BACKGROUND: Oral fluid (OF) is an accepted alternative biological matrix for drug treatment, workplace, and DUID (driving under the influence of drugs) investigations, but establishing the cannabinoid OF detection window and concentration cutoff criteria are important.
METHODS: Cannabinoid concentrations were quantified in OF from chronic, daily cannabis smokers during monitored abstinence. Δ9-tetrahydrocannabinol (THC)3, cannabidiol (CBD), cannabinol (CBN), and 11-nor-9-carboxy-THC (THCCOOH) were determined in daily OF samples collected with the Quantisal™ device. GC-MS limits of quantification (LOQ) were 0.5 μg/L for THC and CBD, 1 μg/L for CBN, and 7.5 ng/L for THCCOOH.
RESULTS: After providing written informed consent for this institutional review board–approved study, 28 participants resided from 4 to 33 days on the secure research unit and provided 577 OF specimens. At the LOQ, THC was generally quantifiable for 48 h, whereas CBD and CBN were detected only at admission. Median THCCOOH detection time was 13 days (CI 6.4–19.6 days). Mean THC detection rates decreased from 89.3% at admission to 17.9% after 48 h, whereas THCCOOH gradually decreased from 89.3% to 64.3% within 4 days. Criteria of THC ≥2 μg/L and THCCOOH ≥20 ng/L reduced detection to <48 h in chronic cannabis smokers. An OF THCCOOH/THC ratio ≤4 ng/μg or presence of CBD or CBN may indicate more recent smoking.
CONCLUSIONS: THC, THCCOOH, CBD, and CBN quantification in confirmatory OF cannabinoid testing is recommended. Inclusion of multiple cannabinoid cutoffs accounted for residual cannabinoid excretion in OF from chronic, daily cannabis smokers and could reduce the potential for positive test results from passive cannabis smoke exposure and lead to greatly improved test interpretation.
Cannabis (marijuana), the most commonly abused illicit drug, was smoked at least once in 2008 by 129–191 million people worldwide (1). During 2009, 16.7 million individuals in the US reported smoking cannabis within the past month (2). Δ9-Tetrahydrocannabinol (THC) was the most commonly detected drug in oral fluid (OF) among drivers testing positive for potentially impairing drugs (3). Of weekend OF samples from nighttime drivers, 6.1% were THC positive in the 2007 National Roadside Survey of Alcohol and Drug Use. The primary psychoactive constituent, THC, is metabolized to 11-hydroxy-THC (11-OH-THC) and 11-nor-9-carboxy-THC (THCCOOH) (4). Further phase 2 metabolic processes include conjugation with glucuronic acid and sulfate, facilitating elimination (5).
OF has gained acceptance over the past decade as an alternative biological matrix for detecting cannabis smoking in testing programs in settings of drug treatment, workplace, pain management, and driving under the influence of drugs (DUID) (6). OF offers advantages including noninvasive and observed specimen collection, which make adulteration more difficult (7–8). Neither specialized collection facilities nor same-sex collectors are necessary. THC in OF derives primarily from deposition into the oral mucosa following ingestion of oral and smoked cannabis, and later, at much lower concentrations, from diffusion from blood (9). Recently, THCCOOH was identified in OF (10,–,12), albeit in low concentrations. This finding is important because THCCOOH results from THC metabolism and is not present in cannabis smoke (13). THCCOOH in OF suggests actual drug use, limiting passive exposure as a source for cannabinoid-positive OF results.
Establishing reliable evidence of recent cannabis exposure is important for drug treatment, workplace, DUID, and accident investigations. Frequently, cannabinoid tests remain positive for extended periods of abstinence in chronic cannabis smokers during continuously monitored abstinence. Extended urinary excretion of 11-OH-THC and THCCOOH for up to 30 days of abstinence in chronic, heavy cannabis smokers has been reported (14–15). In plasma (16) and whole blood (17), THC and THCCOOH are quantifiable for at least 7 days in some chronic cannabis smokers. Establishing cannabinoid OF detection windows, however, will be critical for appropriate interpretation of OF test results.
