Biomarker-based clustering for AF remodelling yielded more physiological changes in action potential duration than the percentage method (e.g., APD20 changed up to 132% vs 500%).
Applying atrial fibrillation remodelling through biomarker-based clustering in computational models maintains non-linear relationships and avoids non-physiological action potential duration values seen with direct conductance changes.
(1) Background: in silico models are increasingly relied upon to study the mechanisms of atrial fibrillation. Due to the complexity associated with atrial models, cellular variability is often ignored. Recent studies have shown that cellular variability may have a larger impact on electrophysiological behaviour than previously expected. This paper compares two methods for AF remodelling using regional populations. (2) Methods: using 200,000 action potentials, experimental data was used to calibrate healthy atrial regional populations with two cellular models. AF remodelling was applied by directly adjusting maximum channel conductances. AF remodelling was also applied through adjusting biomarkers. The methods were compared upon replication of experimental data. (3) Results: compared to the percentage method, the biomarker approach resulted in smaller changes. RMP, APD20, APD50, and APD90 were changed in the percentage method by up to 11%, 500%, 50%, and 60%, respectively. In the biomarker approach, RMP, APD20, APD50, and APD90 were changed by up to 4.5%, 132%, 50%, and 35%, respectively. (4) Conclusion: applying AF remodelling through biomarker-based clustering resulted in channel conductance changes that were consistent with experimental data, while maintaining the highly non-linear relationships between channel conductances and biomarkers. Directly changing conductances in the healthy regional populations impacted the non-linear relationships and resulted in non-physiological APD20 and APD50 values.
Elliott et al. (Sat,) conducted a other in Atrial fibrillation. Biomarker-based clustering for AF remodelling vs. Percentage method was evaluated on Replication of experimental data (changes in RMP, APD20, APD50, APD90). Biomarker-based clustering for AF remodelling yielded more physiological changes in action potential duration than the percentage method (e.g., APD20 changed up to 132% vs 500%).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: