372 Cognitive impairment is associated with increased nightly heart rate variability in a large longitudinal study of sleep in cirrhosis

Objectives/Goals: Heart rate variability (HRV) is the variation in time between each heartbeat and an important biomarker of autonomic nervous system function, with higher HRV generally associated with improved parasympathetic tone and cardiac health. The most reliable measurement is during sleep, but sleep and HRV are both understudied in cirrhosis and HE. Methods/Study Population: We longitudinally assessed sleep and resting cardiovascular biomarkers in a cohort (n=119) of patients with cirrhosis using Oura, a commercially available fitness tracker. Baseline clinical data, psychometric hepatic encephalopathy score (PHES) and 6 months of passive sleep data collection were obtained. 117 participants enrolled, 6 collected no sleep data, 13 did not complete PHES; 10,065 nights (n=98) were available, and 9,692 were analyzed after outlier trimming (top and bottom 1% of data). Mixed effects modeling (MEM) assessed associations. Pre-specified covariates included age, gender, sodium, albumin, bilirubin, INR, and ascites. Missing data was not imputed. Results/Anticipated Results: In univariable analysis, impaired cognition by PHES was associated with increased average HRV (B 5.2 95%CI 0.25, 10.1 p=0.04) but not respiratory rate, heart rate, or lowest measured heart rate. CTP Score (B -1.8 95%CI -3.0, -0.6 p<0.01) and INR (B -6.1 95%CI -11.6, -0.5 p=0.03) were inversely associated with HRV. Higher baseline albumin (B 3.9 95%CI 0.34, 7.4 p=0.03) and sodium (B 1.0 95%CI 0.18, 1.75 p=0.02) were associated with increased HRV, while age, gender, and creatinine did not significantly associate with HRV. In a MEM adjusting for the pre-specified covariates, the association was greater (B 8.85 95%CI 3.85, 13.9 p<0.01), remaining significant after log-normalizing HRV. In a sensitivity analysis adjusting for current alcohol use or prior variceal bleed, the results were unchanged. Discussion/Significance of Impact: In a large cohort in cirrhosis, cognitive impairment was associated with increased HRV, in contrast to other markers of liver disease, where HRV was reduced and unrelated to other cardiovascular measures. This paradoxical relationship, if validated, could be developed as a longitudinal digital biomarker for the detection of cognitive impairment in HE.