Many governmental and private organizations in the United States employ drug testing as part of their drug use-prevention programs. Urine is the biological matrix most commonly tested to identify individuals who use drugs. In 2003 it was estimated that more than 20 million urine specimens were collected for drug testing in United

States programs. Drug testing is also an important objective outcome measure of drug treatment, drug research investigating efficacy of new behavioral therapies, criminal justice, military programs, and emergency, pediatric, and geriatric medicine. A common example is judicial programs that routinely collect urine from individuals on parole. Individuals committing crimes and having a positive urine drug test may be placed in treatment while on parole if the judge believes that drug use contributed to the crime. Parolees are ordered to attend a rehabilitation program, are given a short period of time to eliminate previously self-administered drugs from their bodies, and, as a condition of continued parole, must discontinue use of prohibited drugs. To ensure compliance, treatment managers routinely have the parolee donate urine specimens, and if there is a positive urine test indicating new drug use, the donor may be sent to prison. This example sets the stage for an important social question. If a parolee who was a chronic marijuana user had a sequential set of urine tests during his first week of rehabilitation with decreasing concentrations of THCCOOH from 1000 ng/mL down to 100 ng/mL by the end of the week, and then donated a urine specimen with a concentration of 150 ng/mL, does this increase in urine concentration indicate new use in violation of his parole?

Figure 4 shows a typical urinary excretion profile for THCCOOH in an infrequent marijuana user following smoking of a single marijuana cigarette. As mentioned previously, there is great inter- and intrasubject variability in the urinary excretion of cannabinoids. Many investigators have published studies showing that in a sequential series of urine specimens from individuals who abstained from smoking cannabis, there can occasionally be urine specimens that have higher concentrations of THC-COOH than previous samples (89-91). This could be a result of residual excretion of drug that has been stored in the body following chronic cannabinoid use. Most of these increases in concentration appear to be related to individuals' hydration states that are determined by fluid intake, environmental temperature, levels of activity, disease states, and a multitude of other variables. Urine may be diluted and drug concentrations reduced as a result of ordinary variations in daily activity or purposeful attempts to adulterate the sample by specimen dilution, achieved by simply drinking large quantities of fluid. In controlled studies of cocaine and cannabinoid administration followed by consumption of different amounts of liquids, investigators were able to demonstrate large reductions in urine drug concentrations. In many cases, results fell below cutoff concentrations for a positive test (92).

Manno et al. first suggested that urinary THCCOOH could be normalized to urinary creatinine concentration to account for specimen dilution (91). They recommended a quotient cutoff of >1.5 to identify new drug use. Huestis and Cone addressed this problem by examining more than 1800 urine specimens collected following controlled THC administration (89). They found that the greatest accuracy (85.4%) in predicting new cannabis use occurred when paired specimens collected at least 24 hours apart had a quotient of >0.5 for the [THCCOOH]/[creatinine] in specimen 2 divided by the [THCCOOH]/[creatinine] for specimen 1. If the 1.5 ratio was used, as proposed by Manno, almost 30% of the cases of new drug exposure would be missed. Figure 4 shows that normalizing the THCCOOH concentration to creatinine concentrations makes the excretion pattern more predictable, i.e., it has fewer abrupt changes in the exponential decrease.

The Huestis and Cone study examined infrequent cannabis users and did not address excretion patterns that one would expect from chronic use. As mentioned, chronic users take longer than infrequent users to eliminate marijuana metabolites. This is a result of the disposition of THC into poorly perfused tissues such as fat. With chronic cannabis use, THC concentrations in these poorly perfused compartments increase, forming less accessible depots of THC in the body. Hunt and Jones demonstrated that the slow return of THC from these depots into the plasma was the rate-limiting step in the terminal elimination of THC from the body (36). Fraser and Worth studied a group of 26 chronic marijuana users, testing both the Manno and Huestis criteria for new use and had a false-negative rate of 7.4% with the Huestis guideline and 24% with the Manno rule (93). They extended the study to include 37 chronic marijuana users with at least 48 hours between specimens; with the >0.5 cutoff, new drug use was identified in 80-85% of cases (94). Of course, the smaller the ratio used, the greater the potential for false-positive results. The reasons for conducting the urine test, i.e., treatment or parole, and the impact of the results on the donor guide the choice of which ratio to apply.

