Abstract Body (Do not enter title and authors here): Introduction: Direct mechanical ventricular actuation (DMVA) is a non-blood contacting biventricular support device that uniquely augments diastolic pump function and promotes early recovery of the failing heart. Investigations demonstrate DMVA improves intraventricular dyssynchrony while supporting the failing heart suggesting an effect similar to that of cardiac resynchronization therapy. Optimizing intraventricular synchrony may be dependent on timing device augmentation with the underlying heart rhythm.

Hypothesis: The purpose of this study was to test the hypothesis that a coordination index can be used to optimize synchronous DMVA support of the failing heart.

Methods: Canine (n=9) were instrumented and subjected to repeated periods of ventricular fibrillation (VF) to achieve progressive worsening of ischemic heart failure (HF). DMVA returned the circulation during VF and provided support during HF. DMVA support was initiated using pre-determined pressure/flow profiles based on heart size. DMVA pneumatic drive were tailored to optimize LV pump function using intra-cardiac ECHO imaging. DMVA cycle rates were adjusted between 110-130 bpm to most closely match the underlying native heart rate determined during periodic device removal. Hemodynamics and drive pneumatics were simultaneously recorded using 10 s data acquisitions. The coordination index (CI), a metric for synchrony of peaks was used to analyze the results (see table). Aortic flow waveforms were evaluated for the number of peaks during each DMVA compression. Failure to elicit a pressure or flow response (skipped cycles) was also identified. The CI indices were divided into 5 equal quantiles to characterize synchrony from best (Group 1) thru worst (Group 5). One-way ANOVA with post-hoc Tukey’s HSD was used to assess statistical differences (p<0.05) for all comparisons.

Results: Five groups of equally distributed CI quantiles were grouped from superior to inferior synchrony (Table). Periods of ventricular pacing was suggested evidenced by near-zero CI computations during DMVA support. Improved synchrony correlated with peak systolic and +dP/dt drive drive pneumatics. Lower CI was also linked to absence of skipped cycles. CI during VF support was consistently near-zero.

Conclusions: CI can guide optimal synchronous DMVA support of the failing heart. Future studies will determine if optimizing synchronous support improves intraventricular synchrony and facilitates cardiac recovery.