and the neutral drugs (diazepam, desmethyldiazepam, and methaqualone) from blood, while following the basic method commonly employed by most ...
Journal of Analytical Toxicology, Vol. 5, September/October 1981
A Rapid Procedure for the Screening and Quantitation of Barbiturates, Diazepam, Desmethyld,azepam and Methaqualone H. H o r t o n M c C u r d y , Larry J. L e w e l l e n , Joan C. C a g l e , and Everett T. S o l o m o n s
Department of Toxicology, Division of Forensic Sciences, P.O. Box 1456, Atlanta, Georgia 30371
Abstract A procedure is presented for the concomitant identification end quantitation of certain commonly abused week acid and neutral drugs from whole blood. This procedure offers several advantages over other procedures in that: (A) e means of extraction of the drugs from blood using a very Inexpensive source of diatomaceous earth (Celite ~ 560) is presented, (B) the addition of three internal standards allows the quantitation of several drugs from blood with Just one extraction of the blood sample, (C) the extracts are virtually free of any contaminants from biological sources or from biologically inert drug metabolites, and (D) the barbiturates and desmethyldlazepam are derivatized to enhance their limits of detection and reduce peak tailing frequently observed when using the nitrogenphosphorous detector and gas chromatography (NPD/GC). Some 1000 positive driving-under-theInfluence cases have been quantlteted by this method.
Introduction The analysis o f biological specimens for the presence of weak acid and neutral drugs is a common and integral part of any comprehensive forensic toxicology screening program. In the most commonly used method (1), a portion of the blood sample is extracted at physiological or slightly acidic pH with an organic solvent. The organic solvent is then extracted with an aqueous alkaline solution for the weak acids (barbiturates) leaving behind the neutral and weakly basic drugs (e.g., methaqualone, diazepam, desmethyldiazepam, meprobamate, etc.). The organic layer contains not only the so called "neutral" drugs, but also plasticizers from Vacutainer stoppers, fatty materials, and biologically inactive drug metabolites (particularly those of methaqualone). These various neutral components could serve to interfere with the accuracy, usefulness, and reliability of the analytical method. The procedure we present for the analysis of the barbituates and the neutral drugs (diazepam, desmethyldiazepam, and methaqualone) from blood, while following the basic method
commonly employed by most laboratories, differs markedly in a number of ways. For example, we do not employ the typical separatory funnel liquid/liquid extraction process. This precludes the necessity of washing large amounts of breakable glassware. Nor do we use any of the commercially available extraction apparatuses (e.g., ToxElute| We have, instead, investigated the use of a diatomaceous earth filtering aid (Celite| 560) for the extraction of acid, neutral, and strongly basic drugs from blood and urine. This filtering aid can be purchased economically in 50-pound bulk quantities which reduces the cost of an average commercially available drug extraction device from forty to fifty cents each, to less than two cents. While the expense of the commercial drug extraction devises are obviated, the convenience is retained. Another advantage is that screening and quantitation is performed using a single aliquot of specimen. This is accomplished with the use of three internal standards: (A) hexobarbital for the analysis of barbiturates; (B) ethaqualone for the analysis of methaqualone; and (C) prazepam for the quantitation of diazepam and/or desmethyldiazepam. The addition of these three internal standards gives the analyst very precise quantitative results. In addition, these internal standards for the gas chromatographic (GC) quantitation of the drug(s) do not interfere with the thin layer chromatographic (TLC) screening procedure. After the blood sample has been extracted and the drug(s) tentatively identified by TLC, a relatively specific extraction technique is employed which separates the parent, biologically relevant drugs and metabolites, while omitting the biologically inert and inactive metabolites. Procedures that include inactive metabolites in the quantitative analyses often lead to erroneously high results. This has been found to be particularly true in the case of methaqualone in our laboratory and by others (2,3) where ultraviolet (UV) spectroscopic procedures are employed for quantitation. The extraction method described herein is quite specific for methaqualone and excludes its biologically inert metabolites. It is known that the barbiturates, during the course of GC analysis, can suffer from drastic peak tailing, especially at low concentrations (4,5). Some of the more commonly used liquid phases frequently cause the desmethyldiazepam peak to also tail badly (OV-I) and/or be virtually insensitive to trace amounts of the drug (OV-17) (6). Therefore, both the barbiturates and desmethyldiazepam are derivatized by alkylation
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
253
Journal of Analytical Toxicology, Vol. 5, September/October 1981 before analysis to greatly enhance the limit of detectability and also the quantitative accuracy. The final advantage of the method is the utilization of the nitrogen sensitive detector (NPD) for the gas chromatographic quantitations for maximum sensitivity and selectivity. Because of the high sensitivity of the NPD, it is not necessary to evaporate or concentrate the solution to be analyzed, particularly in the case of the barbiturates. This serves to simplify the method. This procedure has been in use in our laboratory for over a year in the analysis of autopsy and driving-under-the-influence (DUI) cases. It has proven to be especially useful in the analysis of many DUI cases involving methaqualone and methaqualone-diazepam combinations. Some 1000 DUI cases involving methaqualone, either alone or in combination with other drugs, have been analyzed using this method.
