SEARCH

SEARCH BY CITATION

Keywords:

  • Addiction;
  • drug testing;
  • substance abuse;
  • toxicology

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

As substance abusers need to demonstrate abstinence prior to transplant, valid/reliable drug tests are needed. Patients may deny use, fearing surgery will be delayed. Breath, blood and urine tests have brief detection windows that allow patients to evade detection. Routine laboratory tests do not include all substances of abuse. Hair analysis overcomes these barriers, increasing the likelihood that active users will be identified. This study compared results for alcohol, opioids and cocaine based on 445 self-report, breath, urine and hair samples from 42 patients who had been denied a transplant due to recent substance abuse. Compared to hair toxicology, sensitivity for conventional drug tests was moderate for cocaine and opioids, but poor for alcohol. Of positive hair tests, only half were corroborated through other tests. In contrast, specificity was high across tests and substances, with positive findings from conventional tests confirmed through hair toxicology. Based on a 90-day detection window for hair analysis, two negative tests suggest 6 months of continuous abstinence. Hair testing should be considered as an alternative approach for monitoring substance use in the transplant population, either as a routine procedure or when the veracity of findings from conventional tests is in doubt.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

Substance abuse is a common cause of diseases leading to organ failure including alcoholic- and hepatitis C-associated cirrhosis and nephropathy due to chronic use of opioids and/or cocaine (1). Recent drinking or drug use can result in surgery being delayed or denied (1,2), with patients required to demonstrate 6 months of abstinence (3). Preoperative substance use is associated with increased complications, relapse to substance use and impaired adherence (4–7); in contrast, abstinence improves psychological, social, family and occupational functioning, resulting in a better quality of life (2). Sobriety also can improve organ function (8), at times obviating the need for transplantation. Although many transplant programs require 6 months abstinence prior to surgery, there is no consensus regarding how this should be measured. In the absence of a ‘gold standard’ for drug testing in the transplant population, programs rely on self-report in combination with conventional testing methods like self-report, breath, blood and urine testing.

Conventional methods of drug testing (self-report, blood, breath, urine) have significant imitations. While current recommendations call for a nonjudgmental clinical interview to detect substance abuse (1,9), incentives to obscure use are strong (10,11). Stigma and fear of delay or exclusion from the transplant process may lead patients to deny use or attempt to falsify toxicology screens (1). Alcohol testing is particularly difficult. Alcohol is metabolized at the rate of 0.015% of blood alcohol concentration (BAC) per hour. Thus, a person with a BAC of 0.08% (legally drunk in most states) will have no measurable alcohol in the bloodstream within 5 ½ h of the last drink (0.08 divided by 0.015 = 5.33 h); this equation applies to breath testing as well. Thus, sporadic and light users, as well as those who abstain for a single day prior to testing, are unlikely to be identified through breath testing. Alcohol can be detected in urine for several days at most; ethyl glucuronide (EtG) testing extends the window to 3–4 days. A recent study (12) compared self-reported drinking with results from breath testing and urinary EtG in 18 liver transplant patients in drug treatment. No patient ever admitted to drinking and only 1/127 breath tests was positive. In contrast, 9 patients (50%) produced EtG positive samples with 24/49 (49%) of samples testing positive. As some patients refused testing on some occasions, these rates may underestimate drinking in this population. Gamma-glutamyltranspeptidase and carbohydra-deficient transferrin tests are significantly elevated in most chronic drinkers; however, results are nonspecific and difficult to interpret in patients with liver disease (13,14), thus limiting their utility with regard to monitoring. Because these approaches are alcohol-specific, additional tests are required to identify other drugs. While street drugs are readily identified via laboratory and point-of-collection (POC) urine and saliva tests, detection windows are brief and vary by substance. In addition, standard urine screens do not reliably detect synthetic opioids (15) or mixed agonist–antagonists like buprenorphine, necessitating the use of additional assays should opioid analgesics be of interest. The fact that morphine-based opioids like heroin are more likely to be identified than opioid analgesics raises the question of bias, as some populations of drug users are more likely to be abusing illegal drugs.

