Approaches towards the quantification of caffeine and metabolites in fingerprints

• Main Author: Watkinson, Shelley N.
• Other Authors: Bailey, Melanie ; Reddy, Subrayal ; Stevenson, Derek
• Published: University of Surrey 2018
• Subjects: 540
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.736932

A fingerprint offers a convenient matrix for drug testing as samples can be deposited rapidly and securely. In this thesis, different strategies for quantitative analysis of fingerprint residues were explored. The variability in mass of a deposited fingerprint measured by a quartz crystal microbalance (QCM) was found to be up to 100% for multiple donors. Fingerprint depositions could be controlled to 21% for a single donor by washing hands, controlling the deposition pressure and wait time. With the aim of reducing the variability of analyte signals detected, the following approaches were tested for feasibility: (a) Normalising the intensity of endogenous compounds to the measured mass (b) Normalising the intensity of caffeine and its metabolites to endogenous compounds detected in multiple fingerprints using liquid chromatography-mass spectrometry (LC-MS) and in single fingerprints using liquid extraction surface analysis-mass spectrometry (LESA-MS) (c) Developing fingerprints deposited on paper using ninhydrin and normalising to the signal intensity of Ruhemann’s purple, and to a “photo scaling” method developed in this thesis. (d) Normalisation to a “fingerprint reader” developed by Intelligent Fingerprinting Limited, which gives a signal relating to the amount of fingerprint deposited on a glass slide. Using approach (a) there was no clear relationship between the mass of fingerprint measured and the intensity of endogenous compounds measured by LC-MS. Using approach (b) it was possible to reduce (or at least maintain) the variability in replicate fingerprint depositions to below 20% as required of a quantitative method. Using approach (c) normalisation to the mean grey value and integrated density value improved variation in theobromine to below 20% in all samples. Finally using approach (d) normalisation to the “fingerprint reader” reduced the variability below 20%. The thesis therefore concludes that either approach (b), (c), or (d) could be explored further in future work for quantification of fingerprint residues.