MS and GC–MS methods in propofol detection: Evaluation of the two analytical procedures

MS and GC–MS methods in propofol detection: Evaluation of the two analytical procedures

Forensic Science International 256 (2015) 1–6 Contents lists available at ScienceDirect Forensic Science International journal homepage: www.elsevie...

585KB Sizes 1 Downloads 93 Views

Forensic Science International 256 (2015) 1–6

Contents lists available at ScienceDirect

Forensic Science International journal homepage: www.elsevier.com/locate/forsciint

LC–MS/MS and GC–MS methods in propofol detection: Evaluation of the two analytical procedures Fabio Vaiano a, Giovanni Serpelloni a, Martina Focardi b, Alessia Fioravanti a, Francesco Mari a, Elisabetta Bertol a,* a b

Department of Health Science, Forensic Toxicology Division, University of Florence, Florence, Italy Department of Health Science, Forensic Medicine Section, University of Florence, Florence, Italy

A R T I C L E I N F O

A B S T R A C T

Article history: Available online 17 July 2015

Propofol is a short-acting hypnotic agent that is commonly used to induce and maintain anesthesia. Propofol abuse and its involvement in suicide deaths have increased in recent years, especially among healthcare personnel. An example is the suicide of a 61-year-old nurse found with a propofol drip in his left arm. We describe the postmortem concentration of propofol in various tissues (femoral and cardiac blood, bile, urine, brain, and liver) and in the drip. The toxicological analyses were performed through two analytical methods, differing in derivatization reaction and in instrumentation: silylation for gas chromatograph–mass spectrometer (GC–MS), as routinely performed in our laboratory for this kind of analyses (lower limits of quantification–LLOQ–in urine and blood: 0.3 and 5 ng/ml); for liquid chromatograph–tandem mass spectrometer (LC–MS/MS) an innovative azo-coupling derivatization (LLOQ: 0.0004 and 0.1 ng/ml). This latter produces an azo-derivative (molecular composition: C18H22ON2; molecular weight: 282 Da) highly ionizable in electro-spray ion source, both in negative and positive ionizations. These two methods were compared to evaluate the effectiveness of this new LC–MS/MS analysis. An acidic hydrolysis (HCl 6 N, 100 8C, and 1 h) was performed for the biological samples (1 ml or 1 g) irrespective of the analytical method applied. The drip content was extracted adding phosphate buffer (pH 8) and a dichloromethane/ethylacetate 8:2 (v:v) mixture. Derivatization steps were: silylation with N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) + tetramethylammonium hydroxide (TMAH) for GC–MS; regarding LC–MS/MS, azo-coupling reaction with the aryl-diazonium salt (0–5 8C, and 30 min). The analyses were achieved in selected-ion monitoring for GC–MS (m/z, 235,250,73 propofol’’; m/z, 252,267,27 propofol-d17) and in multiple reaction monitoring ([MH]: m/z 283!241,77, azo-propofol; m/z 299!251,77, azo-propofol-d17) for LC–MS/MS. Autopsy showed no significant findings. Propofol concentrations were (LC–MS/MS vs GC–MS, respectively): 15.1 vs 14.5 mg/ ml, drip content; 7.11 vs 6.07 mg/ml, cardiac blood; 9.50 vs 7.19 mg/ml, femoral blood; 0.64 vs 1.07 mg/ ml, bile; 0.042 vs 0.051 mg/ml urine; 4.93 vs 5.89 mg/g, brain; and 7.88 vs 6.80 mg/g, liver. These values are comparable with the ones described in literature for death by acute propofol intoxication; the drip content is compatible with a diluted formulation of propofol available in Italy (20 mg/ml injectable emulsion). The comparison shows an excellent fitting of the data (R2: 0.9362). Toxicological results proved the cause of death as acute propofol intoxication. Furthermore, the new LC–MS/MS method showed an excellent effectiveness and reliability when compared to the routinely used GC–MS method. ß 2015 Elsevier Ireland Ltd. All rights reserved.