Given the limitations of conventional approaches to drug testing, alternative strategies should be explored. Hair as a specimen matrix for xenobiotic analysis is an option (16). Among women with HIV/AIDS, antiretroviral drug concentrations in hair strongly correlate with success of inhibition of viral growth and serve as an independent measure of medication compliance (17). In the addictions field, the goal of a long-term marker for alcohol consumption has been realized through analysis of EtG in hair, with heavy drinkers differentiated from nondrinkers and light drinkers (18,19). In this study (ClinicalTrials.gov Identifier: NCT00249652), a long-term biomarker was introduced because patients were being tested infrequently. This allowed us to compare results obtained from conventional approaches (self-report, breath and urine tests) with those from hair toxicology, an approach that has been studied in other populations with high motivation to conceal substance use, including probationers and pregnant women (20,21). We expected higher drug-positive rates based on hair toxicology compared to conventional monitoring approaches, regardless of testing method and substance.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

Participants

Participants were 42 adults with end-stage hepatic or renal disease who had been denied a transplant due to recent (past 6 months) use of alcohol and/or other drugs. Those who were unable to speak English, were in acute psychiatric crisis (i.e. suicidal, actively psychotic) or were encephalopatic, were excluded. Participants were recruited from Virginia Commonwealth University Medical Center in Richmond, VA (N = 2) and St. Luke’s-Roosevelt Hospital (N = 9), Columbia University Medical Center (N = 12) and the Mount Sinai School of Medicine (N = 19), all in New York City. All study procedures were conducted in accordance with the Institutional Review Boards at each institution.

Procedures

Participants were assessed on a clinical research unit or at their transplant programs. Trained research assistants administered the self-report Addiction Severity Index (ASI) and collected biological specimens at four time points (Weeks 0, 8, 20 and 32). A total of 445 samples were donated by 42 patients. Self-report data were collected for all active participants at each assessment point, although the number and type of biological samples varied. Some patients were unable to provide urine, had insufficient hair to sample or were too ill to participate in face-to-face visits and thus had phone interviews only. Comparisons were undertaken when results for two or more tests were available for the same patient at the same assessment point. Participants were compensated $50 at each research visit.

Measures

Self-report:  The Addiction Severity Index-Lite (22) is the most widely used structured clinical interview in the drug abuse field. It assesses ‘severity’ of alcohol, drug, medical, psychiatric, family–social, employment and legal problems from both the clinician's and patient's perspective. For this study, patients were queried about the number of days during the past 30 on which they used various substances (including some not reported on). Since any substance use is a contraindication for organ transplant, self-report responses were recorded as a binary variable, either ‘positive’ (≥1 day) or ‘negative’ (0 days).

Urine:  Participants provided an unobserved urine sample in a standard specimen cup with a temperature strip to detect adulteration. Research staff conducted POC urinanalysis, using a 10-panel drug screen. Results were recorded as positive or negative based on the manufacturer's cut-off values: amphetamine (1000 ng/mL); barbiturates (300 ng/mL), benzodiazepines (300 ng/mL); cocaine (300 ng/mL); cannabis (50 ng/mL); methadone (300 ng/mL); methamphetamine (1000 ng/mL); opiates (2000 ng/mL), phencyclidine 25 ng/mL); tricyclic antidepressants (1000 ng/mL).

Breath:  Participants were asked to refrain from smoking for 30 min prior to administering the breathalyzer test. The test employs a hand-held machine (LifeLoc Technologies, Inc., Wheat Ridge, CO) similar to those used in law enforcement. After several practice breaths, participants exhaled into a tube attached to the machine in one long sustained breath. The breathalyzer computes the amount of alcohol in expired air into a Blood Alcohol Level (BAL) score. A common reference point is 0.08, the level at which many states consider someone intoxicated. Any BAL above 0 (observed range = 0.007–0.164) was recorded as positive, whereas a BAL of 0 was recorded as negative.

