GC-MS Confirmation of Codeine, Morphine, 6 … · 6-Acetylmorphine, Hydrocodone, Hydromorphone,...

Journal of Analytical Toxicology, Vol. 23, May/June 1999

GC-MS Confirmation of Codeine, Morphine, 6-Acetylmorphine, Hydrocodone, Hydromorphone, Oxycodone, and Oxymorphone in Urine* Robert Meatherall Department of Biochemistry, St. Boniface General Hospital, 409 Tach# Avenue, Winnipeg, Manitoba, Canada, R2H 2A6

Abstract ]

A procedure for the simultaneous confirmation of codeine, morphine, 6-acetylmorphine, hydrocodone, hydromorphone, oxycodone, and oxymorphone in urine specimens by gas chromatography-mass spectrometry (GC-MS) is described. After the addition of nalorphine and naltrexone as the two internal standards, the urine is hydrolyzed overnight with [~-glucuronidase from E. coli. The urine is adjusted to pH 9 and extracted with 8% trifluoroethanol in methylene dichloride. After evaporating the organic, the residue is sequentially derivatized with 2% methoxyamine in pyridine, then with propionic anhydride. The ketone groups on hydrocodone, hydromorphone, oxycodone, oxymorphone, and naltrexone are converted to their respective methoximes. Available hydroxyl groups on the O3 and O6 posilions are converted to propionic esters. After a brief purification step, the extracts are analyzed by GC-MS using full scan electron impact ionization. Nalorphine is used as the internal standard for codeine, morphine, and 6-acetylmorphine; naltrexone is used as the internal standard for the 6-keto-opioids. The method is linear to 2000 ng/ml. for the 6-keto-opioids and to 5000 ng/mL for the others. The limit of quantitation is 25 ng/mL in hydrolyzed urine. Day-to-day precision at 300 and 1500 ng/mL ranged between 6 and 10.9%. The coefficients of variation for 6-acetylmorphine were 12% at both 30 and 150 ng/mL A list of 38 other basic drugs or metabolites detected by this method is tabulated.

Introduction

Much attention has been directed to the confirmation of morphine and codeine in urine by gas chromatography-mass spectrometry (GC-MS). A few methods have been developed specifically for the analysis of 6-acetylmorphine (6-AM). Some have included the analysis of 6-AM with morphine and codeine because all three drugs are often present after heroin use (1). Wasels and Belleville (2) compared these methods in their review article.

GC-MS confirmatory methods proceed by hydrolyzing the glucuronide opioid conjugates with ~-glucuronidase or acid hydrol-

* Presented at the Society of Forensic Toxicologists Annual Meeting, Albuquerque, NM, October 1997.

ysis. The opioids are recovered from the urine by liquid-liquid extraction or by solid-phase extraction at pH 9. After the solvent is evaporated, the residue is derivatized with a silylating reagent or an acid anhydride to convert the opioids to volatile compounds that are amenable to GC. Either a deuterated homologue or an opioid analogue is used as the internal standard. The silyl derivatives (3,4) and the perfluoroesters (3-5) of morphine and codeine are moisture sensitive or unstable, making their preparation and handling more difficult. By contrast, the acetyl (3-5) and propi- onyl (1) esters are stable in the presence of water. They facilitate a postderivatization purification step before injection onto the GC-MS. Kushnir and Urry (6) evaluated propionic anhydride, MBTFA, HFAA, and BSTFA for GC-MS analysis of 6-AM. They concluded that propionic anhydride gave accurate, precise, and sensitive results while providing compatibility with other methods on the same GC-MS instrument. Residual derivatizing reagent in the injector will react with drugs in other methods not intended for derivatization.

Few methods deal with the quantitation of the 6-keto-opioids, hydrocodone, hydromorphone, oxycodone, and oxymorphone, either as a group (7) or combined with codeine and morphine (4,8,9). The enol tautomer of the 6-keto-opioids form esters with acetylating agents and ethers with silylating agents. Two end products are commonly formed from each keto-opioid, one from the end isomer and one from the unreacted keto isomer. Soper et a]. (9) used propionic anhydride in pyridine (1:1) at 40~ for I h to derivatize codeine, morphine, 6-AM, hydrocodone, oxycodone, and hydromorphone. At this temperature, the keto/enol ratios were 9.1, 62.5, and 64.1, respectively, for the 6-keto-opioids hydrocodone, oxycodone, and hydromorphone. These derivatives separate well from each other and allow the detection of 6-AM in the presence of large amounts of morphine. However, the formation of two derivatives in significant proportions is undesirable when performing quantitations.

In the author's experience, the underivatized 6-keto-opioids can be thermally decomposed in the GC injector, which produces day-to-day coefficients of variation (CV) in excess of 25%. In order to quantitate them, a second derivatization step is needed to selectively react with the ketone functional group. Broussard (10) performed sequential derivatization reactions involving hydroxylamine and then BSTFA. Codeine, morphine, hydrocodone,

Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. 177


hydromorphone, and oxycodone derivatives separated from each other. Oxymorphone co-eluted with oxycodone, and the retention time of 6-AM was not mentioned. The method described herein employs sequential derivatization and employs two previously described reactions. Monti et al. (11) used methoxyamine to derivatize the ketone group in naltrexone. Paul et al. (1) used propionic anhydride to convert 6-AM to 6-AM-3-propionate, and later O'Neal et al. (12) used it in their analysis of acetylcodeine, codeine, heroin, 6-AM, morphine, and norcodeine in urine.

