Alternative specimens have been occasionally considered as substitutes for whole blood for postmortem toxicology testing. We studied the applicability of vitreous humor, and evaluated whether it would be suitable to replace (or augment) whole blood for routine drug screening. Results showed that from 51 autopsy cases, we were able to identify an aggregate of 209 findings in whole blood compared with 169 in vitreous. The total number of compounds identified was 71 for whole blood and 60 for vitreous humor. Quantitative analysis showed that whole-blood concentrations of trazodone were several fold higher than vitreous humor concentrations (1.42 ± 0.57 vs. 0.15 ± 0.05 mg/L, respectively) and similar results were also obtained for diazepam (0.37 ± 0.06 vs. 0.13 ± 0.01, respectively). For other drugs such as oxycodone, hydrocodone and doxylamine, a trend suggesting higher concentrations in vitreous humor vs. whole blood was observed; however, this was not significant. Our results are consistent with the limited work of other investigators, and suggest that vitreous humor could be an appropriate matrix for drug screening in postmortem toxicology.

Quantitative serum alcohol concentrations from regional hospitals (from specimens collected at time of hospital admission) were compared to results from whole blood (from specimens collected at the time of hospital admission) concentrations measured at the San Diego County Medical Examiner's Office (SDCMEO). Over a 15 month period (January 2012 to March 2013), the postmortem forensic toxicology laboratory analyzed a total of 2,321 cases. Of these, 280 were hospital cases (antemortem) representing 12% of the overall Medical Examiner toxicology casework. 59 of the 280 hospital cases (or 21%) screened positive for alcohol (ethanol). 39 of these 59 cases were included in the study based on available specimens for quantitative analyses. This investigation indicated that serum hospital ethanol concentrations correlated well (R(2) = 0.942) with ethanol values determined at SDCMEO (generally measured in whole blood). There was an observed negative bias with an average of -14.1%. A paired t-test was applied to the data and it was shown that this observed bias is statistically significant. These differences in ethanol concentrations could result from differences in specimen, analytical techniques, and/or calibration. The potential for specimen contamination is also discussed.

Vitreous humor is a potential alternative matrix for postmortem toxicology drug screens when peripheral blood is unavailable. It is easily and reliably collected and may not suffer from the same postmortem redistribution as seen in blood. Here, we compared the concentrations of 7 acidic drugs (acetaminophen, ibuprofen, naproxen, salicylic acid, carbamazepine, phenobarbital and phenytoin) in peripheral blood and vitreous fluid collected in 89 autopsy cases. Analysis was done by high-performance liquid chromatography with diode-array detection. Overall, we found that vitreous drug concentrations were significantly lower than peripheral blood with median vitreous to peripheral blood (V/PB) ratios ranging from 0.0 to 0.6 (mean, 0.1-0.6). The correlations between the concentrations of over-the-counter analgesics in peripheral blood versus vitreous fluid were poor, with acetaminophen exhibiting the best linearity (R2 = 0.72). The antiepileptic drugs (carbamazepine, phenytoin and phenobarbital) exhibited good correlations between peripheral blood and vitreous humor, with all exhibiting an R2 ≥ 0.95. Overall, we have demonstrated the potential of vitreous fluid as an alternative matrix for the detection of select acidic drugs.