Postmortem Metabolomics of Insulin Intoxications and the Potential Application to Find Hypoglycemia-Related Deaths

Abstract

Postmortem metabolomics can assist death investigations by characterizing metabolic fingerprints differentiating causes of death. Hypoglycemia-related deaths, including insulin intoxications, are difficult to identify and, thus, presumably underdiagnosed. This investigation aims to differentiate insulin intoxication deaths by metabolomics, and identify a metabolic fingerprint to screen for unknown hypoglycemia-related deaths. Ultra-high-performance liquid chromatography-quadrupole time-of-flight mass spectrometry data were obtained from 19 insulin intoxications (hypo), 19 diabetic comas (hyper), and 38 hangings (control). Screening for potentially unknown hypoglycemia-related deaths was performed using 776 random postmortem cases. Data were processed using XCMS and SIMCA. Multivariate modeling revealed group separations between hypo, hyper, and control groups. A metabolic fingerprint for the hypo group was identified, and analyses revealed significant decreases in 12 acylcarnitines, including nine hydroxylated-acylcarnitines. Screening of random postmortem cases identified 46 cases (5.9%) as potentially hypoglycemia-related, including six with unknown causes of death. Autopsy report review revealed plausible hypoglycemia-cause for five unknown cases. Additionally, two diabetic cases were found, with a metformin intoxication and a suspicious but unverified insulin intoxication, respectively. Further studies are required to expand on the potential of postmortem metabolomics as a tool in hypoglycemia-related death investigations, and the future application of screening for potential insulin intoxications.

Postmortem metabolomics as a high-throughput cause-of-death screening tool for human death investigations

“…In addition, postmortem metabolomics in a screening test was able to identify potentially hypoglycemia-related deaths in a large randomly selected cohort. 16 The current study also supports the hypothesis that postmortem metabolomics can effectively discriminate CoD. Herein, using the independent test set of over 1,000 cases, the specificity-optimized model resulted in an average sensitivity of 45% and specificity of 96% for the five CoD groups.…”

“… 19 In the previous oxycodone and insulin intoxication studies, 21 and 12 acylcarnitines, respectively, were observed at lower abundances in intoxication cases compared to controls. 15 , 16 In the current study, acylcarnitines C14, C18:1-OH, and C18:2 were also identified, with the drug intoxication group showing the lowest relative abundance of these, as well as four other acylcarnitines, compared to the other CoD groups. Although the current study identified fewer acylcarnitines that discriminate the drug intoxication groups from the other CoD groups compared to previous studies, this could be attributable to the high degree of heterogeneity within this group.…”

“…Previous postmortem metabolomics investigations have been performed on oxycodone intoxications 15 and insulin intoxications. 16 Moreover, postmortem metabolomics has been utilized in the context of postmortem redistribution of morphine and methadone. 19 In the previous oxycodone and insulin intoxication studies, 21 and 12 acylcarnitines, respectively, were observed at lower abundances in intoxication cases compared to controls.…”

“… version 17; RRID: SCR_014688 R R Foundation of Statistical Computing, Austria version 4.1.2; RRID: SCR_001905 RStudio Posit PBC, MA, USA RRID: SCR_000432 Microsoft Excel Microsoft, WA, USA Microsoft Office 2019; RRID: SCR_016137 MetaboAnalyst McGill University, Canada version 5.0; RRID: SCR_015539 XCMS & CAMERA (customised code) Ward et al. 16 https://www.mdpi.com/2218-1989/13/1/5 Other Agilient 1290 Infinity LC Agilent Technologies Sweden AB, Sweden N/A Agilent 6540 QTOF Agilent Technologies Sweden AB, Sweden N/A Waters Acquity HSS T2 column Waters Sverige AB, Sweden SKU: 186003540 …”

“…Previously, the utility of postmortem metabolomics has been demonstrated by proposing metabolic fingerprints, from femoral blood samples, for various CoD including pneumonia, 14 oxycodone intoxications, 15 and insulin intoxications. 16 Postmortem metabolomics has also identified metabolic fingerprints for hypothermia, using vitreous humor samples, 17 as well as differentiating central-nervous-system-related deaths based on the cerebrospinal fluid metabolome. 18 Postmortem redistribution of exogenous substances has also been investigated using postmortem metabolomics.…”