Bariatric surgery and changes in metformin distribution: About a fatal case.

A 65 year-old woman, suffering from mellitus type 2 diabetes and obesity, died at home, three days after bariatric surgery (Roux-en-Y gastric by-pass: RYGB). Her treatment, including metformin and dapagliflozin, was stopped before surgery and not postoperatively reinstalled. A forensic autopsy, toxicological and histological analyses were performed. No macroscopic or microscopic evidence of digestive perforation or peritonitis was identified, excluding an early surgery complication as the cause of death. Toxicological analysis revealed the presence of Metformin in all matrices tested, with a potentially fatal blood concentration. Death was attributed to lactic acidosis caused by a metformin overdose. With no evidence for suicide by ingestion of metformin, the authors supposed that the bariatric surgery might have caused changes in the absorption of metformin, leading to a rapid overdose and death. The only study in the literature on this subject, demonstrated a significant increase in the bioavailability of metformin following oral administration in gastric bypass patients. Thus, it can be anticipated that a therapeutic dose can become toxic when administrated to a subject who recently modified her digestive equipment. As this represents the first case of metformin overdose following bariatric surgery, further cases will be needed to confirm our initial observations.

[Testing for cannabis and synthetic cannabinoids in human specimens]. / Cannabis et cannabinoïdes de synthèse. À propos de leur détection biologique.

Among the 60 or so various cannabinoids which are present in cannabis sativa indica, terpenophenols are mainly found, including delta-9-trans tetrahydrocannabinol (Δ9-THC) which is the major psychoactive ingredient. Over the last decade, due to the emergence of e-commerce and the continuous investigations by pharmaceutical groups to identify new active molecules, synthetic cannabinoids have been proposed. These compounds, under the generic name "spices" have chemical structures very different from that of Δ9-THC, but share the same CB1 and CB2 receptors. They are mimics of Δ9-THC with much powerful pharmacological effects, including thus more deterrent side-effects, and enhanced windows of detection. All these drugs are controlled. This means that they have to be tested in human biological specimens to document abuse. The aim of this mini-review is to present what has been described in the scientific literature according to the available specimens (blood, urine, saliva, sweat, hair and exhaled breath), focussing on the current advantages and limitations of each test.

First identification of a diuretic, hydrochlorothiazide, in hair: Application to a doping case and interpretation of the results.

An athlete contested an adverse analytical finding involving hydrochlorothiazide, and requested our laboratory for testing his hair. As there is no reference in the literature about identification of hydrochlorothiazide in hair, several volunteers were first enrolled (4 after a single 25 mg administration and 10 with daily therapeutic treatment). A specific method was developed by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS. Hair samples were decontaminated with dichloromethane and 30 mg were incubated in buffer at pH 7.0 for 15 hours at 50°C. Then, 5 mL ethyl acetate was added for extraction. Linearity was observed for hydrochlorothiazide concentrations ranging from 5 to 2000 pg/mg. The limit of quantification was 5 pg/mg. The coefficients of variation (CVs) of repeatability and matrix effect were lower than 20%. Analysis of the 0-2-cm segment of the 4 volunteers having received a single dose, collected 1 month after administration, was negative at the limit of quantification. The hair of the 10 patients (proximal 2 cm) on daily treatment was positive with concentrations ranging from 12 to 1845 pg/mg, with no correlation between daily dose and concentration. The athlete's hair tested positive for hydrochlorothiazide at 36 pg/mg in the segment corresponding to the period of the urinary control. Since a single exposure to hydrochlorothiazide is not detectable in hair and based on the results of the patients on daily treatment, the concentration found in the athlete has been interpreted as corresponding to repeated exposures, where it was not possible to establish the dosage and the frequency.

European guidelines for workplace drug and alcohol testing in hair.

Guidelines for Legally Defensible Workplace Drug Testing have been prepared and updated by the European Workplace Drug Testing Society (EWDTS). They are based on the 2010 version published by Pascal Kintz and Ronald Agius (Guidelines for European workplace drug and alcohol testing in hair. Drug Test. Anal. 2010, 2, 367) and in concordance with the Society of Hair Testing guidelines (Society of Hair Testing guidelines for drug testing in hair. Forensic Sci. Int. 2012, 218, 20-24). The European Guidelines are designed to establish best practice procedures whilst allowing individual countries to operate within the requirements of national customs and legislation. The EWDTS recommends that all European laboratories that undertake legally defensible workplace drug testing use these guidelines as a template for accreditation. Copyright © 2016 John Wiley & Sons, Ltd.