Hair:  Using thinning shears, the hair is cut as close to the scalp as possible. Cutting from all around the head, with the crown as a landmark, allows for a uniform sample and reduces the risk of bias. The goal is to collect one milligram of hair (Figure 1). The collected hair is placed in a sample collection kit that is sealed and sent to the laboratory for analysis. For this study, only participants reporting no coloring/processing of hair within 6 months of collection were sampled in order to minimize concerns about test reliability (23). No participant ever complained about the procurement procedure or its cosmetic implications. Samples were stored at room temperature until the time of analysis. Analyses were conducted at United States Drug Testing Laboratories, Inc. (USDTL, Des Plaines, IL). Two separate aliquots of each hair sample were processed for general drugs of abuse and EtG, a direct ethanol biomarker.

image

Figure 1. Hair procurement; method and sample size.

Download figure to PowerPoint

For the drugs of abuse assay, 20 mg of hair was processed for ELISA initial screening by previously reported methods (24). Following the initial screening, all presumptive positives were confirmed through another aliquot of the sample hair by gas chromatography/mass spectrometry (GC/MS), tandem gas chromatography/mass spectrometry (GC/GC/MS) or liquid chromatography/tandem mass spectrometry (LC/MS/MS) (24,25). For analysis of EtG in hair, 50 mg of hair was processed for LC/MS/MS analysis by a previously reported method (26). Presumptive positives were realiquoted and a second analysis was performed to confirm the first finding. Data from the National Institute on Alcohol Abuse and Alcoholism (NIAAA) grant (PI: D. Lewis, author) showed that teetotalers (known EtG-negative hair) who were exposed to cosmetic products containing ethanol on a daily basis for 1 week did not produce measurable EtG in the hair. Based on study data comparing these patients to a cohort consisting of new admissions to an inpatient alcohol treatment facility, it was concluded that no common treatments caused any false positives for EtG. The current cutoff concentration utilized for hair EtG (2 pg/mg) has been found to correlate well with the Phosphatidylethanol (PEth) in blood cutoff concentration of 20 ng/mL (Plate C. Unpublished data, NIAAA grant 1R43AA016463–01, Hair EtG as a Long-Term Alcohol Biomarker 2008 [Small Business Innovation Research (SBIR) Phase I]). While it takes 50 g of alcohol per day for 2 weeks to reach a 27 ng/mg threshold for EtG in hair (27), the level of drinking required to achieve a PEth above that cutoff is much lower, only 100–200 g of absolute ethanol (the equivalent of 7–14 drinks) over the entire 2-week period (28). PEth concentrations at this level cannot be achieved by casual exposures to liquid medications contained in an alcohol vehicle, foods prepared with alcoholic liquids or multiple hand washings with an ethanol containing sanitizing product (28).

Data Analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

Analyses were performed on 445 samples obtained from 42 patients. Two by two tables of frequency counts were constructed comparing results from hair toxicology with those from more traditional methods of drug testing. Measures of sensitivity and specificity as well as kappa values were calculated using SAS 9.1.3 (29). Sensitivity is calculated as the proportion of cases where presence of substance use is identified through hair toxicology and through ASI self-report, breathalyzer, urine or breath. Specificity is calculated as the proportion of cases where absence of substance use is identified through hair toxicology and ASI self-report, breathalyzer, urine or breath. Kappa coefficient is a measure of test agreement and equals +1 when there is complete agreement. When the observed agreement exceeds chance agreement, kappa is positive, with its magnitude reflecting the strength of agreement.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

Participant characteristics

Forty-two participants [M (SD) = 47.90 (9.7) years of age] contributed data. Thirty-four (81.0%) were male and 8 (19.0%) were female. Race/ethnicities were 14 Caucasian (33.3%), 13 Hispanic (31.0%), 7 African American (16.7%) and 8 other (19.1%). Thirty-one patients (71.4%) needed a liver and 11 (26.2%) a kidney; 10 (90.9%) of renal patients were on dialysis as was 1 liver patient. Indications for hepatic transplant were alcoholic liver disease (ALD)/hepatitis C (HCV) (51.6%), ALD only (29.0%), HCV only (13.0%), hepatitis B (HBV)/HCV (3.2%) and ALD/HBV/HCV (3.2%); for renal transplant, indications were diabetes/hypertension (HTN) (36.4%), HTN only (36.4%), glomerulosclerosis (9.1%), HTN/Alport's Syndrome (9.1%) and HTN/glomerulonephritis (9.1%).

