Mitra™ device for forensic toxicology: A proof of concept study exploring the myth or the reality

Abstract

Aim

Develop an optimized analytical workflow to identify and quantify molecules from 20 μL samples collected using Mitra™ during forensic autopsies.

The use of Volumetric Absorptive Microsampling (VAMS) systems has been previously assessed for both qualitative and quantitative purposes in pharmacology, clinical toxicology, and forensic toxicology. However, the feasibility of a broad, non-targeted toxicological screening including a quantification step has not been reported. Furthermore, the use of such a system by forensic pathologists during autopsies on various matrices has not been described. This study aims to investigate the feasibility of this approach through five real autopsies conducted at the Paris Institute of Forensic Medicine.

Method

The workflow's feasibility was assessed using a blood sample spiked with 100 molecules, collected both with Mitra™ and by conventional methods. After drying for 2 hours at room temperature, the Mitra™ samples were stored at −20 °C, alongside other samples. All Mitra™ samples were desorbed using 300 μL of water. Extraction was performed following previously described procedures for both conventional [1] and Mitra™-collected samples [2]. The testing volume for each matrix was 100 μL for conventional samples and about 7 μL for Mitra™ samples (one-third of the total sample volume). Molecule identification was performed using an LC-HR/MS Orbitrap™ system (Orbitrap™ Exploris 120 ThermoFisher, France) with two complementary chromatographic methods: Accurore™ Pheny Hexyl (RP) and XBridge™ BEH Amide (HILIC). The RP screening used a combined targeted and non-targeted acquisition mode. For quantification, a standard-addition approach was used for molecules identified using 20 μL solutions. For the five autopsy cases, various matrices (cardiac and peripheral blood, urine, vitreous humor, bile) were collected using Mitra™ along with conventional sampling, with four replicates taken for each. All samples were extracted and analyzed according to the same procedure. Additionally, sample stability on Mitra™ was assessed at 24 hours, 1 month, 2 months, and 3 months post-autopsy.

Results

In the blood sample spiked with 100 compounds, toxicological screening using the different analytical approach allowed the identification of 85 molecules using conventional samples and 75 molecules using Mitra™ devices, respectively. Quantification by standard-addition was validated for 10 molecules in blood samples, with biases between −15% and +8% relative to target values. For the five autopsie cases, results obtained from conventional and Mitra™ samples allowed the identification of the same molecules across all matrices. For example, in one real case, the qualitative screening has identified both common psychoactive substances exhibiting broad physicochemical properties as well as novel psychoactive substances like alpha-pyrrolidinohexanophenone. For the quantitative step, comparison of external calibration with standard addition displayed results within a 20% variation between the two methods. Furthermore, whatever the considered matrix, stability was between 88 and 95% for Mitra™ samples analyzed up to 3 months post-collection.

Conclusion

The developed workflow demonstrated that, using a 20 μL Mitra™ device, qualitative results similar to those of conventional screening methods could be achieved. The standard-addition approach proved reliable for quantification, reducing matrix effects and improving precision. This approach was particularly useful to investigate the cause of death in a body-packer case. Noteworthy, only two-thirds of the Mitra™ device was used for qualitative and quantitative analyses in this study, thus leaving room for additional testing on the remaining samples. The stability of Mitra™ samples was confirmed over a 3-month period.

This proof of concept study, conducted on five real autopsy cases, indicates that micro-sampling systems like Mitra™ can be readily used on various matrices in a forensic medicine context. They allow for non-targeted toxicological screenings and reliable quantification, even from low sample volumes.