Contribution of in utero drug exposure when interpreting hair results in young children.

Hair specimen is necessary to complement blood and/or urine analyses as it permits differentiation of a single exposure from chronic use of a drug by segmentation of the hair for a stated growth period. Moreover, due to a frequent long delay between event and police declaration, hair can be the only solution for lack of corroborative evidence of a committed crime. With the exception of lower amount of biological material in children versus adults, there is no specific analytical problem when processing samples from children. The issue is the interpretation of the findings, with respect to the different pharmacological parameters. In some very young children, the interpretation can be complicated by potential in utero exposure. Twenty-four cases from daily practice have been reviewed. Children were less than 1 year old, hair was always longer than 4 cm and the corresponding mothers admitted having used drugs during pregnancy. Drugs involved include methadone, tramadol, diphenhydramine, diazepam, cannabis, heroin, amitriptyline and bromazepam. Analyses were achieved by hyphenated chromatographic validated procedures after hair decontamination and segmentation. The concentrations measured in the hair of children were lower than those observed in subjects using therapeutically (or illegally) these drugs. In that sense, the frequency of exposures appears as un-frequent (low level of exposure), with marked decrease in the more recent period. However, the parents denied any administration in all cases and there was no reason to suspect re-exposure after delivery and no clinical problem during the period between delivery and hair collection during regular visits to the physician was noticed. The pattern of drug distribution was similar in all these cases, low concentrations in the proximal segments and highest concentration in the distal segment (last segment). When considering the concentration in the distal segment as the 100% of the response (highest concentration), after analysis of 4 segments (irrespective of the length of the segment but longer than 1cm), it was observed the following pattern: proximal segment, 5-35% of the response; segment 2, 15-50% of the response; segment 3, 25-60% of the response; and distal segment, 100% of the response. It is proposed to consider 100% in utero contribution to the final interpretation when the ratio concentration of the proximal segment to the concentration of the distal segment is lower than 0.5. This can be applied only when the child is under 1 year old and the hair shaft length is at least 4 cm (to achieve suitable segmentation). It is important, when using this cut-off to have at least 3 or 4 segments to be able to observe the variation in drug concentrations, whatever the length of each segment (>1cm).

Testing for ethanol markers in hair: discrepancies after simultaneous quantification of ethyl glucuronide and fatty acid ethyl esters.

The hair of 97 cases were analysed for ethyl glucuronide (EtG) and fatty acid ethyl esters (FAEE, including ethyl myristate, ethyl palmitate, ethyl oleate and ethyl stearate) according to the Society of Hair Testing guidelines to examine the role of both tests in documenting chronic excessive alcohol drinking, particularly when the results are in contradiction. 27 (27.8%) results were EtG negative and FAEE positive, when applying the SoHT cut-offs, probably due to the use of alcohol-containing hair products. Four cases (4.1%) were EtG positive and FAEE negative that were attributed to the use of herbal lotions containing EtG.

Retrospective diagnosis of an adverse drug reaction in a breastfed neonate: liquid chromatography-tandem mass spectrometry quantification of dextropropoxyphene and norpropoxyphene in newborn and maternal hair.

Dextropropoxyphene (DP) and norpropoxyphene (NP) are commonly used in the treatment of postpartum pain. The drug is widely prescribed in Europe and Canada and has been recently approved for use in the U.S. Its safety during breastfeeding, however, has not been fully established. Very few reports on its effects on neonates have been published. We report here the case of a mother treated with DP (6 capsules a day for 10 days) while she was breastfeeding. On day 7, her baby was lethargic and had difficulties with breastfeeding, which led to early weaning. The correlation between side effects observed in the infant and DP was made retrospectively by measuring DP and NP hair concentrations in the mother-infant pair with liquid chromatography-tandem mass spectrometry. Breastfeeding mothers taking DP expose their infants to high doses of DP and NP. In agreement with previously published reports, these data indicate that acetaminophen and nonsteroidal antiinflammatories are preferable for analgesia during breastfeeding. Breastfeeding should be encouraged under most circumstances, and if the mother takes any treatment for pain, a commonly prescribed drug with pharmacologic data available must be used.

Hair analysis to document a clinical case of TCDD over-exposure.