In the procedure presented here, the urine is first treated with [3-glucuronidase to hydrolyze the opiate glucuronides. After alkalinizing to pH 9.0 with sodium bicarbonate, the drugs are recovered by liquid extraction. The organic is evaporated, and the drugs are subjected to two derivitization steps performed sequentially. First, the ketone group at the 06 position of hydrocodone, hydromorphone, oxycodone, oxymorphone, and naltrexone are reacted at room temperature for 15 min with 2% methoxyamine HC] in pyridine to convert these drugs to their respective methoxime derivatives. Next, propionic anhydride is combined with the pyridine and heated at 56~ for 15 min. The hydroxyl groups at the 03 position of morphine, 6-AM, hydromorphone, oxymorphone, nalorphine, and naltrexone and at the 06 position of codeine, morphine, and nalorphine are converted to their respective propiony! esters. The reaction is summarized in Figure 1 using hydromorphone as the example. After evaporating the excess derivatizing reagents, the reaction product is purified by adding ammonium hydroxide and extracting with an organic solvent. The organic is evaporated, the residue reconstituted in ethyl acetate, and 1 pL injected onto the GC-MS.

Materials and Methods

Morphine, 6-AM, hydrocodone, hydromorphone, oxycodone, oxymorphone, nalorphine and naltrexone hydrochlorides, and codeine phosphate were obtained from Health and Welfare (Ottawa) Canada. Propionic anhydride and methoxyamine hydrochloride were purchased from Aldrich Chemical Co. (Milwaukee, WI). Hydrolysis was performed with [3-glucuronidase, type IX-A from E. Coli, which was purchased from Sigma Chemical Co. (St. Louis, MO). The lyophylized enzyme was reconstituted with 0.1M phosphate buffer (pH 6.8) to give a solution containing a presumed activity of 100,000 units/mL.

Individual stock drug solutions were prepared in methanol at 1000 Bg/mL as the free base concentration. Stock standards were

used to prepare combined working solutions. Linearity and limit of detection experiments were performed on combined urine- based standards containing 10,000, 5000, 2000, 1000, 500, 200, 100, 50, 25, 10, and 5 ng/mL of each analyte. The assay was rou- tinely calibrated with a urine standard containing 600 ng/mL for all analytes except for 6-AM, which was calibrated at 60 ng/mL. Precision and recovery experiments were conducted with two combined urine standards, one containing 300 ng/mL (30 ng/mL for 6-AM) and one containing 1500 ng/mL (150 ng/mL for 6-AM). Aliquots were stored frozen in glass screw cap tubes at-70~ The combined internal standard nalorphine plus naltrexone at 100 IJg/mL was prepared by diluting aliquots of the methanolic stocks into water.

Sample preparation The following were combined in a tefl0n-lined, screw-capped

10-mL glass extraction tube: I mL urine, 5 pL internal standard containing nalorphine plus naltrexone (each 100 ]Jg/mL), 100 luL of 1M phosphate buffer (pH 6.8), and 50 ~L of E. Coli [3-glucuronidase 10,000 units/mL. The urine was hydrolyzed in a 37~ waterbath for 16 h (overnight). The urine was alkalinized by adding 800 gL of 1M NaHCO3 (pH 9) and extracted into 3 mL of methylene dichloride/trifluoroethanol (92:8) on a mixing wheel for 15 rain. After brief centrifuging, the upper aqueous phase was aspirated to waste. Residual water was removed by adding about 100 mg of anhydrous Na2SO4 and briefly vortex mixing. The organic was then transferred to a 5-mL conical tube and evaporated under a gentle nitrogen stream in a 40~ sand bath.

Derivatization The dried residue was reconstituted in 25 pL of pyridine con-

taining 2% methoxyamine HC1. The ketone substituents in hydrocodone, hydromorphone, oxycodone, oxymorphone, and naltrexone react with the methoxyamine at room temperature. After 15 rain, 25 IJL of propionic anhydride was added, and the tube was mixed and heated in a water bath at 56~ for 15 rain. The excess reagents were evaporated under a gentle nitrogen stream in a 40~ sandbath. The derivatized residue was further purified by adding 1 mL of hexane/chloroform (3:1) and 100 lJL of 15% ammonium hydroxide (conc. NH4OH/H20, 1:1) and by vortex mixing for i min. The tube was briefly centrifuged, and the bottom aqueous layer was withdrawn from the cone tip using a 100-pL Hamilton syringe. The organic was evaporated with nitrogen. The residue was reconstituted with 50 pL of ethylace- tate, and 1 IlL was injected onto the GC-MS.

CH=ONH= �9 HCI In pyridine

O

~ ="N-CH= 22~ - 15 rnin N ~

CH,.O-

/-

-CH 3

/

Figure 1. Reaction of hydromorphone with methoxyamine followed by propionic anhydride.

O O CH~CH=C ~ CH~CHzC'O

56~ - 15 min ~'- O

CH=O-N ~

~ N - C H 3

178


GC-MS An ITS40 ion-trap GC-MS controlled by Magnum software

(Finnigan MAT, San Jose, CA) was used in the electron impact ionization mode. Mass spectra were collected by scanning 55--550 amu at 1 scan/s. The mass defect was 100 mamu.

