Utility of device-derived daily activity, a novel digital biomarker, to predict ventricular arrhythmias – data from the CERTITUDE registry

Abstract

Abstract Background While there have been prior studies showing an association between activity and outcomes, there have been no studies examining the temporal relationship between device-derived daily activity, a novel digital biomarker, and ventricular tachyarrhythmias (VT). Purpose In a big dataset with over 55,000 pacemaker, ICD, and CRT devices, we aimed to identify whether changes in activity predict VT, or else VT predict changes in activity. Methods The CERTITUDE registry comprises a de-identified database of over 55,000 U.S. BIOTRONIK pacemaker, ICD, CRT devices, and loop recorders active on Home Monitoring®. Daily data on leads, arrhythmias, and physiological parameters such as activity are captured. Patient activity is reported daily as percentage active during the day, assessed by a one-axis accelerometer at ∼0.4 Hz frequency. Analysis to ascertain temporal changes in device-derived activity associated with treated VT was performed using the first event per device and 7-day activity windows (baseline, pre- and post-event). Baseline period was defined as 31–38 days prior to VT. VT events were categorized by heart rate (≤200 bpm,>200 bpm) and treatment (shock with or without ATP, ATP alone). Differences in activity between baseline, and pre- and post-VT were analyzed using the binomial proportion test. Results A total of 16,475 devices (9732 ICDs, 6743 CRT-Ds) had activity data available for analysis. The cumulative follow-up duration was 18,355 years (5.6 million days with transmission). Of the 2636 VT events analyzed, 1409 had a heart rate >200 bpm, and 593 were treated with shock. Patients with VT events >200 bpm treated with shock had a significant reduction in activity post-VT with a median −8.7% reduction (IQR −24.6%; 7.3%, p<0.001). However, there was no reduction in activity before the VT>200 bpm (p=0.690) (Figure). VT events >200 bpm treated with ATP alone were not associated with reduction in activity before or after the episode. Similarly, VT events ≤200 bpm treated with shock were also associated with a reduction in activity following the event (−5.8%, IQR −29.5, 12.3%, p=0.003), but not prior to the VT event. Conclusions In this report from the CERTITUDE registry, we have shown a temporal decline in device-derived activity following ventricular arrhythmias>200 bpm and ≤200 bpm treated with a shock, but not in patients treated with ATP. Monitoring device-derived activity post-VT events with a shock could provide relevant clinical information and potentially warrant intensified treatment. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): Biotronik