Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication

Transcription

1 Journal of Analytical Toxicology 2012;36: doi: /jat/bks079 Article Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication Olof Beck 1 *, Niclas Stephanson 1,So ren Sandqvist 1 and Johan Franck 2 1 Department of Medicine, section of Clinical Pharmacology, Karolinska Institutet, Stockholm, Sweden and, and 2 Department of Clinical Neuroscience, Division of Psychiatry, Karolinska Institutet, Stockholm, Sweden *Author to whom correspondence should be addressed. olof.beck@karolinska.se It has recently been demonstrated that amphetamine, methadone and tetrahydrocannabinol are detectable in exhaled breath following intake. Exhaled breath, therefore, constitutes a new possible matrix for drugs-of-abuse testing. The present work aims to further explore this possibility by a study on patients treated for acute intoxication with abused drugs. Fifty-nine patients (44 males, age range 24 74) were included in the study, and breath, plasma and urine samples were collected following recovery, together with interview data. Analyses of breath and plasma samples were conducted with liquid chromatography mass spectrometry methods. Urine was screened using immunochemical reagents and positive findings confirmed with liquid chromatography mass spectrometry methods. The following analytes were investigated: methadone, amphetamine, methamphetamine, 3,4- methylenedioxymethamphetamine, codeine, 6-acetylmorphine, diazepam, oxazepam, morphine, benzoylecgonine, cocaine, buprenorphine and tetrahydrocannabinol. In 53 of the studied cases, recent intake of an abused substance prior to admission was reported. In 35 of these (66%), the breath analysis gave a positive finding. Identifications were based on correct chromatographic retention time and product ion ratios obtained in selected reaction monitoring mode. Generally, data from breath, plasma, urine and self-report were in agreement. Detected substances in breath included amphetamine, methamphetamine, buprenorphine, 6-acetylmorphine, morphine, codeine, methadone, tetrahydrocannabinol, diazepam, oxazepam and cocaine. Problem analytes with low detection rates were benzodiazepines and tetrahydrocannabinol. This study gives further support to the possibility of developing exhaled breath into a new matrix for drugs-of-abuse testing by extending the number of analytes that are documented to be detectable in breath. Introduction Human breath contains aerosol particles that are formed from the respiratory tract lining fluid during normal breathing (1 3). These aerosol particles carry non-volatile components containing diagnostic information and are often studied as the breath condensate fraction (4, 5). In this aerosol fraction, both lipids and peptides of endogenous origin have been demonstrated (6, 7). It has also been demonstrated that exogenous compounds, amphetamine, methadone and tetrahydrocannabinol (THC) are present in the exhaled breath (8 13). has been demonstrated to be carried by the aerosol particles (13), which is to be expected because methadone is a non-volatile substance under physiological conditions. In the case of THC, it can be assumed to be present in the lung as a result of contamination from smoking. However, regarding amphetamine and methadone, their presence in lungs must be a consequence of distribution from blood. The surfactant phase covering the airways is lipophilic, as it is composed of 90% lipids (14), and may represent a pharmacokinetic compartment for exogenous compounds with its own unique elimination kinetics. Following the original demonstration that amphetamine and methamphetamine are detectable in human exhaled breath following intake (8), the authors have been exploring exhaled breath as a possible matrix for abused drug testing. In the field of drug testing, there has been an interest in finding an alternative to urine and blood (15). The most developed and promising matrix today is oral fluid (16). Exhaled breath might become a further alternative, offering new, unique features. The promising findings have triggered the present study to further explore the possibility of developing exhaled breath as a matrix for abused drug testing. The present work studied the potential of exhaled breath analysis in a population of patients recovering from acute intoxication at an emergency ward and compared this analysis with data from plasma and urine analysis and from self-report. Materials and Methods Chemicals and materials, amphetamine, methamphetamine, 3,4-methylenedioxymethamphetamine (MDMA), codeine, 6-acetylmorphine (), diazepam, oxazepam, morphine, benzoylecgonine, cocaine, buprenorphine, THC, MDMA-d5 and THC-d3 were obtained as ampouled methanol solutions from LGC Standards AB (Bora s, Sweden). Methanol and acetonitrile of liquid chromatography mass spectrometry (LC MS) grade were from Fisher Scientific AB (Gothenburg, Sweden). 