Background: Ablation Index (AI) is a marker of lesion quality involving contact force (CF), time and power in a weighted formula. Despite its common use in pulmonary vein isolation, data on left atrial linear lesion (LALL) formation guided by AI are limited. 

Objective: To compare the safety and efficacy of AI guided ablation of an index LALL in the treatment of atrial tachyarrhythmia compared to standard CF guidance. 

Methods: A total of 313 consecutive patients with symptomatic atrial tachyarrhythmia (atrial fibrillation (AF), atrial tachycardia (AT)) undergoing their first time LALL ablation (anterior line, left atrial isthmus line, roof line or box lesion) guided by either AI or CF from 01.07.2015 until 31.08.2020 were enrolled. Target values for anterior line and left atrial isthmus line were 550, while target values for roof line and box lesion were 380.

Procedures and follow- ups were performed as per institutional standard. The median follow-up was 25.20 ± 14.30 months in the AI guided group and 39.90 ± 22.63 months in the CF only group. In case of atrial tachyarrhythmia relapse, a further re-ablation was offered to the patients. Thereby, LALL were checked for persistent block via standard pacing maneuvers during the re-ablation.

Results: A total of 261 anterior lines (CF: n= 48; AI: n= 213), 44 left atrial isthmus lines (CF: n= 9; AI: n= 35), 131 roof lines (CF: n= 17; AI: n= 114) and 66 box lesions (CF: n= 3; AI: n= 63) were performed in 313 patients (CF: n= 53; AI: n= 260). RF time per application was shorter in AI guided LALL compared to CF only guided LALL (anterior line: AI: 23.13 ± 6.93 s vs. CF: 21.39 ± 5.40 s, p= 0.06; left atrial isthmus line: AI: 21.23 ± 7.92 s vs. CF: 19.15 ± 6.25 s, p= 0.40; roof line: AI: 20.96 ± 7.64 s vs. CF: 17.07 ± 6.70 s, p= 0.03; box lesion: AI: 21.26 ± 6.07 s vs. CF: 17.72 ± 4.74 s, p= 0.21) (Figure 1). 

The rate of acute block for left atrial isthmus line was significantly higher in AI than CF only guided LALL (AI: n= 6 (67%) vs. CF: n= 35 (100%), p= 0.01) (Figure 2). 

No significant difference in the rate of complications was shown between both groups (CF: n= 2 (3.77%) vs. AI: n= 20 (7.69%), p= 0.31). Tamponade has occurred two times in the AI guided LALL, while no tamponade has occurred the CF only guided LALL.

Follow- up was available in 265 patients (85% of each group, p= 0,96).

An arrhythmia recurrence occurred in 182 of 265 patients, 148 in the AI guided (paroxysmal AF: n= 20 (13,51%), persistent AF: n= 50 (33,78%), AT: n= 78 (52,70%)) and 34 in the CF only guided group (paroxysmal AF: n= 3 (8,82%), persistent AF: n= 10 (29,41%), AT: n= 21 (61,76%)). Arrhythmia-free survival did not differ among both groups (p= 0.45). A total of 104 patients (CF: n= 19 (55.88%) vs. AI: n= 85 (57.43%), p= 0.87) underwent re-do AF/AT ablation. There was no significant difference in the rate of persistent bidirectional block along the ablation lines during the re-do procedure.

Conclusion: AI guided LALL for left atrial tachyarrhythmias is associated with shorter RF times and higher rates of acute block of mitral isthmus lines compared to CF only guided ablation while safety and long-term efficacy are comparable.