Exploring the association between epigenetic age and intensive care unit delirium: a pilot feasibility study

Epigenetic clocks (ECs) measuring epigenetic age (EA) and epigenetic age acceleration (EAA) are biomarkers of biological aging, but their association with intensive care unit (ICU) delirium remains underexplored. This is a pilot study utilizing blood samples from participants enrolled in Pharmacological Management of Delirium. Serum samples were collected within 48 h of ICU admission. DNA isolated from serum clots was analyzed in triplicate for DNA methylation (DNAm). EA and EAA were computed for the Horvath, Hannum, PhenoAge, Horvath Skin & Blood, Telomere Length, Best Linear Unbiased Predictor, Elastic Network (EN), GrimAge1, GrimAge2, and DunedinPACE ECs from DNAm data. Principal-component clocks were also assessed. Coma, delirium, and delirium severity were assessed twice daily. LOS was assessed using electronic medical records. Spearman correlations were computed for relationships between EA/EAA and delirium outcomes using SAS. A convenience sample of 20 ICU patients with delirium was included. Mean age was 66.7 years (SD = 11.3), 12% were female, and 50% were Black. The median delirium/coma-free days (DCFD) by day 8 were 3 (IQR 0, 6.5). The intra-class correlation coefficients for EA ranged from 0.893 to 0.999, indicating good reliability and minimal variability across the replicates. EN EAA was moderately inversely correlated with mean CAM-ICU-7 scores by day 8 (Spearman r = -0.54, p = .01) and discharge (r = -0.48, p = .03). No other correlations between other EA/EAAs and delirium, ICU or hospital LOS reached statistical significance. This pilot study demonstrates the feasibility of using ECs from serum clots. Larger studies are needed to assess the relationship between EA/EAA and delirium.