Substance use: Table 1 shows drug detection windows and thresholds by substance and testing method. Table 2 provides an overview of the data, including the number and type of samples collected at each of four assessment points. Changes in substance use status over time (‘cross-over’) are displayed in Table 3. Some participants provided data only at baseline. Of those with repeated measures available, some consistently tested negative or positive, whereas others ‘crossed over’ from positive to negative status or vice versa. Positive-to-negative conversion rates ranged from 14% (opiate self-report and urine toxicology) to 60% (cocaine self-report) whereas negative-to-positive conversion rates ranged from 0% (cocaine self-report and hair toxicology) to 23% (opiate self-report and hair toxicology). At Week 0, 15 participants (35.7%) reported using alcohol during the preceding 30 days, 12 (28.6%) cocaine, 3 (7.1%) heroin and 6 (14.3%) opioid analgesics.

Table 1.  Drug detection windows and thresholds
AnalyteBreath*Urine (based on industry standard cutoffs)Hair** (based on industry standard cutoffs)
  1. *According to NIAAA, ‘standard drink equivalents’ are:

  2. Beer (12 oz; 5% ethyl alcohol [ETOH]): 12 oz. = 1; 16 oz. = 1.3; 22 oz. = 2; 40 oz. = 3.3.

  3. Malt Liquor (8–9 oz; 7% ETOH): 12 oz. = 1.5; 16 oz. = 2; 22 oz. = 2.5; 40 oz. = 4.5.

  4. Wine (5 oz; 12% ETOH): a 750 mL (25 oz.) bottle = 5.

  5. Hard Liquor (1.5 oz; 40% ETOH): a mixed drink = (1–3); a pint (16 oz.) = 11; a fifth (25 oz.) = 17; 1.75 L (59 oz.) = 39.

  6. **Hair window is measured in growth interval. The hair sample length of 1.5 inches represents a mean growth period of 3.5 months with 82% of individuals in a range of 2.7 to 3.9 months. (35).

Cocaine/metabolites 24–48 h (300 ng/mL)Up to 90 days [liquid chromatography/tandem mass spectrometry (LC/MS/MS) cutoff 100 pg/mg]
Opiates 24–48 h (2000 ng/mL),Up to 90 days (LC/MS/MS cutoff 100 pg/mg)
Ethyl glucuronide1 h per standard drink (blood alcohol level >0.00) Up to 90 days (LC/MS/MS cutoff 2.0 pg/mg)
Table 2.  Frequency counts by substance, assessment time, test method and toxicology result
SubstanceAssessment timeTest resultTest method
 Self-report  Hair  Breath  Urine 
  1. Note: positive = self-reported use or positive toxicology result; negative = self-reported nonuse or negative toxicology result; weeks = correspond to assessment time points for treatment outcomes study; empty cells = test method not available for that substance.

AlcoholWeek 0Positive1564 
 Negative271338 
Week 8Positive972 
 Negative22928 
Week 20Positive582 
 Negative19621 
Week 32Positive653 
 Negative16619 
CocaineWeek 0Positive1210 10
 Negative3019 29
Week 8Positive68 6
 Negative2514 22
Week 20Positive46 6
 Negative2012 15
Week 32Positive34 5
 Negative1910 15
OpiatesWeek 0Positive98 9
 Negative3321 30
Week 8Positive98 6
 Negative2314 22
Week 20Positive85 6
 Negative1712 15
Week 32Positive63 4
 Negative1611 16
Table 3.  Frequency counts of substance use patterns by test method and substance
SubstanceTest methodBaseline assessment onlyRepeated measures test results
 Consistent outcomeCross-over
NegativePositiveNegativePositivePositive to positiveNegative to negative
  1. Note: negative = negative toxicology result or no self-reported use of substance; positive = positive toxicology result or self-reported use of substance.

