Pharmacoeconomic modeling of lorazepam, midazolam, and propofol for continuous sedation in critically ill patients

Study Objective. To compare the expected costs of short-, intermediate-, and long-term sedation (< 24, 24-72, and > 72 hrs, respectively) with propofol, lorazepam, and midazolam in an intensive care unit. Methods. Decision-analysis models were constructed for each sedative and each duration by using institutional costs associated with drug administration and adverse events (including personnel time). Costs were A expressed in 2002 U.S. dollars. Adverse events were agitation, hypertriglyceridemia and/or pancreatitis, hypotension, nutritional changes, ventilator-associated pneumonia, and prolonged awakening and/or extubation. MEDLINE and EMBASE databases were searched to obtain durations of sedation, the incidence of outcomes, and cost estimates of outcomes. The ability to maintain specific levels of sedation was assumed equivalent among the sedatives. Univariate sensitivity analyses were conducted to determine the cost-driving variables, and probabilistic sensitivity analyses were conducted by using second-order Monte Carlo simulations. Results. Weighted mean durations of sedation from 50 studies were 13.46 (short term), 45.27 (intermediate term), and 119.78 (long term) hours. Expected costs for sedation with lorazepam, midazolam, and propofol, respectively, were $497, $294, and $272 short term; $932, $587, and $674 intermediate term; and $1604, $1737, and $2033 long term. Propofol was least costly in 86% of the short-term simulations, midazolam was least costly in 97.5% of the intermediate-term simulations, and lorazepam was least costly in 84% of the long-term simulations. The most important cost-driver for all sedatives was drug cost. Prolonged extubation after sedation was an important cost-driver for lorazepam and midazolam, especially as sedation was lengthened. Conclusion. Propofol, midazolam, and lorazepam had the lowest expected costs for short-, intermediate-, and long-term sedation, respectively. Many factors aside from drug costs influenced the cost of sedation.