Keywords: Propofol LC–MS/MS Azo-coupling GC–MS Propofol postmortem distribution

1. Introduction Propofol (2,6-diisopropylphenol) is a short-acting intravenous hypnotic-amnesiac agent and is extensively used for the induction

* Corresponding author at: Department of Health Science, Forensic Toxicology Division, University of Florence, Largo Brambilla, 3 50134 Firenze, Italy. Tel.: +39 055415525; fax: +39 0557946171. E-mail address: elisabetta.bertol@unifi.it (E. Bertol). http://dx.doi.org/10.1016/j.forsciint.2015.07.013 0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.

and maintenance of general anesthesia [1–4]. It is widely used in surgical procedures because of its rapid onset, low toxicity, and long duration of narcotic effects [5–7]. However, hypotension, cardiac arrhythmia, apnea, and respiratory failures are the main side effects [8–10]. Since other side effects includes euphoria, sexual hallucinations, and disinhibition [11–14], its recreational use is recently increased fueling the debate about its potential of abuse and dependence. In recent years, many cases of addiction and deaths have been reported (just as the death of the performer Michael Jackson), providing evidences about the risks of propofol

2

F. Vaiano et al. / Forensic Science International 256 (2015) 1–6

use/abuse out of medical control [12,15–17]. Accordingly, Korea has been the first country to classify propofol as psychotropic agent in 2011. Thus, the spread of this substance for recreational or suicide purposes, and the related deaths, has been a great impulse for forensic toxicologists to validate new and more sensitive analytical procedures. Gas chromatography-mass spectrometry (GC–MS) is the most used technique for propofol detection due to its high separation capacity and detection sensitivity toward volatile compounds [18–21]. In particular, the headspace GC–MS is the simplest and fastest analysis (very low preparation time). Anyway, novel methods have been described for liquid chromatography– tandem mass spectrometry (LC–MS/MS), especially for phase II metabolites [2,22,23] as marker of propofol administration. Direct determination by means of LC–MS/MS systems is affected by the low ionization efficiency (IE) of propofol since its nonpolar nature and the lack of group is easy to ionize in electrospray ion source (ESI). Chemical derivatizations may be a useful solution to overcome this hurdle. Introduction of ionizable moieties such as charged (i.e., quaternary ammonium) or chargeable groups [24–26], may increase the IE and, moreover, may provide specific fragmentations by collision-induced dissociation (CID), making easier and more effective the analyte detection. Propofol derivatization methods are exclusively focused on the hydroxyl group [27,28]. With a previous paper [29], we were the first to shift attention to the phenolic ring taking advantage of its reactivity toward the electrophilic aromatic substitution (EAS). In that study, we presented a new derivatization procedure for LC–MS/MS detection of propofol in urine and blood by means of the EAS with aryldiazonium salt (ArN2+). The reaction is known as azo-coupling (AC) reaction (Fig. 1) and the derivative (molecular composition: C18H22ON2 and molecular weight: 282 Da) presents an azo-group (–N5 5N–) in para position. This moiety provides not only an increase in ionizability, both in negative (NIM) and positive ion modes (PIM) in ESI, but also specific fragmentation pathways (Figs. 2 and 3). The two methods were widely validated and were more sensitive than the ones described in literature (up to 200-fold than the analysis with dansyl derivatization [27] in blood and 6000-fold than detection of propofol-glucuronide in urine [20]). In this study, the new LC–MS/MS analysis was compared to the routine method in GC–MS, in order to investigate its effectiveness and reliability. Therefore, several parameters, such as time consuming, analytical procedure, quantitative findings, and their statistical processing, were evaluated. Analyses were performed on real specimens (femoral and cardiac blood, bile, urine, brain, liver, and drip content) collected from a case of suicide by propofol intoxication. In addition, toxicological interpretation about the postmortem concentrations and the cause of death were discussed.

