Aims: Alcohol abuse is a major risk factor for premature death. Confirming the role of alcohol consumption in cause-of-death investigations has, however, remained difficult, due to lack of reliable biomarkers.
Methods: We compared ethyl glucuronide (EtG) and carbohydrate-deficient transferrin (CDT) assays from serum, urine, cerebrospinal fluid and vitreous humor in a forensic autopsy population with either a positive (n = 38) or negative (n = 22) history of alcohol abuse based on detailed medical and police records and forensic toxicological investigations.
Results: A positive blood alcohol concentration (median 1.15‰, range 0-3.3‰) was found in 26/38 (68%) of the cases with a documented history of alcohol abuse. EtG concentrations (mean ± SD) in urine (339 ± 389 mg/l, P < 0.001), vitreous humor (4.2 ± 4.8 mg/l, P < 0.001), serum (6.9 ± 8.9 mg/l, P < 0.01) and cerebrospinal fluid (1.7 ± 2.7 mg/l, P < 0.01) were significantly higher among the cases with a positive history of alcohol use than those in the alcohol-history negative group, whereas in corresponding comparisons CDT was significantly different only in cerebrospinal fluid (4.3 ± 2.1 vs. 2.3 ± 0.6%, P < 0.05). The highest sensitivities (92%) in detecting ante-mortem alcohol use were obtained for urine and vitreous humor EtG assays.
Conclusion: Our data indicate that measurements of EtG in urine or vitreous humor show the highest diagnostic accuracies in post-mortem investigations of excessive alcohol consumption and can be recommended for routine applications.