Few controlled THC administration studies have been performed to define the detection window of cannabinoids in OF, and no investigators have monitored cannabinoid OF excretion in chronic, daily smokers during continuously monitored abstinence. Huestis and Cone (18) described OF THC detection up to 24 h after controlled smoked cannabis administration with a 1.0 μg/L limit of detection for RIA analysis. Niedbala et al. (19) reported a longer detection window of up to 72 h by GC-MS/MS (gas chromatography–tandem mass spectrometry) with a 0.2-μg/L cutoff. Because of its highly lipophilic nature, THC accumulates in tissues, especially with chronic, daily smoking, and is slowly excreted over a much longer time frame (20).
The objectives of this research were to examine THC, THCCOOH, cannabidiol (CBD), and cannabinol (CBN) OF concentrations in chronic, daily cannabis smokers during sustained abstinence, and to establish OF cutoff concentrations for maximizing the diagnostic sensitivity of OF test results, reduce the potential of passive smoke exposure, and account for residual cannabinoid excretion in chronic, daily cannabis smokers.
Materials and Methods
Male cannabis smokers, ages 18–65 years, were recruited for participation in a positron emission tomography imaging study evaluating CB1 cannabinoid–receptor binding after daily cannabis smoking and after approximately 30 days of continuously monitored abstinence. This study also provided an opportunity to establish the window of cannabinoid detection and cannabis OF pharmacokinetics in chronic, daily cannabis smokers. Participants self-reported cannabis smoking for at least 1 year, and had smoked at least 5 days per week for the previous 6 months; urine specimens were positive for cannabinoids on admission. Exclusion criteria included history or presence of clinically significant illness or head trauma with 10 min unconsciousness, recent radiation exposure, and consumption of >6 alcoholic drinks per day more than 4 times per week. Current physical dependence on any substance other than cannabis, nicotine, or caffeine, and interest in or participation in drug abuse treatment within 60 days preceding study entry also were exclusionary. This study was approved by the National Institute on Drug Abuse institutional review board, and participants provided voluntary written informed consent. Participants resided on the Johns Hopkins Behavioral Pharmacology Research Unit under continuous surveillance to ensure cannabis abstinence. Participants and their belongings were searched for drugs on admission, and they were not allowed to leave the unit or receive visitors. Some participants left before 30 days; reasons included family emergencies, homesickness, job offers, and discharge for behavioral issues and protocol noncompliance.
OF SAMPLE COLLECTION
OF was collected with the Quantisal™ device, an absorptive cellulose pad mounted on a polypropylene stem with a volume adequacy indicator. The pad was placed into the participant's mouth until the designated volume [1.0 (0.1) mL] was collected, then put into a plastic tube containing 3 mL buffer, yielding a 1:4 OF dilution. The tube was capped and refrigerated for at least 24 h. Pads were squeezed dry with a serum separator before decanting into a Nunc® cryotube and storage at −20 °C before analysis. OF samples were collected on admission (day 0) and once each 24 h thereafter, until discharge up to 33 days later.
OF samples were analyzed for THC, CBD, CBN, and THCCOOH by use of a previously published 2-dimensional GC-MS (2D-GC-MS) method that uses 2 analytical systems with different ionization techniques (12). Briefly, deuterated internal standards were added to 1-mL calibrators and QC samples [0.25 mL blank authentic OF, 0.75 mL Quantisal buffer mixture (Immunalysis Corporation), and 1 mL participant samples (0.25 mL authentic OF mixed with 0.75 mL Quantisal buffer], followed by 1 mL cold acetonitrile. After mixing and centrifugation, solid-phase extraction was performed. THC, CBD, and CBN were eluted with 3 mL of a mixture of 60 mL hexane, 30 mL acetone, and 20 mL ethyl acetate, followed by THCCOOH elution into separate tubes with 3 mL of a mixture of 75 mL hexane, 25 mL ethyl acetate, and 2.5 mL glacial acetic acid. Eluates were evaporated to dryness under nitrogen and derivatized at 65 °C for 40 min with 20 μL of N,O-bis (trimethylsilyl) trifluoroacetamide + 1% trimethylchlorosilane for THC, CBD, and CBN, and 20 μL of 1,1,1,3,3,3-hexafluoro-2-propanol and 40 μL of trifluoroacetic anhydride for THCCOOH. Limits of quantification (LOQ) were 0.5 μg/L for THC and CBD, 1 μg/L for CBN (electron impact 2D-GC-MS), and 7.5 ng/L for THCCOOH (negative chemical ionization 2D-GC-MS). Calibration curves had r2 > 0.990, and each calibrator quantified within ±15% of target, except at the LOQ (±20%), compared to the full calibration curve. Intraassay imprecision was 2.2%–6.6%, and interassay imprecision was <5.2%. Analytical recovery was within 13.8% of target.
