Alcohol elimination was examined in 10 patients involved in accidents while intoxicated. The influence of trauma, particularly polytrauma resulting in haemorrhage shock and its therapeutic treatment were analysed. The blood alcohol concentrations were determined according to the usual forensic criteria (2 alcohol dehydrogenase and 2 gas chromatography measurements). Observation periods ranged from 3 hours 45 minutes to 12 hours 35 minutes, with blood being drawn at intervals ranging from 45 minutes to 185 minutes (on average 70 minutes). Results of two patients (delta 60 = 0.22/1000/h and 0.28/1000/h) who only had 3 venous drawn and results of a deceased patient from whom only four arterial samples could be obtained (beta 60 = 0.21/1000/h) were disregarded when working out the average values. The blood alcohol curve plateaued in the case of the deceased patient as well as in the case of a patient whose hepatic circulation was curtailed for approximately half an hour during surgery. The blood alcohol curves for the remaining patients for uniformly linear with beta 60 values between 0.17/1000/h and 0.21/1000/h (mean = 0.18/1000/h +/- 0.01) in arterial samples and 0.18/1000/h and 0.21/1000/h (mean = 0.18/1000/h +/- 0.01) in venous samples. Given our results and the existing literature, we feel that retrograde calculations of the BAC can be justified in patients with polytrauma, despite the small number of patients included in the study. Naturally, the usual forensic criteria have to be taken into account, as well as individual situations. Examples that can be mentioned here are liver failure or curtailment of hepatic circulation during surgery.