Using dominant eigenvalue analysis to predict formation of alternans in the heart

Area of Science:

• Cardiac Electrophysiology
• Computational Biology
• Medical Physics

Background:

• Alternation of action potential duration (alternans) at the cellular level precedes whole-heart ventricular fibrillation.
• Traditional methods for predicting alternans, like restitution curve slope analysis, have proven unsuccessful.
• A novel dominant eigenvalue analysis technique has shown theoretical promise for predicting alternans in isolated cardiac myocytes.

Purpose of the Study:

• To validate the applicability of dominant eigenvalue analysis for predicting alternans formation at the whole-heart level.
• To compare eigenvalue analysis in regions with and without alternans.
• To establish a predictive tool for cardiac alternans in whole hearts.

Main Methods:

• Optical mapping in Langendorff-perfused rabbit hearts (n=4) paced at decreasing cycle lengths to induce action potential duration alternans.
• Identification of B(onset), the basic cycle length at the local onset of alternans.
• Generation of 2D eigenvalue maps using principal component analysis of action potentials after perturbations.
• Calculation of mean eigenvalues in regions with (1:1(alt)) and without (1:1) alternans at B(onset).

Main Results:

• A significant difference (p<0.05) in mean eigenvalues (λ[over ¯]) was observed between regions with and without alternans at B(onset).
• The dominant eigenvalue technique successfully distinguished between regions prone to and resistant to alternans formation.
• This method provides a localized prediction of alternans onset within the heart.

Conclusions:

• Dominant eigenvalue analysis is a viable technique for predicting local alternans formation in the whole heart.
• This method offers a significant advancement over traditional approaches for predicting cardiac arrhythmias.
• The findings support the use of eigenvalue analysis as a predictive tool for ventricular fibrillation precursors.