AlcoholSelf-report 8317 (89%)5 (42%) 2 (11%)7 (58%)
Breath10127 (96%)2 (67%)1 (4%)1 (33%)
Hair 34 5 (83%)8 (73%) 1 (17%)3 (27%)
CocaineSelf-report 92 21 (100%)4 (40%)0 (0%)6 (60%)
Urine10018 (95%)6 (60%)1 (5%)4 (40%)
Hair104 12 (100%)7 (78%)0 (0%)2 (22%)
OpiatesSelf-report 8220 (80%)6 (86%)5 (20%)1 (14%)
Urine 8221 (96%)6 (86%) 1 (4%)1 (14%)
Hair11212 (80%)3 (50%) 3 (20%)3 (50%)

Table 4 compares drug detection rates by test method and substance. Self-report sensitivity for alcohol was fair (0.36), but specificity was high (0.85). In all but 4 of 26 instances (15%), self-reported alcohol use was corroborated by hair toxicology; in contrast, 21 of 43 self-reports of alcohol abstinence (49%) were disconfirmed by a positive hair test result. A similar pattern was observed for breath testing, where sensitivity was low (0.12), but specificity was high (1.0); all cases detected as using alcohol by breath testing also were detected through hair toxicology. For alcohol, Kappa values showed only slight agreement for both self-report (0.18) and breath testing (0.12) when compared to hair toxicology. For cocaine, sensitivity was moderate for self-report (0.50) and urine toxicology (0.44), while specificity was high (1.0 and 0.98, respectively). There was one cocaine positive urine sample for which hair analysis did not detect cocaine metabolites. Kappa values for cocaine were moderate for self-report (0.57) and urine toxicology (0.49). Finally, sensitivity was moderate (0.58) for self-reported use of opioids, with specificity high (0.93); 7 of 17 self-reports of opioid use (41%) were not confirmed through hair toxicology. Urine toxicology similarly revealed moderate sensitivity (0.52) and high specificity (0.95) for opioids. Only 3 of 13 opioid positive urine samples (23%) were not detected using hair toxicology; in contrast, 9 of 19 opiate-positive hair tests (47%) corresponded to negative urine tests. Kappa values showed moderate agreement with self-reported opioid use (0.56) and urine toxicology (0.52) with hair toxicology results.

Table 4.  Frequency counts, sensitivity and specificity for hair toxicology by ASI self-report, breathalyzer and urine toxicology results for alcohol, cocaine and opiate use
Alcohol use
HairSelf-reportBreath
NegativePositiveNegativePositive
  1. Note: negative = negative toxicology result or no self-reported use of substance; positive = positive toxicology result or self-reported use of substance.

Negative22 4270
Positive2112294
Sensitivity0.360.12
Specificity0.851.0
 
Cocaine use
HairSelf-reportUrine
NegativePositiveNegativePositive
 
Negative54 053 1
Positive14141512
Sensitivity0.500.44
Specificity1.00.98
 
Opiate use
HairSelf-reportUrine
NegativePositiveNegativePositive
 
Negative42 745 3
Positive1010 910
Sensitivity0.580.52
Specificity0.930.95

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References

Although substance abusers are required to demonstrate abstinence prior to receiving a transplant, no consensus exists about how this should be measured. In the absence of a ‘gold standard’, programs rely on self-report and conventional biomarkers like breath, blood and urine testing. Because these tests have brief detection windows, they are reasonably good measures of current substance use, but poor measures of abstinence. Frequent, random drug testing increases the chances that episodic users will be detected, although most programs lack the capacity to oversee rigorous monitoring programs. In situations where close monitoring is necessary, repository tissue analysis (hair or nails) may be an option (30,31).

