Can measurements of heroin metabolites in post-mortem matrices other than peripheral blood indicate if death was rapid or delayed?

BACKGROUND: In heroin-related deaths, it is often of interest to determine the approximate time span between intake of heroin and death, and to decide whether heroin or other opioids have been administered. In some autopsy cases, peripheral blood cannot be sampled due to decomposition, injuries or burns. The aim of the present study was to investigate whether measurements of heroin metabolites in matrices other than peripheral blood can be used to differentiate between rapid and delayed heroin deaths, and if morphine/codeine ratios measured in other matrices can separate heroin from codeine intakes.

METHODS: In this study, we included 51 forensic autopsy cases where morphine was detected in peripheral blood. Samples were collected from peripheral and cardiac blood, pericardial fluid, psoas and lateral vastus muscles, vitreous humor and urine. The opioid analysis included 6-acetylmorphine (6-AM), morphine, morphine-3-glucuronide (M3G), morphine-6-glucuronide (M6G) and codeine. Urine was only used for qualitative detection of 6-AM. 45 heroin-intake cases were divided into rapid deaths (n=24), based on the detection of 6-AM in blood, or delayed deaths (n=21), where 6-AM was detected in at least one other matrix but not in blood. An additional 6 cases were classified as codeine-intake cases, based on a morphine/codeine ratio below unity (<1) in peripheral blood, without detecting 6-AM in any matrix.

RESULTS: The median morphine concentrations were significantly higher in the rapid compared with the delayed heroin deaths in all matrices (p=0.004 for vitreous humor and p<0.001 for the other matrices). In the rapid heroin deaths, the M3G/morphine concentration ratios were significantly lower than in the delayed deaths both in peripheral and cardiac blood (p<0.001), as well as in pericardial fluid (p<0.001) and vitreous humor (p=0.006), but not in muscle. The morphine/codeine ratios measured in cardiac blood, pericardial fluid and the two muscle samples resembled the ratios in peripheral blood, although codeine was less often detected in other matrices than peripheral blood.

CONCLUSIONS: Measurements of heroin-metabolites in cardiac blood, pericardial fluid and vitreous humor provide information comparable to that of peripheral blood regarding rapid and delayed heroin deaths, e.g. M3G/morphine ratios <2 indicate a rapid death while ratios >3 indicate a delayed death. However, considerable overlap in results from rapid and delayed deaths was observed, and measurements in muscle appeared less useful. Furthermore, matrices other than peripheral blood can be used to investigate morphine/codeine ratios, but vitreous humor seems less suited.