The purpose of this study was to compare blood fentanyl concentrations in fentanyl-related deaths with fentanyl concentrations found incidentally at autopsy, as well as with fentanyl concentrations found in hospitalized patients receiving fentanyl. Between the years 1997 to 2005, 23 fentanyl-positive postmortem cases were identified. Nineteen of 23 (82.6%) cases were deemed to be drug overdoses. Fentanyl alone was responsible for 8 of the 19 (42.1%) overdose deaths. Mean and median fentanyl concentrations were 36 (SD 38) microg/L and 22 microg/L, respectively, range 5-120 microg/L. Seven of the cases were accidental, one undetermined. The remaining 11 of the 19 (57.9%) cases were mixed drug overdoses. Fentanyl concentrations in these cases were 31 (SD 46) microg/L, range 5-152 microg/L. All of the mixed drug overdoses were determined to be accidental. Four cases where fentanyl was considered an incidental postmortem finding were determined to be natural deaths. In hospitalized inpatients (n = 11) receiving fentanyl 2 of the patients receiving fentanyl for chronic pain for more than 3 months had concentrations of 8.5 microg/L and 9.9 microg/L. The other nine inpatient concentrations were less than 4 microg/L. In conclusion, blood fentanyl concentrations found in cases where fentanyl alone was determined to be the cause of death were similar to cases where fentanyl was part of a mixed drug overdose. There was also considerable overlap between fentanyl concentrations in fentanyl-related overdose deaths compared to hospitalized patients being treated for chronic pain. Fentanyl concentrations in postmortem cases must be interpreted in the context of the deceased's past medical history and autopsy findings.