Left ventricular systolic dysfunction induced by ventricular ectopy: a novel model for premature ventricular contraction-induced cardiomyopathy

Abstract

Background: Premature ventricular contractions (PVCs) commonly coexist with cardiomyopathy. Recently, PVCs have been identified as a possible cause of cardiomyopathy. We developed a PVC-induced cardiomyopathy animal model using a novel premature pacing algorithm to assess timeframe and reversibility of this cardiomyopathy and examine the associated histopathologic abnormalities.

Methods and results: Thirteen mongrel dogs were implanted with a specially programmed pacemaker capable of simulating ventricular extrasystoles. Animals were randomly assigned to either 12 weeks of bigeminal PVCs (n = 7) or no PVCs (control, n = 6). Continuous 24-hour Holter monitoring corroborated ventricular bigeminy in the PVC group (PVC, 49.8% versus control, < 0.01%; P<0.0001). After 12 weeks, only the PVC group had cardiomyopathy, with a significant reduction in left ventricular ejection fraction (PVC, 39.7 ± 5.4% versus control, 60.7 ± 3.8%; P < 0.0001) and an increase in left ventricular end-systolic dimension (PVC, 33.3 ± 3.5 mm versus control, 23.7 ± 3.6 mm; P < 0.001). Ventricular effective refractory period showed a trend to prolong in the PVC group. PVC-induced cardiomyopathy was resolved within 2 to 4 weeks after discontinuation of PVCs. No inflammation, fibrosis, or changes in apoptosis and mitochondrial oxidative phosphorylation were observed with PVC-induced cardiomyopathy.

Conclusions: This novel PVC animal model demonstrates that frequent PVCs alone can induce a reversible form of cardiomyopathy in otherwise structurally normal hearts. PVC-induced cardiomyopathy lacks gross histopathologic and mitochondrial abnormalities seen in other canine models of cardiomyopathy.

Figure 1

Intracardiac electrograms and markers before and after the premature pacing algorithm is enabled. Initial tracing demonstrates normal sinus rhythm (cycle length close to 500 ms,120 bpm). After premature pacing algorithm is enabled, a VP event occurs 250ms (240 bpm) following every VS event reproducing ventricular bigeminy. Furthermore, a compensatory pause can be noted after paced event. VS, Ventricular sensed event; VP, ventricular paced event.

Figure 2

Examples of the programmability of the Premature Pacing Algorithm in a single canine. Panel (a), premature pacing algorithm programmed to 75% ventricular pacing burden (3 VP out of 4 beats) with 200ms fixed coupling interval. Panel (b), premature pacing algorithm programmed to 50 % ventricular pacing burden (1 VP out of 2 beats, ventricular bigeminy) with 300ms fixed coupling interval. An adaptive coupling interval cannot be programmed due to the lack of at least 2 intrinsic R-R intervals. Panel (c), the premature pacing algorithm set for 20 % PVC burden (1 VP out of 5 beats, ventricular pentageminy) and 55% adaptive coupling interval. The initial tracing demonstrates an average cycle length of 2 prior R-R = 450ms, thus VP is delivered at 247ms (55 % of average R-R). Later portion of same panel demonstrates a longer average intrinsic R-R interval (660ms) with subsequent VP delivered at 55% (363ms). Panel (d) depicts 25% PVC burden and 65% adaptive coupling interval. The initial tracing has an average R-R interval=1080ms, thus VP is delivered at 702ms, whereas, the last segment has an average R-R interval=650ms with a VP triggered at 409ms. VP, ventricular paced event; CI, coupling interval; R, Intrinsic ventricular sensed event.

Figure 3

Two-lead ECG tracings and ventricular rate histograms in 24-hour holters at baseline and after premature pacing algorithm is enabled. Panel A (disabled premature pacing algorithm) shows normal ECG tracings (without PVCs) and normal bell-curve shape of intrinsic heart rate distribution with a median R-R interval of 600ms (100 bpm). Panel B (enabled premature pacing algorithm) confirms ventricular bigeminy with one PVC or “V” after every normal beat or “N”. Histogram demonstrates a dominant R-R interval of 250ms due to bigeminal PVCs (R-PVC interval). In contrast, to baseline, the median intrinsic R-R interval (PVC-R interval) increased to 750ms (80 bpm), which can be explained due to a compensatory pause as noted in Figure 1.

Figure 4

LV ejection fraction in PVC (n=7) and control groups (n=6) during 3-month follow up. A 4-week recovery period in the non-euthanized PVC group canines (n=3) demonstrates normalization of LV ejection fraction (bars represent standard deviation).