Pith. sign in

REVIEW 3 major objections 5 minor 38 references

Cardiac Mechano-Electrical Dynamical Instability

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A new instability, MEDI, turns the heart's own contraction into a driver of spiral-wave arrhythmias.

desk verdict New and plausible computational mechanism for MEF-driven instability, but the 2D evidence leaves the causal chain from 1D biexcitability incomplete. read the letter →

arxiv 1908.05144 v1 pith:QM4TTJHE submitted 2019-08-14 physics.bio-ph physics.med-ph

L. D. Weise , A. V. Panfilov This is my paper · ORCID
classification physics.bio-phphysics.med-ph
keywords mechano-electricalfeedbackstretch-activatedchannelscardiacarrhythmiaspiralwavesbiexcitabilityactionpotentialdurationwavebreakcalcium-drivenupstroke
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

MEDI is a dynamical instability of excitation waves in cardiac tissue in which the tissue's own contraction feeds back into its electrical wave. The paper argues that stretch-activated currents accelerate the wave front until it collides with the tail of the preceding wave; at those collision sites a short diastolic interval does not shorten but lengthens the next action potential, because the upstroke switches from a sodium-driven to a calcium-driven form (biexcitability). This abnormal $\mathrm{APD}(\mathrm{DI})$ relation creates positive feedback — longer action potentials shorten the diastolic interval, which lengthens the next action potential — until the wave breaks and spiral waves form or hypermeander. If correct, MEDI provides a mechanical route to arrhythmia onset that operates at longer pacing periods than the classical alternans instability, and it identifies stretch-activated channels and L-type calcium current as control points.

What carries the argument

The machinery that carries the argument is the electromechanical coupling chain: a reaction–diffusion equation for transmembrane voltage, a spring–mass mechanical model coupled to it, and a stretch-activated current $I_{sac}$ that adds a depolarizing conductance when local stretch $\lambda>1$. To isolate the wave front–tail interaction, the authors use a moving obstacle in a one-dimensional cable that forces the first wave to propagate at a controlled speed, producing controlled collisions; varying the forced conduction velocity tunes $\mathrm{DI}$ at the collision point and reveals the transition between a steep sodium-driven upstroke and a slow calcium-driven upstroke. Biexcitability — the coexistence of these two propagation modes in the same tissue — is the named phenomenon that converts short $\mathrm{DI}$ into long $\mathrm{APD}$, and blocking $I_{CaL}$ suppresses MEDI.

What would settle it

Record transmembrane voltage and calcium current at a collision site in a stretched cardiac tissue preparation while imposing a controlled front-tail interaction: if action potential duration still falls when diastolic interval shortens below the threshold at which the sodium upstroke fails, MEDI will not occur; if it rises as the upstroke becomes calcium-driven, the instability is confirmed.

Watch

Extended reading notes

Core claim

The central discovery is that mechano-electrical feedback can create its own dynamical instability. In the authors' computational model of human cardiac tissue, stretch-activated current $I_{sac}=G_s(\lambda-1)/(\lambda_{\max}-1)(V-E_s)$ accelerates the wave front, producing wave front–tail collisions. At the collision site, short $\mathrm{DI}$ produces longer $\mathrm{APD}$ (opposite to standard restitution) because the upstroke is no longer sodium-driven but is carried by L-type calcium current $I_{CaL}$; the longer $\mathrm{APD}$ shortens the next $\mathrm{DI}$, and this positive feedback grows beat to beat until the wave front blocks and curls into two counter-rotating spiral waves. The same mechanism makes spiral wave tips drift rapidly along the line of maximum $\mathrm{APD}$, producing hypermeandering trajectories. The paper concludes that MEDI is a consequence of MEF and can cause both formation and hypermeandering of spiral waves.

Load-bearing premise

The load-bearing premise is that the abnormal one-dimensional relation — short diastolic interval produces longer action potential via the calcium-driven upstroke — also holds at the two-dimensional collision site in the full electromechanical model, where the paper does not directly measure it and where APD itself alters the strain distribution and conduction velocity.

