OBJECTIVES: Cost-consequence analysis of a RP (remifentanil + propofol) regimen versus a conventional, commonly used MF (midazolam + fentanyl) regimen for sedation of mechanically ventilated postoperative cardiac surgery patients. METHODS: We conducted a prospective, single-blinded, randomised cost-consequence study with 80 patients in one German intensive care unit (ICU). The RP group received remifentanil (6- max. 60 µg kg-1 h-1) and - if sedation at maximal remifentanil dose was insufficient - propofol (0.5-4.0 mg kg-1 h-1). The MF group received midazolam (0.02-0.2 mg kg-1 h-1) and fentanyl (1.0-7.0 µg kg-1 h-1). Direct costs for drugs, material (variable costs only), and staff were considered (hospital's perspective, 2003 prices). Sensitivity and scenario analyses were performed with a decision-analytic model. As the remifentanil dose in the study RP regimen (baseline) was higher than in routine practice we simulated a “routine practice” scenario: We lowered the mean remifentanil infusion rate from 41.2 µg kg-1 h-1 to 9 µg kg-1 h-1, increased the propofol infusion rate from 2.2 mg kg-1 h-1 to 4 mg kg-1 h-1 and assumed that this scenario would have rendered the same reduction (24%) in staff costs compared to MF regimen and identical material and drug utilisation (except RP) as at baseline. RESULTS: Compared to MF regimen, RP regimen (baseline) led to a significantly shorter mechanical ventilation time (3.5h) and earlier discharge from ICU (18.3 h) and equal average net costs of €1700 for the ICU stay per patient. The routine practice scenario rendered 53% lower RP medication costs than baseline thus yielding net savings of €200 per patient. These results are sensitive to staff and drug cost variations. CONCLUSIONS: Analysis indicates that RP regimen dominates MF regimen in the investigated setting as it reduces the mechanical ventilation time and hence the risk of ventilator-associated morbidity at equal costs (baseline) or even savings (scenario).