Table 3.
Variable | 25th Percentile | 75th Percentile | Difference in Mean Delirium/Coma-Free Days (95% Confidence Interval) | p |
---|---|---|---|---|
Age at enrollment | 47.8 | 66.0 | −0.6 (−1.6 to 0.3) | .19 |
Acute Physiology and Chronic Health Evaluation II acute physiology score component | 11.0 | 20.0 | −0.4 (−1.6 to 0.8) | .48 |
Dexmedetomidine vs. lorazepama | Lorazepam | Dexmedetomidine | 2.1 (0.6 to 3.6) | .007 |
Baseline kynurenine/tryptophan ratio | 0.1 | 0.6 | − 2.1 (− 3.2 to − 1.0) | −.001 |
Linear regression was used to study the role of kynurenine/tryptophan in delirium/coma-free days after adjusting for age, severity of illness, and study drug. The difference in means represents the difference in delirium/coma-free days between patients at the 75th percentile of kynurenine-to-tryptophan ratios of our population vs. the 25th percentile. Thus, a patient at the 75th percentile for kynurenine/tryptophan levels had 2.1 fewer days without acute brain dysfunction than a patient at the 25th percentile. This comparison between percentiles is more clinically relevant than the traditionally used one-unit change.
Patients in this cohort received sedation with either dexmedetomidine or lorazepam. Patients using dexmedetomidine (as compared to lorazepam) had 2.1 more days alive and free from delirium and coma.