Editorial extensions

If this is right

  • MEDI offers a mechanism for wave break and spiral-wave formation that requires no anatomical heterogeneity and no alternans; it appears in the model for a broad range of stretch-activated conductances and pacing periods.
  • Because MEDI occurs at longer stimulation periods than classical $\mathrm{APD}$ alternans, it could explain arrhythmia onset at heart rates where restitution-based predictions say the tissue should be stable.
  • In the spiral-wave simulations, MEDI causes rapid drift of the spiral core along the collision line, a hypermeandering motion the paper links to polymorphic ventricular tachycardia.
  • The dependence on $I_{CaL}$ means the instability is not a generic stretch effect but specifically requires the calcium-driven upstroke, giving a targeted point for intervention.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the same positive feedback transfers to three-dimensional hearts, MEDI would be a candidate driver of vortex filament breakup, a scenario the authors explicitly leave as an open next step.
  • A testable extension would be to apply localized stretch during pacing in a tissue preparation to reproduce controlled front-tail collisions and look for the abnormal $\mathrm{APD}(\mathrm{DI})$ slope at the collision site.
  • The same logic could apply to other excitable media with deformation-sensitive currents, such as smooth muscle, where mechanical state modulates excitability.
  • A practical consequence is that drugs or mutations that reduce $I_{CaL}$ availability may suppress MEDI even if stretch-activated channel conductance is unchanged, because the calcium-driven upstroke is the necessary switch.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript reports a computational discovery of a 'mechano-electrical dynamical instability' (MEDI) in a two-dimensional electromechanical model of human cardiac tissue. The authors couple the Ten Tusscher-Panfilov epicardial ionic model to a discrete mechanical lattice with stretch-activated current Isac = Gs (λ-1)/(λmax-1) (V-Es), and simulate periodic pacing with a period of 300 ms after conditioning at 340 ms. They observe repeated wave front-back collisions, gradual local prolongation of APD, shortening of DI, and eventual wave break with formation of spiral waves. A parameter scan in (Gs, T) maps stable propagation, alternans, and MEDI regions. To explain the mechanism, the authors use a one-dimensional constantly stretched cable with a moving obstacle to force a controllable S1 conduction velocity, and show that during front-tail collisions short DI produces abnormally long APD, driven by a shift from sodium-dominated to ICaL-dominated upstroke ('biexcitability'); blocking ICaL abolishes MEDI in this 1D setup. They further show that in 2D, MEDI causes rapid spiral drift at larger Gs. The authors acknowledge that no analytical theory is provided and that the full 2D model is more complex because APD affects spatiotemporal strain and hence conduction velocity.

Significance. If the proposed mechanism is correct, MEDI is a genuinely new dynamical instability that is distinct from alternans and could contribute to wave break and spiral-wave hypermeander during cardiac arrhythmia. The paper has several concrete strengths: the simulation evidence is internally consistent; a useful negative control (blocking ICaL in the 1D setup) supports the role of biexcitability; the robustness scan over Gs and pacing period gives a predictive parameter range; and the spiral-drift results connect the instability to an observable arrhythmia-relevant phenotype. However, the significance of the central claim depends on whether the 1D reduced mechanism actually drives the 2D instability, which the manuscript does not yet establish.