Based on this valuable scientific information, we can answer the question about whether the individual on parole in our example had smoked marijuana between donating the specimen containing 100 ng/mL THCCOOH and the specimen with 150 ng/mL THCCOOH. The answer is that we cannot tell if he used cannabis in violation of his conditions for parole. Additional information is needed to differentiate between new cannabis use and residual drug excretion. This spike in urine concentration would not be unusual for an individual who had complied with his treatment protocol. If the treatment center had collected the specimens at least 24 hours apart and had measured creatinine concentrations, we would have additional information to provide a more definitive answer. If the outcome of the evaluation could be used to place the individual, who was a former chronic cannabis user, in prison for continuing use after entering his rehabilitation program, the higher ratio of 1.5 might be a better choice for evaluating his urine tests. This would achieve better specificity, rather than sensitivity. In addition, more frequent monitoring may be useful if urine specimens are being collected more than 48 hours apart.

7.3. Oral Fluid

Oral fluid is composed of saliva and secretions from the nasopharyngeal area and mouth. Mechanisms of drug entry into oral fluid are not fully understood. Scientists have determined that passive diffusion from blood and tissue depots and direct entry into oral fluid following smoked, oral, sublingual, or snorted routes of drug administration are the primary sources. In rare cases (e.g., lithium), active transport mechanisms also may contribute. Some of the factors affecting how much drug enters oral fluid from the blood are the lipophilicity of the drug, the degree of plasma protein binding, the drug's pKa, and pH differences between blood and oral fluid. In general, if the drug is not extensively bound to plasma proteins, is lipophilic, and is present in an unionized state, passive diffusion is the primary mechanism for drug entry into oral fluid. The lower pH in oral fluid as compared with blood can result in ion trapping of drugs with a higher pK (e.g., codeine), which has concentrations three to four times

Pka Difference Needed For Ion Trapping
Fig. 7. General drug effects and detection time ranges in various matrices following occasional cannabinoid use. (Personal communication from Edward J. Cone, PhD.)

higher in oral fluid (95). In general, detection times for drugs in oral fluid range from a few hours to 1 or 2 days following use (see Fig. 7).

There are few data on the disposition of cannabinoids in oral fluid following controlled cannabis administration. Scientists have known that THC is present in oral fluid since the 1970s (96,97), and in the 1980s Gross et al. found that they could detect THC in saliva with RIA for 2-5 hours in 35 subjects who smoked one marijuana cigarette containing 27 mg THC (98). However, the specificity of this assay was low, with frequent false-positive results. One of the first studies to examine cannabinoid concentrations in oral fluid after intravenous administration of radiolabeled THC found no radioactivity in the oral fluid, indicating that THC in oral fluid after smoking was a result of direct contamination of the oral mucosa and oral fluid in the mouth, and not from passive diffusion from plasma (99). Another study examined oral fluid following the smoking of 1.75 and 3.55% marijuana cigarettes by six participants (100). Specimens were collected by expectoration before and periodically up to 72 hours after smoking. All specimens were analyzed for cannabinoids using specific RIAs for THC and THCCOOH, with cutoff concentrations of 1.0 and 2.5 ng/mL, respectively. THC was detected in oral fluid for up to 24 hours after the higher dose. No specimens were positive for THCCOOH by RIA. In addition, one participant's specimen set was analyzed by GC/MS for THC, 11-OH-THC, and THCCOOH with LOQs of 0.5 ng/ mL. This analysis confirmed that no measurable 11-OH-THC or THCCOOH was present throughout the time course in any of the oral fluid specimens. Niedbala et al. studied 18 subjects who were administered single doses of marijuana by smoked (2025 mg) or oral (20-25 mg) routes (101). Urine and oral fluid specimens (Intercept collection device, OraSure Technologies, Inc., Bethlehem, PA) were collected at intervals up to 72 hours. Oral fluid was screened with a cannabinoid enzyme immunoas-

