Dynamic assessment of atria substrate through pacing
Wong et al. reported significant variation in left atrial voltage, burden of LVAs and electrogram fractionation during atrial pacing, depending on the pacing site and atrial rate (156). The prognostic implication of these changes was no examined. Kim et al., previously demonstrated steeper action potential restitution curves associated with AF (157). Williams et al. evaluated electrophysiological responses to incremental atrial pacing in a cohort of patients with PAF (158). Patients displayed highly variable changes in electrogram voltage and fractionation, however greater rate-dependent activation dispersion was associated with increased AF vulnerability. Finally, Hunter and colleagues assess conduction velocity dynamics in patients undergoing ablation for PsAF (84). Most sites with early rate-dependent conduction slowing were located in regions of low voltage, many of which harbored rotational drivers for AF.
Some groups have implemented electrophysiology study-guided LA substrate modification in PsAF with favorable results. In patients without LVAs on the posterior wall of the LA, Yamaji et al. performed posterior wall isolation if regions of the LA demonstrated pro-arrhythmic repolarization properties or AF/AT was inducible during electrophysiological testing (159). Patients who underwent posterior wall isolation displayed greater freedom from AF/AT than those without LVAs who had PVI alone, suggesting that atrial voltage assessment in isolation may underestimate the risk of AF recurrence.
Finally the majority of studies in the current era of substrate modification have focused on endocardial evaluation of the LA. Recent high-resolution EAM has demonstrated significant burden of LVAs in the right atrium as well as other pro-arrhythmic markers such as conduction slowing and block (160). Comprehensive substrate modification in the future will likely involve concomitant assessment of the right atrium. A recent analysis of atrial LVAs have shown that these may be located exclusively in the epicardial layer and correspond to normal bipolar voltages when measured endocardially (161). Conventional bipolar mapping and novel omnipolar technology have a limited field of view, and thus are less sensitive to epicardial voltage when mapping endocardially, and vice versa. Unipolar measurements do seem to improve the detection of such epicardial LVAs to some degree, but a large proportion of potentially arrhythmogenic areas may still be missed (162).
In summary, novel approaches to atrial substrate assessment are moving towards a genuinely individualized ablation strategy, guided by multimodal assessment of atrial electrophysiology. Progress on this front will require a reassessment of historical definitions of low voltage, appreciating the influence of clinical and technical factors. Moreover, it is clear that low voltage in isolation is unlikely to capture the arrhythmogenic potential of the tissue fully, and further functional evaluation may be required to identify sites critical to AF maintenance correctly. Finally, efficacious treatment of AF requires not only accurate substrate characterization, but also means of delivering durable ablation lesions safely.