major comments (3)
  1. [Mechanism explanation, Figures 3-4 and Figure 1] The central conclusion that MEDI is caused by biexcitability-mediated abnormal APD(DI) restitution rests entirely on the 1D constantly stretched cable with a moving obstacle (Figures 3 and 4), in which electromechanical coupling is disabled by construction. At the 2D collision site the authors do not measure APD as a function of DI, nor do they identify the upstroke type (INa- vs ICaL-dominated) during the growing APD in Figure 1. The authors themselves state near the end of the paper that 'in the full model it is more complex, as APD affects also the spatiotemporal strain distribution in the medium, and thus affects CV'. Under strain-dependent CV, the inter-beat interval at a collision site is not T - APD, because arrival-time differences from the pacing site are themselves changing; the claimed positive feedback could therefore be weakened or reversed. A direct 2D measurement of APD(DI) at the collision site, or a 2D ICaL-block control, is needed to support the transfer.
  2. [Figure 2 and protocol description] The robustness diagram in Figure 2 is obtained by a dynamic protocol in which the pacing period is decreased by 2.5 ms after every ten beats, and the boundary between stable propagation and instability is identified during the ramp. The claim that MEDI occurs in a large parametric space (Gs between roughly 15 and 75 S/F and T between roughly 220 and 400 ms) would be stronger if the authors reported whether MEDI develops under constant pacing at the final period after many beats, and whether the onset depends on the ramp rate. This matters because the title and abstract present MEDI as an instability of a steady period, not of a particular sweep protocol.
  3. [Equation (2) and the 1D constantly stretched cable] The 1D 'constantly stretched cable' (λ = λmax) used to establish the mechanism has a uniform, time-independent stretch-activated current; it therefore cannot represent the dynamic strain feedback that the 2D model includes, where λ varies both in space and time as each wave triggers contraction. The agreement between Figure 3 and Figure 1 is qualitative and established by visual comparison, without a quantitative metric. Please provide at least one quantitative comparison (e.g., APD increase per beat or DI trajectory) and specify which features of the 2D instability are captured by the 1D model.
minor comments (5)
  1. [Throughout] There are several typographical errors and style inconsistencies: 'it's back' should be 'its back'; 'PloS One' appears in the bibliography instead of the journal's preferred 'PLoS ONE'; and the reference list contains inconsistent capitalization.
  2. [Main text and figure captions] The manuscript contains placeholder links such as [35], [36], [37], and [39] ('LINK TO SUPPLEMENTAL MOVIE', 'LINK TO SUPPLEMENTAL FIGURE', and 'LINK TO SI'); these should be resolved to actual URLs or removed before publication.
  3. [Figure 2 caption] The caption states 'Crosses: measurements' but the crosses are not explained in the main text; please clarify what measurements the crosses denote and how they were obtained.
  4. [Methods, APD and DI definition] The definition of APD and DI recorded at -60 mV appears only in footnote [38]; this definition should be stated in the main text or methods, as it is essential for interpreting all quantitative claims.
  5. [Mechanical model description] The sentence 'To solve the mechanical model we assumed elastostatics, and used Verlet integration' seems contradictory; if a dynamic relaxation scheme is used, this should be stated explicitly, and the relation to elastostatics clarified.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the 2D MEDI is a simulation output with independent controls, and the 1D mechanism study is an extrapolation rather than a fitted prediction.

full rationale

The central claim is a new dynamical instability observed in a 2D electromechanical simulation, not a quantity fitted to a target. Figure 1 shows APD growth, DI decrease, wavebreak, and spiral formation under periodic stimulation; Figure 2 maps the instability regime as a function of Gs and stimulation period T, providing an independent parameter scan. The 1D moving-obstacle setup in Figures 3 and 4 is used to expose the mechanism, but the authors do not fit the 1D model to reproduce 2D APD values; they choose a forced conduction velocity that qualitatively reproduces the 2D behavior and then study the dependence on that velocity. The ICaL-block control in the 1D setup is an additional mechanistic check, and no equivalent control is claimed for 2D. The paper cites the authors' prior work for the moving-obstacle method and for the biexcitability transition, but the current simulations themselves reproduce the collision, the replacement of the sodium upstroke by an ICaL-driven upstroke, and the APD elongation (Figure 4B), so the argument does not reduce to self-citations. The feedback explanation uses the identity DI=T-APD, which is an approximation for periodic point stimulation; at a moving collision site the inter-beat interval also depends on conduction velocity differences, and the authors explicitly concede that in the full model APD affects the spatiotemporal strain distribution and thus CV. This is a genuine limitation in transferring the 1D mechanism to 2D, but it is a modeling assumption, not a circular definition or a fitted parameter renamed as a prediction. No equation in the paper is equivalent to its own input by construction, and the instability itself is a simulation output rather than a restatement of an input relation.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central result rests on a specific ionic model, a phenomenological stretch-activated current, fixed-boundary mechanics, and the transfer of a 1D mechanism to 2D. No free parameters are fitted to a target output; Gs, Es, and lambda_max are chosen from prior literature or scanned as controls. No invented physical entities are introduced.