acid, hydrochloric acid (HCl), sodium nitrite (NaNO2), formic acid, and n-hexane (Hex) were obtained from J.T. Baker (Deventen, Netherlands). Sodium hydroxide (NaOH) was supplied by Carlo Erba Reagenti (Milano, Italy). LC–MS CHROMASOLV1 MeOH was purchased from Sigma-Aldrich (St. Louis, MO, USA). Sterile water for injection (H2O) was obtained from B. Braun (Milano, Italy). Aniline was acquired from Riedel-de Hae¨n (Seelze, Germany). N,Obis(trimethylsilyl)trifluoroacetamide (BSTFA) and Tetramethylammonium hydroxide (TMAH) were provided by Supelco (Bellefonte, PA, USA). Propofol and propofol-d17 (internal standard, IS) methanolic standards were purchased from Chemical Research 2000 s.r.l. (Rome, Italy) and diluted to the appropriate concentrations with MeOH. 2.2. Case presentation A 61-year-old man was found dead by his son in the house where he lived. The corpse was on the bed with an almost empty 100 ml drip (labeled ‘‘NaCl 0.9%’’) still inserted, with an intravenous tube, in a vessel on the elbow of the left arm. The drip content was about 5 ml of a lactescent liquid. The subject saved on his computer a farewell letter explaining the reasons for his suicide. Autopsy did not reveal any remarkable findings. Femoral and cardiac blood, bile, urine, brain, liver, and drip content samples were collected and sent to our laboratory for the toxicological analysis. 2.3. Sample preparation 2.3.1. Drip content A total of 10 ml of drip content was added with 20 ml of phosphate buffer (pH 8), 10 ml of a 10 mg/ml solution of IS and then a liquid-liquid extraction (LLE) was performed with 100 ml of a DCM/AcOEt 8:2 (v:v) mixture. 2.3.2. Biological samples A protein precipitation was achieved on 1 ml of blood specimens (cardiac and femoral) with 2 ml of MeOH in presence of 100 ng of IS. Since propofol is extensively metabolized (especially to glucuronide-conjugated form) an acidic hydrolysis was performed on 1 ml or 1 g of the other biological specimens (bile, urine, brain, and liver) by adding 300 ml HCl 6 M and 10 ml of IS (50 ng/ml) at 100 8C for 1 h. After cooling at room temperature, the mixture was neutralized with 300 ml of NaOH 6 M and 20 ml of phosphate buffer (pH 8) were then added. A LLE was achieved with 3 ml of a DCM/AcOEt 8:2 (v:v) mixture. 2.4. Derivatization

2. Material and methods 2.1. Chemical and reagents Dichloromethane (DCM), diethyl ether (Et2O), and methanol (MeOH) were purchased from Panreac Quimica S.L.U. (Castellar del Valle`s, Spain). Ethyl acetate (AcOEt), sodium acetate, glacial acetic

2.4.1. Silylation for GC–MS Organic layers from LLE and 100 ml of supernatant from deproteinated blood were added with 20 ml of TMAH and then dried under gentle stream of nitrogen (N2) at 40 8C. The residue was incubated at 80 8C for 15 min with 50 ml of BSTFA. A total of 1 ml was injected in GC–MS system.

Fig. 1. Mechanism for the azo-coupling reaction.

F. Vaiano et al. / Forensic Science International 256 (2015) 1–6

3

Fig. 2. Fragmentation patterns of azo-propofol in positive ionization mode.