Dioxins and related compounds (furans) are persistent environmental contaminants that cause adverse biological effects. Their influence on humans is still unclear, except for accidental high-dose exposure. Chronic exposure to these compounds seems to be involved in cancer, endocrine disruption and neurobehavioral effects. For several years, a large concern about the potential health risks of dioxins is emerging in Europe and United States. The case of a 50-year-old man victim of an acute over-exposure to tetrachloro dibenzo-p-dioxin (TCDD or Seveso dioxin) is reported here with particular focus on hair investigations. The developed method involved the decontamination of the hair strand using picograde level methylene chloride, the homogenization of hair segments with scissors and their extraction in presence of 13C12-marked dioxin congeners under reflux of toluene using a Soxhlet, 8h at 130 degrees C. After reduction of the toluene fraction to 1 ml and addition of purification marker (37Cl4-2,3,7,8-TCDD), dioxins purification was achieved using three successive columns: silica, alumina/sodium sulfate and carbon/Celite columns. Finally, the toluene eluent was evaporated and the extract injected in the analytical system. After chromatographic separation, detection was achieved in single ion monitoring mode using a high resolution mass spectrometer operating in electron impact ionization mode (40 eV, minimal resolution of 10,000). The analysis of the first hair segment (0-6 cm) revealed the presence of 2,3,7,8-TCDD at 65 fg/mg when the distal one remained negative (LOQ=0.3 fg/mg). All other congeners (n=16) were in the range of those determined in the general population (0.62 and 0.89 fg/mg in the two hair segments, respectively). The extremely low dioxin levels generally found in hair specimens (low fg/mg range) lead us to analyze them using the very sensitive and specific gas chromatography-high resolution mass spectrometry apparatus. From the best of our knowledge, this is the first case of over-exposure to Seveso dioxin through hair analysis reported in the literature.

Quantitative determination of taxine B in body fluids by LC-MS-MS.

A new specific and sensitive LC-MS-MS method for the detection of taxine B and isotaxine B, the main toxic pseudo-alkaloids from yew (Taxus sp.), in biological samples (blood, urine, gastric content) was developed. Biological samples were prepared for LC-MS-MS by means of solid-phase extraction (SPE) procedure and yielded a recovery of 86%. Chromatographic separation was achieved using an RP(18) column. Detection of taxine B and isotaxine B was performed using multiple reaction monitoring with m/z 584.2 as precursor ion, i.e. [M+H](+), of both isomers and m/z 194.3 and m/z 107.1 as product ions after collision-induced dissociation. Docetaxel was applied as internal standard. The method was fully validated for the analysis of blood samples. Linearity was proven in the range from 0.1-500 ng/g. The limit of detection and the limit of quantitation are 0.4 and 2 ng/g, respectively. The method was applied to the determination of taxine B and isotaxine B in four fatal cases (two humans, two horses) with suspected yew intoxication. Blood levels were 105, 168, 174 and 212 ng/g.

Use of drugs of abuse in less than 30-year-old drivers killed in a road crash in France: a spectacular increase for cannabis, cocaine and amphetamines.

A collaborative study was conducted in France in order to determine the prevalence of cannabinoids, opiates, cocaine metabolites and amphetamines in blood samples from drivers killed in road accidents in 2003 and 2004 and to compare these values with those of a previous study performed during the period 2000-2001 involving 900 drivers. Blood samples were provided from 2003 under 30-year-old drivers, killed in a traffic accident. Drugs of abuse were determined by gas chromatography-mass spectrometry using the same analytical procedures in all the 12 laboratories. The most frequently observed compounds were by far cannabinoids, that tested positive in 39.6% of the total number of samples. Delta9 tetrahydrocannabinol (THC), the most active of the principle constituents in marijuana (cannabis sativa), was detected in the blood of 28.9% drivers and was the single drug of abuse in 80.2% of the positive cases. It was associated with amphetamines in 7.4% and with opiates and cocaine in 1.9 and 4.8%, respectively. Amphetamines were present in 3.1% of the total number of samples, cocaine metabolites in 3.0% and opiates in 3.5%. When comparing these results with those of a previous study performed 3 years before, a significant increase is observed for THC (28.9% versus 16.9%), cocaine metabolites (3.0% versus 0.2%) and amphetamines (3.1% versus 1.4%). This study demonstrates the critical necessity of implementing in France as soon as possible systematical roadside testing for drugs of abuse.

Identification of alprazolam in hair in two cases of drug-facilitated incidents.