free parameters (4)
  • Gs (stretch-activated maximal conductance) = 50 S/F (2D main), 10.5 S/F (1D), 25-75 S/F (spiral tests); scanned range 0-100 S/F
    The central control parameter in Eq. (2). MEDI only appears once Gs exceeds a threshold, so the existence of the instability is conditioned on this hand-chosen conductance. It is scanned, not fitted.
  • Es (reversal potential of stretch-activated channels) = 0 mV
    Set to 0 mV from a measured range of -20 to 0 mV. The sign and magnitude of Isac, and thus the wavefront acceleration, depend on this choice.
  • lambda_max (maximal normalized sarcomere length) = 1.1
    Chosen as in Ref. [14]. It sets the scaling of Isac with stretch in Eq. (2).
  • Forced conduction velocity in moving-obstacle setup = 17.65 cm/s (1D MEDI reproduction); 4.81 to 17.5 cm/s (Figure 4); no-ICaL wavebreak window [12.4, 15.26] cm/s
    The velocity is an externally imposed parameter in the 1D protocol. The value 17.65 cm/s is chosen to reproduce the 2D phenomenon, and the APD(DI) relation is mapped over a range of forced velocities.
assumptions (6)
  • domain assumption The Ten Tusscher-Panfilov human epicardial ionic model adequately reproduces the excitation dynamics relevant to the instability.
    Used as I_ion in Eq. (1); all APD, CV, restitution, and biexcitability behavior depends on this ionic model. The paper does not test other ionic models.
  • domain assumption The linear, time-independent stretch-activated current model with Es = 0 mV and lambda_max = 1.1 captures cardiac mechano-electrical feedback.
    Eq. (2) is a phenomenological model taken from prior literature. Its linearity and parameter values are assumed, not derived or validated in this paper.
  • domain assumption Fixed boundaries mimic isovolumic cardiac phases and are sufficient for MEDI to develop in the 2D model.
    Fixed boundaries are necessary for contraction to produce stretch (lambda > 1) and hence Isac. Free boundaries or 3D geometry could change or abolish the feedback loop.
  • domain assumption A 1D constantly stretched cable with a moving obstacle captures the local collision dynamics of the full 2D electromechanical model.
    Used to measure the abnormal APD(DI) relation in Figures 3-4. In the full model, stretch is spatiotemporally varying and coupled to APD, so the 1D reduction is an idealization.
  • domain assumption Biexcitability, the coexistence of fast sodium-driven and slow calcium-driven wave propagation, is a property of the model in the simulated collision regimes.
    Invoked to explain APD elongation. The paper relies on prior work and on the ICaL block control, but does not directly visualize the sodium-to-calcium switch in the full 2D model.
  • standard math The explicit Euler and Verlet integration schemes are stable and accurate at the chosen space and time steps.
    The paper reports space step 0.25 mm and time step 0.02 ms, but provides no convergence tests or error estimates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cardiac Mechano-Electrical Dynamical Instability." pith.science (2026). https://pith.science/paper/QM4TTJHE

@misc{pith2026190805144,
  author       = {Pith},
  title        = {Pith review of: Cardiac Mechano-Electrical Dynamical Instability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QM4TTJHE}},
  note         = {Machine review of arXiv:1908.05144}
}
read the original abstract

In a computational study we reveal a novel dynamical instability of excitation waves in the heartmuscle. The instability manifests itself as gradual local increase in the duration of the actionpotential which causes formation and hypermeandering of spiral waves. The mechanism is causedby stretch-activated currents that cause wave front-tail collisions and beat to beat elongation of theaction potential duration due to biexcitability. We discuss the importance of the instability for theonset and dynamics of cardiac arrhythmias.

Figures

Figures reproduced from arXiv: 1908.05144 by the authors.

Figure 1
Figure 1. FIG. 1: MEDI. Periodical wave initiation in left, upper cor [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: Wave patterning as a function of period of stimula [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3: Stepwise APD increase and wave block in constantly [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: FIG. 4: APD elongation is caused by biexcitability. Electro [PITH_FULL_IMAGE:figures/full_fig_p003_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5: MEDI causes rapid spiral drift. (A) spiral tip trajec [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 37 canonical work pages

  1. [1]

    M. A. Allessie, F. I. M. Bonke, and F. J. G. Schopman, Circ. Res. 39, 168 (1976)

    work page 1976

  2. [2]

    Gerisch, Wilhelm Roux’ Archiv f¨ ur Entwick- lungsmechanik der Organismen 156, 127 (1965)

    G. Gerisch, Wilhelm Roux’ Archiv f¨ ur Entwick- lungsmechanik der Organismen 156, 127 (1965)

    work page 1965

  3. [3]

    Agladze and O

    K. Agladze and O. Steinbock, The Journal of Physical Chemistry A 104, 9816 (2000)

    work page 2000

  4. [4]

    B. P. Belousov, Collection of short papers on radiation medicine for 1958 (Med. Publ., Moscow, 1959), pp. 145– 147, in Russian

    work page 1958

  5. [5]

    Zaikin and A

    A. Zaikin and A. Zhabotinsky, Nature 225, 535 (1970)

    work page 1970

  6. [6]

    Winfree and S

    A. Winfree and S. Strogatz, Nature 311, 611 (1984)

    work page 1984

  7. [7]