Materials and M e t h o d s The polypropylene tubes (27 x 135 mm) used for the extractions are the same as those obtained from any of the several commercial sources (e.g., ToxElute| from Fisher Scientific, Pittsburgh, Pennsylvania). The cleaned tubes were repacked by first inserting a cotton ball (Fisher Scientific) into the bottom of the tube followed by approximately 6 g (or a 15 mL beaker-full) of Celite| 560 (Johns-Mansville Corp., Atlanta, Georgia). No prerinsing or washing of the column was required. The time required to pack twenty extraction columns with Celite| 560 was generally less than two minutes. The polypropylene tubes may be reused indefinitely.
Thin Layer Chromatography The TLC plates were Analtech Silica Gel GF, 250 micron, 20 x 20 cm, obtained from Analtech, Newark, Delaware. The mobile phase was chloroform/acetone (9:1). The solvents were prepared daily and allowed to equilibrate in a lined TLC tank for one-half hour before use.
Gas Chromatography The gas chromatograph was a Hewlett-Packard Model 5830A (Hewlett-Packard, Avondale, Pennsylvania), equipped with a nitrogen-phosphorous sensitive detector and a glass column, 120 cm x 2 mm i.d., containing 3% OV-I on 80/100 mesh chromosorb W-HP. Oven temperature was 165 ~ C for barbiturates and 225 ~ C for methaqualone, diazepam and desmethyldiazepam. Helium (carrier) was 30 mL/min. Hydrogen and air flows were 3 and 50 mL/min, respectively and collector voltage was 16V.
Solvents and Chemicals The solvents used (methylene chloride, hexanes, and N,Ndimethylacetamide) were of certified ACS quality and were obtained from Fisher Scientific. Ethaqualone (3-ortho-ethylphenyl-2-methyl-4(3H)-quinazolinone) was obtained from Alfred Bader Library of Rare Chemicals, Milwaukee, Wisconsin through Aldrich Chemical Company. Hexobarbital and other barbiturates were purchased through Applied Science Division, Milton Roy Company, State College, Pennsylvania. Diazepam and desmethyldiazepam were gifts
254
from Hoffmann LaRoche, Inc., Nutley, New Jersey and prazepam was a gift from Warner-Lambert Company, Morris Plains, New Jersey. Trimethylanilinium hydroxide (TMAH, 0.2 mol/L) was prepared in our laboratory according to a previously reported procedure (5). To 5 mL of the TMAH solution was added 43 mg of tetrabutylammonium hydrogen sulfate (TBAHS, Aldrich Chemical Company) to make 0.025 mol/L TBAHS in 0.2 mol/L TMAH. Methaqualone was a gift from William H. Rorer, Inc., Research Division, Fort Washington, Pennsylvania. All other chemicals were obtained from Fisher Scientific.
Reference Solutions Stock solutions of methaqualone, diazepam, desmethyldiazepam, amobarbital, secobarbital, pentobarbital, phenobarbital, butabarbital, and butalbital were prepared in ethanol at concentrations of 1 mg/mL. To each of three 100 mL volumetric flasks were added a sufficient amount of methaqualone and each of the barbiturate stock solutions to prepare 2, 5, and 10 mg/L solutions and a sufficient amount of the diazepam and desmethyldiazepam stock solutions were added to each of the flasks to prepare 0.5, 1.0, and 2.0 mg/L solutions. The ethanol in each flask was evaporated to dryness and drugfree ~vhole blood added to bring each flask to volume.