The use of a drug to modify a person's behaviour for criminal gain is not a recent phenomenon. However, the recent increase in reports of drug-facilitated crimes (sexual assault, robbery) has caused alarm by the general public. Among the drugs that can be used, alprazolam (Xanax), an anxiolytic benzodiazepine, has been seldom observed. To document two cases involving this drug, we have developed an approach based on hair testing by LC-MS/MS. After pH 8.4 buffer incubation and extraction with methylene chloride/diethyl ether (80/20, v/v), hair extracts were separated on a XTerra MS C18 column using a gradient of acetonitrile and formate buffer. Alprazolam and diazepam-d5, used as internal standard, were detected by electrospray tandem mass spectrometry. In the first criminal case, alprazolam tested positive in two consecutive 2 cm hair segments at 4.9 and 2.4 pg/mg, from a 12-year-old girl, assaulted by her father who had sedated her three or four times. In the other case, alprazolam was detected in four consecutive 1cm hair segments at 3.1-0.4 pg/mg, obtained from an adolescent who had been forced to prostitute herself.

Evidence of addiction by anesthesiologists as documented by hair analysis.

Chemical dependency is a disease that can affect all professions. Among the health care professionals, anesthesiologists represent a specific group. Numerous factors have been proposed to explain the high incidence of drug abuse among anesthesiologists. These include: easy access to potent drugs, particularly narcotics, highly addictive potential of agents with which they are in contact, and easy diversion of these agents since only small doses will initially provide an effect desired by the abuser. Opioids are the drugs of choice for anesthesiologists, and among them fentanyl and sufentanil are the most commonly used. Alcohol is mostly abused by older anesthesiologists. Propofol, ketamine, thiopental and midazolam are also abused. In fact, all but quaternary ammonium drugs can be observed. Signs and symptoms of addiction in the hospital workplace include: unusual changes in behavior, desire to work alone, refusal of lunch relief or breaks, volunteer for extra cases, call, come in early and leave late, frequent restroom breaks, weight loss and pale skin, malpractice, behind on charts .... Toxicological investigations are difficult, as the drugs of interest are difficult to test for. In most cases, half-lives of the compounds are short, and the circulating concentrations weak. It is, therefore, necessary to develop tandem mass spectrometry procedures to satisfy the criteria of identification and quantitation. In most cases, blood and/or urine analyses are not useful to document impairment, as these specimens are collected at inadequate moments. Hair analysis appears, therefore, as the unique choice to evidence chronic exposure. Depending the length of the hair shaft, it is possible to establish an historical record, associated to the pattern of drug use, considering a growth rate of about 1cm/month. An original procedure was developed to test for fentanyl derivatives. After decontamination with methylene chloride, drugs are extracted from the hair by liquid/liquid extraction after incubation in pH 8.4 phosphate buffer. Fentanyl derivatives are analyzed by GC-MS/MS. The following cases are included in this paper: Case 1: 50-year-old anesthetist, positive for fentanyl (644 pg/mg); Case 2: 42-year-old anesthetist, positive for fentanyl (101 pg/mg) and sufentanil (2 pg/mg); Case 3: 40-year-old anesthetist, positive for codeine (210 pg/mg), alfentanil (30 pg/mg) and midazolam (160 pg/mg); Case 4: 46-year-old nurse, found dead, positive for alfentanil (2 pg/mg) and fentanyl (8 pg/mg). In these cases, the combination of an alternative specimen (hair) and hyphenated analytical techniques (tandem mass spectrometry) appears to be a pre-requisite.

Dioxins and furans determination in postmortem blood by gas chromatography--high-resolution mass spectrometry.