Statistical calculations were performed by use of SPSS 18.0 (SPSS). Normal distribution of data was tested with the Shapiro-Wilk test and normal Q-Q plot. Comparative analyses were done with Wilcoxon signed-rank tests. Correlations (r) were determined with regression and ANOVA. The relationship was considered significant for P < 0.05. Survival analysis employed the Kaplan-Meier method; the conditional probability of being positive at each time point was estimated by the product of the probability of being positive at time t and the probability of being positive beyond time t among those positive at time t. Concentrations below the LOQ were considered to be 0 for statistical analyses and median calculations. Body mass index (BMI) approximates total body fat, calculated as BMI = 703 × [weight (lb)/height2 (in2)]. Data are presented as mean (SD) unless specified otherwise.
Twenty-eight male chronic, daily cannabis smokers (ages 19–43 years) provided 577 OF samples during monitored abstinence over 4–33 days (Table 1); 82% of participants reported smoking at least 13 of the last 14 days. Participants reported smoking, on average, 10.6 (6.3) joints or blunts (range 1–30) per day for 10.6 (5.8) (range 4–28) years. Participant BMI ranged from 16.4 to 32.8.
Of 577 OF specimens, THC was greater than the LOQ (0.5 μg/L) in only 48 samples (8.3%), whereas THCCOOH was detected (LOQ 7.5 ng/L) in 212 (36.7%). THC and THCCOOH were quantifiable in 89% participants' specimens at admission; 3 were negative for both THC and THCCOOH. In contrast, CBD and CBN were positive only at admission in the OF of 5 and 14 study participants, respectively, except in 1 case discussed below. THC was present in the highest concentrations (1.0–204.6 μg/L) at admission in all but 2 study participants, whose maximum concentrations, of 3.0 and 2.3 μg/L, occurred on days 21 and 28, respectively. All OF samples from another 2 study participants were THC and THCCOOH negative for 14 and 19 days, respectively. In 24 of 28 study participants, THC was not detected beyond 48 h, always with concurrent THCCOOH. However, in 4 study participants, THC was present in 9 of 106 (7.5%) OF samples during 4–28 days abstinence. In these samples, THC concentrations were ≤3.0 μg/L after 24 h, except in 1 case. In 5 of the 9 samples, there was no measureable THCCOOH. Maximum THCCOOH concentrations (10.1–255.6 ng/L) also occurred within 48 h, except in 3 of 28 study participants whose maximum concentrations occurred after 3, 4, and 7 days of abstinence. There was 1 OF sample collected 18 days after admission that contained 13.2 μg/L THC, 3.5 μg/L CBN, and 17.0 ng/L THCCOOH. Despite all security measures, including searches, continuous staff presence, and no visitors, illicit use while on the research unit was suspected. This sample was excluded from all calculations related to detection windows.
Large intersubject variability is illustrated in Fig. 1, documenting THC and THCCOOH OF concentration time courses in 4 study participants who were THC negative within 24 h (Fig. 1, A and B) and in another 4 study participants who produced sporadic THC-positive samples after 24 h (Fig. 1, C and D).
