Insulin preparations, a group of drugs used to treat diabetes, are of great forensic interest because they are difficult to detect in blood while they are sometimes intentionally administered in large doses for suicide or murder. In this study, we validated a quantification method for insulin glargine and insulin lispro using liquid chromatography-tandem mass spectrometry (LC-MS/MS) in the blood sample from a case of suicidal overdose with insulin.

Case: A woman in her 30s had a history of mental illness and diabetes. A day before her death, she informed her partner that she had self-administered large doses of insulin preparations and prescription drugs.

Bovine insulin was used as the internal standard (IS). The blood sample (250μL) was added to the diluent and IS solution, followed by 250μL each of hexane and N,N-dimethylformamide, and the mixture was centrifuged (10000g×10min). The supernatant was collected and 900μL of 5% ammonium solution was added. The extract was analyzed using Nexera 30A, LCMS8060 (Shimadzu), L-column3 ODS C18 metal-free column (150×2.1mm, 3μm), ionization method: electrospray ionization+, multiple reaction monitoring conditions: insulin glargine [M+6H]6+ 1011.5>1179.4, insulin lispro[M+6H]6+ 968.9>217.0, bovine insulin (IS) [M+6H]6+ 956.3>1120.9.

The autopsy revealed no evidence of fatal injuries. Coronary artery stenosis was observed as a finding of endogenous disease, but there were no significant histopathological changes in the myocardium. Biochemical tests revealed the following: blood glucose level, 6mg/dL; glucose in the vitreous region, 6mg/dL; glycated hemoglobin (HbA1c), 11.2%; blood insulin level estimated using the chemiluminescent enzyme immunoassay method, 1.03μU/mL (6.27pmol/mL); and C-peptide concentration, 0.6ng/mL (16.8pmol/L). A drug screening test and simple quantification using LC-MS/MS detected many prescription drugs, but none of them were at intoxicating concentrations.

The insulin analysis showed insulin glargine concentrations of 429μU/mL in femoral blood and 1362μU/mL in cardiac blood, and insulin lispro was detected in femoral and cardiac blood below the lower limit of quantification (<50μU/mL). The cardiac/peripheral blood ratio, a measure of postmortem redistribution, was 3.2.

Although the therapeutic ranges of insulin glargine and insulin lispro have not been defined, the femoral and cardiac blood concentrations of insulin lispro in this case was lower than the serum concentration (155μU/mL) listed in the package insert. However, while the femoral and cardiac blood insulin glargine concentrations were considerably higher than the serum concentration (15μU/mL) listed in the package insert. Considering that insulin is easily degraded in whole blood due to hemolysis, it is possible that insulin concentrations were even higher at the time of administration and death than those observed in autopsy specimens on the third postmortem day. However, since insulin lispro is a fast-acting analog of insulin (1–3hours), it may have contributed to hypoglycemia but was not the direct cause of death, and hypoglycemia due to insulin glargine overdose was the most likely cause of death in this case. We quantified insulin lispro and insulin glargine in a case of suicide to determine insulin glargine's involvement in causing death. In future, it is necessary to examine postmortem degradation and redistribution of insulin and consider the appropriate concentration in postmortem samples of insulin poisoning.