2.4.2. Azo-coupling for LC–MS/MS Organic layers from LLE were dried under gentle stream of nitrogen (N2) at 40 8C and then dissolved in 100 ml of MeOH. Both this latter mixture and 100 ml of supernatant from deproteinated blood were added with 1 ml of NaOH 2 M and cooled at 0–5 8C. Working solutions of aniline chloride (ArNH3Cl) and ArN2+ were freshly prepared as described in the previous study [29]. A 10 ml of the ArN2+ (100 nmol) were added dropwise to the cooled samples and a color change to light yellow suggested that the reaction was occurring. The reaction was maintained under magnetic stirring in an ice-salt bath for 30 min. After warming at room temperature, the derivatization compound was isolated by LLE with 5 ml of a

Fig. 3. Fragmentation patterns of azo-propofol in negative ionization mode.

DCM/EtOAc 8:2 (v:v) mixture. After centrifugation at 4000 rpm for 5 min, the lower organic layer was transferred to a tube and dried under a gentle stream of N2 at 40 8C. The residue was dissolved in 100 ml of LC–MS CHROMASOLV1 MeOH. A 5 ml aliquot was injected into the LC–MS/MS system. Regarding the drip content, these same analytical steps were applied on 10 ml of sample, using a different volume of ArN2+ (100 ml), since a higher propofol concentration was expected. 2.5. GC–MS The GC–MS instrument consisted in an Agilent 7890A GC system equipped with an Agilent 7683B series autosampler (Agilent Technologies, Palo Alto, CA, USA), and interfaced to a single quadrupole Agilent 5975C mass spectrometer (Agilent Technologies). The column used was an Agilent HP-5MS, 30 m length, 0.25 mm i.d., and 0.25 m film thickness (Agilent Technologies). The gas carrier (He) flow was constant at 1 ml/min. The oven temperature was initially set at 80 8C and programmed to 225 8C at 30 8C/min for 10 min. Injector and transfer line temperatures were respectively 300 8C and 230 8C. The injection volume was 1 ml in splitless mode. Electron ionization and selected ion monitoring (SIM) acquisition mode were used and the ions of interest were 235, 250, and 73 for TMS-propofol and 252, 267, and 73 for TMS-IS. A total of 1 ml of the derivatized extracts was injected in splitless mode.

F. Vaiano et al. / Forensic Science International 256 (2015) 1–6

4

2.6. LC–MS/MS Analysis was conducted using an HPLC Agilent 1290 Infinity system (Agilent Technologies, Palo Alto, CA, USA) interfaced with an Agilent 6460 Triple Quad LC/MS (Agilent Technologies), equipped with an ESI. The column used was a Zorbax SB-C18 Rapid Resolution HT (2.1  50 mm and 1.8 mm, Agilent Technologies), heated at 30 8C. The flow rate was always 0.4 ml/min. ESI configuration was: gas temperature 325 8C; gas flow rate 10 l/min; nebulizer 20 psi; and capillary 4000 V. Mobile phases consisted of 5 mM aqueous formic acid (A) and methanol (B). Starting from 10% B, gradient elution was carried out by increasing B to 90% at 3 min, then running isocratically for 1 min; post-run time was 1.5 min. Multiple reaction monitoring (MRM) acquisition was applied in NIM adopting the transitions: 281!176, 161 for azo-propofol; 297!192, 174 for azo-IS. Data acquisition and elaboration were performed using the Agilent MassHunter Workstation software package. 3. Results Urine sample were first screened for the common drugs of abuse and their metabolites (cocaine, opiates, cannabinoids, benzodiazepines, amphetamines, methadone, and barbiturates) by means of the EMIT1 Siemens VIVA-E drug testing system (Siemens, Newark DE) according to the manufacturer’s instructions. No positive result was observed. Propofol concentrations were (LC–MS/MS vs GC–MS, respectively): 7.11 vs 6.07 mg/ml, cardiac blood; 9.50 vs 7.19 mg/ml, femoral blood; 0.64 vs 1.07 mg/ ml, bile; 0.042 vs 0.051 mg/ml urine; 4.93 vs 5.89 mg/g, brain; and 7.88 vs 6.80 mg/g, liver (Table 1). The drip content was 15.1 and 14.5 (LC–MS/MS vs GC–MS, respectively) mg/ml. In order to estimate the reliability of the new method against the routine one, the concentrations in all the biological specimens were statistically processed. The coefficient of determination (R2) was 0.9362 indicating an excellent data fitting (Fig. 4). Moreover, no statistically significant difference (t-test, p < 0.01) between the quantitative outcomes of the two methods was found. The main validation parameters for both the methods are resumed in Table 2. Postmortem distribution was investigated calculating the ratios of heart to femoral blood concentrations that were 0.74 for LC–MS/ MS and 0.84 for GC–MS. 4. Discussion 4.1. Comparison between LC–MS/MS and GC–MS analyses In order to investigate the real likelihood of the new LC–MS/MS method as routine procedure, a comparison with our GC–MS method was performed in a real case analysis. As described above, the two sample treatments differed only for the derivatization reaction, i.e., silylation for GC–MS and AC for LC–MS/MS. AC