Internal Standards Approximately 15 mg of ethaqualone, 45 mg of hexobarbital, and 25 mg of prazepam were added to a one liter flask, dissolved in about 100 mL of ethanol and diluted to volume with water. One milliliter of internal standard solution was added per 5 mL of blood sample to be analyzed. Although 5 mL was the usual sample size used, it was possible to use a much smaller volume (e.g., l - 2 mL).
P roced u re Five milliliters of blood were measured into a 10 mL graduated cylinder, one milliliter of internal standard solution added and the mixture vortexed. This mixture was poured into a polypropylene tube (27 x 135 mm) containing approximately 6 g of Celite | 560. The graduated cylinder was rinsed with 2 mL of H20 and added to the Celite| 560 and blood sample mixture. To the tube was added 50 mL of methylene chloride (CH2CIz) in 10 mL portions (or all at once if desired) and the eluent collected into a 50 mL beaker. The methylene chloride was evaporated to near dryness and divided into approximately equal portions for TLC and for the quantitative analysis procedure. A thin layer plate was spotted, developed, dried thoroughly and observed under short wave UV light. The presence of methaqualone was noted at an approximate Rf of 0.63 and also the ethaqualone internal standard spot just above methaqualone with Rf of 0.66. The plate was then sprayed with HgSO4 solution and the plate observed under long wave UV light for the presence of diazepam (Rf -- 0.49) and desmethyldiazepam (Rf = 0.20) as yellow fluorescent spots. The plate was then sprayed with diphenylcarbazone solution for the presence of barbiturates, ethchlorvynol (Rf = 0.66) and glutethimide (Rf -- 0.60). The TLC plate, after heating in a 150~ C oven for one minute and while still warm, was sprayed with 1~ vanillin in concentrated H2SO4 solution
Journal of Analytical Toxicology, Vol. 5, S e p t e m b e r / O c t o b e r 1981
for the presence o f meprobamate (Rf = 0.05). Those cases which contain any combination of barbiturates, methaqualone, diazepam, and desmethyldiazepam were carried through the following extraction and derivatization procedure and were quantitated directly. While other neutral drugs such as ethchlorvynol or meprobamate can be detected by this procedure, it was necessary to quantitate these drugs by other appropriate methods. To the CHzClz portion reserved for quantitation of the positive cases, was added 5 mL of CH2Clz and transferred to a screw-capped test tube (16 x 125 mm). The organic layer was extracted for barbiturates with 3 mL 0.1 mol/L KOH. This mixture was centrifuged and the aqueous layer transferred to another tube. The organic layer was saved for the analysis of methaqualone, diazepam, and desmethyldiazepam. The alkaline aqueous layer was acidified with concentrated HCI (promptly, to avoid the possible hydrolysis of hexobarbital), 5 mL CH2C12 was added, vortexed, centrifuged and the aqueous layer aspirated and discarded. To the CH2CIz layer, a small portion of anhydrous Na2SO 4 was added, vortexed, centrifuged, and the CHzC12 decanted into another clean tube. The methylene chloride layer was evaporated to dryness. To the residue was added 200/~L of N,N-dimethylacetamide, 25 /~L TBAHS in T M A H solution, and 50 /~L ethyl iodide (or alternatively butyl iodide). The mixture was vortexed briefly and allowed to stand at room temperature for 2-3 minutes. To the mixture, 3 mL of hexanes was added, vortexed briefly and then the hexanes washed with 3 mL o f 0.1 m o l / L KOH. After centrifuging, approximately 1 /aL o f the hexanes solution was injected onto the gas chromatograph. The peak height ratios were compared to those o f the working standard curve (2,5, and 10 mg/L) prepared simultaneously. If the presence of barbiturates was negative by TLC, the initial KOH extract was discarded. The CH2Cl~ layer in either case was evaporated to dryness. The addition of a boiling bead and gentle shaking of the tube facilitated the evaporation of methylene chloride without bumping. Desmethyldiazepam was derivatized by the addition of 200/~L N,N-dimethylacetamide, 100/~L TBAHS in T M A H solution, and 100/~L 1-iodopropane to the residue and heated on a 100 ~ C water bath for 5 or 6 minutes. To the cooled tube, 3 mL of hexanes was added and vortexed. The hexanes layer was washed with 3 mL of 