ABSTRACT
Background: The trajectory of cardiac physiology throughout a COVID-19 infection remains poorly described. This study aimed to characterize physiologic changes associated with acute COVID-19 infection among patients with cardiac implantable electronic devices (CIEDs). Methods: The study population included 286 patients with a Boston Scientific CIED across a large health system. CIED sensor data from March 2020-February 2021 were linked to clinical data from electronic health records. Three cohorts were created: COVID-positive (n=20), COVID-negative (n=166), and an untested control (n=100) to account for testing bias as institutional protocols only allowed symptomatic patients to be tested in the early months of the study. The magnitude of sensor changes from baseline (30-60 days before a COVID-19 test) to event (15-day window around a test) was compared between cohorts using one-way ANOVA. A date during the study window was randomly selected to serve as the event for control cases. Results: Comparing COVID-positive vs. COVID-negative vs. control, there was a greater change from baseline to event in respiratory rate (15% vs. 2% vs. 0.6% [p<0.0001]), activity (-44% vs.-12% vs.-6% [p<0.0001]), temperature (1% vs. 0.1% vs.-0.3% [p<0.0001]), and HeartLogic Index, a composite index using multiple sensors (227% vs.-11% vs. 2% [p=0.004]). Average sensor changes show prominent differences surrounding a COVID-19 test for COVID-positive compared to COVIDnegative and control cases (Figure). Conclusions: Physiologic changes associated with COVID-19 infection are detectable using CIEDs. COVID-positive patients have significant increases in respiratory rate, temperature and HeartLogic index and significant decreases in activity compared to COVID-negative and untested patients. Future studies in a larger population will help characterize the utility of CIEDs in infection surveillance and physiologic worsening that should prompt medical attention.