A DB-1 column (15 m x 0.25 mm x 0.25-pro methylsilicone film, J&W Scientific, Folsom, CA) was used for the chromatographic separation. A 1-m x 0.5-mm retention gap deactivated with 5% phenylmethylsilicone was joined at the column inlet with a VuUnion | (Restek Corp., Bellefonte, PA) and to the on- column injector. The helium carrier gas flow was 60 cm/s. The oven was initially held at 140~ for I rain, rapidly heated at 50~ to 220~ gradually heated at 10~ to 290~ where the temperature was held for 0.5 rain. The septum programmable injector was initially held at 130~ for 0.5 min, then rapidly heated at 100~ to 290~ where it was held for 8 rain.

Results

Figure 2 is the total ion chromatogram from a 1500-ng/mL urine standard. There is good chromatographic separation between each of the opioids except for oxycodone and codeine, which elute at 322 and 327 s, respectively, and are 75% baseline resolved. A small amount of codeine might be difficult to quantitate in the presence of a large amount of oxycodone, although the two drugs do not have major common ions. The large peak at 300 s is a phthalate originating from the pyridine. This urine standard has few extraneous peaks. Oftentimes urine matrix peaks are observed at 360, 375, 415, and 440 s. Hydrocodone and oxycodone display two distinct peaks separated by 4 s. For each drug, both components have identical mass spectra; the early eluting peak represents approximately 20% of the total area. The two components are probably the syn- and the anti- methoxime isomers. The corresponding isomers have been described (13) for the methoxime pentafluoropropionate derivative of naltrexone (11). The two isomers are not resolved for the later eluting 6-keto-opi-

oids: hydromorphone, oxymorphone, and naltrexone. When performing quantitations, the peaks corresponding to their quanti- ration ions were observed to have broader bases, which supports the concept that the isomers also exist but are just poorly separated from each other. In performing the quantitations, the area of both isomers were included in the peak integration. During the method development, the methoxyamination and the propionylation reactions were systematically varied from 2 to 40 rain at 20, 37, 56, 70, and 100~ The ratio of the isomers remained constant for each of these conditions.

The mass spectra of the opiate derivatives are shown in Figure 3. All spectra are unique, having at least three characteristic ions in addition to a predominant parent molecular ion. The M +, ion ranged in intensity from 20% for morphine and hydromorphone to 100% for codeine and oxycodone. The corresponding molecular structures of the derivatives are superimposed on the mass spectra. The assignment of the molecular weight, and hence the proposed structure, was supported by mass spectra (not shown) that were obtained using chemical ionization. The reagent gas was 5% ammonia in methane. Spectra were collected at m/z 50 to 650 at 1 scan/s. The mass spectra have previously been reported for morphine O3pO6p (14) and for 6-AM O3p (1,14) and agree with the spectra shown in Figure 3.

The retention times and EI mass spectra for an additional 38 basic drugs and metabolites are listed in Table I. Where multiple derivatives are possible, they are indicated by (minor) to mean that they are present at less than 10% of the (major) derivative product. Heroin and the oxycodone and codeine derivatives elute within 8 s. Each has its characteristic mass spectrum with no common ions; however, it would be difficult to identify a small amount of any of the three in the presence of a large amount of either of the other two. After heroin use, a large amount of morphine and a small amount of 6-AM appear in the urine. Because morphine elutes 30 s later than 6-AM, large amounts do not interfere with the detection of 6-AM.

The performance characteristics are summarized in Table II. The retention times and ions used for the quantitations are listed under each analyte. The quantitation ion is not always the base

1BBy.

" 4

TO1

I l l W l l e W V

1

3 4

6

- J~ .J'r',,

9 8

i i | i i I w i v ~ I i i i = ! | i i w

1BB 2BB 36B 4BB SBB Retention time (s)

Figure 2. Total ion chromatogram (55-550 amu) from a spiked urine standard. Peak identification: 1, hydrocodone 1500 ng/mL; 2, oxycodone 1500 ng/mL; 3, codeine 1500 ng/mL; 4, hydromorphone 1500 ng/mL; 5, 6-acetylmorphine 150 ng/mL; 6, oxymorphone 1500 ng/mL; 7, morphine 1500 ng/mL; 8, nalorphine IS 500 ng/mL; 9, naltrexone IS 500 ng/mL.

179


peak ion. The ion was selected imperically to provide the more precise result. Nalorphine and naltrexone quantitation ions are 367 and 370 amu, respectively. Natorphine was used as the internal standard for codeine, morphine, and 6-AM. Naltrexone was used as the internal standard for hydrocodone, hydromorphone, oxycodone, and oxymorphone because of the common ketone group at the C6 position. The day-to-day precision data were gathered over 4 months. A single-point calibration using the 600-ng/mL standard (containing 60 ng/mL of 6-AM) was used for each run. Except for 6-AM, the CVs for all six opiates were below 11% at both 300 ng/mL and 1500 ng/mL. The CV for 6-AM was 12% at both 30 and 150 ng/mL. For comparison, the CVs for hydrocodone, hydromorphone, oxycodone, and oxymorphone were also calculated using nalorphine rather than naltrexone as the internal standard. At 300 ng/mL, the CVs increased to 9.6, 11.3, 13.0, and 11.3%, respectively. At 1500 ng/mL, the CVs increased to 10.6, 13.0, 11.7,

and 12.8%, respectively. The modest increases suggest that the methoxyamination reaction is well controlled. In addition, the extraction and propionylation conditions do not discriminate the keto-opioids. Both nalorphine and naltrexone are opiate antago- nists and are administered predominantly by medical personnel in emergency rooms. If hospital records show that either antagonist was administered before urine collection or if the ratio of nalorphine to naltrexone was different than expected, an alternative internal standard is necessary.

