A cost analysis of alfentanil plus propofol vs morphine plus midazolam for the sedation of critically ill patients

Morphine + midazolam and alfentanil + propofol are regimens offering well tolerated and effective sedation for critically ill patients. However, morphine + midazolam is associated with accumulation in these patients, resulting in prolonged recovery characteristics. Alfentanil + propofol, although more expensive, has a shorter elimination half-life, is not associated with accumulation problems and results in a rapid recovery. This study compared sedation quality, patient recovery characteristics and the cost of alfentanil + propofol and morphine + midazolam for sedating critically ill patients in the intensive care setting. 26 patients were randomly allocated to receive sedation with alfentanil + propofol (n = 17) or morphine + midazolam (n = 9). Outcome measures were the times until extubation, intensive care unit (ICU) transfer and final hospital discharge. Cost analysis assessed both drug-related costs, including drug acquisition and administration, and non-drug-related costs, including bed occupancy. Age, gender, diagnosis, Acute Physiological and Chronic Health Evaluation (APACHE) II scores and sedation quality did not differ significantly between groups. The times to extubation and until patients were fit for transfer from ICU were significantly shorter for patients sedated with alfentanil + propofol than for those sedated with morphine + midazolam. The total costs (at the time of the study pound 1 was equivalent to $US 1.59) for ICU hospital stay per patient for alfentanil + propofol and morphine + midazolam were pound 3063 and pound 9511, respectively, because the shorter recovery characteristics of alfentanil + propofol led to a reduction in ICU stay. Corresponding costs for total hospital stay were pound 6063 and pound 13 735, respectively. In conclusion, alfentanil. + propofol has a better pharmacoeconomic profile than morphine + midazolam for sedating critically ill patients in the ICU setting.