As expected, daily OF THC detection rates decreased from 89.3% on admission to 32.1% at 24 h and 17.9% at 48 h after abstinence initiation. Median THC OF concentration after 24 h abstinence (0.0 μg/L) was significantly less than median THC OF concentration at admission (9.3 μg/L; P < 0.001). In contrast, THCCOOH OF excretion was prolonged. Daily THCCOOH detection rates slowly decreased from 89.3% to 78.6% to 64.3% at admission and after 48 and 96 h of abstinence, respectively. Daily THCCOOH concentrations significantly decreased from admission to 24 h (P < 0.001), with a median percent change of −52.2%. However, over the maximum 33 days abstinence period, THCCOOH OF concentrations often increased; 75% of chronic, daily cannabis smokers had at least 1 increase during abstinence. For all samples collected during the first, second, third, and fourth weeks of abstinence, 62.8%, 32.2%, 19.5%, and 8.2% of samples contained THCCOOH concentrations ≥7.5 ng/L. For 18 samples collected on days 29–33, THCCOOH was detected in only 1 sample, with no sample positive after day 29. Table 2 summarizes OF cannabinoid concentrations for all study participants.
At admission, THC concentrations were significantly correlated to CBD (r = 0.645; P < 0.001) and CBN (r = 0.963; P < 0.001) concentrations, but not to THCCOOH concentrations (r = 0.255; P = 0.190). Correspondingly, correlations between THCCOOH and CBD (r = 0.250; P = 0.199) or CBN (r = 0.301; P = 0.120) were not significant. After 24 h of monitored abstinence, the OF THC and THCCOOH correlation became significant (r = 0.428; P = 0.023).
THC and THCCOOH OF detection windows were determined by Kaplan-Meier survival analysis to account for people who withdrew before achieving negative cannabinoid OF results (Fig. 2). Of 28 study participants, 6 had sporadically positive THCCOOH results interspersed with up to 4 negative results. No OF sample was positive after 5 consecutive negative OF THCCOOH results, thus indicating the last THCCOOH detection time. If participants withdrew before meeting this criterion, their data were censored. Participants with all negative results from admission were excluded. Median THC and THCCOOH detection windows were 24 h (95% CI 4.8–43.2 h) and 13 days (95% CI 6.4–19.6 days), respectively. Correlations between THC and THCCOOH last detection times and potential effect modifiers, including BMI, number of joints or blunts smoked per day, number of days smoked in the last 14 days, and lifetime smoking years, were not significant (P > 0.05). Two study participants who were THC negative from admission and 14 study participants who did not satisfy the THCCOOH last detection time criterion were excluded from THC and THCCOOH analyses, respectively.
THCCOOH/THC ratios (ng/μg) were monitored over 48 h of abstinence. Further evaluation was not possible owing to the limited number of THC-positive samples after this time. At admission, the median THCCOOH/THC ratio was 3.1 (n = 25; range 0.4–48.7), increasing to 43.2 (n = 9; range 5.0–144.9) after 24 h abstinence, and 67.4 (n = 5; range 24.0–155.5) after 48 h.
Cannabinoid OF cutoffs higher than our method LOQ were proposed by the Substance Abuse and Mental Health Services Administration (SAMHSA) and used in the Driving under the Influence of Drugs, Alcohol and Medicines (DRUID) European Union program. At the 2 μg/L SAMHSA THC OF confirmation cutoff, only 5.2% of our samples were THC-positive, with 82% of chronic, daily cannabis smokers' OF positive on admission. With the lower DRUID confirmation cutoff (1 μg/L), we identified 11 additional positive samples (7.1%), with 89% THC-positive at admission. No sample was THC positive thereafter, except for the 9 sporadic positive samples mentioned earlier. With the SAMHSA THC cutoff, 2 of 28 study participants were THC positive for at least 24 h and none thereafter, and with the DRUID cutoff, 6 of 28 study participants were THC positive at least 24 h, and only 1 at least 48 h.
With our low LOQ, cannabinoid OF detection windows were large in this population of chronic, daily cannabis smokers. Seven alternative cutoffs were evaluated (Fig. 3). Of the 7 cutoffs, 4 decreased the detection window substantially and increased identification of recent cannabis smoking: (a) THC ≥2 μg/L + THCCOOH ≥20 ng/L; (b) THCCOOH/THC ratio ≤4 ng/μg; (c) THC ≥2 μg/L + CBD ≥0.5 μg/L; and (d) THC ≥2 μg/L + CBN ≥1.0 μg/L.
