'Slow pathway' ablation in patients with atrioventricular nodal reentrant tachycardia: do we understand what we are doing?

midseptal area of the inter-atrial septum. But what is the anatomical substrate of the slow pathway? Inferior atrionodal connections with nodal-type electrophysiology, as well as the inferior nodal extension have been proposed as possible substrates. And what is the mechanism of ‘slow’ conduction through this pathway? Apart from normal conduction through just a longer path, slowing of conduction due to the characteristics of depolarizing membrane currents, electrical cell-to-cell uncoupling, and discontinuous conduction due to branching tissue architecture have been discussed as possible mechanisms. As pointed out above, the risk of atrioventricular nodal damage during catheter ablation in patients with atrioventricular nodal reentrant tachycardia and normal atrioventricular conduction parameters is low. However, in patients with a prolonged PR interval at baseline, one would intuitively assume that the risk of atrioventricular block is higher. A prolonged PR could indicate impaired or absent antegrade function of the fast pathway, and radiofrequency ablation targeting of the slow pathway would subsequently lead to atrioventricular block. This reasoning, however, is far too simplistic. The conduction delay can take place anywhere in the conducting axis between the atrium and the His bundle without affecting the limbs of the reentrant circuit. In addition, more than two inputs to the atrioventricular node certainly exist. Indeed, several studies have suggested that slow pathway ablation can be safely performed in patients with atrioventricular nodal reentrant tachycardia and prolonged PR intervals. The prospective study by Li et al. in this issue is a welcome addition. Li et al. describe the acute results of radiofrequency catheter ablation and the follow-up in a study population of 346 patients presenting with typical and atypical atrioventricular nodal reentrant tachycardia. Out of this group, 18 patients had a prolonged PR interval at baseline. Interestingly, impaired conduction was not restricted to the atrioventricular node (A-H 156 28 vs 83 16 ms, P<0·0001; Wenckebach cycle length 378 72 vs 320 60 ms, P=0·003), but extended also to the His-Purkinje system (H-V 52 7·4 vs 45·8 9·1 ms, P<0·001) and intraventricular conduction (QRS 121 28 vs 103 17 ms, P=0·01). Patients See page 89 for the article to which this Editorial refers