Background: The concentration-effect relationship of piritramide, a synthetic opioid analgesic predominantly used for postoperative analgesia and analgosedation, has not been reported so far. Methods: Twenty-four patients of both genders aged 58.1 (11.7) yr (mean (SD)) received inhalational anaesthesia for abdominal surgery. Postoperative pain was assessed with a visual analogue scale (VAS). Analgesia was provided with piritramide, infused at a rate of 7 mu g.kg(-1).min(-1) until analgesia was considered sufficient NAS<25) or up to a maximum dose of 0.2 mg/kg. The plasma concentrations of piritramide were determined by gas chromatography. An inhibitory fractional sigmoid E-max-model was used to describe the relation between effect site concentration and perceived pain. Results: The equilibration half-life between plasma and effect site concentrations (T-1/2(k(eo))) was 16.8 min (median; range: 4.4-41.6 min). The steady-state plasma concentration required to produce 50% of maximum analgesia (E-50) was 12.1 ng/ml (range: 2.9-29.8 ng/ml) and correlated with initial pain intensity. The slope factor gamma was 1.9 (range: 0.5-6.1) and increased with age. Clinically relevant respiratory depression did not occur Due to the relevant large equilibration half-life of the effect compartment, the context-sensitive half-time of the effect site concentrations after short-time administration (<2 h) clearly exceeded those of alfentanil, sufentanil, and fentanyl. Conclusions: The analgesic effect of piritramide was adequately described by an inhibitory fractional E-max-model. in order to overcome the pronounced hysteresis, piritramide should initially be administered as an intravenous bolus of at least 5 mg.