The extraction efficiencies were determined by spiking five drug-free urines at 300 ng/mL (30 ng/mL for 6-AM) and another five urines at 1500 ng/mL (150 ng/mL for 6-AM). The urines were extracted without the addition of internal standards. The internal standards were added to the extracts prior to evaporating the solvent. No corrections were made for solvent loss during sample preparation. Urine matrix extracts representing

18@x I'[),drocodone Scan 291 297

82 11(: 105 Z14 249 271 313

i . I . i . i . l . i . I , i . l , I , I , i . i , J . i , i , ' I t I , l , i ' I ' i . l . l ' l ' I ' i . I , p l ' I . i . 1 . 1 . 1 ' I ' i , i . p i , i . i , i , i . i , ~ , i , i . i , i , 58 188 158 Z88 ZS8 300 358 4 8 e 4 5 8 58e 558

n V z

i e ~ O~codo.e S~tu 322

76 313

115 238 2 5 6 287

58 lO0 1S8 288 2S8 300

m/z

) . r

3 4 4

. i , l , 1,1, i , i , l , i . l , P l . l , l , p i , J, l , l , i , i , i ' I 3SO 488 4S8 560 ~;~6

M" 1 0 8 x C o d e i n e S ~ t n 327 355

N.c.,

r

Z29

115 102 Z l j ~ 2 B

i, i , i , i , l , i , [ , i , ) ,1 ,1, I . 1,1,1 , i , 1,4. i . l , l r l , l . l . l , l . l . i , i , i , I , I , I , i , i , I , I . I , i . i . I , [ , t , p I. I , i , i , i . i , I $8 188 158 Z ~ ZS8 388 358 488 458 58B 5,~6

m/z leez Hydromorpbone Scum 348 314 9

2S? 339 378

OZ 189 2J.4 Z421 157 b~J~J,I, J . ~ k , tk ,a ac xJ,l~. ,~ J~ i,~ �9 t J~ 480 44Z 494 SZ8 l ' ) ' [ ' , ' , ' ~ " , ' , ' ~ ' , ' ~ ' ~ " , ' ; " ~ ' l ' H ' , �9 I ' i ' ; ' i ' " ' l ' " F ' ) * " l ; ~ ' ~ " t ' t ' ] " ' ' " ; " l " ' ' " " " l ' ' ' ' ' ' ' ' ' ~ 5B 108 158 ZOO 2~8 38D 3SB 468 451] 598 $58

m/z leS~ (PAcetyImo rpb ine Scan 366 327

" j . l , l , l , l r l , [ r l , l , i . p l , [ , i , l , l , l , l , l , l , lU 'P lqr l . IH,~ l . [ , I , I " I , I , I , I , ~ . l , I , I , l , l , l , l . l , I , I , l ' I

~40 [ 0 9 IBO ZOO ZSO 3~J 3~0 4~8 4S8 $90 ~B9

r ~

lm~ Oz3'uerpboae So. 38s 33e c ~ ' - ~

o N N ~ ,

I ' i , i . i , i ' 1 , i , i , i , i ' 1 ' i . i . i . i , i , I , i . i , I ' 1 ' i , i , l , i ' 1 ' i , i , i �9 i , i , i , i , l , i " 1 " i , i , i , i , I . i . i . 1 - 1 " 1 ' J , l , i . i "1 5 8 1 8 4 9 1 5 8 Z B B 2 5 8 3 8 4 9 3 5 8 4 0 0 4 S O 5 8 8 $ 5 8

m/z 106",r Morpbiae Scan 39T 341 Q

/ N . ~ ,

ltll,, l-LL " ~ ' , ' . , ; , . r ;v, . . . . . . . i.,: . . . . . . ,~,.,.,~,.,.~.,.~_,..,442 . . . . . . . . . , . . . . . . . . . , so ~ge ~se zoo aso 3oe 3so 4Be ,so sse sse

m/z le@x N l l o rph ine ~ t i s 44S 367

I ~ l ~ l ~ I ~ P ~ ' ~ l ~ P l ~ [ ; ~ i ` l ~ ) ~ l ~ ` ~ ` ~ l q ~ 1 ~ l ~ l ` l ~ l ~ l ` l ~ . 1 ~ t ~ t ~ l . ~ ` l ` l ` ~ ` P ~ i . ~ ` ~ 58 108 158 ~86 Z58 388 3S8 488 458 SOe

m/z

~,BO~ NlltrelOne Scat) 491 378 4 M-zb

118 cM, I I Z43 273 339 395 l

149 173 ~ .1 L �9 481 58g . , ~ , u , . I L ~ . . . . . . . . . . . . . . . L L J , ,, h,ai~. ; i , i . i . l . l , i , f , i , i . i , I , i , i , p I , j . i . p i . P i , i , i , i , i , [ . i , i , i . i - ] , i , i . i , i , 1 . 1 . i , i . t , I , I . I , i . r , I , 4 , [ , i , I . I

58 190 158 204b ~ 388 358 400 458 500 $58

m/z

Figure 3. Mass spectra of the opioid derivatives.

