Life-threatening abnormalities in the electrical rhythm of the heart are usually treated with the application of a large electric shock. An approach involving a significantly smaller shock energy may be equally effective. See Letter p.235
Electricity can kill, but it can also revive individuals experiencing life-threatening cardiac arrhythmias when it is applied through external or implanted cardiac defibrillator devices. Sudden cardiac death is the result of fibrillation — irregular contractions of the heart muscle caused by multiple electrical waves rotating throughout the heart1. Although these 'rotor' waves are unstable, their spatial and temporal organization provides hope that defibrillation-shock energies can be significantly decreased through the design of better devices. Then, not only may the size and cost of defibrillator devices be reduced, but so may the pain experienced by patients during a shock2. On page 235 of this issue, Luther et al.3 describe a significant advance in this direction by showing the efficacy of a sequence of low-energy electric shocks.
Fibrillation is not driven by a particular heart region, but is sustained by rotor waves across the heart. This means that halting fibrillation requires an approach that affects the entire heart. Traditionally, this is done using two widely spaced metal electrodes to deliver a high-energy shock, which creates a large electric field throughout the heart. The current flowing between the two electrodes follows the paths of least resistance, and so crosses cell membranes, thereby changing — both increasing and decreasing — the electrical potential across the membranes (Vm) in affected regions and creating what are known as virtual electrodes4. A successful defibrillation shock quickly restores Vm to resting levels across the heart (Fig. 1a,b).
One approach to reducing the strength of the defibrillation shock is to use several metal electrodes placed throughout the heart5,6. At each of the electrode locations, a propagating wave induced by electric stimulation can alter the dynamics of rotor waves, if the pacing rate is faster than the fibrillation frequency (overdrive pacing)7,8. This approach, however, has not proven very effective in the clinic9,10, partly because of the challenges involved in implanting multiple electrodes in the heart.
Virtual electrodes (ΔVm) can be generated in the heart by the interaction between the electric field and the tissues of the heart and thorax. They are affected by membrane resistance and capacitance; by tissue conductivity; by the orientation of muscle fibres; and by the detailed geometry of electrical inhomogeneities such as those seen at the boundaries between the heart and lungs, or between the heart tissue, blood vessels and scar tissue. The equations representing these phenomena are well understood and involve one of the four fundamental forces of physics — electromagnetism. Nonetheless, rigorous simulations of the effect of an electric field on the heart are possible only if all the relevant geometry, conductivities and boundary conditions can be accurately described.
To make matters even more complicated, successful defibrillation is not a result of just the interaction between the applied electric field and the tissue. It is also influenced by the pre-shock state of the transmembrane potential (Vm,pre) and the stability of the rotor waves. The combined effects of tissue heterogeneity and the unpredictable nature of fibrillation mean that the outcome of the shock treatment varies between individuals and even in the same individual during a cardiac arrest. For traditional defibrillation, the post-shock transmembrane potential (Vm,post) resulting from a single, brief, high-energy shock determines whether the shock is successful, and this depends largely on whether any sustainable rotor waves remain after the shock (Fig. 1b).
Although any spatial heterogeneity in the conductivity of the normal or diseased heart is known to contribute to the generation of virtual electrodes11,12,13, it is not known which heterogeneities are most relevant for re-establishing synchronous contractions. Luther et al.3 show that the coronary arteries can have a large role in determining the precise spatial patterns of virtual electrodes throughout the heart — a finding that is also supported by another study14.
Armed with the knowledge that, with increasing shock strength, the tissue regions experiencing virtual electrodes increase and the time taken to excite the entire heart (the global activation time) decreases15, Luther and colleagues build on an elegant scaling argument16. Their results relate the global activation time obtained in their experiments to the size distribution of the branching structure of the coronary vasculature. The branching structure was computed from polymer casts obtained from the same experiments. The authors propose that during a shock, the specific geometry (size, orientation and so on) of the blood-filled coronary vasculature, and its difference in electrical conductivity from that of the surrounding heart, lead to the formation of virtual electrodes throughout much of the heart. Indeed, Luther et al. show that shock strengths significantly lower than those required for traditional defibrillation can produce many virtual electrodes throughout the heart of beagle dogs, and that these electrodes generate numerous propagating waves without the need for multiple metal electrodes.