say (Intercept Micro-Plate EIA, OraSure Technologies, Inc.) with a cutoff concentration of 1.0 ng/mL and confirmed for THC by GC tandem MS, cutoff concentration of 0.5 ng/mL. Urine was screened by cannabinoid immunoassay (Abuscreen Online, Roche Diagnostics, Inc., Indianapolis, IN) and GC-MS for THC-COOH, cutoff concentrations of 50 and 15 ng/mL, respectively. Oral fluid specimens tested positive following marijuana smoked consecutively for average periods of 13 hours. The average time of the last positive test was 31 hours. There was great individual variation, with one subject having the last positive specimen at 2 hours and another at 72 hours. The decrease in oral fluid THC concentrations during the first 2 hours appeared to parallel those published by others for plasma THC, but no plasma was collected in this study for direct comparison. Urine specimens were consecutively positive following smoking for an average of 26 hours. The average time for the last positive reading was 42 hours with ranges up to 72 hours, the last collection. In the oral ingestion study, each of three subjects ate one brownie that had been cooked with plant material containing 20-25 mg of THC. THC was present in oral fluid following this method of oral ingestion, but concentrations peaked at 1-2 hours, were low, 3-5 ng/mL, and declined rapidly to negative, typically at 4 hours.

In recent studies oral fluid has been collected in a wide variety of devices designed by different manufacturers. Unfortunately, the recovery of cannabinoids from these devices is frequently unknown, a fact that significantly affects the devices' sensitivity in detecting cannabinoid use. Another problem area is the immunoassay reagent used to screen oral fluid specimens for cannabinoids. Many of the manufacturer's reagents target THC-COOH in their antigen-antibody reactions, making the sensitivity of these tests for cannabinoid exposure unacceptably low. Kintz et al. examined oral fluid (Salivette), blood, forehead wipes, and urine from 198 injured drivers and found 22 positive by urine testing for THC-COOH (102). Fourteen of these patients were also positive for THC in oral fluid, with no specimens positive for 11-OH-THC or THC-COOH at the limits of detection for their method. Samyn et al. collected urine from drivers who failed field sobriety tests at police roadblocks (103). For drivers who had a positive urine test, blood specimens were collected and, following informed consent, oral fluid (Salivette) and sweat specimens were collected. Oral fluid specimens and plasma were collected from 180 drivers and analyzed by GC-MS with cutoff concentrations of 5.0 and 1.0 ng/mL, respectively. The predictive value of oral fluid compared with plasma was 90%. In a different approach, Cone et al. examined 77,218 oral fluid specimens submitted to a large drug-testing laboratory (104). Using an oral fluid screening cutoff concentration for cannabinoids of 3 ng/mL and a confirmation THC cutoff concentration of 1.5 ng/mL, they found a cannabis positive rate of 3.22%, which was similar to the positive rate of 3.17% for large urine drug-testing laboratories using federally mandated cutoff concentrations. These studies have shown that measurement of THC in oral fluid compares favorably with sweat and urine testing for detecting cannabis use. Others have not found a good correlation between cannabinoid tests for oral fluid and other body fluids (105-109). Some of this variability in performance may be related to differences in cutoff concentrations, different screening specificities, binding of THC by the collection devices, and large intersubject differences of cannabinoid concentrations in biological fluids. The Substance Abuse Mental Health

Services Administration, Department of Health and Human Services (SAMHSA), which regulates federal workplace drug testing in the United States, is currently proposing a screening cutoff of 4 ng/mL for cannabinoids and a confirmation cutoff of 2 ng/mL THC for oral fluid (110).

Menkes et al. reported that the logarithm of salivary THC concentrations correlated with subjective effects and heart rate (111). Based on all of the available data and the ease of collection of oral fluid, many states and countries are considering the use of oral fluid testing for identification of drugged drivers. A large-scale roadside evaluation of the effectiveness of oral fluid monitoring for identifying drug-impaired drivers is being conducted currently in Europe and the United States (112,113).