1 8 0

Journal of Analytical Toxicology, Vol. 23, May/june 1999

Table I. Retention Times and Electron Impact Ions for Selected Basic Drugs*

Retention time (s) Drugs t Molecular weight (ainu) Mass/charge (relative intensity)

125 methylecgonine p 255 143 meperidine 247 155 lidocaine 234 185 EDDP* 278 205 methadone 346 212 dextromethorphan 271 215 propoxyphene 339 215 cocaine 303 235 pemoline 2pw 288 257 levorphanol O3p 313 258 alprenolol ?p ? 265 timolol p 372 270 pentazocine p 341 275 diazepam 284 277 oxprenolol 2p 377 283 thebaine 311 288 6-acetylcodeine 341 291 hydrocodone O6x 328 294 clenbuterol 2p (minor) 389 305 dihydrocodeine 357 313 dihydromorphine p (minor) 343 317 metoprolol 2p 379 317 morphine p (minor) 340 322 oxycodone Osx 344 327 codeine O6p 355 330 heroin 369 338 ethylmorphine O6p 369 343 oxycodone 06xO14p (minor) 400 348 hydromorphone 06xO3p 370 353 propranolot 2p 371 366 6-acetylmorphine 03p 383 369 nalorphine p (minor) 367 369 dihydromorphine O3pO6p (major) 399 376 temazepam Op 357 385 oxymorphone 06xO3p 386 395 pindolol ?p ? 397 morphine 03pO~ 397 398 oxymorphone 06xO3p014 p (minor) 442 410 clenbuterol 4p (major) 501 412 apomorphine 2p 379 423 nadolol 3p 477 431 naloxone O6xO3p 412 436 atenolol ?p ? 440 naloxone O6xO3,O14p (minor) 468 445 nalorphine 03pO6p 423 459 norcodeine O6pNp 397 491 naltrexone 06xO3p 426 500 naltrexone 06xO3p014 p (minor) 482 521 6-~naltrexol O3pO6p (major) 455 522 etorphine p 467 542 6-~naltrexol O3pO0pO14p (minor) 511 550 normorphine O3pO6pNp 439 558 nalbuphine O3pO6p 469 584 anileridine p 408

255(13) 182(100) 96(26) 94(39) 82(99) 248(29) 247(15) 246(50) 232(14) 218(26) 172(76) 89(47) 71(100) 235(13) 86(100) 80(13) 58(10) 278(22) 277(75) 276(100) 262(33) 220(19) 120(28) 310(3} 165(3) 72(100) 271(72) 270(100) 214(31) 171(35) 150(57) 89(45) 61(66) 91(9) 61(11) 58(100) 303(15) 182(68) 94(34) 82{100) 288(2) 232(100) 176(90) 313(97) 312(100) 245(72) 200(37) 150(96) 57(99) 288(8) 228(100) 172(7) 140(11) 98(18) 373(30) 357(30) 186(55) 112(62) 56(100) 342(87} 341(16) 340(57) 326(40) 273(100) 258 (54) 110(95) 70(70) 285(44) 284(30) 283(100) 257(46) 256(93) 255(43) 221(45) 304(30) 228(100) 98(10) 312(32) 311(100) 296(67) 255(19) 342(28) 341(100) 282(95) 229(30) 328(61) 313(8) 298(18) 297(100) 271(16) 240(13) 390(24) 389(40) 299(17) 86(100) 57(55) 466(82) 434(100) 358(33) 357(100) 300(53) 284(51) 344(50) 343(100) 326(26) 287(77) 379(1) 306(27) 229(11) 228(100) 154(16) 342(42) 341(74) 340(20) 284(10) 268(100) 267(23) 215(20) 345(38) 344(100) 313(19) 356(45) 355(100) 298(11) 282(87) 229(22) 370(22) 369(32) 368(22) 327(100) 310(56) 268(62) 370(26) 369(100) 340(15) 312(16) 296(83) 243(25) 401(14) 400(100) 343(34) 295(18) 230(28) 370(17) 339(22) 314(100) 283(90) 257(18) 298{11 ) 229(12) 228(100) 98(13) 383(27) 327(100) 324(37) 284(10) 268(50) 215(20) 204(24) 368(22) 367(I 00) 366(15) 310(9) 294(I00) 241(20) 400(33) 399(74) 369(52) 343(I00) 326(32) 286(30) 273(32) 272(I 5) 271(I00) 257(28) 256(20) 387(15) 386(32) 355(8) 331 (I 7) 330(I 00) 299(I 5) 287(22) 228(I00) 397(22) 342(]8) 341(100) 284(8) 268(54) 218(15) 215(15) 442(30) 387(20) 386(I00) 385(21) 355(9) 329(I 3) 281 (I 3) 216(I 5) 466(82) 434(I00) 380(37) 379(85) 378(86) 322(87) 266(I00) 57(64) 478(36) 462(92) 183(I 5) 112(30) 86(I 00) 413(34) 412(62) 381 (I 8) 356(I 00) 325(20) 305(I 2) 228(] 00)205(20) 154(32) 98(42) 72(53) 468(I 8) 412(I00) 411 (26) 395(29) 394(20) 363(25) 307(22) 423(28) 367(I 00) 350(49) 294(36) 244(I 8) 398(30) 397(25) 223(98) 101 (I00) 57(90) 427(38) 426(94) 395(I 8) 371 (26) 370(I00) 339(14) 483(I 2) 482(44) 451(7) 427(22) 426(I00) 409(I 2) 456(34) 455(82) 454(34) 414(35) 399(100) 382(18) 358(15) 467(8) 450(13) 380(100) 297(17) 250(33) 164(33) 511 (59) 455(100) 439(38) 398(17) 440(36) 209(50) 101(100) 57(90) 470(6) 415(20) 414(100) 57(39) 409(10) 247(16) 246(100) 172(5)