Dioxins and related compounds (furans) are persistent environmental contaminants that cause adverse biological effects. Their influence on humans is still unclear, except for accidental high-dose exposure. However, chronic exposure to these compounds seems to be involved in cancer, endocrine disruption, and neurobehavioral effects. For several years, a large concern about the potential health risks of dioxins is emerging in Europe and United States. Dioxin levels in biological specimens are extremely low and require very sensitive and specific methods of analysis. In this study, gas chromatography coupled to high-resolution mass spectrometry was used to evaluate dioxin body burden of two women deceased from generalized cancer. Fat fraction of blood specimens was obtained after precipitation with ethanol and extraction of both liquid and solid phases spiked with labeled 13C12-dioxin analogues. Organic phases were grouped, washed, and evaporated to weigh the lipid content. Lipids were dissolved in hexane, hydrolyzed with concentrated sulfuric acid, and discarded during water washes. Dioxins purification was achieved using three successive columns: silica, alumina/sodium sulfate, and carbon/Celite. Finally, the toluene eluent was evaporated and the extract injected in the analytical system. After chromatographic separation, detection was achieved in single ion monitoring mode using a high-resolution mass spectrometer operating in electron impact ionization mode (40 eV, minimal resolution of 10,000). Dioxin levels were expressed in pg TEQ/g of fat as defined by the World Health Organization. Quantification limits for each dioxin congener ranged from 2.5 to 12.0 pg/g fat with a relative extraction recovery always higher than 60%. Dioxin concentrations in the blood of the two deceased women were 35.0 and 42.7 pg TEQ/g fat, respectively. These concentrations are largely lower than those observed after accidental releases, but in the range of those observed in the general European population. Therefore, it was not possible to correlate dioxin body burden of the two women as a potential contributor of their cancer pathology. Nevertheless, knowledge of dioxin body burden in the French population would be of interest for an accurate interpretation of these results.

Testing for zopiclone in hair application to drug-facilitated crimes.

The use of a drug to modify a person's behaviour for criminal gain is not a recent phenomenon. However, the recent increase in reports of drug-facilitated crimes (sexual assault, robbery) has caused alarm in the general public. The drugs involved can be difficult to detect due to low dosages or chemical instability. They possess amnesic properties and can be quickly cleared from the body fluids. In these situations, blood or even urine can be of poor interest. This is the reason why this laboratory developed an original approach based on hair testing by LC-MS/MS. Zopiclone (Imovane), due to its short half-life associated with rapid hypnotic activity, is considered as a compound of choice to sedate victims. To document the detection of zopiclone in hair, we first tested specimens obtained from two volunteers who had ingested a single 7.5 mg Imovane tablet, and from repetitive consumers of zopiclone. After pH 8.4 buffer incubation and extraction with methylene chloride/diethyl ether (80/20 (v/v)), hair extracts were separated on a Xterra MS C18 column using a gradient of acetonitrile and formate buffer. Zopiclone and diazepam-d5, used as internal standard, were detected by tandem mass spectrometry. A single exposure to zopiclone was detectable in the first hair segment of two volunteers at concentration of 5.4 and 9.0 pg/mg, respectively. Hair from repetitive consumers tested positive for zopiclone at concentrations of 37 and 66 pg/mg. Hair analysis was applied to two authentic criminal cases. In the first one, zopiclone tested positive in the corresponding hair segment at 4.2 pg/mg, in accordance with a single exposure to the drug. In the other expertise, zopiclone was detected in the two segments analyzed, at 21.3 and 21.5 pg/mg, making unlikely the hypothetical single exposure to zopiclone.

Windows of detection of zolpidem in urine and hair: application to two drug facilitated sexual assaults.

A LC-MS/MS method for the detection of zolpidem in hair was developed to detect this drug after a single dose in possible drug facilitated sexual assaults. To determine the window of detection of zolpidem in both urine and hair, three volunteers received a 10 mg dose. Urine specimens were collected each 12 h for 144 h. Hair was sampled 3-5 weeks after exposure. Hair and urine extracts were separated on a Xterra MS C18 column using a gradient of acetonitrile and formate buffer. For each compound, detection was related to two daughter ions. Zolpidem was detected for up to 60 h in urine with peak concentrations obtained at 12 h. A single exposure to zolpidem was detected in hair at concentrations ranging from 1.8 to 9.8 pg/mg. Hair analysis was applied to two possible criminal cases. In the first case, zolpidem tested positive in the corresponding hair segment at 4.4 pg/mg. In the other case, zolpidem was detected in all the segments analyzed, demonstrating likely previous drug use in addition to recent exposure associated with a positive blood result.

Evaluation of the One-Step ELISA kit for the detection of buprenorphine in urine, blood, and hair specimens.