0.1 m o l / L KOH, centrifuged and the hexanes layer was then transferred to a clean tube. Methaqualone, diazepam, and the propyl derivative of desmethyldiazepam plus the internal standards were extracted with 3 mL of 1 m o l / L HCI. The solutions were vortexed, centrifuged, and the hexanes layer aspirated and discarded. To the aqueous layer was added 3 mL of CHzClz, the solutions vortexed and the aqueous layer aspirated and discarded. The CHzCI2 was dried with a small amount of anhydrous Na2SO 4 as above. The CH2C12 was decanted into a clean test tube and the solution evaporated to dryness. The residue was dissolved in 100/~L of ethanol and 1 /~L was injected onto the gas chromatograph. The peak area ratios were compared to those of the working standard curves prepared simultaneously: 2, 5, and 10 mg/L for methaqualone and 0.5, 1.0, and 2.0 m g / L for diazepam and desmethyldiazepam.
Results and Discussion The use of Celite| 560 diatomaceous earth in place o f the
considerably more expensive commercial extraction devices has proven to be equally reliable and simple. The difference in quantitative data obtained from the commercial apparatus as opposed to those obtained from our " h o m e m a d e " version was insignificant. Table I compares the performance of a commercial extraction column that uses Chromasorb W diatomaceous earth 40/60 mesh as the extraction media to that of the modification using Celite | 560 diatomaceous earth. The Celite | 560 has also performed well in the extraction o f the strongly basic and amphoteric drugs from blood and/or urine. This procedure takes advantage of the fact that hydrochloride salts of certain drugs such as methaqualone and diazepam, as well as the chosen internal standards, are methylene chloride soluble. Derivatization of desmethyldiazepam also renders this drug methylene chloride soluble as its hydrochloride salt. The biologically inactive metabolites of methaqualone are not extracted since their hydrochloride salts are not soluble in methylene chloride and thus cannot interfere with the subsequent gas chromatography of the drugs.
Table I. Comparison of quantitative results of three drugs from a commercial extraction column containing Chromosorb W diatomaceous earth vs Celite | 560 "homemade" version Drug Methaqualone Diazepam Pentobarbital
Conc. (mg/L)
n
2.0 0.5 2.0
10 8 10
Commercial Column* CV 2.02 0.51 2.00
8.5% 7.2% 7.1%
Celite~ 560"*
CV
1.99 0.52 1.97
5.0% 6.8% 7.6%
r *JETUBEV(#1020), Har-Len& Associates,Pittsburgh,PA.; meanconcentration (mgl L). **Mean concentration(mg/L).
The scheme for the extraction of the weak acid and neutral drugs is shown in Figure 1. A typical gas chromatogram of methaqualone, diazepam and the p r o w l derivative of desmethyldiazepam is shown in Figure 2. The limit of detection for the barbiturates is greatly enhanced by derivatization by alkylation for detection by G C / N P D . Ethylation with ethyl iodide rather than methylation with methyl iodide or any other alkyl halide derivative gave better separation of the various barbiturates from the internal standard, hexobarbital. However, derivatization of the barbiturates with butyl iodide gave good separation between butabarbital and butalbital at al slightly higher column temperature (190 ~ C) with the internal standard eluting between the am| and pent| peaks. The use of the NPD for barbiturates also avoids any evaporation and concentration step of the very volatile alkyl derivatives. Table II shows the precision data obtained for the barbiturates from an average of twenty determinations over a period of several months. Table III shows the precision data obtained for methaqualone (n = 25) and diazepam and desmethyldiazepam (n = 20) over the same period of time. The limits of detection for the drugs were well within the ranges necessary for detection of therapeutic dosages. The barbiturates, as well as methaqualone, were easily detected as low as 0.2 mg/L. The limit of detection for diazepam and desmethyldiazepam as previously reported (6) was 0.01 mg/L. 255
Journal of Analytical Toxicology, Vot. 5, September/October 1981
Conclusion
Residue after TLC
5 mL CH2Cl2 16 X 125 mm screw-capped test tubes 3 mL 0.1 mol/L KOH
CH2CI2
V, C & T (aqueous) Ev CH2CI 2 jf
Acidify Aqueous
200/~L DMA 100/~L TBAHS/TMAH 100/zL 1-1odopropane cap tube & heat 5 mln.