2-Propanol of normapur grade was from VWR International (Westchester, PA). Formic acid of analytical grade was from Merck (Darmstadt, Germany). The Milli-Q water was of ultra-pure quality (.18 MV/cm) and prepared in-house. The 47 mm C18 Empore disc was from Varian (Palo Alto, CA). The Technostat polymer particle filter (type 25) of 32 mm diameter was from Lindpro AB (O rebro, Sweden). Preparation of standard solutions The ampouled solutions of analytes and internal standards were diluted to a concentration of 100 ng/ml using methanol. These solutions were further diluted with methanol and used to fortify blank filters and plasma to prepare calibrators and quality controls. # The Author [2012]. Published by Oxford University Press. All rights reserved. For Permissions, please journals.permissions@oup.com

2 Study subjects Fifty-nine patients undergoing recovery from acute intoxication (44 males, 15 females, age range years) were recruited from the drug addiction emergency clinic at Beroendecentrum Stockholm. All subjects gave informed consent for participation. The history of drug use was assessed by interviewing the subjects using two structured questionnaires, AUDIT (for alcohol) and DUDIT (for illicit drugs) (17 19). The patients scored a median of 4 (range 0 36) in the AUDIT and 29 (range 0 39) in the DUDIT questionnaires. In the AUDIT questionnaire, the limit for harmful drinking is 8. The low median AUDIT score and high DUDIT score reflect a more limited use of alcohol and a heavy use of illicit drugs in the studied patients. Recent drug intake was further investigated by analysis of plasma and urine samples. In many of the cases, blood sampling or urine collection was not possible due to clinical circumstances. The urine and EDTA plasma samples were collected following the exhaled breath sampling and were stored at 2808C. Ethical approval was obtained from the Stockholm Regional Ethics Review Board (No. 2008/ ). Sampling of exhaled breath Two alternative sampling procedures were employed. In the first, compounds present in the exhaled breath were collected for 10 min by suction through a 47 mm Empore C18 disc using a membrane pump to assist the flow ( pump capacity, 300 L/min). The subjects were asked to breath more deeply than normal into an alcometer mouth piece (Palmenco AB, Stockholm, Sweden) mounted in the sampling device holding the filter (11). The subjects were asked to rinse the mouth with a mouthful of water before the sampling. It was estimated that all exhaled breath was passed through the filter during the sampling period time. Following sampling, the filter was dismantled using a pair of tweezers and stored at 2208C. The sampling device was carefully cleaned between uses with bacterial disinfectant and 70% ethanol. In the alternative, second procedure, breath samples were collected using a polymer particle filter in a similar way as described previously, but with the following modifications. The sampling time was 3 min and no pump assistance was needed. A special device and mouthpiece was constructed and is shown in Figure 1. Figure 1. Schematic illustration of the breath sampling device used for the polymer particle filter. The mouth piece was constructed to only let microparticles pass onto the filter compartment to protect from oral fluid contamination. This was accomplished by the 3 ledges that block any larger particle from passing through. Total length of sampling device was 12 cm. Analysis of exhaled breath samples Following storage, the Empore filter was cut into smaller pieces and prepared for analysis, as has been described before (11). The final dry residue was dissolved in 50 ml of 50% methanol. Polymer filters were prepared for analysis by placement into a 12 ml glass test tube with 7 ml of methanol. Following the addition of 100 ml internal standard solution (1.0 ng MDMA-d5 and 5.1 ng THC-d3), the extraction of analytes from the surface was performed in an ultrasound bath for 5 min at room temperature (þ228c). The supernatant was transferred to a new 12 ml glass tube, an additional wash (no ultrasound) of the filter was performed with 1 ml of methanol and the combined supernatants were evaporated to dryness. The final dry residue was dissolved in 75 ml of 50% methanol. After injection for THC analysis, the remaining