The aim of this study was to establish criteria for judging whether postmortem blood drug concentrations could be used for toxicologic evaluation. The study involved 11 autopsy cases in which no obvious decay was observed. Twelve drugs were examined with a cumulative frequency detection of 16. The drug concentrations in cerebrospinal fluid and pericardial fluid were averaged and the ratio between this average concentration and the concentration in femoral venous blood was determined. The mean ratio was close to 1. From these results, the following criteria were drawn: (1) when the ratio of average cerebrospinal fluid/pericardial fluid drug concentration to blood drug concentration is within a range of 0.6-1.4, the postmortem blood drug concentration is usable for toxicologic evaluation; and (2) when the ratio is outside this range, the average cerebrospinal fluid/pericardial fluid concentration should be used as an alternative to drug concentration in postmortem blood. For further substantiation of the criteria, a recommendation was made for more cases and drugs to be examined in relation to postmortem interval and time between drug use and death. Tables and graphs