This report is the first to present THC, CBD, CBN, and THCCOOH concentrations measured in OF from chronic, daily cannabis smokers during extended observed abstinence. Generally, THC was negative by 48 h of abstinence, whereas CBD and CBN were detected only at admission. In contrast, THCCOOH detection times were much longer, up to 29 days, with a median of 13 days. We observed large interindividual variation in cannabinoid OF concentrations and detection times. This variation appeared to be influenced primarily by magnitude and time since last cannabis smoking, and cannabinoid body stores after chronic smoking (9). Minor contributors could include OF flow rate and pH (21–22), as well as THC metabolism variations (23,–,25).
At admission, THC concentrations were significantly correlated with CBD and CBN concentrations, likely owing to the fact that all 3 analytes are present in cannabis smoke, and most of the cannabinoids present immediately after cannabis smoking are due to oral mucosal contamination. However, concentrations of THCCOOH, which is not present in cannabis smoke, did not correlate with concentrations of other cannabinoids. After 24 h abstinence, when oral mucosal contamination cleared, THC and THCCOOH concentrations were significantly correlated, suggesting equilibrium with blood concentrations. Results of several studies substantiate a temporal correlation between OF and blood THC concentrations after the initial high OF concentrations dissipate (18, 26,–,28). These data support oral mucosal contamination during smoking rather than tissue storage as the main THC source close to the time of smoking (29). Occasional OF samples with low THC-positive concentrations unaccompanied by THCCOOH after multiple consecutive negative results most likely reflected release of residual THC stores, as was recently reported to occur in whole blood (17), plasma (16), and urine (14), and now also in OF.
In 21 study participants, maximum THC concentrations were observed on admission, with decreases over time. Within 24 h, only 9 study participants' OF samples were still positive, by 48 h, only 5 were positive, and by 72 h, none were positive. Of the samples from the remaining 7 study participants, OF samples from study participants M and Y, who reported smoking fewer joints or blunts per day than the group mean, were all negative, although the urine samples from these individuals were positive on admission. Participant F was THC positive at admission and 48 h later, but not at 24 h; OF THCCOOH was positive until day 11. Likewise, participant X was THC positive at admission and day 4 but negative in between; however, his OF was THCCOOH negative after 24 h. Participant U was negative for THC at admission and all for all samples except for low concentrations on days 25 and 28. As the residential unit was secure with 24 h surveillance, and all participants and belongings were searched before entry, these results are not easily explained other than as residual drug excretion of chronic users. Similarly, participant V had 6 positive results with concentrations ranging from 1.1 to 3.0 μg/L on days 0, 12, 15, 21, 24, and 28, interspersed with negative results. For study participants V and X, corresponding expectorated OF and/or blood specimens showed a similar trend (data not shown). All specimens with positive results after several negative specimens were reanalyzed and the results were found to be reproducible. Furthermore, 9 of 28 study participants had 1–4 negative OF THCCOOH results interspersed with positives. These sporadic positives could be due to a variable rate and amount of THC released from storage, metabolism to THCCOOH, and cannabinoid diffusion from blood into OF. Participant Q was THC positive at admission and suddenly positive on day 18 after 17 consecutive negative specimens. Expectorated OF specimens also were collected, and these samples also were positive at admission and on day 18 with negative specimens in between. Because this individual was also positive for CBN and THCCOOH with concentrations too high to indicate residue from intake before the admission, we suspect new cannabis intake, although we do not know how this could have occurred.