Table 1 Comparison between quantitative results obtained through LC–MS/MS and GC–MS analyses.

Drip content Femoral blood Cardiac blood Bile Urine Brain Liver

LC–MS/MS

GC–MS

15.1 mg/ml 9.50 mg/ml 7.11 mg/ml 0.63 mg/ml 0.042 mg/ml 4.93 mg/g 7.88 mg/g

14.5 mg/ml 7.18 mg/ml 6.07 mg/ml 1.07 mg/ml 0.051 mg/ml 5.89 mg/g 6.80 mg/g

Fig. 4. Correlation between quantitative results for the two methods.

greatest drawback was a longer time consuming mainly due to the related LLE: silylation required a mean time of 45 min to be carried out against the 90 min for AC. Although this disadvantage may represent an important limitation to its use, it is well-balanced by a faster chromatographic run (Fig. 5), since each LC–MS/MS run lasts 5.5 min against 20 min for GC–MS. This means that the entire LC– MS/MS procedure is expected to be faster when more than four samples are analyzed, as in our case and all the quantitative analysis, where a calibration curve and a blank sample are required. Hence, we would recommend this method also for routine purposes. Moreover, AC procedure is very easy to perform and do not require particular manual skills. Crucial point is the temperature control as it should not exceed 5 8C to avoid the hydrolysis of ArN2+ to phenol and N2. An ice-salt bath is enough to keep the temperature within the range for all the reaction time (30 min). Another key step is the isolation of azo-propofol by means of a LLE that may directly affect the analysis. However, each single variable (extractive mixture, volume ratio, vortex time, etc.) has been planned in the validation phase to obtain the best recovery rate (>85%, Table 2), with the fewest chromatographic interferences. Moreover, sensitivity was not compromise by LLE at all, as demonstrated in the validation phase. Sensitivity plays an important role especially in forensic toxicology, when more and more cases require detection of very low amounts of analyte (i.e., hair analysis for single exposure). In this case, the high propofol concentrations in all the specimens did not allow to test sensitivity in a real analysis of various matrices. However, comparing the values obtained in the previous validation phases (Table 2), the LC– MS/MS analysis turned out more sensitive both in urine and blood detection, with lower limit of quantification (LLOQ) estimated in 0.0004 and 0.1 ng/ml (urine and blood, respectively); for GC–MS, the LLOQ values were 0.3 and 5 ng/ml (urine and blood, respectively). Unfortunately, the method validations for bile, liver, and brain were only partially achieved because of the lack of these tissues enough to perform all necessary tests. Nevertheless, matrix effects were estimated comparing the slopes from aliquots of the real samples (added with standard amounts of analyte, after quantitative analysis), with water solutions (at the equivalent total concentrations); the matrix effects were (LC–MS/MS vs GC–MS, respectively): +18% vs +7%, brain; +23% vs +11%, liver; and +21% vs +15%, bile. The high concordance and the absence of significant differences (R2: 0.9362; t-test, p < 0.01) between the two procedures pointed out the actual reliability of the new method, also in term of quantitative results. According all these evaluations, this LC–MS/MS procedure proved to be suitable as routine method for propofol detection because of its great sensitivity, lesser time consuming (when more than four samples are analyzed), and being user-friendly.