* Not detected: oxazepam, buprenorphine, acebutolol, labetolol. Bolded drugs are those under discussion in the text. + x refers to methoxyamination; p refers to propionylation; (major) > 90% peak area. (minor) < 10% peak area. * EDDP, methadone metabolite, 2-ethyl-l,5,dimethyl-3,3--diphenylpyrrolinium. s The mass spectrum using chemical ionization gave m/z= 288(80) 232(100) 176(100).

181


t88"x

. . . . ' . . . . I . . . . i . . . . I . . . . i . . . . I ' " ' ' . . . . I . . . . ' . . . . I . . . . ' . . . .

. . . . ' . . . . I . . . . i . . . . I . . . . ' . . . . I . . . . ' . . . . . . . . ' ' ' " I . . . . I . . . .

. . . . ' . . . . I . . . . ' . . . . I . . . . ' . . . . I . . . . I . . . . I ' ' ' I ' " ' ' I . . . . ' ' " ' '

223~ norcodr f . . . . . . . . . f . . . . . �9 . . . . ~.,-_ . . . . . . ] A_ A,_

I R R 2 8 8 3 8 8 4 1 ~

Retention "rime (m)

Figure 4. Total ion and extracted ion chromatograms of a urine sample from a woman who ingested an unknown quantity of codeine. The urine was diluted 10-fold with drug-free urine before analysis; codeine = 10 x 2811 = 28,110 ng/mL, morphine ; 10 x 204 ; 2040 ng/mL. Peak identification: 1, codeine; 2, morphine; 3, nalorphine 15 500 ng/mL; 5, naltrexone IS 500 ng/mL; 4, norcodeine.

complete recovery were prepared as follows: two of the drug-free urines were extracted without the addition of internal standards; one extract was spiked with 300 (30 for 6-AM) ng of each analyte, and the second was spiked with 1500 (150 for 6- AM) ng of each analyte. Internal standards were added to both, and the solvent was evaporated. All residues were derivatized, purified, and injected onto the GC-MS. The peak areas for each injection were normalized to the appropriate internal standard. The extraction efficiencies were calculated as the ratios of the normalized peak areas in the extracted to those in the spiked blank extract. The range of extraction efficiencies are given in Table II for each opioid at the two concentrations. All were above 80% except for morphine, which had an overall recovery between 62 and 81%. This can be attributed to its polar amphotedc prop- erty. The extraction efficiency of 6-AM varied more widely than the other opiates. The 10-fold lower drug concentrations and the poorer precision at these concentrations are likely responsible for this finding.

Table II. Performance Characteristics

Codeine Morphine 6-AM* Hydrocodone Hydromorphone Oxycodone Oxymorphone

Retention time (s) 327 397 366 291 348 332 385 Quantitation ion (amu) 282 341 327 297 314 344 330

Internal standard Nalorphine (445 s, 367 ainu)

Day-to-day precision 300 ng/mL 30 ng/mL N=18 Mean (ng/mL) 283 303 27.8 288 CV (%) 8.6 6.7 12.0 7.6

1500 nglmL 150 nglmL

Mean (ng/mL) 1482 1457 148 1702 CV (%) 8.3 7.9 12.0 9.1

300 ng/mL 30 ng/mL

Extraction efficiency (%) 93-98 75-81 54-I 04 84-96 range (N = 5)

1500 ng/mL 150 ng/mL

range (N = 5)

Limit of quantitation (ng/mL)

Limit of detection (ng/mL) Linearity (ng/mL)

Naltrexone (491 s, 370 amu)

300 nglmL

287 293 289 8.7 10.9 8.3

1500 nglmL

1556 1857 1680 6.0 10.2 9.0

1500 nglmt

91-100 104-114 103-117

1500 ng/mL

80-91 62-77 72-I 02 83-I 02 85-98 83-I 09 88-I 04

25 25 25(I 0) I 25 25 25 25

I0 I0 10(10) I I0 I0 10 10 5000 5000 5000 2000 2000 2000 2000

* 6-AM = 6-acetylmorphine. + In unhydrolyzed urine.

182

J o u r n a l o f A n a l y t i c a l T o x i c o l o g y , V o l . 2 3 , M a y / J u n e 1 9 9 9

The upper limit of linearity was determined by assaying the standards above 600 ng/mL in a single run. Linearity was accept- able if the calculated value was within • 20% of the target value. Codeine, morphine, and 6-AM were linear to 5000 ng/mL, whereas the four 6-keto-opioids were linear to 2000 ng/mL.

The limits of quantitation and detection were determined by analyzing the urine standards at 200,100, 50, 25,10, and 5 ng/mL. The limit of quantitation was the lowest drug concentration that gave a mass spectral fit of greater than 800 out of a possible 1000, a signal-to-noise ratio (S/N) > 10 for the quantitation ion, and a calculated concentration within 80 to 120% of the spiked value. It was 25 ng/mL for all analytes. The limit of detection was the lowest

1 8 8 ~

TO1 4

1 2

2 9 7 ~ hydrocodonr

. . . . ' . . . . I ' ~ " -" I A A . . . . . .

