Article révisé par les pairs
Résumé : Purpose: To investigate heart rate and EEG variability and their coupling in patients with sepsis and determine their relationship to sepsis severity and severity of sepsis-Associated brain dysfunction. Methods: Fifty-Two patients with sepsis were prospectively identified, categorized as comatose (N = 30) and noncomatose (N = 22), and compared with 11 control subjects. In a 30-minute EEG and electrocardiogram recording, heart rate variability and EEG variability (measured by the variability of relative power in a modified alpha band = RAP) and their coupled oscillations were quantified using linear (least-square periodogram and magnitude square coherence) and nonlinear (Shannon entropy and mutual information) measures. These measures were compared between the three groups and correlated with outcome, adjusting for severity of sepsis. Results: Several measures of heart rate variability and EEG variability and of their coupled oscillations were significantly lower in patients with sepsis compared with controls and correlated with outcome. This correlation was not independent when adjusting for severity of sepsis. Conclusions: Sepsis is associated with lower variability of both heart rate and RAP on EEG and reduction of their coupled oscillations. This uncoupling is associated with the severity of encephalopathy. Combined EEG and electrocardiogram monitoring may be used to gain insight in underlying mechanisms of sepsis and quantify brainstem or thalamic dysfunction.