The mechanisms that cause these phenomena are complex and often intricate. Certain organs, which concentrate the molecules before death, may release them very early in the vascular sector. The gastrointestinal tract, liver, lungs and myocardium are mainly concerned. Cell autolysis also plays a part in drug release. Furthermore, micro-organisms (mainly bacteria and yeasts) which colonize the organism during putrefaction may cause neoformation and/or the degradation of certain molecules. Lastly, it appears that the physicochemical and pharmacokinetic profile of xenobiotics, notably their lipophilic nature, their ionization state and their volume of distribution may be factors likely to influence redistribution phenomena. Some recommendations concerning anatomic sampling sites, sampling methods and sample storage make it possible to limit these phenomena.

Hydrogen sulfide (H(2)S) poisonings are classically reported in occupational settings. We describe an unusual domestic case of fatal acute poisoning by H(2)S inhalation. A mother and her infant daughter were found dead in the kitchen of their home. The emergency medical team described a strong smell of rotten eggs, suggesting acute H(2)S poisoning. Autopsies revealed only multiorgan congestion. H(2) S was measured in blood and lung tissue samples by gas chromatography/mass spectrometry. Body fluids were negative, but H(2) S was found in the lungs of both the mother and the child at concentrations of 1.46 and 1.92 mg/kg, respectively, concentrations described in the literature as potentially lethal. Expert surveys of the premises suggested a complex mechanism involving both defective maintenance of the pipes and drains of the building and faulty assembly of the sink siphon, which led to stagnation of waste water and formation of a pocket of H(2)S.

The success of high-dose buprenorphine (HDB) as substitution therapy for major opioid dependence is related to its partial agonist effect on opioid receptors, which in theory makes it very safe to use. However, numerous deaths directly attributable to buprenorphine have been described in the literature. These deaths are generally related to misuse of HDB with intravenous administration and/or concomitant use of benzodiazepines, and they usually occur in patients on HDB substitution therapy for opioid dependence. We present three deaths attributed to HDB which arose from uncommon mechanisms and led to unusual forensic situations. The first death was that of a patient admitted to hospital after simultaneous prescription of HDB, clonazepam, oxazepam, and cyamemazine. The second death followed forcible administration of a very low dose of HDB to a patient with post-hepatitis C cirrhosis and heart failure. The third death was subsequent to an HDB overdose, probably with suicidal intent, in a young woman who had not been prescribed the drug as opiate substitute. Such deaths raise the question of the mechanisms involved and draw attention to the resulting unusual forensic situations.

If femoral blood is not available at autopsy, toxicological analyses, in particular blood ethanol measurements, are carried out on cardiac blood. This is known to be subject to major redistribution. We aimed to determine whether subclavian blood can be equated with a peripheral blood sample and could be used if femoral blood is not available. The study was based on 50 medicolegal autopsies in which we compared ethanol concentrations between subclavian blood, the different heart blood compartments (right and left cardiac blood), and femoral blood. Mechanisms that could lead to variations in concentration, i.e., postmortem redistribution and/or endogenous production, were also taken into account in interpreting the results. Ethanol concentrations were determined by headspace gas chromatography with a flame ionization detector. In each case, we recorded the circumstances of death, resuscitation attempts if any, degree of putrefaction, chest or abdominal trauma, and/or inhalation of gastric fluid in the airways. Ethanol concentrations in subclavian blood were found to be close to those in peripheral blood (p = 0.948) and were not influenced by the degree of putrefaction (r = 0.017, p = 0.904), gastric ethanol concentration (r = -0.011, p = 0.940), inhalation of gastric contents in the airways (p = 0.461), or cardiac resuscitation attempts (p = 0.368). We discuss the possible explanations for these findings and stress the value of sampling subclavian blood when femoral blood is not obtainable at autopsy.