    D. P. Zipes and H. J. J. Wellens, Circulation 98, 2334 (1998), 01659

    work page 1998

  8. [8]

    Garfinkel, Journal of electrocardiology 40, S70 (2007)

    A. Garfinkel, Journal of electrocardiology 40, S70 (2007)

    work page 2007

Show all 38 references
  1. [9]

    P. Kohl, P. Hunter, and D. Noble, Prog. Biophys. Molec. Biol. 71, 91 (1999)

    1999

  2. [10]

    Ten Tusscher and A

    K. Ten Tusscher and A. Panfilov, Am. J. Physiol. Heart Circ. Physiol. 291, H1088 (2006)

    2006

  3. [11]

    L. D. Weise, M. P. Nash, and A. V. Panfilov, PLoS ONE 6(7), e21934 (2011)

    2011

  4. [12]

    Niederer, P

    S. Niederer, P. Hunter, and N. Smith, Biophys. J. 90, 1697 (2006)

    2006

  5. [13]

    Niederer and N

    S. Niederer and N. Smith, Prog. Biophys. Mol. Biol. 96, 90 (2008)

    2008

  6. [14]

    R. H. Keldermann et al., Am J Physiol Heart Circ Physiol 299, H134 (2010)

    2010

  7. [15]

    Hodgkin and A

    A. Hodgkin and A. Huxley, J. Physiol. 117, 500 (1952)

    1952

  8. [16]

    L. D. Weise and A. V. Panfilov, PLoS ONE 8, e59317 (2013)

    2013

  9. [17]

    Verlet, Phys

    L. Verlet, Phys. Rev. 159, 98 (1967)

    1967

  10. [18]

    P. Kohl, P. Hunter, and D. Noble, Progress in Biophysics and Molecular Biology 71, 91 (1999)

    1999

  11. [19]

    Skouibine, N

    K. Skouibine, N. Trayanova, and P. Moore, Math. Biosci. 166, 85 (2000)

    2000

  12. [20]

    P. Kohl, K. Day, and D. Noble, Can. J. Cardiol. 14, 111 (1998)

    1998

  13. [21]

    L. D. Weise and A. V. Panfilov, Phys. Rev. Lett. 108, 228104 (2012)

    2012

  14. [22]

    Panfilov, R

    A. Panfilov, R. Keldermann, and M. Nash, Phys. Rev. Lett. 95, 258104 (2005)

    2005

  15. [23]

    Panfilov, R

    A. Panfilov, R. Keldermann, and M. Nash, Proc. Natl. Acad. Sci. U.S.A. 104, 7922 (2007)

    2007

  16. [24]

    L. D. Weise and A. V. Panfilov, Phys. Rev. Lett. 119, 108101 (2017)

    2017

  17. [25]

    Vandersickel et al

    N. Vandersickel et al. , PloS One 9, e84595 (2014)

    2014

  18. [26]

    M. G. Chang et al. , Heart Rhythm: The Official Journal of the Heart Rhythm Society 9, 115 (2012)

    2012

  19. [27]

    Qu and J

    Z. Qu and J. N. Weiss, Annual Review of Physiology 77, 29 (2015)

    2015

  20. [28]

    Grill, V

    S. Grill, V. Zykov, and S. M¨ uller, Phys. Rev. Lett. 75, 3368 (1995)

    1995

  21. [29]

    Dierckx et al

    H. Dierckx et al. , New Journal of Physics 17, 043055 (2015)

    2015

  22. [30]

    R. A. Gray et al. , Circulation 91, 2454 (1995)

    1995

  23. [31]

    Gray et al

    R. Gray et al. , Science 270, 1222 (1995)

    1995

  24. [32]

    N. A. Trayanova, J. Constantino, and V. Gurev, Ameri- can journal of physiology. Heart and circulatory physiol- ogy 301, H279 (2011)

    2011

  25. [33]

    Christoph et al

    J. Christoph et al. , Nature 555, 667 (2018)

    2018

  26. [34]

    Magome et al

    N. Magome et al. , Tissue Engineering Part A 17, 2703 (2011)

    2011

  27. [36]

    LINK TO SUPPLEMENTAL MOVIE

  28. [37]

    LINK TO SUPPLEMENTAL FIGURE

  29. [38]

    APD and DI are recorded at −60mV

  30. [39]

    DI plot for dynamic restitution proto- col and collisions LINK TO SI

    compare APD vs. DI plot for dynamic restitution proto- col and collisions LINK TO SI

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.