In other reported studies, OF THC concentrations >1000 μg/L were documented in study participants immediately after they had smoked (18, 30). Our THC concentrations after at least 24 h abstinence (range 0.5–16.8 μg/L) were not much lower than THC concentrations detected 2 h after a single smoked dose, 3.5 μg/L (18), 0.3–10 μg/L (19), and 9–13 μg/L (31). Kauert et al. (32) also observed a sharp decline during the first 1–2 h after smoking, followed by a slower decrease. These observations suggest that a few hours after cannabis smoking, OF THC concentrations drop to values at which recent use and residual excretion are no longer distinguishable in chronic cannabis smokers. On the other hand, another potential biomarker for recent cannabis exposure, THCCOOH, is problematic because of its prolonged excretion. Dietz et al. (33) reported that even cannabis-naïve individuals excreted THCCOOH in urine for up to 4 days after a single 5-mg intravenous THC dose. Our data documented a significant THCCOOH decrease within 24 h of abstinence, but after that time, gradual decreases with detection for up to 29 days in these chronic, daily cannabis smokers. This finding highlights the importance of detecting multiple cannabinoids, rather than THC alone, to identify recent use. Thus, it is worth noting that 6 of 9 sporadically THC-positive samples after 48 h were negative for THCCOOH; in the remaining 3 sporadically THC-positive samples, THCCOOH was <20 ng/L. Combining a THCCOOH cutoff of 20 ng/L with the current SAMHSA THC cutoff of 2 μg/L decreased the cannabinoid last detection time to 24 h, yielding a maximum detection window of 48 h owing to once a day sampling. We also documented that 15 of 23 SAMHSA THC-positive participants at admission had THCCOOH/ THC ratios between 0 and 4.0 ng/μg at the method's LOQs. Such a ratio occurred in no other samples after admission except the 1 case for which new intake was suspected. The impact of THCCOOH/THC ratios awaits further research.
One limitation of the present study was that the exact times and amounts of last cannabis smoking were unknown, except for self report of the study participants. However, this limitation does not change our conclusions that in chronic cannabis smokers at the method's low analytical LOQ THC can be detected in OF for at least 48 h, with a few occasional positives up to 28 days, and THCCOOH can be detected for at least 29 days. Furthermore, these data document that detection of cannabinoids in OF from chronic cannabis smokers may not reflect recent use. Neurocognitive impairment in chronic cannabis users was demonstrated on some measures for at least 7 (34), and up to 28 days of abstinence (35). Further research is needed to determine if cognitive and performance impairment persists during sustained abstinence in chronic, daily cannabis smokers, and whether this impairment is mirrored in detection windows of THC and THCCOOH in OF. Another limitation of the present study was the rather homogenous demographics of the participants, consisting of all males and 89% African Americans.
In the present study, OF cutoffs of ≥2 μg/L THC and ≥20 ng/L THCCOOH restricted the cannabis detection window in OF from chronic cannabis smokers to within 48 h. These cutoffs eliminated the low residual THC OF concentrations that occurred randomly upon release of THC from tissues stores far into sustained abstinence, a finding similar to recently reported observations in other biological matrices (36). In addition, the presence of THCCOOH at or above these concentrations minimizes the possibility of passive contamination, because THCCOOH is not present in cannabis smoke. A THCCOOH/THC ratio (ng/μg) of 0–4.0 or the presence of CBD and CBN observed only at admission may indicate more recent cannabis smoking. High intersubject variability and possible residual THC and THCCOOH excretion into OF following chronic cannabis smoking argue against the use of THC concentration alone to determine recent cannabis use. Confirmation of multiple cannabinoids in OF would improve test interpretation in treatment, workplace, DUID, and pain-management programs.
We acknowledge the clinical staff of the National Institute on Drug Abuse, Intramural Research Program and the Behavioral Pharmacology Research Unit.
This manuscript was previously presented at the American Academy of Forensic Sciences 63rd Annual Scientific Meeting, Chicago, IL, February 2011.
↵3 Nonstandard abbreviations:
- oral fluid;
- 11-nor-9-carboxy THC;
- driving under the influence of drugs;
- 2-dimensional GC-MS;
- limit of quantification;
- body mass index;
- Substance Abuse and Mental Health Services Administration;
- Driving under the Influence of Drugs, Alcohol and Medicines.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest:
Employment or Leadership: None declared.
Consultant or Advisory Role: None declared.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: Intramural Research Program of the National Institute on Drug Abuse, National Institutes of Health.
Expert Testimony: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
- Received for publication February 28, 2011.
- Accepted for publication May 16, 2011.
- © 2011 The American Association for Clinical Chemistry