Translational Study Tracking Dynamic Changes in Cardiac Pattern Variability and Systemic Inflammation in Critically Ill Patients

Abstract

Sepsis is a dynamic complex condition with high patient mortality and health‐care system cost for which the understanding of underlying pathophysiology is still evolving. We hypothesize that pattern variability of centrally controlled physiological rhythms, like cardiac and respiratory rhythms, decreases with an increase in systemic inflammation, such as sepsis. In our lab, we have previously demonstrated in an animal sepsis model that respiratory pattern predictability will increase with disease onset and progression. We describe a complementary IRB approved clinical study, in which physiological waveform data were recorded continuously from consented adult patients (n=43 to date) admitted to the medical intensive care unit (MICU). Serum was collected at intervals determined clinically and subsequently quantified using Luminex assays. The electrocardiogram data was analyzed for heart rate variability (HRV) indices, including Coefficient of Variation, Poincare Plots and Power Spectral Density (high frequency power (HFP)) for the hour preceding and succeeding each serum sampling. Of the 15 patients analyzed, we identified 18 intervals between successive blood draws with a significant change in IL‐6 serum concentration, defined as a 20% relative change accompanied by >20 pg/mL absolute change. In each of the 6 time intervals where IL‐6 was significantly increased there was an associated decrease in HRV and/or HFP. In 12 time intervals IL‐6 decreased; 7 of these were associated with an increase in HRV and/or HFP. Previous studies and our own studies on humans and rats conclude that decreases in the variability, or increases in the predictability, of physiological rhythms correlate to diseased state. Our working hypothesis is that neural inflammation in the brainstem alters central regulation of cardio‐respiratory rhythms and their coupling. As sepsis has had consistent and widespread negative impact on patients, the heath‐care system, and healthcare economy, it is imperative to identify and describe biometrics that serve as early warnings to clinicians.Support or Funding InformationNIH U01 EB021960‐01A1 and VA Research Service Merit Award I01BX004197‐01A1