F. Vaiano et al. / Forensic Science International 256 (2015) 1–6

5

Table 2 Main validation results for LC–MS/MS and GC–MS detection of propofol in urine and blood. Detection

LC–MS/MS Urine Blood GC–MS Urine Blood a

LLOQa

Accuracy range

Intra-day precision range

Inter-day precision range

ng/ml

%MRE

%RSD

%RSD

0.0004 0.1

4.8–1.7 5.8–9.6

1.3–7.9 0.4–3.2

0.3 5

8.4–6.3 3.0–9.7

2.5–6.9 1.9–7.8

Recovery

Matrix effect

2.1–8.1 1.8–2.7

>85% >87%

+22% +27%

3.9–10.9 4.3–9.9

>89% >93%

+7% +11%

LLOQ: lower limit of quantification; %MRE: mean relative error percentage; %RSD: relative standard deviation percentage.

4.2. Toxicological interpretation In literature, several cases of death by self-administration of propofol have been reported, mainly involving health care personnel, as also observed in our case. It is peculiar that in most of them, the postmortem blood concentration was within the therapeutic range (1.3–6.8 mg/ml [30]). This is due to the pharmacokinetics of propofol: despite an injection of propofol overdose, its rapid metabolism, and redistribution do not allow to exceed the therapeutic range [31]. However, this is not enough to avoid the death, since even such concentrations can cause prolonged apnea, extreme hypoxia, and hypotension that could be lethal without a medical support [3,5,11,17,32,33]. Therefore, interpretation of blood and tissue concentrations of propofol

should take into account this aspect. Blood concentrations were in the range or slightly beyond it also in our case (LC–MS/MS vs GC–MS: 7.11 vs 6.07 mg/ml, cardiac blood; and 9.50 vs 7.19 mg/ml, femoral blood), but the continuous administration through a catheter without the medical control led to a fatal poisoning. The ratios of heart to femoral blood concentrations (0.74, LC–MS/MS; and 0.84, GC–MS) were within the range published in literature (0.45–3.66) [34]. These values can give an approximate indication that post-mortem redistribution occurred. Regarding drip content, it was qualitatively (lactescent solution, presence of excipients such as triglycerides and fatty acids) and quantitatively (average concentration of 14.8 mg/ml) compatible with a diluted formulation available in Italy (20 mg/ml injectable emulsion).

Fig. 5. Comparison between the GC–MS (A) and the LC–MS/MS (B) chromatograms.

6

F. Vaiano et al. / Forensic Science International 256 (2015) 1–6

5. Conclusions In conclusion, in this real case the new LC–MS/MS procedure showed an excellent effectiveness and reliability when compared to the routinely used GC–MS method. Although AC reaction and the LLE presented a higher time consuming, on the other hand the LC–MS/MS analysis is shorter than the GC–MS run, making faster the entire procedure. Moreover, the new method presents a higher sensitivity that is a key factor especially in forensic toxicology. It must be also considered the high concordance between the quantitative results. In addition, we emphasized that death by propofol administration can occur even if the drug concentrations do not exceed the therapeutic levels. Therefore, a mere interpretation of concentrations may be of limited significance when circumstantial features, such as administration way and presence of medical support, are not considered.