A solid-phase enzyme immunoassay involving microtiter plates was recently proposed by International Diagnostic Systems corporation (IDS) to screen for buprenorphine in human serum. The performance of the kit led us to investigate its applicability in other biological matrices such as urine or blood, and also hair specimens. Low concentrations of buprenorphine were detected with the ELISA test and confirmed by HPLC/MS (buprenorphine concentrations measured by HPLC/MS: 0.3 ng/mL in urine, 0.2 ng/mL in blood, and 40 pg/mg in hair). The intra-assay precision values were 8.7% at 1 ng/mL of urine (n = 8), 11.5% at 2 ng/mL in serum (n = 8), and 11.5% at 250 pg/mg of hair (n = 8), respectively. The immunoassay had no cross-reactivity with dihydrocodeine, ethylmorphine, 6-monoacetylmorphine, pholcodine, propoxyphene, dextromoramide, dextrometorphan at 1 and 10 mg/L, or codeine, morphine, methadone, and its metabolite EDDP. A 1% cross-reactivity was measured for a norbuprenorphine concentration of 50 ng/mL. Finally, the immunoassay was validated by comparing authentic specimens results with those of a validated HPLC/MS method. From the 136 urine samples tested, 93 were positive (68.4%) after the ELISA screening test (cutoff: 0.5 ng/mL) and confirmed by HPLC/MS (buprenorphine concentrations: 0.3-2036 ng/mL). From the 108 blood or serum samples screened, 27 were positive (25%) after the ELISA test with a cutoff value of 0.5 ng/mL (buprenorphine concentrations: 0.2-13.3 ng/mL). Eighteen hair specimens were positive (72%) after the screening (cutoff: 10 pg/mg) and confirmed by LC/MS (buprenorphine concentrations: 40-360 pg/mg). The ELISA method produced false positive results in less than 21% of the cases, but no false negative results were observed with the immunological test. Four potential adulterants (hypochloride 50 mL/L, sodium nitrite 50 g/L, liquid soap 50 mL/L, and sodium chloride 50 g/L) that were added to 10 positive urine specimens (buprenorphine concentrations in the range 5.3-15.6 ng/mL), did not cause a false negative response by the immunoassay.

Entomotoxicology for the forensic toxicologist: much ado about nothing?

We present a series of 29 necropsies in which organic compounds (including benzodiazepines, barbiturates, antidepressants, phenothiazine, opiates, cannabinoids, meprobamate, digoxin and nefopam) were detected in arthropod larvae sampled on human corpses. No correlation was observed between drug concentrations in the larvae versus human samples. When tested, inter-larvae and inter-site variations of drug concentrations (i.e., within larvae when analyzed separately, and within anatomic sites when larvae were grouped according to their site of sampling) were enormous and not reproducible from one case to another, confirming that arthropod larvae are unreliable for quantitative toxicological analysis. Since drugs identified in maggots are always detectable in the cadaver too, we conclude that larvae analysis is of almost no interest for practical forensic casework.

Fatal overdosage with cisplatin.

This paper presents a case of fatal overdosage due to an accidental massive administration (750 mg instead of 170 mg) of cisplatin, an anticancer agent, to a 63-year-old patient suffering from lymphoma. Platinum was measured in various postmortem samples by means of inductively coupled plasma mass spectrometry. Heart and peripheral blood concentrations of platinum were 1515 and 1253 micro g/L, respectively. Concentrations in urine and bile were 1038 and 501 micro g/L, respectively. Renal dialysis was started immediately after the end of cisplatin perfusion, when the mistake was noticed, but the patient deceased at day 16, presenting renal and hepatic insufficiency, ototoxicity, and pancytopenia.

Testing for kavain in human hair using gas chromatography-tandem mass spectrometry.

A sensitive, specific and reproducible method for the quantitative determination of kavain in human hair has been developed. The sample preparation involved a decontamination step of the hair with methylene chloride. The hair sample (about 50 mg) was incubated in 1 ml of methanol for 1 h, in an ultrasonic bath, in presence of 20 ng of methaqualone-d7 used as internal standard. The methanolic solution was evaporated to dryness, and the residue reconstituted by adding 30 microl of methanol. A 2 microl aliquot of the extract was injected onto the column (Optima5-MS capillary column, 5% phenyl-95% methylsiloxane, 30 m x 0.25 mm i.d. x 0.25 mm film thickness) of a Hewlett-Packard (Palo Alto, CA) gas chromatograph (5890). Kavain was detected by its parent ion at m/z 230 and daughter ions at m/z 111 and 202 through a Finnigan TSQ 700 MS/MS system. The assay was capable of detecting 30 pg/mg of kavain (limit of detection (LOD)). Linearity was observed for kavain concentrations ranging from 100 to 2000 pg/mg with a correlation coefficient of 0.998. Intra-day precision at 400 pg/mg was 13.7%. The analysis of a segment of hair, obtained from an occasional consumer, revealed the presence of kavain at the concentration of 418 pg/mg. A higher concentration (1708 pg/mg) was detected in the corresponding pubic hair.