3 mL CH2CI 2 V, C, & A (aqueous)
Anhydrous Na2SO4, V, C Decant CH2Cl 2, & Ev CH2Cl 2
3 mL Hexenes & V ~f
f
3 mL 0.1 mol/L KOH
200/~L DMA 25/~L TBAHS/TMAH 50/d. Ethyl Iodide
A method is presented for a rather comprehensive screening and quantitation technique for certain commonly abused weak acid and neutral drugs from blood. Furthermore, we have shown conclusively that drug extraction devices that employ a diatomaceous earth, whether purchased premade or "homemade" extraction columns using Celite| 560 diatomaceous earth can be used exceedingly well for quantitation as well as qualitative work. Thus, with the addition of the appropriate internal standards, both qualitative and quantitative analysis is accomplished with just one extraction of the blood specimen. This method has been particularly useful in the analysis of driving-under-the-influence cases. Our results of some 1000 positive DUI cases using this method will be the subject of a separate article.
V, C, & T (Hexanes) /
l
Table II. Precision data for the various barbiturates over a period of 4 months
3 mL 1 mol/L HCI
/
V & let stand for 2-3 min. ,~
V, C, & A (Hexanes)
!
Barbiturate
3 mL Hexanes, V & C
Concentration mg/L
n
Mean Value mg/L
CV
2 2 2 2 10 10 10 10
20 20 20 20 20 20 20 20
1.98 1.99 1.99 2.08 10.3 10.3 10.4 10.2
6.6 % 7.0% 7.0% 6.5 % 6.1% 6.2% 5.3% 5.0%
3 mL CH2Cl 2 3 mL 0.1 mol/L KOH
V, C, & A (aqueous) Anhydrous Ns2SO4, V, C &decant& EvCH2CI2
V&C
Inject 1 pLof Hexanes Layer on GC/NPD (Barbiturates)
100 p.L EtOH & Inject 1 ,~L onto GC/NPD (Methaqualone, Dlazepern and Desmethyldlazepam)
Figure 1. Schematicfor the extractionand derivatizationof the weak acid and neutral drugs. Abbreviationsused are: V = Vortex, C = Centrifuge,A = Aspirate (and discard), T = Transfer(to a clean tube), Ev = Evaporate(to dryness), and DMA = N,N-Dimethylacetamide.
Amobarbital Pentobarbital Secobarbital Phenobarbital Amobarbital Pentobarbital Secobarbital Phenobarbital
Table III. Precision data obtained for the neutral drugs over a period of 4 months Drug
ME
ME
I
Concentration (mg/L)
n
Mean Value mg/L
CV
25 25 20 20 20 20
2.00 9.95 0.51 2.00 0.52 2.03
7.3% 6.3% 6.9% 4.3% 9.3% 9.3%
II Methaqual0ne Methaqual0ne 0iazepam Diazepam Desmethyldiazepam Desmethyldiazepam
Dz
2 10 0.5 2.0 0.5 2.0
Acknowledgement
t
DM Dz
The authors gratefully acknowledge the assistance of Lynette M. McCurdy in the preparation of this manuscript.
