Condensation of marijuana smoke on the hair surface can be a source of an external contamination in hair analysis and may have serious consequences for the person under investigation. Δ9-tetrahydrocannabinolic acid A (THCA-A) is found in marijuana smoke and in hair analysis, but is not incorporated into the hair through the bloodstream. Therefore it might be a promising marker for external contamination of hair and could facilitate a more accurate interpretation of analytical results. In this study, three participants were exposed to the smoke of one joint every weekday over three weeks. Inhalation was excluded by an alternative breathing source. Hair samples were obtained up to seven weeks after the last exposure and analyzed for THCA-A, Δ9-tetrahydrocannabinol (THC) and cannabinol (CBN) by liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. Additionally 30 hair samples from various regions of the head were obtained seven weeks after the exposure from one participant. The obtained results show that the degree of contamination depends on the hair length, with longer hair resulting in higher THC and CBN concentrations (1300 pg/mg and 530 pg/mg at the end of the exposure period) similar to the ones typically found after daily cannabis consumption. THCA-A could be detected in relatively low concentrations. Analysis of the distribution of the contamination showed that the posterior vertex region was affected most. The relatively low THCA-A concentrations in the samples suggest that most of the THCA-A found in forensic hair samples is not caused by sidestream marijuana smoke, but by other sources.

Hair analysis for cannabinoids is extensively applied in workplace drug testing and in child protection cases, although valid data on incorporation of the main analytical targets, ∆9-tetrahydrocannabinol (THC) and 11-nor-9-carboxy-THC (THC-COOH), into human hair is widely missing. Furthermore, ∆9-tetrahydrocannabinolic acid A (THCA-A), the biogenetic precursor of THC, is found in the hair of persons who solely handled cannabis material. In the light of the serious consequences of positive test results the mechanisms of drug incorporation into hair urgently need scientific evaluation. Here we show that neither THC nor THCA-A are incorporated into human hair in relevant amounts after systemic uptake. THC-COOH, which is considered an incontestable proof of THC uptake according to the current scientific doctrine, was found in hair, but was also present in older hair segments, which already grew before the oral THC intake and in sebum/sweat samples. Our studies show that all three cannabinoids can be present in hair of non-consuming individuals because of transfer through cannabis consumers, via their hands, their sebum/sweat, or cannabis smoke. This is of concern for e.g. child-custody cases as cannabinoid findings in a child’s hair may be caused by close contact to cannabis consumers rather than by inhalation of side-stream smoke.

For analysis of hair samples derived from a pilot study ('in vivo' contamination of hair by sidestream marijuana smoke), an LC-MS/MS method was developed and validated for the simultaneous quantification of Δ9-tetrahydrocannabinolic acid A (THCA-A), Δ9-tetrahydrocannabinol (THC), cannabinol (CBN) and cannabidiol (CBD). Hair samples were extracted in methanol for 4 h under occasional shaking at room temperature, after adding THC-D(3), CBN-D(3), CBD-D(3) and THCA-A-D(3) as an in-house synthesized internal standard. The analytes were separated by gradient elution on a Luna C18 column using 0.1% HCOOH and ACN + 0.1% HCOOH. Data acquisition was performed on a QTrap 4000 in electrospray ionization-multi reaction monitoring mode. Validation was carried out according to the guidelines of the German Society of Toxicological and Forensic Chemistry (GTFCh). Limit of detection and lower limit of quantification were 2.5 pg/mg for THCA-A and 20 pg/mg for THC, CBN and CBD. A linear calibration model was applicable for all analytes over a range of 2.5 pg/mg or 20 pg/mg to 1000 pg/mg, using a weighting factor 1/x. Selectivity was shown for 12 blank hair samples from different sources. Accuracy and precision data were within the required limits for all analytes (bias between -0.2% and 6.4%, RSD between 3.7% and 11.5%). The dried hair extracts were stable over a time period of one to five days in the dark at room temperature. Processed sample stability (maximum decrease of analyte peak area below 25%) was considerably enhanced by adding 0.25% lecithin (w/v) in ACN + 0.1% HCOOH for reconstitution. Extraction efficiency for CBD was generally very low using methanol extraction. Hence, for effective extraction of CBD alkaline hydrolysis is recommended.