Factors Predisposing to Coma and Delirium

BACKGROUND: Delirium and sedative-induced coma are described as incremental manifestations of cerebral dysfunction. Both may be associated with sedative or opiate doses and pharmacokinetic or pharmacogenetic variables, such as drug plasma levels (exposure), drug metabolism, and/or their transport across the blood-brain barrier. OBJECTIVES: To compare biological and drug treatment characteristics in patients with coma and/or delirium while in the ICU. PATIENTS AND MEASUREMENTS: In 99 patients receiving IV fentanyl, midazolam, or both, we evaluated drug doses, covariates likely to influence drug effects (age, body mass index, and renal and hepatic dysfunction); delirium risk factors; concomitant administration of CYP3A and P-glycoprotein substrates/inhibitors; ABCB1, ABCG2, and CYP3A5 genetic polymorphisms; and fentanyl and midazolam plasma levels. Delirium and coma were evaluated daily. In patients with only coma (n=15), only delirium (n=7), and neither ever (n=14), we measured plasma levels of tumor necrosis factor-α, interleukin (IL)-1β, IL-1RA, IL-6, IL-8, IL-10, IL-17,macrophage inflammatory protein-1β, and monocyte chemotactic protein-1. RESULTS: Time to first coma was associated with fentanyl and midazolam doses (p=0.03 and p=0.01, respectively). The number of days in coma was associated with the number of days of coadministration of CYP3A inhibitors (r=0.30; p=0.006). Plasma levels of fentanyl were higher in patients with clinical coma (3.7±4.7 vs. 2.0±1.8 ng/mL, p=0.0001) as were midazolam plasma levels (1050±2232 vs. 168±249 ng/mL, p=0.0001). Delirium occurrence was unrelated to midazolam administration, cumulative doses, or serum levels. Days with delirium were associated with days of coadministration of P-glycoprotein inhibitor (r=0.35; p=0.0004). Delirious patients had higher levels of the inflammatory mediator IL-6 than comatose patients (129.3 vs. 35.0 pg/mL, p=0.05). CONCLUSIONS: Coma is associated with fentanyl and midazolam exposure; delirium is unrelated to midazolam and may be linked to inflammatory status. These data suggest that iatrogenic coma and delirium are not mechanistically linked.