Identifying insulin overdose as a cause of death is difficult in the forensic setting as there are no real-world data on expected concentration of insulin in a diabetic subject on insulin therapy due to the assay-dependent cross-reactivity of insulin analogues and their metabolites in vivo . Despite this, pathologists are often asked to provide interpretation of post-mortem bloods regarding whether excessive insulin analogue was administered. We aimed to define expected trough insulin concentrations in subjects on a variety of routine insulin therapies and regimens. Audit of our laboratory data over the past seven years showed an average of 32 requests per year for insulin and C-peptide testing, with 33% of samples deemed unsuitable due to haemolysis. Routine blood samples from diabetic inpatients were analysed using Abbott, Beckman and Roche platforms, and C-peptide, insulin analogue dose, timing and estimated glomerular filtration rate (eGFR) were recorded. Of the seven analogues studied, Abbott showed dose- and eGFR-dependent cross reactivity with Ryzodeg. Abbott and Beckman results showed no significant difference except among Levemir and Ryzodeg users. Despite its specificity for human insulin, the Roche assay detected recent NovoRapid use and glargine metabolite. In conclusion, post-mortem insulin interpretation needs consideration of assay cross-reactivity, analogue type, eGFR and time since last dose.
Thomas et al. (Sun,) studied this question.