With this in mind, this study compared results from conventional drug tests with those from hair toxicology. As expected, hair testing generated more drug-positive results than other approaches for all substances. In most instances, negative hair samples were complemented by negative findings from other tests (high specificity). In contrast, positive hair samples were only inconsistently associated with positive findings on other tests (low–moderate sensitivity). For cocaine and opioids, 50% of positive hair samples were not corroborated by either self-report or urinanalysis and 56% and 47% (respectively) were not confirmed by urinanalysis. Even larger discrepancies were observed for alcohol, with 64% of positive hair tests not self-reported and 88% not confirmed by breath testing. Discrepancies in drug-positive rates likely reflect motivational factors (for self-report) and brief detection windows (for breath, urine and self-report). As there is no incentive for admitting to substance use, a positive self-report in the absence of a positive hair test likely reflects a ‘false negative’ hair test; in contrast, a negative hair test, in conjunction with a negative self-report, should be considered a ‘true negative’ finding.

Transplant programs should develop testing protocols that are consistent with their treatment philosophies. Some programs may prefer not to identify casual or episodic users, believing this needlessly interferes with surgery; others may want detailed information about each patient's drug use so that they can intervene when necessary. Because hair testing identifies twice as many instances of drinking or drug use as conventional methods, programs using this technology should develop relationships with drug treatment programs and practitioners with whom they can collaborate. Programs that adopt hair testing may find it takes longer for patients to meet predetermined abstinence requirements. The number of patients who fail to qualify for surgery may also be higher. Intermittent users are most likely to be affected. On the other hand, hair testing has the potential to help good surgical candidates advance to surgery on an accelerated schedule. Because of its wide detection window, transplant clinicians have the opportunity to ‘look back’ (engage in retrospective review). If a negative hair sample (along with a negative self-report) is obtained upon initial evaluation and again 12 weeks later, then it is highly unlikely that any substances have been used during the preceding 6 months. Such patients potentially could be ‘fast-tracked’ for surgery.

This study has several limitations. As this is the first study of its kind in the transplant population, more research is needed to confirm findings. We were unable to study cannabis due to a small number of samples testing positive; future studies should include cannabis and other drug users if possible. While no patient ever refused hair testing or complained about the procedure, we were unable to obtain samples from a few patients who were bald or on dialysis. We did not attempt to sample hair from patients with less than 6 months of natural hair grown because results can be affected by frequent washing, dying, straightening and/or curling (32). Although increased detection rates have been found among individuals with darker hair (33), we were unable to consider this variable in this study. Despite considerable interest in determining a dose/response relationship for drugs of abuse, the physiology of hair makes this impossible unless the population is homogeneous with regard to hair color and texture. In the general population, variations in melanin content and hair washing render drug concentrations valuable in only the most rudimentary way of establishing that sufficient drug is present to detect. Accordingly, results of hair testing should be seen as positive or negative rather than as a continuous variable. On the other side of the kinetics, response is the washout rate. Only cocaine has been studied to any extent, with the half-life estimated at 1–1.5 months for both cocaine and benzoylecgonine. The washout of EtG in hair (and nails) currently is under investigation (NIAAA grant: 2R44AA016463–02, Hair EtGas a Long-Term Alcohol Biomarker; C. Plate PI) (34,35). Another consideration is cost. While a panel that tests for cocaine/metabolytes, opiates, cannabis (THC)/metabolytes and EtGcosts $30–50 in urine, $50–80 in oral fluid (saliva) and $80–120 in hair, different testing schedules more than compensate for any differences. Furthermore, existing Current Protocol Terminology (CPT) toxicology codes may be used to bill for hair testing, with charges on a per-drug basis.

