Catheter positioning for radiofrequency ablation of extranodal slow pa
thways is often guided by local electrogram recordings. To determine t
he predictors of successful ablation sites, we reviewed data from 32 s
uccessful and 104 unsuccessful sites. Univariate predictors of a succe
ssful site included the occurrence of junctional rhythm during ablatio
n (P < 0.001), shorter time to onset of junctional rhythm (P = 0.05),
the presence of a discrete slow pathway potential (P < 0.001), a small
er ratio of the amplitude of the atrial:ventricular electrogram (P = 0
.04), later timing (P = 0.001) and longer duration (P < 0.001) of the
atrial slow pathway electrogram, and the duration of (P < 0.001), and
maximal voltage used during ablation (P < 0.001). By multivariate anal
ysis junctional rhythm (P < 0.001), a discrete slow pathway potential
(P = 0.003), a longer duration of the atrial slow pathway electrogram
(P = 0.01) and the duration of ablation (P = 0.02) were predictors of
success. Because ablations at unsuccessful sites were often aborted at
10-30 seconds, a separate analysis was performed using only the 41 un
successful sites where the duration of ablation was greater-than-or-eq
ual-to 30 seconds. The results were nearly identical. Thus, the occurr
ence of junctional rhythm during ablation and the morphology and durat
ion of the atrial slow pathway electrogram may serve as guides for slo
w pathway ablation site selection.