Some organizations are interested in oral fluid testing of employees before beginning safety-sensitive work, because collection is easy and devices can give a quick screening result on-site. We will take this setting for a question regarding oral fluid testing. If a woman reports to a worksite to operate the reactor in a nuclear power station and her oral fluid screens positive for THC, is the manager justified in assigning her less sensitive duties until the test can be confirmed by a more specific method? If the woman had signed a pre-employment agreement not to use impairing drugs within 24 hours of reporting to work, did she violate her agreement, an act that could result in termination of her employment? The easy answer is that we cannot prove that she used cannabis based on a screening test. The result must be confirmed by a second method based on a different scientific principle of identification; however, it is instructive to examine the reliability of the result because many organizations would remove this person from safety-sensitive duties based on a positive screening test. The suspect employee would be returned to normal duties if the presumptive positive test was not confirmed by further laboratory testing. If the nuclear power facility had a drug policy outlining the terms and conditions for drug testing and ramifications of a positive screening and confirmation test and the woman had been informed of these regulations, then removal from a safety-sensitive position is a prudent action to take. Can we determine when the cannabinoid exposure occurred to answer the second part of the question? As mentioned above, with an oral collection device and screening and confirmation cutoffs of 1 and 0.5 ng/mL, respectively, Niedbala et al. found typical detection times of less than 24 hours, but some subjects produced a positive oral fluid specimen 72 hours after smoking (101). If the confirmatory test is positive and the cutoff concentrations and methodology are the same as those used in the controlled clinical study, we may be able to limit the window of drug exposure to within the past few days. It would be important to know the collection device and the laboratory's procedures, in particular the cutoff concentrations used. Unfortunately, data from well-controlled clinical studies to aid our interpretation are limited. Oral fluid collection devices and testing methodologies differ, and their performance may not have been evaluated in controlled studies. We cannot state definitively that she violated her agreement and used cannabis within 24 hours prior to reporting for work.

There is another interesting point to consider in the interpretation of oral fluid results. Suppose the woman states that she did not use illegal drugs but that she was passively exposed to marijuana smoke when her boyfriend and two of his friends smoked cannabis in her small kitchen. Could this explain the positive oral fluid test?

Although there are limited data in the literature, Niedbala et al. reported that two subjects who did not smoke cannabis but were in the room when others smoked had some positive screening but no confirmed oral fluid cannabinoid tests (101). Subsequent studies that are not yet published but were presented at the International Association of Forensic Toxicologists meeting in 2003 in Melbourne, Australia, and at a conference for Medical Review Officers (personal communication from S. Niedbala of OraSure Technologies, Inc.) conveyed the potential for passive exposure to marijuana smoke resulting in positive screening and confirmation tests. These results occurred when considerable smoke was present in small spaces, and oral fluid specimens were negative within 45 minutes of the end of exposure. This situation may be analogous to research that documented the possibility of a positive urine drug test following extensive passive exposure to marijuana smoke in a sealed experimental room (114). Although a positive test was produced in this experimental setting, participants complained of noxious smoke and irritation to the eyes. Other research conducted under more realistic passive smoke conditions indicated that production of a positive urine test with currently mandated federal guideline cutoffs is highly unlikely (115,116). A passive inhalation defense has rarely been accepted for a positive urine cannabinoid test. Additional research is needed to characterize the potential for positive oral fluid cannabinoid test from passive exposure. Perhaps the selection of appropriate oral fluid screening and confirmation cutoff concentrations can eliminate a positive oral fluid test from passive exposure. We lack appropriate data to answer the question of passive exposure of oral fluid at this time and must admit that additional controlled drug administration and naturalistic studies of drug in oral fluid are needed before we can definitively address the woman's claim of passive exposure.

Continue reading here: Pharmchem Sweat Drug Patch Test

Was this article helpful?

0 0