liquid was diluted with 50 ml of 0.1% formic acid. For each filter type, separate calibration samples were prepared by fortifying blank filter with methanol solutions of analytes. These were prepared by adding ml of methanol solutions containing ng/ml of analytes. After drying, the discs were prepared for analysis as described previously. The calibrators were in the ranges of 10 10,000 pg/filter (10, 25, 100, 500, 1,000, 5,000 and 10,000). Calibration curves were constructed using linear regression analysis, with weighting factor 1/x. MDMA-d5 was used as internal standard for all analytes except THC. Mass spectrometry analysis system The LC MS-MS system consisted of a Waters Acquity ultraperformance liquid chromatograph (UPLC) with a vacuum degasser, binary pump and sample manager at ambient temperature connected to a Xevo TQ tandem mass spectrometer with MassLynx/Target Lynx Software version 4.1 (Waters, Milford, MA). The electrospray interface (ESI) was used, with the instrument operating in the positive ion mode. Nitrogen was used as nebulizer, desolvation and cone gas, and argon as collision gas. For all analytes except THC, the LC system was operated in a gradient mode with a flow rate of 700 ml/min (Table I). Table I Gradient Systems Used in the LC MS-MS Method Time (min) Solvent A* (%) Solvent B (%) Mobile phase composition used for THC Mobile phase composition for other compounds *Solvent A consisted of 0.1% formic acid (ph 2.85) and Solvent B consisted of methanol (THC) or acetonitrile. Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication 639

3 Chromatography was performed using a 1.7 mm mm (inner diameter) ethylene bridged hybrid (BEH) C18 column (Waters), preceded by a 0.2 mm column filter (Waters). Solvent A consisted of 0.1% (26.5 mmol/l) formic acid (ph 2.85) and Solvent B was 100% acetonitrile. The injection volume was 5 ml and the column oven temperature was 608C. The total run time of the method was 4.0 min. The following conditions were used in the mass spectrometer: source temperature, 1508C; desolvation gas temperature, 6008C; capillary voltage, 0.5 kv; multiplier voltage, 515 V; extractor voltage, 3.0 V; cone gas flow, 25 L/h; desolvation gas flow, 1,200 L/h; ion energy 1, 0.4 V; ion energy 2, 0.7 V; entrance and exit potential, 0.5 and 20.5 V, respectively; collision gas flow, 0.15 ml/min. The selected ions, cone voltage, collision energy and dwell time used for each compound are presented in Table II. Detection by the mass spectrometer was divided into five unique time periods, with each time period dedicated to specific compounds (Table II). The monitoring time for all compounds was min. For THC, the LC system was operated in a gradient mode with a flow rate of 400 ml/min (Table I). Chromatography was performed using a 1.7 mm mm (inner diameter) BEH Phenyl column (Waters), preceded by a 0.2 mm column filter (Waters). Solvent A consisted of 0.1% (26.5 mmol/l) formic acid (ph 2.85) and Solvent B was 100% methanol. The injection volume was 5 ml and the column oven temperature was 608C. The total run time of the method was 4.0 min. The following conditions were used in the mass spectrometer: source temperature, 1508C; desolvation gas temperature, 6508C; capillary voltage, 1.0 kv; multiplier voltage, 515 V; extractor voltage, 4.0 V; cone gas flow, 25 L/h; desolvation gas flow, 1,100 L/h; Table II Mass Spectrometric Parameters Used in the LC MS-MS Method for Breath and Plasma Compound Precursor ion (m/z) Product ion (m/z) Cone (V) Collision energy (ev) Dwell time (s) Monitoring time (min) MDMA MDMA MAM MAM Benzoylecgonine Benzoylecgonine Cocaine Cocaine MDMA-d THC THC THC-d ion energy 1, 0.4 V; ion energy 2, 0.7 V; entrance and exit potential, 0.5 and 20.5 V, respectively; collision gas flow, 0.15 ml/min. The selected ions, cone voltage, collision energy and dwell time used for THC are presented in Table II. The monitoring time for THC was min. Method validation For each filter type, replicates of calibration curves were documented at different occasions. Recovery of the extraction of analytes from filters was studied by preparing fortified (1 or 2 ng/filter) filters in triplicate and comparing with reference samples directly prepared in test tubes. The internal standard was added after elution of filters to separate recovery from any matrix effect. Figure 2. Chromatograms from the selected reaction monitoring of a 500 pg/filter calibrator, (A) displays most analytes, (B) displays THC. The monitored transitions are indicated in the figure. In 2A only one transition per analyte is shown due to space constraints. 640 Beck et al.

