Main content area

Determination of cocaine in postmortem human liver exposed to overdose. Application of an innovative and efficient extraction/clean up procedure and gas chromatography–mass spectrometry analysis

Magalhães, Elisângela Jaqueline, Queiroz, Maria Eliana Lopes Ribeiro de, Penido, Marcus Luiz de Oliveira, Paiva, Marco Antônio Ribeiro, Teodoro, Janaína Aparecida Reis, Augusti, Rodinei, Nascentes, Clésia Cristina
Journal of chromatography 2013 v.1309 pp. 15-21
acetonitrile, cattle, cocaine, gas chromatography-mass spectrometry, humans, ionic strength, liver, overdose, pH, sodium chloride, solvents, temperature
A simple and efficient method was developed for the determination of cocaine in post-mortem samples of human liver via solid–liquid extraction with low temperature partitioning (SLE–LTP) and analysis by gas chromatography coupled to mass spectrometry (GC–MS). The extraction procedure was optimized by evaluating the influence of the following variables: pH of the extract, volume and composition of the extractor solvent, addition of a sorbent material (PSA: primary–secondary amine) and NaCl to clean up and increase the ionic strength of the extract. A bovine liver sample that was free of cocaine was used as a blank for the optimization of the SLE–LTP extraction procedure. The highest recovery was obtained when crushed bovine liver (2g) was treated with 2mL of ultrapure water plus 8mL of acetonitrile at physiological pH (7.4). The results also indicated no need for using PSA and NaCl. The complete analytical procedure was validated for the following figures of merit: selectivity, lower limit of quantification (LLOQ), calibration curve, recovery, precision and accuracy (for within-run and between-run experiments), matrix effect, dilution integrity and stability. The within-run and between-run precision (at four levels) varied from 2.1% to 9.4% and from 4.0% to 17.0%, respectively. A maximum deviation of 11.62% for the within-run and between-run accuracies in relation to the nominal concentrations was observed. Moreover, the LLOQ value for cocaine was 50.0ngg⁻¹ whereas no significant effects were noticed in the assays of dilution integrity and stability. To assess its overall performance, the optimized method was applied to the analysis of eight human liver samples collected from individuals who died due to the abusive consumption of cocaine. Due to the existence of a significant matrix effect, a blank human liver was used to construct a matrix-matched analytical curve. The concentrations of cocaine found in these samples ranged from 333.5 to 5969ngg⁻¹.