Oxycodone (OC) is an opioid with strong analgesic effects widely used to treat acute and chronic pain. Interpretation of OC concentrations in postmortem cases is complicated due to tolerance and overlapping concentrations for fatal and non-fatal levels. In this study, our aim was to develop and validate a method for OC and its three metabolites: noroxycodone (NOC), oxymorphone (OM) and noroxymorphone (NOM) in postmortem femoral blood. Our goal was to define reference concentrations for intoxications and non-intoxications and investigate metabolic ratios in different causes of death. A rapid LC-MS-MS method using protein-precipitated postmortem blood was developed. Lower limit of quantitation was 0.005 μg/g blood for all analytes; upper limit of quantitation was 1.0 μg/g for OC and NOC and 0.25 μg/g for OM and NOM. The method displayed high precision (3.3-7.7%) and low bias (-0.3 to 12%). In total, 192 cases were analyzed and concentrations ranged from 0.005 to 13 μg/g for OC, 0.005 to 2.0 μg/g for NOC, 0.005 to 0.24 μg/g for OM, and 0.005 to 0.075 μg/g for NOM. We found a significant difference in OC concentration between the cases where OC contributed and those where it did not. In spite of that, we do not recommend the use of a specific blood concentration to distinguish fatal intoxications. Instead, the percentiles from our data set suggest that concentrations >0.2 μg/g are likely to have contributed to toxicity, but that concentrations as high as 0.3 might be tolerated without toxic effects. In addition, we also found that a low NOC/OC ratio could point toward an acute fatal intoxication. In conclusion, the OC concentration alone may not be sufficient to diagnose a fatal intoxication.