4 Intra-day imprecision and accuracy in quantifications were estimated by repetitive analysis of the samples prepared from fortified filters at two concentrations. Limit of detection [LOD, signal-to-noise ratio (S/N) ¼ 3] and limit of quantification (LOQ, S/N ¼ 10) were estimated by using diluted calibrator extracts. The LOD was estimated at the qualifier transition and the LOQ at the quantifier transition. In addition, the lower limit of quantification (LLOQ) was experimentally determined by using calibrators at concentration levels of 10 and 25 pg/filter. Matrix effect was studied by a post-column infusion of the analytes while injecting blank matrix extracts. Stability during storage at 2208C was studied for one month. Urine analysis Urine was screened for amphetamines, opiates, cannabis, cocaine, benzodiazepines, buprenorphine and methadone using CEDIA immunoassay reagents applied on an Olympus 640 instrument according to the manufacturer s instructions. Confirmations were made with in-house LC MS, LC tandem mass spectrometry (MS-MS) and gas chromatography (GC) MS methods. Plasma analysis Analysis of plasma samples (0.2 ml) was conducted using protein precipitation with 0.6 ml of acetonitrile. Following centrifugation, the supernatant was evaporated to dryness under nitrogen and redissolved with 150 ml of 25% methanol in 0.1% formic acid for the basic substances and with 75 ml 50% methanol in 0.1% formic acid for THC. The same internal standards and LC MS-MS conditions were used as for breath extracts. Calibrators were prepared from blank plasma. Results Method performance and validation The analytical method was divided into two different chromatographic and mass spectrometric procedures, with one directed only for THC. This was performed to achieve optimal analytical performance for THC. All analytes were separated Table III Summary of Method Validation Results for Breath Analysis (n ¼ 6, Intra-Day Experiment) Analyte LOD* (pg on column) LOQ (pg on column) QC low (ng/filter, mean) both by chromatography and by MS-MS. The chromatograms from a calibrator sample are shown in Figure 2. All calibration curves (n ¼ 3 4) showed a linear relationship between analyte concentration and analyte/internal standard peak area ratio. The correlation coefficients were typically The mean values of triplicate determinations of extraction recovery of analytes from fortified filters were %. The data for estimated LOD, LOQ and imprecision in quantification are shown in Table III. The LLOQ [coefficient of variation (CV),20%] was determined to be 25 pg/filter for diazepam and oxazepam, and 10 pg/filter for the other analytes. Following infusion of a blank matrix extract, a transient drop ( 10 s) was observed in response for infused analytes (Figure 3). However, no matrix effect could be observed near the retention times of any analytes as compared with injecting mobile phase A. Analyte recovery after storage at 2208C for one month was generally.90%, but 88% for THC. Clinical data Among the 59 studied patients, at least one compound was detected in the breath sample in 35 cases (59%) (Table IV). In the 53 cases with self-reported recent drug intake (within 24 h of sampling), the breath analysis detected substances in 66% of the cases. The detected substances included amphetamine, methamphetamine, buprenorphine,, morphine, codeine, methadone, THC, diazepam, oxazepam and cocaine. The analytical identification was based on correct relative retention time to internal standard (+1%) and by a correct (+20%) product ion ratio. Figure 4 shows examples of chromatograms from the identification of buprenorphine, oxazepam, diazepam, cocaine and. In general, the analytical data corresponded accurately for breath, plasma and urine. The detection of an analyte in breath corresponded to high concentrations in plasma and urine. In six cases, no recent intake (24 h) was self-reported and no analyte was found in the analysis of the breath sample. However, the results from plasma analysis were positive in two of these cases (Cases 34 and 36). QC low (accuracy, %) QC low (CV, %) QC high (ng/filter, mean) QC high (accuracy, %) MDMA Cocaine Benzoylecgonine THC QC high (CV, %) *LOD (S/N ¼ 3) estimated at qualifier transition. LOQ (S/N ¼ 10) estimated at quantifier transition. Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication 641

