Abstract
Diabetic ketoacidosis (DKA) is a frequent and life-threatening complication, whose diagnosis remains challenging in forensic practice. We aimed at assessing the performance of a commercially available blood glucose and ketone monitoring device (BGMD) in measuring glucose and ketone levels in post-mortem vitreous (VH) and blood samples, in order to determine if such a device can be used for screening lethal cases of DKA at autopsy. VH and blood samples were collected in cases of unexplained causes of death at autopsy. Glucose and β-hydroxybutyrate (BHB) were measured in VH and BHB in blood using the BGMD. The values were compared to those obtained with validated enzymatic methods. Values ≥ 10 mmol/L were considered to be elevated for glucose, and BHB values ≥ 2.5 mmol/L were considered to indicate ketoacidosis. There was a strong and significant correlation between VH glucose and blood BHB concentrations measured with the BGMD and the validated method (r = 0.78 and r = 0.80, p < 0.0001, respectively), whereas no correlation was found for VH BHB values (r = 0.19, p = 0.19). The sensitivity and specificity of the BGMD were both excellent (1.0) to detect elevated VH glucose levels with a threshold of 14.4 mmol/L, and to detect elevated blood BHB levels with a threshold of 2.85 mmol/L. In contrast, the specificity of the BGMD to detect high BHB levels in VH was poor (0.50) with an optimal threshold of 2.5 mmol/L. We showed that a commercially available BGMD is suitable for identifying cases of lethal DKA and other metabolic disorders at autopsy, through the investigation of vitreous glucose and blood BHB. We therefore recommend the systematic use of a BGMD for screening these conditions in cases of unexplained deaths.