Background: The optimal way to map localised drivers in persistent atrial fibrillation (AF) remains unclear. The study aim was to apply a novel vector mapping approach called Stochastic Trajectory Analysis of Ranked signals (STAR) in AF and evaluate its ability to identify localised drivers using global and sequential mapping. In this study, global STAR maps created with whole-chamber basket catheters were considered the gold standard for comparison of sequential STAR mapping in regards to acute electrophysiological endpoints and long-term outcomes.
Methods: Patients having persistent AF ablation were included. Early sites of activation (ESA) identified on global STAR maps created using 5 minutes of unipolar recordings obtained using a whole-chamber basket catheters were used to guide AF ablation post-pulmonary vein isolation (PVI). All patients also had 30-seconds of unipolar recordings obtained with a PentaRay catheter. A minimum of 10 recordings were obtained to ensure adequate left atrial coverage. Each sequential PentaRay recording was then superimposed on the same geometry to produce an ‘amalgamation map’ of separate recordings, which made up a sequential STAR map. ESA that were associated with a study-defined ablation response (AF termination or CL slowing ≥30ms) as per the global STAR maps were correlated offline to ESA identified on the sequential STAR map. The ESA identified using each mapping modality were superimposed on the same geometry to ensure they had the same anatomical. Sites that were within 5 mm of each other were said to co-locate. Two observers who were blinded to the procedural data reviewed and interpreted the sequential STAR maps to confirm the anatomical location of the ESA.
Results: Thirty-five patients were included, of which 32 patients underwent STAR mapping guided ablation (3 patients terminated to sinus rhythm on PVI). In the 32 patients, 92 ESA were identified on the global STAR maps of which 83 (90.2%) were targeted with ablation. A study-defined ablation response was seen with 73 out of the 83 ESA (88.0%, 24 AF termination and 49 CL slowing). The 32 patients had on average 12.8 ± 2.7 sequential recordings performed. Sixty-two out of the 73 (84.9%) ESA were also identified on the sequential STAR maps with a kappa of 0.81 (95% CI 0.60–1.00) indicating strong inter-observer agreement. Figure 1A-C demonstrates a global (A) and sequential STAR map (B) of which both demonstrates two ESA mapped to the anterior LAA and anteroseptum. CL slowing was observed at both ESA. The electrograms indicate CL slowing from 145–182 ms on ablation at the ESA at the anteroseptum (C). ESA that were also identified on sequential STAR maps showed greater consistency (p<0.001), were seen pre- and post-PVI (p<0.001) and were more likely to be associated with AF termination on ablation (p=0.007). The sensitivity and specificity of sequential mapping for the detection of ESA with an ablation response was 84.9% (95% CI 74.6–92.2) and 90.0% (95% CI 55.5–99.8) respectively. During an average follow-up of 19.4 ± 3.7 months (all patients reached >12 months follow-up), 28 patients (80%) were free from AF/AT, off anti-arrhythmic drugs.
Conclusions: STAR mapping consistently identified ESA in all patients and the ablation response was compatible with ESA being driver sites. Mechanistically important ESA were successfully identified using sequential recordings.