5 Figure 3. Post-column infusion experiment to investigate matrix effect on ionization efficiency: morphine (A); THC (B). For morphine, a solution of 80 ng/ml was infused at a rate of 50 ml/min and a blank patient breath extract was injected. For THC, a solution of 1 mg/ml was infused at a rate of 50 ml/min and a blank patient breath extract was injected. For comparison, the chromatograms from the 1 ng/ml calibrator are overlayed to show the retention times. In 14 cases, recent amphetamine use was self-reported. These reports were supported by results from plasma and urine analysis. In all cases, amphetamine was detected in breath, sometimes together with methamphetamine. In three other cases, amphetamine was also detected. Despite not being self-reported, this was supported by the presence of amphetamine in plasma (Cases 6, 51 and 57). Heroin use was self-reported in eight cases. In four of these, was detected in breath, whereas morphine and codeine were detected with in one case. In only one of these cases, a plasma sample was provided and both morphine and codeine were detected, but not. was present in urine in high concentrations in three cases and in breath in one case (Case 5). intake was self-reported in seven cases, which was confirmed by its presence in plasma in six cases. In the non-confirmed case (Case 17), no buprenorphine was detected in breath. Out of the six confirmed cases, buprenorphine was detected in breath in four. Cocaine was self-reported and detected in breath in only one case. Unfortunately, no plasma sample was available for investigation. was self-reported in 10 cases. Most were taking methadone on a daily basis. In all cases in which plasma was available, the self-reported use was confirmed. In nine of the 10 cases, methadone was detected in breath. In the missing case (Case 6), methadone was present in plasma. In one case (Case 2), methadone was detected in breath despite the lack of self-reported intake. No plasma sample was available in this case, but urine data confirmed the use of methadone. Two benzodiazepines, diazepam and oxazepam, were included as analytes. and oxazepam were detected in breath in several cases, but in rather low amounts. In several cases, diazepam and oxazepam were not detected in breath, despite self-reported intake and their presence in plasma. THC was detected in breath in only one case, although recent use was self-reported in nine. Five cases were selected to be investigated for the presence of THC only. In one of those (Case 12), THC was detected. In this case, no plasma sample was available for analysis. A summary of results based on self-reported intake is presented in Table V. Discussion The results of this study offer further support to the possibility of using exhaled breath as a specimen for drugs-of-abuse testing. The following analytes were detected for the first time in exhaled breath: buprenorphine, 6-acetylmorphine, morphine, codeine, diazepam, oxazepam and cocaine. and methamphetamine were previously documented under similar experimental conditions, and methadone and THC have previously been studied under more controlled circumstances (9 13). The new data coincide with the previous studies. In the present study, two different procedures for breath sampling were used. One used the silica based C18 modified filter, which needs pumping assistance due to high flow resistance. During the course of the work, the more practical polymer based filter was made available, which enabled elimination of the pumping. This was based on the demonstration that both procedures offered the same result for methadone (13). Also, the sampling time was reduced to 3 min to make it more convenient for the subject being sampled. Contamination from oral fluid was of concern in this work and has been considered in earlier studies (10, 13). Both devices were constructed to include oral fluid traps to reduce risk for contamination. The present result for amphetamine, together with a previous result (8), demonstrates a high detection rate for this analyte in breath. Previously, all 12 subjects with recent intake were detected (8), and in this study, all 14 subjects with recent intake were detected. Additionally, for methadone, buprenorphine and heroin intake, the detectability seems to be good. For heroin, the results indicate that is the primary analyte, which is also the case for oral fluid testing (16). The single case obtained for cocaine is promising, but does not allow for any firm conclusion at this point. Problem analytes with low detectability appear to be benzodiazepines and THC, and this may call for even more sensitive analytical methods to be applied in the future. In the case of benzodiazepines, this can be related to the high degree of protein binding in blood, and in the case of THC, it may be related to the rapid elimination in blood after intake (20). A previous study showed that THC is detected in breath 1 2 h after smoking cannabis (12). 642 Beck et al.