To summarize, conventional drug tests (like self-report, breath and urine testing) are relatively good measures of recent substance use; however, their ability to reliably detect sporadic users or to verify continuous abstinence is limited. Whether or not a patient who is using receives a positive test result depends on when he/she uses, when testing occurs and what procedure is used. Unless substance use happens to occur within the window of detection for a particular substance, it will not be identified. This could lead treatment staff to falsely assume abstinence. In contrast, hair testing has the potential to identify substance use that occurs at any point in the window. Since hair testing produces different information than conventional drug tests, we recommend combining these approaches in order to generate a more complete picture of each patient's drug use status.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Data Analysis
  6. Results
  7. Discussion
  8. References
  • 1
    DiMartini AF, Dew MA, Trzepacz PT. Organ transplantation. In: LevensonJL, ed. Textbook of Psychosomatic Medicine. Washington , DC : APPI, 2005: 675700.
  • 2
    Lim JK, Keeffe EB. Liver transplantation for alcoholic liver disease: Current concepts and length of sobriety. Liver Transpl 2004; 10: S31S38.
  • 3
    Levenson JL, Olbrisch ME. Psychosocial evaluation of organ transplant candidates. A comparative survey of process, criteria, and outcomes in heart, liver, and kidney transplantation. Psychosomatics 1993; 34: 314323
  • 4
    Shapiro PA, Williams DL, Foray AT, Gelman IS, Wukich N, Sciacca R. Psychosocial evaluation and prediction of compliance problems and morbidity after heart transplantation. Transplantation 1995; 60: 14621466.
  • 5
    Bunzel B, Laederach-Hofmann K. Solid organ transplantation: Are there predictors for post transplant noncompliance?: A literature overview. Transplantation 2000; 70: 711715.
  • 6
    Dew MA, DiMartini AF, Steel J et al. Meta-analysis of risk for relapse to substance use after transplantation of the liver or other solid organs. Liver Transpl 2008; 14: 159172.
  • 7
    Gedaly R, McHugh PP, Johnston TD et al. Predictors of relapse to alcohol and illicit drugs after liver transplantation for alcoholic liver disease. Transplantation 2008; 86: 10901095.
  • 8
    Lucey MR. Management of alcoholic liver disease. Clin Liver Dis 2009; 13: 267275.
  • 9
    Weinrieb RM, Lucey MR. Treatment of addictive behaviors in liver transplant patients. Liver Transpl 2007; 13: S79S82.
  • 10
    Day E, Best D, Sweeting R et al. Detecting lifetime alcohol problems in individuals referred for liver transplantation for nonalcoholic liver failure. Liver Transpl 2008; 14: 16091613.
  • 11
    Weinrieb RM, Van Horn DHA, McLellan AT, Lucey MR. Interpreting the significance of drinking by alcohol-dependent liver transplant patients: Fostering candor is the key to recovery. Liver Transpl 2000; 6: 769776.
  • 12
    Erim Y, Böttcher M, Dahmen U, Beck O, Broelsch CE, Helander A. Urinary ethyl glucuronide testing detects alcohol consumption in alcoholic liver disease patients awaiting liver transplantation. Liver Transpl 2007; 13: 757761.
  • 13
    Rinck D, Frieling H, Freitag A et al. Combinations of carbohydrate-deficient transferrin, mean corpuscular erythrocyte volume, gamma-glutamyltransferase, homocysteine and folate increase the significance of biological markers in alcohol dependent patients. Drug Alcohol Depend 2007; 89: 6065.
  • 14
    DiMartini A, Day N, Lane T, Beisler A, Dew MA, Anton R. Carbohydrate deficient transferring in abstaining patients with end-stage liver disease. Alcohol Clin Exp Res 25: 17291733.
  • 15
