Pulmonary vein measurements on pre-procedural CT/MR imaging can predict difficult pulmonary vein isolation and phrenic nerve injury during cryoballoon ablation for paroxysmal atrial fibrillation

Abstract

Objective: We tested the hypothesis that pulmonary vein (PV) measurements on pre-procedural CT/MR imaging can predict difficulty in isolation and phrenic nerve (PN) injury during cryoballoon ablation for paroxysmal atrial fibrillation (AF).

Methods: Consecutive patients with paroxysmal AF who had pre-procedural CT/MRI and underwent cryoballoon ablation as part of a randomized trial were studied. Imaging was anonymized for blinded analysis of: (1) maximum ostial diameter, (2) minimum ostial diameter, (3) ostial area and (4) ratio of maximum over minimum ostial diameter (eccentricity index). Veins that required more than 2 freezes of at least 200 s duration to isolate or not isolated were defined as difficult to isolate. Loss of PN pacing during right-sided ablation was defined as PN injury. Logistic regression was used to analyze the predictive effect of the measurements on the 2 outcomes.

Results: 148 PVs in 38 patients (aged 60 ± 11 years, 76% male) were analyzed. Left inferior PV (LIPV) was most difficult to isolate with 23 out of 37 PVs (62%), and PN injury occurred in 3 of 38 (8%) right superior PV (RSPV). Greater eccentricity index predicted difficulty in isolating LIPV, OR 40.33 (95% CI 1.40 to 1160, p = 0.03) and smaller eccentricity index predicted PN injury in RSPV, OR 0.01 (95% CI 0.01-0.16, p = 0.001).

Conclusions: Eccentricity index measured from pre-procedural CT/MR imaging can predict difficulty of PV isolation and PN injury during cryoballoon ablation for paroxysmal AF.

Keywords: Atrial fibrillation; CT; Cryoballoon; MRI.