6 Table IV Results from the 59 Investigated Cases Following Recovery from Acute Intoxication Subject Exhaled breath pg/min Plasma ng/ml Urine ng/ml Self-reported recent intake 1 No finding Heroin -3-glucuronide -6-glucuronide -6-glucuronide 3 No finding No finding 24 No intake 4 2, , glucuronide -6-glucuronide -6-glucuronide detected 7 detected detected THC 1,100 1, , , ,400 18,000.50,000 38,000 1,300 15, , Heroin, diazepam 10* No finding THC Cannabis 11 40, benzodiazepines 12* Tetrahydrocannabinol Cannabis 13* No finding No finding 9 Cannabis 14* No finding 6 Cannabis 15* No finding 145 Cannabis 16 5, No finding glucuronide -6-glucuronide 170 4, , buprenorphine 18 1, , heroin -3-glucuronide Cocaine 80 Benzoylecgonine Cocaine , , , , No finding , , , Heroin glucuronide -6-glucuronide -6-glucuronide.10, ,300 1, ,000, heroin (continued) Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication 643

7 Table IV Continued Subject Exhaled breath pg/min Plasma ng/ml Urine ng/ml Self-reported recent intake , No finding 1, No finding 920 Cannabis 34 No finding No finding , , glucuronide -6-glucuronide -6-glucuronide glucuronide -6-glucuronide -6-glucuronide ,100.50,000 15, , , , No intake, benzodiazepines No intake 38 No finding No finding 11 No intake 39 No finding , , , heroin, cannabis 35 Benzoylecgonine -3-glucuronide -6-glucuronide -6-glucuronide 42 No finding 4, No finding 2, glucuronide -6-glucuronide -6-glucuronide.10, ,900 50,000 10, , ,000.50,000.50,000 1,670 2,4000 1,500 THC Heroin, benzodiazepines 44 No finding No finding No finding 64 No intake 46 1, , No finding 3, No finding Temazazepam -3-glucuronide -6-glucuronide -6-glucuronide 49 detected 1, , ,300 36, Cannabis 50 2, , (continued) 644 Beck et al.

8 Table IV Continued Subject Exhaled breath pg/min Plasma ng/ml Urine ng/ml Self-reported recent intake , No finding No finding 42 No intake 54 No finding 3, No finding 410 Heroin 57 1,200 Benzoylecgonine Tetrahydrocannabinol , No finding *Only analyzed for cannabis , Benzoylecgonine -3-glucuronide -6-glucuronide -6-glucuronide 5, ,400 46, ,200 4, Cannabis, morphine Figure 4. Chromatograms from the analysis of patient breath and plasma samples. detected from Subject 22 in Table IV: breath, 29 pg/min (A); plasma, 2.2 ng/ml (B); oxazepam (detected) and diazepam (3.4 pg/min) detected in breath from Subject 9 in Table IV (C); (3.6 pg/min) detected in breath from Subject 2 in Table IV (D); cocaine (80 pg/min) detected in breath from Subject 19 in Table IV (E). Detection of Drugs of Abuse in Exhaled Breath from Users Following Recovery from Intoxication 645

9 Table V Summary of Results for Patient Samples According to Admitted Recent Intake Admitted recent intake n Detection rate in breath (%) Confirmed in plasma Not confirmed or sample missing Confirmed in plasma Not confirmed or sample missing 1 0 Benzodiazepines, diazepam or oxazepam Confirmed in plasma 6 67 Not confirmed or sample missing 1 0 Cannabis Confirmed in plasma 2 0 Not confirmed or sample missing 8 12 Cocaine Confirmed in plasma 0 Not confirmed or sample missing Heroin Confirmed in plasma () 0 Not confirmed or sample missing 7 43 Confirmed in plasma Not confirmed or sample missing No intake 6 0 It must also be considered that the time between intake and sampling is rather long in this study, as some time elapsed from when subjects were taking their dose to the actual sampling time. The subjects needed to be transferred for medical care and to recover from intoxication before entering the study. Sampling in a clinical setting is typically performed at close to 24 h after last intake. Recent developments in the field of drugs-of-abuse testing have primarily focussed on oral fluid testing, and at this point, the sampling procedure, analytes, reporting limits and method development can be standardized and applied for a number of clinical and forensic applications (16). However, there is still room for additional alternatives to be developed and evaluated. Exhaled breath might become such an alternative, offering unique features compared to urine, blood, hair and oral fluid. Breath sampling has the advantages of instant application, constant functionality, and safety with respect to potential adulteration. The detection time can correlate to the time interval that a subject is under the influence of the drug. Recent research on exhaled breath aerosol particles has demonstrated that they are formed deep in the lung during normal breathing and can be sampled selectively using impaction technology (3). The particles are formed from