The aim of the study was to test the hypothesis that the requirement of sufentanil is reduced in elderly patients when the opiate is primarily used to facilitate mechanical ventilation in a medical intensive care unit. A further aim was to study whether elderly patients developed withdrawal symptoms after discontinuing prolonged sufentanil administration. We have studied prospectively two groups of patients requiring mechanical ventilation for more than 96 h; group 1 age < 60 years (n=316 or 68%) and group 2 age > 70 years (n=150 or 32%). In all patients sufentanil and midazolam were administered continuously in order to facilitate ventilatory support. After an initial intravenous bolus injection of sufentanil 3.0–8.0 μg kg−1, the dosage was adjusted to the patients needs (0.75–1.0 μg−1 kg−1 h) using a modified Ramsey score by accepting between 3b and 4a as the end point. The amount of sufentanil administered and side effects were recorded at 24-h intervals. Seventy-two hours following the start of sedation with sufentanil/midazolam the dose of sufentanil required for sedation increased significantly (P <0.05) in both groups when compared with the first 24 h. There was no statistical difference between the two groups in sufentanil requirement at any time during the study. This suggests that tachyphylaxis develops to a similar degree in patients in both age groups. In addition, weaning in the elderly was characterized by a similar degree of withdrawal-like symptoms suggesting that independent of age, there are similar receptor related reactions once the opiate is withdrawn.