Successful ablation of a concealed accessory pathway in the coronary sinus in a patient with mechanical mitral valve

Patients treated with cardiac surgery often present with supraventricular arrhythmias, classically atrial fibrillation and atrial flutter. Accessory pathways (APs) are a rare finding in these patients, except for congenital heart disease (eg, Ebstein anomaly). Only a few case reports of AP ablation at the site of previous valve surgery have been published. Ablation in this setting is challenging and carries the risk of injuring the prosthetic valve. We report a successful ablation of a concealed left-side AP in the coronary sinus (CS) in a patient with previous mechanical mitral valve replacement for rheumatic valve disease.

A 47-year-old woman was referred to our hospital for management of recurrent supraventricular tachycardia. She had a history of rheumatic valve disease and she had undergone mitral valve replacement with a mechanical prosthesis for severe mitral regurgitation at the age of 36 years. Her recurrent palpitations were initially managed with amiodarone and beta blockers, although without complete tachycardia suppression.

Her baseline electrocardiogram (ECG) showed sinus rhythm, nonspecific ST-T abnormalities, QT prolongation, and no evidence of preexcitation (Figure 1A ). During palpitations, the 12-lead ECG demonstrated a regular narrow QRS tachycardia at a rate of 130 beats per minute (Figure 1B).

An electrophysiological study was performed under conscious sedation using a 3-dimensional mapping system (CARTO system; Biosense Webster, Diamond Bar, CA). The electrophysiological study was performed on uninterrupted anticoagulation and the international normalized ratio was 2.5 on the day of the procedure. A decapolar catheter was placed in the CS, a quadripolar catheter in the right ventricular apex, and a 4 mm nonirrigated ablation catheter in the His bundle region. Incremental ventricular pacing showed a retrograde eccentric pattern of atrial activation in the CS, earliest at the level of CS electrodes 5–6. Retrograde AP refractory period was obtained at 600/370 ms. Incremental atrial pacing did not show ventricular preexcitation. Programmed atrial pacing induced narrow QRS tachycardia with a ventricular-atrial (VA) time of 90 ms. VA linking was noticed after atrial overdrive pacing during tachycardia, ruling out atrial tachycardia. Ventricular overdrive pacing (Figure 2) elicited a VAV response with a postpacing interval minus tachycardia cycle length (PPI−TCL) of 60 ms. The stimulus-atrial minus ventriculoatrial interval difference was 60 ms. A premature ventricular contraction, delivered from the right ventricular apex 50 ms before the His depolarization, advanced the subsequent atrial deflection by 50 ms. A diagnosis of orthodromic atrioventricular (AV) reentrant tachycardia through a concealed left-sided AP was made.

Mapping the mitral annulus using either the transseptal approach or the retrograde approach would have been extremely challenging owing to difficulties in obtaining true annular signals and the risk of catheter entrapment or interference with the mechanical valve. We therefore decided to map the CS with a PentaRay catheter (Biosense Webster, Diamond Bar, CA) (Figure 3B ). During tachycardia, the earliest atrial signal with VA fusion was observed close to CS electrodes 5–6. A CS venogram was performed and no CS diverticulum was seen (Figure 3A). We decided to start ablating using a nonirrigated catheter, carefully monitoring the ECG for ST-T changes. Ablation at the site of earliest atrial electrogram with a power of 20 watts and a temperature limit of 60°C terminated the tachycardia after 4 seconds; however a premature ventricular contraction showed the pathway was still conducting retrogradely (Figure 3C). No ST-T changes on the ECG and no impedance rise were observed during ablation, so 40 additional seconds of ablation were performed at the same location. Subsequently, VA block was observed during ventricular incremental pacing and tachycardia was no more inducible with programmed and incremental atrial pacing. The tachycardia did not reoccur during a 10-month follow-up period after catheter ablation.

The prevalence of manifest APs in the general population is estimated to be at 0.25%, and around 20% of the APs, depending on the clinical series, demonstrate retrograde-only conduction. So, except for congenital heart disease patients, the odds of finding a concealed AP in a patient who underwent valve surgery are low. During surgery, direct trauma secondary to stitching and debridement, tissue remodeling, and ischemia have been identified in pathologic studies as the mechanism of postoperative AV nodal conduction block. During mitral valve surgery, the same factors might impair conduction through an AV AP anatomically close to the endocardial side of the mitral annulus. Despite this, we report a case of a concealed left-side AP in a 47-year-old woman with previous mechanical valve replacement. Only anecdotal cases of successful AP ablation at the site of previous valve surgery have been reported. Epicardial AV pathways or valve deployment on the atrial aspect of the annulus, rather than on the true annulus, might explain intact AP conduction after valve surgery. Interestingly, surgical creation of AV connections has been reported after cardiac surgeries in congenital heart disease patients^{,  ,}  and might be conceivable after rheumatic valve surgery. In patients with a mitral mechanical valve, few isolated cases of endocardial ablation of an AP have been reported, making this approach potentially feasible. Nonetheless, a transprosthetic approach carries life-threatening risks of catheter entrapment, thrombosis, and prosthesis damage.^{,  ,  ,}  Adverse events have been reported mainly in patients with caged-ball or tilting-disc valves^,; however complications of crossing a bileaflet mechanical valve might be unreported, and therefore this approach is not recommended. Mapping above or below the prosthesis with a transseptal or transaortic approach may be an option, especially if the prosthesis is located slightly toward the atrial side. However, it is difficult to determine the exact position of the mechanical valve based on imaging alone, and so planning the right approach is extremely challenging. In a case series of left-side AP ablation in patients with mechanical mitral valve, a transaortic approach was successful in 2 cases and CS ablation and bipolar ablation obtained AP suppression in another 2 patients. Aslani and colleagues reported catheter ablation of an AP within the Marshall vein in a patient with mechanical mitral valve after a failed transseptal approach. In patients with mechanical mitral valve the chances of an AP lying deep in the atrioventricular groove might be higher than the general population. This factor, together with the risks related to catheter manipulation through or close to the mechanical prosthesis, might justify mapping the CS as the first step. In the early seventies, mapping and ablation in the CS was proposed as a first-line approach for left-side pathways, aiming to avoid all the risks related to the arterial access, transaortic, and transseptal approach. Nonetheless, ablating in the CS carries risks as well, including coronary arterial injury that could manifest as acute myocardial ischemia or chronic coronary stenosis. Appropriate precautions must be adopted to perform a successful and reasonably safe ablation in the CS. An irrigated ablation catheter is likely to have superior efficacy and less likelihood of char and thrombus formation, particularly in a low-flow structure as the CS. We were successful in this case with ablation with a nonirrigated catheter and did not need to switch to an irrigated ablation catheter. Ablations should be started at low power (10–15 W) and the power increased gradually to 25 W, until conduction block occurs across the AP. Careful monitoring of impedance and prompt termination of ablation in case of impedance rise (≥5 Ω) or ST-T changes is warranted. Ideally a preablation coronary angiogram should be performed, and ablation aborted in case of proximity (less than 5 mm) to the coronary arteries.

In conclusion, this case emphasizes successful ablation of a concealed left-side AP in the CS in a patient with previous mitral valve replacement with a mechanical prosthesis. Mapping in the CS and ablation in case of satisfactory signals may represent a first-line strategy in these patients. A transseptal/transaortic route may be pursued only in case of failure.

We would like to thank Omar Ayaz for editing the figures of this manuscript.