3 1 4 ~ hydromorphonr

. . . . ' . . . . I . . . . i . . . . I . . . . ' . . . . I . . . . ' . . . . ] . . . . ~ ' ' I ' I . . . . ' . . . .

. . . . i . . . . I . . . . I ' ; i : ' I ' % ' ' I r . . . . " 1 ' ' o f I . . . . I ' ~ 'O"g ' I ' ' I . . . . i . . . .

leo ZOO 3oo 4 s o s e e

Retention time ( e )

Figure 5. Total ion and extracted ion chromatograms of a urine sample from a woman who ingested hydrocedone. The urine was diluted fivefold with drug-free urine before analysis; hydrocedone = 5 x 812 = 4060 ng/mL, hydromorphone = 5 x 316 = 1580 ng/mL. Peak identification: 1, hydrocodone; 2, hydromorphone; 3, nalorphine IS 500 ng/mL; 4, naltrexone IS 500 ng/mL.

1 8 8 ~ 2

3 4 5

TOI 1

. . . . j . . . .

. . . . ' . . . . f . . . . ' . . . . I . . . . ' . . . . I ' ' ~ ' ' ' " I . . . . ' . . . . I . . . . ' . . . .

. . . . ' . . . . I . . . . ' . . . . I . . . . ' . . . . I . . . . ' ' ' " ' 1 . . . . " . . . . I " " , . i . . . .

1 8 8 Z 8 8 3 9 8 4 8 e 5 8 8

Retention time (e )

Figure 6. Total ion and extracted ion chromatograms of a urine sample from a patient enrolled in an abused-drug-treatment program. The urine was diluted threefold with drug-free urine prior to analysis; oxycoclone = 3 x 693 = 2079 ng/mL; oxymorphone = 3 x 1949 = 5847 ng/mL Peak identification: 1, oxycodone; 2, oxymorphone; 3, nalorphine IS 500 ng/mL; 5, naltrexone IS 500 ng/mL. Peak 4 tentative identification: noroxycodone.

drug concentration which gave a mass spectral fit of greater than 700 and a S/N > 3. It was found to be 10 ng/mL for all analytes. Urine hydrolyzed with ~-glucuronidase often produced a chromatogram with an elevated background. When this occurs, it is best to confirm the presence of 6-AM from an unhydrolyzed urine. Analysis of the unhydrolyzed urine not only produced fewer excip- lent peaks but also.reduced the morphine peak. The limit of quantitation for 6-AM on unhydrolyzed urine was 10 ng/mL.

The extract stability was assessed by reinjecting the same 600-, 300-, and 1500-ng/mL standards every third day for two weeks. The 600-ng/mL standard was used as the single-point calibrator and the concentrations were calculated for the 300- and 1500-

ng/mL standards. The extracts were stored capped at room temperature. Over two weeks, the calculated concentrations were within • 20% of the target concentrations for all the opiates. At the end of two weeks, the extracts were allowed to evaporate to dryness. The residue was reconstituted, and the standards were reinjected. All of the peak areas were halved, and the calculated concentrations of 6-AM, oxycodone, and oxymor-

f phone were outside the target concentration by more than • 20%. The extracts are therefore stable at room temperature for at least two weeks, provided they are kept in solution.

Figure 4 is a chromatogram of a urine extract obtained from a woman 8 h after she ingested an unknown quantity of ~lenol 3 | Each tablet Con- tained 30 mg of codeine phosphate and 300 mg of acetaminophen. The urine was diluted 10-fold with drug-free urine before analysis. Extracted ion chromatograms for m/z 282, 341,367, and 223 representing codeine, morphine, nalorphine, and norcodeine, respectively, are displayed beneath the total ion chromatogram. The codeine and morphine concentrations were 28,100 and 2040 ng/mL, respectively, after correcting for the dilution. Similar codeine and morphine concentrations have been measured in the urine of indi- viduals 10 to 12 h after oral ingestion of either single or multiple 30-mg doses of codeine (15). Norcodeine was identified from its retention time and mass spectrum but was not quantitated. The dipropyl derivatives of norcodeine and morphine have the same molecular weight of 397 but are separated by 62 s. The two mass spectra are dis- similar except for their M § ions. In contrast, the TMS derivatives of norcodeine and morphine, also having the same molecular weight, co-elute, and share some common ions (16,17). Normorphine was not detected on the diluted urine but was identified when the urine was analyzed undiluted.

A 50-year-old woman had ingested five doses of Novahistex DH | elixir cough suppressant at 6-h intervals. Each dose contained 5 mg hydrocodone bitartrate and 20 mg phenylephrine HCl. Urine was collected 6 h after the fifth dose. The chromatogram in Figure 5 was obtained from the

183

analysis of the urine after it had been diluted fivefold with drug- free urine. Extracted ion chromatograms for quantitation ions m/z 297, 314, and 370 ainu, corresponding to hydrocodone, hydromorphone, and naltrexone, are displayed beneath the total ion chromatogram. After correcting for the dilution, the concentration of hydrocodone was calculated to be 4060 ng/mL, whereas hydromorphone was 1580 ng/mL. Six hours after four healthy subjects consumed either 10- or 20-rag single doses of hydrocodone, their urinary concentrations of hydrocodone and hydromorphone were previously reported to range 1840-5786 and 157-910 ng/mL, respectively (7). The hydromorphone metabolite was proportionately higher in the described case, presumably because of its accumulation after multiple hydrocodone doses.

