The goal of this study was to determine how the postmortem interval and duration of storage of blood at 4 degrees C affect endogenous gamma-hydroxybutyric acid (GHB) levels in blood. Forty-three autopsy cases of non-users of GHB were involved. The postmortem interval ranged from 8 to 132 h. Blood samples were collected and stored without any preservatives at 4 degrees C for 1 day up to 15 months until analysis. In some cases, samples were also stored at -20 degrees C for 10 days to 7 months to determine GHB levels at autopsy. Blood GHB concentrations were measured by headspace gas chromatography after GHB was converted to gamma-butyrolactone. Blood GHB concentrations ranged from 0 to 43.0 microg/ml and averaged 9.80 microg/ml. A positive correlation was observed between concentration and postmortem interval (r = 0.571) but no correlation was found between concentration and storage interval at 4 degrees C. In 14 blood samples stored at -20 and 4 degrees C for 10 days, GHB concentrations were 4.55+/-3.88 and 6.06+/-4.27 microg/ml, respectively. In another eight blood samples stored at -20 and 4 degrees C for 1-7 months, GHB concentrations were 3.77+/-2.76 and 5.49+/-2.97 microg/ml, respectively. A large portion of endogenous GHB detected in blood of corpses may be produced during the interval between death and autopsy, rather than during storage of blood at 4 degrees C until analysis. In an additional experiment, it was suggested that glycolysis by bacteria may enhance endogenous GHB production.