardia at the coronary sinus ostium, approximately 5 mm below the His-bundle electrogram recording site with stable haemodynamics (Figures 2( b ), 2( c )). Radiofrequency energy (25W) using a 4-mm tip catheter (Thermocool smarttouch®, Biosense Webster, Diamond Bar, California, USA) for this site terminated the tachycardia. After the termination of the tachycardia, the accelerated junctional rhythm did not occur during the application of radiofrequency energy. Finally, we confirmed the disappearance of ventriculoatrial conduction after ablation. The tachycardia was no longer inducible and has not recurred for more than 7 years of follow-up.
( a ) The computed tomography and fluoroscopic images showing successful ablation site. (Upper) coronal view of the atrioventricular valve position. ABL = successful ablation site; CS = coronary sinus; orange circle = two patent atrioventricular annulus (MA = mitral annuls; TA = tricuspid annulus); His = His potential recording site. (Middle) fluoroscopic image in anterior view. (Lower) fluoroscopic image in left lateral view. ABL = the ablation catheter placed at the successful ablation site via aorta; Eso = the oesophageal catheter; RV = the catheter placed at the right ventricle via aorta. ( b , c ) Surface and intracardiac electrograms during tachycardia before ( b ) and after ( c ) verapamil administration. After verapamil administration, prolongation of the atrioventricular interval without change in the ventriculoatrial interval was observed. Both intervals were measured using the oesophageal catheter, which clearly recorded atrial and ventricular potentials before and after verapamil administration. The ablation catheter recorded earliest atrial excitation with a quasi-slow pathway potential (red circle in ( c ). ABL = the ablation catheter placed at the successful ablation site via aorta; Eso = the oesophageal catheter; RV = the catheter placed at the right ventricle via aorta.

Discussion
Catheter ablation for atrioventricular nodal reentrant tachycardia associated with complex CHD after total cavopulmonary connection Fontan surgery is challenging, because of the lack of anatomical indication of the slow pathway as well as the difficulty in catheter access. Moreover, persistent tachycardia after Fontan surgery can easily lead to low cardiac output and elevated central venous pressure associated with increased ventricular filling pressure. Therefore, the risk of haemodynamic collapse is very high in cases requiring slow pathway mapping during tachycardia. This is the first report demonstrating the usefulness of verapamil for stabilisation of haemodynamics during tachycardia. The optimal dosage of verapamil for this purpose is unknown. In our case, although a transient ventriculoatrial block was observed, 1 mg (0.06 mg/kg) of verapamil was effective. Haemodynamics stability during tachycardia induced within minutes after ventriculoatrial block recovery was maintained without blood pressure decrease due to the prolonged tachycardia cycle length. Verapamil prolonged the tachycardia cycle length by extending the atrioventricular interval without affecting the ventriculoatrial interval during tachycardia (Figures 2(b), 2(c)). It remains unclear whether this is simply a result of conduction delay of the fast pathway or indicates a transition of the reentrant circuit from fast-slow to slow-slow atrioventricular nodal reentrant tachycardia. Half-life of verapamil is less than 15 minutes; hence, the risk of verapamil use during electrophysiological study is not high under conditions where a ventricular pacing backup is ensured. Therefore, verapamil would be useful for activation mapping during supraventricular tachycardias associated with the atrioventricular node in patients after Fontan surgery. More fortunately, verapamil also contributed to 1:1 ventriculoatrial conduction during tachycardia, although it is not clear whether this was associated with a slowing of the tachycardia rate or a direct effect on the conductivity of the upper common pathway. When the upper common pathway complicates the diagnosis of tachycardia, administering verapamil may have diagnostic value and could be worth attempting. Landiolol, an ultra-short-acting beta-1 blocker, was another option to avoid haemodynamic collapse. Although it is generally easy to adjust to optimal dosage, we did not select it in this case due to the possibility of adverse effects such as an asthma attack and an increase in pulmonary vascular resistance because of his history of asthma and relatively high central venous pressure after Fontan surgery.
Catheter ablation for atrioventricular nodal reentrant tachycardia in a normal structural heart is usually performed within the triangle of Koch during sinus rhythm. On the other hand, electrophysical features of atrioventricular nodal reentrant tachycardia and anatomical location of the slow pathway associated with complex CHDs are still unknown.Reference Papagiannis, Daniel and Ulrich1 Especially, the cardiac conduction system in criss-cross hearts with complex anatomical atrioventricular connections, as in the present case, has not been reported. Therefore, atrial activation mapping during tachycardia is usually required. In our case, the coronary sinus was relatively enlarged because the persistent left superior caval vein perfused the right atrium with venous blood from the left upper body via the coronary sinus until bilateral bidirectional Glenn surgery at 6 months of age. The relationship between persistent left superior caval vein and supraventricular tachycardia has been reported.Reference Uhm, Choi and Baek2,Reference Hwang, Park and Kim3 Uhm et al. reported that a slow pathway at the coronary sinus ostium was significantly more frequent in patients with persistent left superior caval vein rather than in those without persistent left superior caval vein.Reference Uhm, Choi and Baek2 Although the location of the slow pathway in patients with complex CHD is unknown, we should start mapping from the coronary sinus ostium position, especially in cases with persistent left superior caval vein.
Acknowledgements
None.
Financial support
This research received no specific grants from any funding agency or from commercial or not-for-profit sections.
Competing interests
None.
Ethical standard
The authors assert that all procedures contributing to this study comply with the ethical standards of relevant guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. Our Institutional Review Board has confirmed that no ethical approval is required because this is a case report.