References
s
r8
i
,
,
t
4 5678 123 Time (rain.) Figure 2. Chromatogramsof (I) blood referencesolutioncontaining 5 mg/L methaqualone(M), 1.0 mg/L diazepam(Dz), and 1.0 mg/L desmethyldiazepam (DMDz, prowl derivative)with internal standards ethaqualone(E) and prazepam(P); and (11)actual 0Ul case containingthe samedrugs. 256
1. E.G.C. Clarke (Ed). Isolation and Identification of Drugs, Vol. 1. Pharmaceutical Press, London, pp. 24-26, 1974. 2. D.N. Bailey and P.I. Jatlow. Methaqualone overdose: Analytical methodology and significance of serum drug concentrations. C/in. Chem. 19:615-20 (1973). 3. L. Kazyak, R.M. Anthony, and I. Sunshine. Methaqualone
Journal of Analytical Toxicology, VoI. 5, September/October 1981 metabolites and their significance in acute intoxication. J. Anal. Toxicol. 3:67-71 (1979). 4. B. Welton. Analysis of nanogram levels of barbiturates. Chromatographia, 3:211-15 (1970). 5. E. Brochmann-Hanssen and T.O. Oke. Gas chromatography of barbiturates, phenolic-alkaloids, and xanthine
bases: Flash-heater methylation by means of trimethylanilinium hydroxide. J. Pharm. Sci. 58:370-71 (1969). 6. H.H. McCurdy, J.C. Harrill, and E.L. Slightom. Diazepam and Desmethyldiazepam quantitation in whole blood by nitrogen sensitive detector after derivatization. J. Anal. Toxicol. 3:195-98 (1979).
Manuscript received May 29, 1981; revision received July 29, 1981.
i
i :ii
~: ~
i
~
:
............
9 : :Letters to the Editor :i To the Editors, We must address the provocative information disclosed by Dr. Bergman et al., in a recent article: The Detection of Tetrahydrocannabinol in Blood: A Comparative Study (5:85-89). This is essential to all research workers concerned with marijuana. The authors' attentive and unbiased evaluation of two cannabinoid antisera is to be commended. However, it is unfortunate that some of their conclusions may produce confusion rather than clarity. The title of Table III, "Cross-Reactivity of Immunalysis and Guildhay Antisera to Drug-Positive Blood Samples" certainly is misleading. From anecdotal information that "No history of cannabis use surrounded any of the samples" the authors' conclude: A. there is major crossreactivity of the marijuana antisera with (all) the listed drugs, and B. /',9-THC RIA is a qualitative test. It has been well established in the scientific literature that RIA is a quantitative procedure. Life and death decisions world-wide depend on serum RIA quantification performed b~_qualified technologists. It has also been established that a THC antiserum (lmmunatysis) fails to cross react even with a cannabinoid, differing in structure by one hydroxyl group (1), Cannabidiol. The numerous drugs in the authors' Table III are structurally very different from 69-THC. Therefore, a priori it would seem reasonable that these dissimilar molecules in vast excess would not be recognized by Ag-THC antibodies. Appreciating the significance raised by this important work, it was essential to learn experimentally if increasing amounts of suspected interfering drugs alter known RIA mass/volume quantification. We performed routine crossreactivity experiments with: Amitriptyline, Propoxyphene and Oxazepam. We also tested these drugs with our 9SM (9 substituted cannabinoid) antiserum (2). The latter hasproved useful to confirm marginal (2-5 ng/mL) 6g-THC blood concentration, being 6-to 10-fold higher (2). Massive (3000 ng/mL) excess (200-to 600-fold) of these heterologous molecules failed to interfere with binding of A9-THC to homologous antiserum (Figure 1). In far more rigorous experiments, hemolyzed blood with a fixed amount of ~9-THC (15 ng/mL) and up to 3000-fold (48 Fg/mL) beyond the authors' concentrations of heterologous drugs were tested. These data confirm unambiguously absence of interference with the ~9-THC RIA (Figure 2). Decomposition and particulate matter in blood samples described by the authors as an "endogenous or putrificative component" could interfere with any RIA by adsorption of ligand or degradation of antibody. Unless treated to remove such contaminants, such samples may be unsuitable for RIA per se.
"
99
~_:--:-_-."":'~;'~:'~;'~'::.=:"_L".~;'~"A .............. 9 OXAZEPAM* (IO2ng/ml) c- . . . . . . . . . . . . . . .
-c--. . . .
-PROPOXYPHENE (IO2ng/ml} "~AMITRIPTYLINE (IO2 ng/ml)
.
