BackgroundJunctional ectopy may occur during radiofrequency (RF) catheter ablation of the slow pathway in patients with atrioventricular nodal reentrant tachycardia (AVNRT). The purpose of the present study was to characterize this junctional ectopy quantitatively.Methods and ResultsThe subjects of this study were 52 consecutive patients with AVNRT who underwent slow pathway ablation and 5 additional patients included retrospectively because they had developed high-degree atrioventricular (AV) block during the procedure. A combined anatomic and electrogram mapping approach was used for slow pathway ablation, and AVNRT was successfully eliminated in all patients. In the group of 52 consecutive patients, the incidence of junctional ectopy was significantly higher during 52 effective applications of RF energy than during 366 ineffective applications (100% versus 65%,P< .001). Compared with ineffective RF energy applications, successful RF energy applications had a significantly longer duration of individual bursts of junctional ectopy (7.1 ± 7.1 versus 5.0 ± 7.0 seconds [± SD],P< .05), a greater total number of junctional beats during the applications (24 ± 16 versus 15 ± 8,P< .01), and a greater total span of time during which junctional ectopy occurred (19 ± 15 versus 11 ± 12 seconds,P< .01). Four of the 52 patients plus an additional 5 patients developed transient AV block lasting 34 ± 37 seconds. In 1 of the 9 patients who had transient AV block, third-degree AV nodal block requiring a permanent pacemaker recurred 2 weeks later. In each of the 9 patients who developed AV block, there was ventriculoatrial (VA) block in association with junctional ectopy during the RF energy application immediately preceding the AV block. Among 48 patients who did not develop AV block, 17 patients had at least one episode of VA block during junctional ectopy. The positive predictive value of VA block during junctional ectopy for the development of AV block was 19% in the consecutive series of 52 patients. Among 31 patients who always had 1:1 VA conduction in association with junctional ectopy, 12 had poor VA conduction in the baseline state, with a VA block cycle length of at least 500 milliseconds during ventricular pacing.ConclusionsIn patients with AVNRT undergoing slow pathway ablation, junctional ectopy during the application of RF energy is a sensitive but nonspecific marker of successful ablation. The bursts of junctional ectopy are significantly longer at effective target sites than at ineffective sites. VA conduction should be expected during the junctional ectopy that accompanies slow pathway ablation, even when there is poor VA conduction during baseline ventricular pacing. VA block during junctional ectopy is a harbinger of AV block in patients undergoing RF ablation of the slow pathway. If energy applications are discontinued as soon as VA block occurs, the risk of AV block may be markedly reduced.