Possibilities for discrimination between chewing of coca leaves and abuse of cocaine by hair analysis including hygrine, cuscohygrine, cinnamoylcocaine and cocaine metabolite/cocaine ratios.

Abstract

Contrary to the illegal use of any form of manufactured cocaine, chewing of coca leaves and drinking of coca tea are allowed and are very common and socially integrated in several South American countries. Because of this different legal state, an analytical method for discrimination between use of coca leaves and abuse of processed cocaine preparations is required. In this study, the applicability of hair analysis for this purpose was examined. Hair samples from 26 Argentinean coca chewers and 22 German cocaine users were analysed for cocaine (COC), norcocaine (NC), benzoylecgonine (BE), ecgonine methyl ester (EME), cocaethylene (CE), cinnamoylcocaine (CIN), tropacocaine (TRO), cuscohygrine (CUS) and hygrine (HYG) by hydrophilic interaction liquid chromatography (HILIC) in combination with triplequad mass spectrometry (MS/MS) and hybrid quadrupole time-of-flight mass spectrometry (QTOF-MS). The following concentrations (range, median, ng/mg) were determined in hair of the coca chewers: COC 0.085-75.5, 17.0; NC 0.03-1.15, 0.12; BE 0.046-35.5, 6.1; EME 0.014-6.0, 0.66; CE 0.00-13.8, 0.38; CIN 0.005-16.8, 0.79; TRO 0.02-0.16, 0.023; CUS 0.026-26.7, 0.31. In lack of a reference substance, only qualitative data were obtained for HYG, and two metabolites of CUS were detected which were not found in hair of the cocaine users. For interpretation, the concentrations of the metabolites and of the coca alkaloids in relation to cocaine were statistically compared between coca chewers and cocaine users. By analysis of variance (ANOVA) significant differences were found for all analytes (α = 0.000 to 0.030) with the exception of TRO (α = 0.218). The ratios CUS/COC, CIN/COC and EME/COC appeared to be the most suitable criteria for discrimination between both groups with the means and medians 5-fold to 10-fold higher for coca chewers and a low overlap of the ranges between both groups. The same was qualitatively found for HYG. However, these criteria cannot exclude cocaine use in addition to coca chewing. In this regard screening for typical cutting agents can be helpful and led to the detection of levamisole (21×), lidocaine (6×) and paracetamol (3×) in the 22 samples from German cocaine users, whereas no levamisole, lidocaine (3×) and paracetamol (1×) were found in hair from the Argentinean coca chewers. These criteria have to be confirmed for South American cocaine consumers including smokers of coca paste and may be different because of different composition of the drug and other use habits.