    Levy S, Sherritt L, Vaughan BL, Germak M, Knight JR. Results of random drug testing in an adolescent substance abuse program. Pediatrics 2007; 119: 843848.
  • 16
    Baumgartner A, Jones P, Baumgartner W, Black T. Radioimmunoassay of hair for determining opiate-abuse histories. J Nucl Med 1978; 20: 748752.
  • 17
    Gandhi M, Ameli N, Bacchetti P et al. Protease inhibitor levels in hair strongly predict virologic response to treatment. AIDS 2009; 23: 471478.
  • 18
    Skopp G, Schmitt G, Pötsch L, Drönner P, Aderjan R, Mattern R. Ethyl glucuronide in human hair. Alcohol 2000 MayJun; 35: 283285.
  • 19
    Concheiro M, Cruz A, Mon M et al. Ethylglucuronide determination in urine and hair from alcohol withdrawal patients. J Anal Toxicol 2009; Apr; 33: 155161.
  • 20
    Marques P, Tippetts A, Branch D. Cocaine in the hair of mother-infant pairs: Quantitative analysis and correlations with urine measures and self-reports. Am J Drug Alcohol Abuse 1993; 19: 159175.
  • 21
    Mieczkowski T, Newel R. Patterns of concordance between hair assays and urinanalysis for cocaine: Longitudinal analysis of probationers in Pinellas County Florida, in NIDA Research Monograph (#167). The Validity of Self-Reported Drug Use: Improving the Accuracy of Survey Estimates. pp. 161199 (1997).
  • 22
    McLellan AT, Cacciola JS, Zanis D. The Addiction Severity Index-“Lite” (ASI-“Lite”). Center for the Studies of Addiction, University of Pennsylvania, Philadelphia VA Medical Center, 1997.
  • 23
    Jurado C, Kintz P, Menéndez M, Repetto M. Influence of the cosmetic treatment of hair on drug testing. Int J Leg Med 1997; 110: 159163.
  • 24
    Moore C, Deitermann D, Lewis D, Feeley B, Niedbala RS. The detection of cocaine in hair specimens using micro-plate enzyme immunoassay. J Forensic Sci 1999; 44: 609612.
  • 25
    Negrusz A, Moore C, Deitermann D et al. Highly sensitive micro-plate enzyme immunoassay screening and NCI-GC-MS confirmation of flunitrazepam and its major metabolite 7-aminoflunitrazepam in hair. J Anal Toxicol 1999; Oct; 23: 429435.
  • 26
    Bean P. New findings with ethyl glucuronide testing in hair samples. Am Biotech Lab 2008; 26: 2021.
  • 27
    Morini L, Politi L, Acito S, Groppi A, Polettini A. Comparison of ethyl glucuronide in hair with carbohydrate-deficient transferrin in serum as markers of chronic high levels of alcohol consumption. Forensic Sci Int 2009; 188: 140143.
  • 28
    Stewart SH, Reuben A, Brzezinski WA et al. Preliminary evaluation of phosphatidylethanol and alcohol consumption in patients with liver disease and hypertension. Alcohol Alcohol 2009 SepOct; 44: 464467. Epub 2009 Jun 17.
  • 29
    SAS version 9.1.3 (2002–2003). Cary, NC: SAS Institute.
  • 30
    Crosby RD, Carlson GA, Specker SM. Simulation of drug use and urine screening patterns. J Addict Dis 2003; 22: 8998.
  • 31
    DuPont RL, Griffin DW, Siskin BR, Shiraki S, Katze E, Random drug tests at work: The probability of identifying frequent and infrequent users of illicit drugs. J Addict Dis 1995; 14: 117.
  • 32
    Tanaka S, Lio R, Chinaka S, Takayama N, Hayakawa A. Identification of reaction products of methamphetamine and hydrogen peroxide in hair dye and decolorant treatments by high performance liquid chromatography/mass spectrometry. Biomed Chromatog 2001; 15: 4549.
  • 33
    Rollins D. Role of melanin in drug incorporation into hair. Society of Hair Testing, Arlington Heights , IL ; 2004.
  • 34
    Garcia-Bournissen F, Moller M, Nesterenko M, Karaskov T, Koren G. Pharmacokinetics of disappearance of cocaine from hair after discontinuation of drug use. Forensic Sci Int 2009 Aug 10; 189: 2427.
  • 35
    Kintz P, Bundeli P, Brenneisen R, Ludes B. Dose-concentration relationships in hair from subjects in a controlled heroin-maintenance program. J Anal Toxicol 1998 MayJun; 22: 231236.