References [1] A. Bienert, Z. Zaba, S. Dyderski, M. Ogrodowicz, L. Drobnik, Long-term stability of propofol in human plasma and blood in different storage conditions, Acta Pol. Pharm. 62 (2005) 95–97. [2] S. Cohen, F. Lhuillier, Y. Mouloua, B. Vignal, P. Favetta, J. Guitton, Quantitative measurement of propofol and in main glucuroconjugate metabolites in human plasma using solid phase extraction-liquid chromatography-tandem mass spectrometry, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 854 (2007) 165–172. [3] R.R. Kirby, J.M. Colaw, M.M. Douglas, Death from propofol: accident, suicide or murder? Anesth. Analg. 108 (2009) 1182–1184. [4] G. Klausz, K. Ro´na, I. Kristo´f, K. To¨ro, Evaluation of a fatal propofol intoxication due to self administration, J. Forensic Leg. Med. 16 (2009) 287–289. [5] E.F. Kranioti, A. Mavroforou, P. Mylonakis, M. Michalodimitrakis, Lethal self administration of propofol (Diprivan). A case report and review of the literature, Forensic Sci. Int. 167 (2007) 56–58. [6] P.E. Marik, Propofol: therapeutic indications and side-effects, Curr. Pharm. Des. 10 (2004) 3639–3649. [7] P.E. Marik, Propofol: an immunemodulating agent, Pharmacotherapy 25 (5 Pt 2) (2005) 28S–33S. [8] B.K. Burow, M.E. Johnson, D.L. Packer, Metabolic acidosis associated with propofol in the absence of other causative factors, Anesthesiology 101 (1) (2004) 239–241. [9] J.C. Salengros, C.E. Velghe-Lenelle, R. Bollens, E. Engelman, L. Barvais, Lactic acidosis during propofol-remifentanil anesthesia in an adult, Anesthesiology 101 (1) (2004) 241–243. [10] B.M. Schramm, B.A. Orser, Dystonic reaction to propofol attenuated by benztropine (cogentin), Anesth. Analg. 94 (2002) 1237–1240. [11] A. Roussin, M. Mirepoix, G. Lassabe, V. Dumestre-Toulet, V. Gardette, J.L. Montastruc, M. Lapeyre-Mestre, Death related to a recreational abuse of propofol at therapeutic dose range, Br. J. Anaesth. 97 (2006) 268. [12] A. Roussin, J.L. Montastruc, M. Lapeyre-Mestre, Pharmacological and clinical evidences on the potential for abuse and dependence of propofol: a review of the literature, Fund. Clin. Pharmacol. 21 (5) (2007) 459–466. [13] S.C. Sweetman, Martindale the Complete Drug Reference, 34th ed., Pharmaceutical Press, London, 2005, pp. 1305–1307.

