Three forensic cases with unexpected benzodiazepines are described. Case A involved a 41-year old man who was hospitalized with a severe head injury and died after four days of intensive treatment. In the autopsy samples, desalkylflurazepam was detected. In the samples collected at hospital admission, nor desalkylflurazepam nor one of its possible orally taken parent compounds (flurazepam or ethylloflazepate) were present. In case B, the urine of a 22 year old woman who was believed to be the victim of a drug facilitated sexual assault (DFSA) was confirmed positive for the presence of diazepam. The victim claimed not to have used diazepam. Case C involved a 20-month old toddler who was administered to the hospital with signs of weakness and sedation after a stay with the grandmother. Besides the presence of lorazepam in the urine, GC-MS analysis indicated the presence of delorazepam, which would be suggestive for the intake of cloxazolam.

All cases have undergone the Systematic Toxicological Analysis (STA) in our lab, which consists of screening procedures (immunological screening on urine, screening by GC-MS without and with derivatization on enzymatically hydrolyzed urine and blood, screening by HPLC-PDA on blood) and targeted analyses (ethanol by headspace-GC-FID, GHB and “classic” drugs of abuse by GC-MS). The libraries used for identification are both commercial (Maurer/Pfleger/Weber 4^th edition 2011, Rösner Designer Drugs 2013) and in-house built. Furthermore, most compounds identified by screening were quantified in blood using the most suitable inhouse technique (GC-MS, LC-MSMS or HPLC-PDA).

In case A, no alcohol, pharmaceuticals or drugs of abuse were detected in the samples collected at hospital admission. In the hospital, the patient was administered atracurium, levetiracetam, lidocaine, midazolam, propofol, remifentanyl and thiopental. In the autopsy samples, all pharmaceuticals (mostly together with one or more metabolites) were detected by our STA, except for remifentanyl. Moreover, desalkylflurazepam, the metabolite of flurazepam and ethylloflazepate, was detected. This phenomenon could initially not be explained, since both pharmaceuticals are taken orally, and the patient became comatose soon after hospital admission. The presence of desalkylflurazepam could be explained by a recent paper of Vogt et al. (Drug Test. Analysis 2013, 5, 745-747) who unravelled that desalkylflurazepam is not a metabolite of midazolam, but an intermediate in the synthesis of midazolam which is sometimes still present in the formulation Dormicum®. In case B, laboratory analysis revealed the presence in urine of acetaminophen, aripiprazole, citalopram, diazepam, tetrazepam and valproic acid (most of them together with one or more of their metabolites). All parent compounds were also detected in blood, except for diazepam. The presence of diazepam in urine could be explained by the metabolism of tetrazepam and not by an (involuntary) consumption of diazepam itself. In case C, a positive benzodiazepine screening on urine was confirmed by the presence of lorazepam in urine. Only after injection of an acetylated extract in the GC-MS apparatus, a library hit for delorazepam appeared. Further chromatographic analysis revealed that this compound was not delorazepam (different retention time), but most probably a lorazepam (metabolite) artefact. Therefore, the possibility of administration of cloxazolam (which metabolizes quickly to delorazepam) could be ruled out.

In a forensic context, one has to take care in the interpretation of the origin of benzodiazepine findings. The presence of synthesis-intermediates in pharmaceutical formulations, minor metabolism pathways and GC-artefacts can significantly complicate elucidation of the events.