Nonlinear analysis of beat-to-beat variability of action potential time series data identifies dynamic re-entrant substrates in a hypokalaemic mouse model of acquired long QT syndrome

Abstract Background Previous studies have quantified repolarization variability using time-domain, frequency-domain and nonlinear analysis in mouse hearts. Here, we investigated the relationship between these parameters ventricular arrhythmogenicity a hypokalaemia model of acquired long QT syndrome. Methods Left monophasic action potentials (MAPs) were recorded during right regular 8 Hz pacing normokalaemia (5.2 mM [K + ]), modeling LQTS (3 ]) or with 0.1 heptanol Langendorff-perfused Results During normokalaemia, mean APD was 33.5 ± 3.7 ms. Standard deviation (SD) APDs 0.63 0.33 ms, coefficient variation 1.9 1.0% root square (RMS) successive differences 0.3 Low- high-frequency peaks 0.6 0.5 2.3 0.7 Hz, respectively, percentage powers 38 22 61 23%. Poincaré plots n +1 against revealed ellipsoid morphologies SD along line-of-identity (SD2) to perpendicular (SD1) ratio 4.6 1.1. Approximate sample entropy 0.49 0.12 0.64 0.29, respectively. Detrended fluctuation short- long-term slopes 1.62 0.27 0.60 0.18, Hypokalaemia provoked tachycardia six seven hearts, prolonged (51.2 7.9 ms), decreased SD2/SD1 (3.1 1.0), increased approximate (0.68 0.08 1.02 0.33) short-term slope (1.23 0.20) (ANOVA, P < 0.05). Heptanol prevented VT all hearts studied without further altering above observed hypokalaemia. Conclusion Reduced SD2/SD1, may reflect arrhythmic risk exerts anti-arrhythmic effects affecting variability.