[14] M.N.G. Dukes, J.K. Aronson, Z. Baudoin, General anaesthetics and anaestheticgases, in: M.N.G. Dukes, J.K. Aronson (Eds.), Meyler’s Side Effects of Drugs,, 14th ed., Elsevier, Amsterdam, 2000, p. 330. [15] C. Wilson, P. Canning, E.M. Caravati, The abuse potential of propofol, Clin. Toxicol. (Phila) 48 (3) (2010) 165–170. [16] J.P. Zacny, J.L. Lichtor, W. Thompson, J.L. Apfelbaum, Propofol at a subanesthetic dose may have abuse potential in healthy volunteers, Anesth. Analg. 77 (1993) 544–552. [17] S. Iwersen-Bergmann, P.R. Ro¨sner, H.C. Ku¨hnau, M. Junge, A. Schmoldt, Death after excessive propofol abuse, Int. J. Legal Med. 114 (2001) 248–251. [18] P. Favetta, C. Dufresne, M. De´sage, O. Paı¨sse´, J.P. Perdrix, R. Boulieu, J. Guitton, Detection of new propofol metabolites in human urine using gas chromatography/mass spectrometry and liquid chromatography/mass spectrometry techniques, Rapid Commun. Mass. Spectrom. 14 (2001) 1932–1936. [19] V. Cirimele, P. Kintz, S. Doray, B. Ludes, Determination of chronic abuse of the anaesthetic agents midazolam and propofol as demonstrated by hair analysis, Int. J. Legal Med. 116 (2001) 54–57. [20] S.Y. Lee, N.H. Park, E.K. Jeong, J.W. Wi, C.J. Kim, J.Y. Kim, M.K. In, J. Hong, Comparison of GC/MS and LC/MS methods for the analysis of propofol and its metabolites in urine, J. Chromatogr B Analyt. Technol. Biomed. Life Sci. 900 (2012) 1–10. [21] F.T. Peters, j. Jung, T. Kraemer, H.H. Maurer, Fast, simple and validated gas chromatographic-mass spectrometric assay for quantification of drugs relevant to diagnosis of brain death in human blood plasma samples, Ther. Drug Monit. 27 (2005) 334–344. [22] H.S. Kim, J.C. Cheong, J.I. Lee, M.K. In, Rapid and sensitive determination of propofol glucuronide in hair by liquid chromatographic and tandem mass spectrometry, J. Pharm. Biomed. Anal. 85 (2013) 33–39. [23] J. Kim, S. In, Y. Park, M. Park, E. Kim, S. Lee, Quantitative analysis of propofolglucuronide in hair as a marker for propofol abuse, Anal. Bioanal. Chem. 405 (2013) 6807–6814. [24] J.G. Bollinger, W. Thompson, Y. Lai, R.C. Oslund, T.S. Hallstrand, M. Sadilek, F. Turecek, M.H. Gelb, Improved sensitivity mass spectrometric detection of eicosanoids by charge reversal derivatization, Anal. Chem. 82 (2010) 6790–6796. [25] S.J. Barry, R.M. Carr, S.J. Lane, W.J. Leavens, S. Monte´, I. Waterhouse, Derivatization for liquid chromatography/electrospray mass spectrometry: synthesis of pyridinium compounds and their amine and carboxylic acid derivatives, Rapid Commun. Mass Spectrom. 17 (2003) 603–620. [26] T. Santa, Derivatization reagents in liquid chromatography/electrospray ionization tandem mass spectrometry, Biomed. Chromatogr. 25 (2011) 1–10. [27] F. Beaudry, S.A. Gue´nette, A. Winterborn, J.F. Marier, P. Vachonm, Development of a rapid and sensitive LC-ESI/MS/MS assay for the quantification of propofol using a simple off-line dansyl chloride derivatization reaction to enhance signal intensity, J. Pharm. Biomed. Anal. 39 (2005) 411–417. [28] D. Thieme, H. Sachs, G. Schelling, C. Hornuss, Formation of the N-methylpyridinium ether derivative of propofol to improve sensitivity, specificity and reproducibility of its detection in blood by liquid chromatography-mass spectrometry, J. Chromatogr. B 877 (2009) 4055–4058. [29] F. Vaiano, F. Mari, F.P. Busardo`, E. Bertol, Enhancing the sensitivity of the Lc-MS/MS detection of propofol in urine and blood by azo-coupling derivatization, Anal Bioanal. Chem. 406 (2014) 3579–3587, http://dx.doi.org/10.1007/s00216-013-7573-y. [30] R.J. Levy, Clinical effects and lethal and forensic aspects of propofol, J. Forensic Sci. 56 (2011) S142–S147. [31] P.E. Marik, Propofol: therapeutic indications and side-effects, Curr. Pharm. Des. 10 (2004) 3639–3649. [32] P. Kintz, M. Villain, V. Dumestre, V. Cirimele, Evidence of addiction by anesthesiologists as documented by hair analysis, Forensic Sci. Int. 153 (2005) 81–84. [33] M. Strehier, J. Preuss, H. Wollersen, B. Madea, Lethal mixed intoxication with propofol in a medical layman, Arch. Kriminol. 217 (2006) 153–160. [34] E. Han, S. Jung, S. Baeck, S. Lee, H. Chung, Deaths from recreational use of propofol in Korea, Forensic Sci. Int. 233 (2013) 333–337, http://dx.doi.org/10.1016/j.forsciint.2013.10.008.