Through their in vitro and in vivo animal experiments, the researchers3 show that the application of multiple, brief shocks significantly reduces the energy required for defibrillation by launching numerous propagating waves from many virtual electrodes across the heart (Fig. 1c). Surprisingly, the interval between the application of each low-energy antifibrillation pacing (LEAP) stimulus was longer than the average rotor period (underdrive pacing).
These intriguing findings lead to equally intriguing questions, an obvious one being whether this phenomenon will translate to humans. Moreover, the curious researcher might wonder whether even lower defibrillation energies — which are always desirable — could be used. However, there may be a theoretical minimum energy for various reasons, including the fact that a minimum electric field (around 1 volt per centimetre) is required to excite cardiac tissue17. In a given patient, the success of LEAP defibrillation will depend on the relationship of rotor-wave density and location and the spatial distribution and size of heterogeneities in tissue conductivity.
In light of Luther and co-workers' study, it seems prudent to factor coronary vasculature into whole-heart simulations of defibrillation. But deciding how to do this appropriately requires a detailed understanding of the geometry, conductivity and path of the electrical current near blood vessels. Such details can vary widely between individuals.
Although provocative, the new work3,14 does not directly show the exact mechanism involved, nor does it outline the precise experimental pattern of virtual-electrode polarization resulting from vasculature-induced heterogeneities. Factors to explore include the effects of vessel shape and of vessel-wall and blood conductivities on the generation of virtual electrodes18. That said, it is exciting that LEAP can reduce the defibrillation threshold for both atrial and ventricular fibrillation in vitro and can terminate atrial fibrillation in vivo (by means of coil electrodes inside the heart). Indeed, LEAP is an important development showing that a significant decrease in the required shock strength results from a combination of dynamic control and the interaction of the electric field with the heart structure — purportedly, the coronary vasculature.
Gray, R. A., Pertsov, A. M. & Jalife, J. Nature 392, 75–78 (1998).
Dosdall, D. J. & Ideker, R. E. Heart Rhythm 4, S51–S56 (2007).
Luther, S. et al. Nature 475, 235–239 (2011).
Wikswo, J. P., Lin, S. F. & Abbas, R. A. Biophys. J. 69, 2195–2210 (1995).
Pak, H.-N. et al. Am. J. Physiol. Heart Circ. Physiol. 285, H2704–H2711 (2003).
Hosfeld, V. D. et al. J. Biol. Phys. 33, 145–153 (2007).
Allessie, M. et al. Circulation 84, 1689–1697 (1991).
Davidenko, J. M. et al. Circ. Res. 77, 1166–1179 (1995).
Ditto, W. L. et al. Int. J. Bifurc. Chaos 10, 593–601 (2000).
Kok, L. C. & Ellenbogen, K. A. Cardiol. Clin. 22, 71–86 (2004).
Fishler, M. G. J. Cardiovasc. Electrophysiol. 9, 384–394 (1998).
Sobie, E. A., Susil, R. C. & Tung, L. Biophys. J. 73, 1410–1423 (1997).
Trayanova, N. A., Roth, B. J. & Malden, L. J. IEEE Trans. Biomed. Eng. 40, 899–908 (1993).
Bishop, M. J. et al. IEEE Trans. Biomed. Eng. 57, 2335–2345 (2010).
Maleckar, M. M. et al. Am. J. Physiol. Heart Circ. Physiol. 295, H1626–H1633 (2008).
Pumir, A. et al. Phys. Rev. Lett. 99, 208101 (2007).
Ideker, R. E., Zhou, X. & Knisley, S. B. Pacing Clin. Electrophysiol. 18, 512–525 (1995).
Cartee, L. A. & Plonsey, R. IEEE Trans. Biomed. Eng. 39, 76–85 (1992).
About this article
Royal Society Open Science (2017)
Heart Rhythm (2017)
Hysteresis and drift of spiral waves near heterogeneities: From chemical experiments to cardiac simulations
Physical Review E (2016)
Detection Algorithm of Phase Singularity Using Phase Variance Analysis for Epicardial Optical Mapping Data
IEEE Transactions on Biomedical Engineering (2016)
Circulation: Arrhythmia and Electrophysiology (2015)