the respiratory tract lining fluid that contains lipids and proteins and constitutes the surfactant phase. The turn-over rate is rather rapid (14), not allowing for a prolonged detection time to be expected. The discovery in a previous study (13) that the particles can be selectively collected, not only by impaction technology, but also by simple filtration, offers other applications than drugs-of-abuse testing for this sampling technique. Exhaled breath sampling of non-volatile compounds has potential applications in the diagnostic medical field (5). One challenge still remaining is to standardize the sampling procedure for quantitative applications. Acknowledgments We thank Inger Engman-Borg for excellent clinical assistance. This work was supported by grants from the Swedish Research Council, the Stockholm County Council and by financial support from the Swedish Prison and Probation Service. References 1. Papineni, R.S., Rosenthal, F.S. (1997) The size and distribution of droplets in the exhaled breath of healthy subjects. Journal of Aerosol Medicine, 10, Almstrand, A.C., Ljungstro m, E., Lausmaa, J., Bake, B., Sjövall, P., Olin, A.C. (2009) Airway monitoring by collection and mass spectrometric analysis of exhaled particles. Analytical Chemistry, 81, Almstrand, A.C., Bake, B., Ljungstro m, E., Larsson, P., Bredberg, A., Mirgorodskaya, E. et al. (2010) Effect of airway opening on production of exhaled particles. Journal of Applied Physiology, 108, Montuschi, P. (2007) Analysis of exhaled breath condensate in respiratory medicine: methodological aspects and potential clinical applications. Therapeutic Advances in Respiratory Disease, 1, Buszewski, B.B., Kesy, M., Ligor, T., Amann, A. (2007) Human exhaled air analytics: Biomarkers of disease. Biomedical Chromatography, 21, Almstrand, A.C., Josefson, M., Bredberg, A., Lausmaa, J., Sjövall, P., Larsson, P. et al. (2012) TOF-SIMS analysis of exhaled particles from patients with asthma and healthy controls. European Respiratory Journal, 39, Bredberg, A., Gobom, J., Almstrand, A.C., Larsson, P., Blennow, K., Olin, A.C. et al. (2012) Exhaled endogenous particles contain lung proteins. Clinical Chemistry, 58, Beck, O., Leine, K., Palmskog, G., Franck, J. (2010) s detected in exhaled breath from drug addicts: A new possible method for drugs-of-abuse testing. Journal of Analytical Toxicology, 34, Beck, O., Sandqvist, S., Eriksen, P., Franck, J., Palmskog, G. (2010) Method for determination of methadone in exhaled breath collected from subjects undergoing methadone maintenance treatment. Journal of Chromatography B, 878, Beck, O., Sandqvist, S., Eriksen, P., Franck, J., Palmskog, G. (2011) Determination of methadone in exhaled breath condensate by liquid chromatography-tandem mass spectrometry. Journal of Analytical Toxicology, 35, Beck, O., Sandqvist, S., Bo ttcher, M., Eriksen, P., Franck, J., Palmskog, G. (2011) Study on the sampling of methadone from exhaled breath. Journal of Analytical Toxicology, 35, Beck, O., Sandqvist, S., Dubbelboer, I., Franck, J. (2011) Detection of THC in exhaled breath collected from cannabis users. Journal of Analytical Toxicology, 35, Beck, O., Sandqvist, S., Franck, J. (2011) Demonstration that methadone is being present in the exhaled breath aerosol fraction. Journal of Pharmaceutical and Biomedical Analysis, 56, Presneill, J.J., Nakata, K., Inoue, Y., Seymour, J.F. (2004) Pulmonary alveolar proteinosis. Clinics in Chest Medicine, 25, Gallardo, E., Queiroz, J.A. (2008) The role of alternative specimens in toxicological analysis. Biomedical Chromatography, 22, Bosker, W.M., Huestis, M. (2009) Oral fluid testing for drugs of abuse. Clinical Chemistry, 55, Saunders, J.B., Aasland, O.G., Amundsen, A., Grant, M. (1993) Alcohol consumption and related problems among primary health care patients: WHO collaborative project on early detection of persons with harmful alcohol consumption I. Addiction, 88, Saunders, J.B., Aasland, O.G., Babor, T.F., De La Fuente, J.R., Grant, M. (1993) Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption II. Addiction, 88, Berman, A.H., Palmstierna, T., Ka llmen, H., Bergman, H. (2007) Evaluation of the drug use disorders identification test (DUDIT) in criminal justice and detoxification settings and in a Swedish population sample. Journal of Substance Abuse Treatment, 32, Huestis, M.A., Cone, E.J. (2004) Relationship of delta 9- tetrahydrocannabinol concentrations in oral fluid and plasma after controlled administration of smoked cannabis. Journal of Analytical Toxicology, 28: Beck et al.