TL;DR
This study uses computational modeling to investigate how variations in pulmonary vein (PV) electrical properties and atrial fibrosis influence the initiation and maintenance of atrial fibrillation (AF). The findings suggest that measuring these factors could help predict patient-specific susceptibility to AF and the likelihood of successful pulmonary vein isolation (PVI) treatment.
What Problem Does This Paper Address?
Persistent atrial fibrillation (AF) is a common cardiac arrhythmia, and while pulmonary vein isolation (PVI) is a standard treatment, its success rates are not optimal, and predicting which patients will benefit most remains a challenge. Current understanding suggests that electrical "drivers" (often rotors or focal sources) are responsible for maintaining AF, and these drivers are frequently located near the pulmonary veins. However, it's difficult to clinically disentangle the individual contributions of various factors, such as the electrical properties of the PVs (action potential duration - APD, conduction velocity - CV) and the presence of atrial fibrosis, to AF susceptibility and the effectiveness of PVI. This paper aims to address this by using computational modeling to explore how these factors interact and influence AF dynamics and PVI outcomes.
Methods
The researchers employed a bilayer biatrial computer model based on patient-specific geometries. This model, while computationally less intensive than volumetric models, captures transmural heterogeneity and detailed atrial structure. They incorporated a sophisticated human atrial ionic model, modified to represent electrical remodeling in persistent AF.
Key methodological aspects include:
- Varied PV Electrophysiology: They systematically altered PV action potential duration (APD) and conduction velocity (CV) by modifying ionic conductances (specifically IK1 for APD) and tissue conductivity. These variations were applied homogeneously or as ostial-distal gradients.
- Fibrosis Modeling: Interstitial fibrosis was modeled in the PVs with a density that varied along the vein, increasing from the ostium to the distal boundary. For some simulations, they incorporated fibrosis based on late gadolinium enhancement (LGE)-MRI data from patients.
- PV Fiber Direction: They explored various arrangements of endocardial and epicardial fibers within the PVs, reflecting known inter-patient variability.
- Pacing Protocol: To test arrhythmia inducibility, they used a clinically motivated protocol of extrastimulus pacing from each of the four PVs, simulating PV ectopics.
- Phase Singularity (PS) Analysis: To characterize arrhythmia dynamics, they analyzed phase singularity trajectories, which represent re-entrant waves. PS density maps were used to identify driver locations and quantify their distribution in the PV and left atrial (LA) regions.
- Pulmonary Vein Isolation (PVI) Simulation: PVI was simulated by setting tissue conductivity to near zero in the PV regions after AF initiation.
Datasets
The study utilized patient-specific geometries derived from MRI data for twelve patients. For modeling atrial fibrosis, they incorporated average LGE-MRI distributions from populations of patients with paroxysmal AF (34 patients) and persistent AF (26 patients), as well as patient-specific LGE-MRI data.
Key Insights and Ideas
- PVs are more than just ectopic beat sources: The study demonstrates that PV properties significantly influence not just the initiation of AF via ectopic beats but also the maintenance of re-entrant circuits.
- APD gradient matters: While shorter PV APD generally increased susceptibility, the way it was shortened (homogeneous vs. ostial-distal gradient) had different effects on inducibility and driver location.
- Fibrosis anchors rotors: In cases with significant atrial fibrosis, rotors tended to stabilize in fibrotic regions, suggesting that fibrosis plays a crucial role in maintaining AF, especially in persistent AF.
- PV PS density as a predictor: A higher PV phase singularity (PS) density before PVI was associated with successful AF termination or conversion to macroreentry in the simulations, suggesting it could be a valuable clinical metric for predicting PVI success.
- Patient-specific modeling is key: The importance of individual patient geometry and fibrosis distribution was highlighted, indicating that a one-size-fits-all approach to PVI might not be optimal.
Key Results
- APD and Inducibility: Shorter PV APD increased arrhythmia inducibility, while longer PV APD was protective. Inducibility was significantly higher for ostial-distal gradients of decreasing APD compared to homogeneous changes.
- CV and Inducibility: Slower CV at the LA/PV junction increased inducibility, but homogeneous CV slowing or slowing at the distal PV had less impact.
- Fibrosis Impact: The inclusion of PV fibrosis increased PS density in the PV region and led to more stable reentry around the LA/PV junction.
- PVI Outcome: Simulation outcomes after PVI were classified into termination, macroreentry, or AF sustained by LA rotors. A higher PV PS density ratio pre-PVI was significantly associated with termination or macroreentry, suggesting higher likelihood of PVI success.
- Anatomy Influence: Arrhythmia inducibility varied significantly between patient geometries, with larger PV areas generally increasing inducibility.
Limitations and Future Work
The authors acknowledge several limitations:
- PV Branching: PV branching structures were not explicitly modeled, with PVs being trimmed to a single rim.
- Endocardial vs. Epicardial ERP: They modeled endocardial and epicardial ERP identically, despite some evidence of differences.
- APD Modeling: APD changes were primarily modeled by altering IK1, not considering other ionic conductances or parametrization methods.
- Bilayer Model: The use of a bilayer model instead of a volumetric one might affect rotor drift.
- Simultaneous Lesion Application: PVI lesions were applied simultaneously in the model, unlike the sequential application in clinical practice.
- Long-Term Recurrence: Long-term AF recurrence was not modeled.
- Focal Beats: The study focused on reentry and PV ectopics, not AF sustained by focal beats.
Future work suggested includes:
- Clinical Validation: Conducting retrospective clinical studies to validate the PV PS density metric for predicting PVI success.
- Patient-Specific Tuning: Using measurements like PV ERP, CV, and LGE-MRI to tune model properties for individual patients.
- Advanced Modeling: Incorporating more detailed PV anatomy, endocardial/epicardial differences, and other ionic conductances.
- Investigating Other Factors: Exploring the role of connexins and cell morphology.
Summary updated