Figure 6 is a chromatogram from a patient's urine that screened positive with EMIT II Opiate assay. The patient was enrolled in an abused-drug-treatment program and was known to abuse Percocet | which contains oxycodone HC1 and acetaminophen. The urine was diluted threefold with drug-free urine before GC-MS analysis, After correcting for the dilution, oxycodone and oxymorphone were determined to be 2079 and 5847 ng/mL, respectively. The extracted ion chromatograms corresponding to the quantitation ions for oxycodone, oxymorphone, and naltrexone are displayed below the total ion chromatogram. Peak 11 eluted at 456 s. The mass spectrum consisted of the following ions (intensity): 386(60), 355(60), 299(100), and 57(95). This com- pound is probably the derivatized metabolite noroxycodone O6xNp, having a molecular weight of 386.

Discussion

The sequential derivatization procedure accommodates the analysis of opioids commonly requiring GC-MS confirmation in urine. It is necessary to convert the ketone group to the methoxime prior to propionylation to prevent defivitization of the enol isomer. This would produce two compounds for each of the ketone opioids, the enol derivative plus the unreacted ketone. The methoximes are thermally stable and have good chromatographic properties, which allows them to be quantitated well. Un- fortunately, the syn- and the anti- isomers of hydmcodone and oxycodone are separated by 4 s on a DBq capillary column. Quantitation may be done equally well by selecting the larger, late- eluting isomeric peak or by integrating the combined area of both isomeric peaks as was done here. The inclusion of the methoxyamination step does not affect the quantitation of codeine, morphine, 6-AM, and nalorphine. Pyridine readily dissolves methoxyamine HC1 and functions as an efficient vehicle for the conversion of ketones to their methoximes. Optimal reaction conditions were investigated by allowing the reaction to proceed for 5, 10, 15, 20, 30, and 40 min at 20, 37, 56, and 70~ while maintaining the propionylation at 56~ for 15 rain. Heating did not increase the product yield. The reaction was complete by 10 rain, and the products remained stable. A 15-min reaction time was arbitrarily chosen to minimize the possibility of between-tube variations.

The propionic anhydride may be added to the reaction mixture without removal of the methoxyamine HC1. The propionic anhydride reacts with or destroys the excess methoxyamine. After the

184


propionylation step, when the propionic anhydride is dried, there is no white residue in the conical tube as there is if propionic anhydride is omitted entirely. No large peak corresponding to a reaction product between the excess methoxyamine and propionic anhydride was identified in the chromatogram, even when the ini- tial oven temperature was reduced to 80~ The reaction was further investigated by combining 25 laL of 2% methoxyamine HCl in pyridine with 25 taL of propionic anhydride. After 15 rain, 10 laL of pyridine containing the seven opioids, each at 100 lag/mL, was added to the mixture, which was then heated for 15 rain. The reaction product was evaporated and purified as described in the method. Only propionylated derivatives and unreacted hydrocodone were identified in the chromatogram.

Pyridine is a catalytic solvent for reactions with propionic anhydride. Optimal conditions were investigated for the reaction conditions by allowing the propionylation to proceed for 5,10,15, 20, 30, and 40 rain at 37, 56, and 70~ while maintaining the methoxyamine reaction at room temperature for 15 min. Product yield for each of the opioids was the same after heating at 56~ for 10 min as for heating at 37~ for 20 min. The reaction at 70~ was difficult to control. Even after 5 min, rea/:tion products incorpo- rating propionylation of the hydroxyl groups at C14 of oxycodone, oxymorphone, and naltrexone began to appear. Formation of these reaction products was simultaneously heralded by a loss of the desired reaction products, which do not include propionylation at the C14 hydroxyl group. Approximately equal quantities of each pair of C14 hydroxylated versus unhydroxylated products were formed when heated at 70~ for 20 rain. These three opioid derivatives that are propionylated at C14 can potentially interfere with the other target opiate derivatives. Oxycodone O6xO14 p elutes 5 s ahead of hydromorphone O6xO3p. Oxymorphone O6xO3pO14p co-elutes with morphine O3pO6p. Natrexone O6xO3pO14p elutes 5 s ahead of nalorphine O3pO6p. The milder propionylation condition, 56~ for 15 min, was selected to provide derivatives that separated well from one another. Based on peak areas, less than 2% propionylation occurred at the C-14 position.

O'Neal et al. (12) used propionic anhydride to derivatize codeine, 6-AM, morphine, and norcodeine in urine extracts. After the excess propionic anhydride was evaporated, they reported low recoveries when ethyl acetate, isopropanol or chloroform were used as the reconstituting solvent. They attributed the problem to co-extracted urinary components which prevented the dissolu- tion of the drugs. They solved the problem by using a mixture of toluene/hexane/isoamyl alcohol (78:20:2) for reconstitution. Low recoveries were not observed in the method described here. After the propionic anhydride was evaporated, the residue was further purified by extracting it into hexane/chloroform (3:1), a relatively nonpolar solvent, from 15% ammonium hydroxide. The offend- ing matrix element probably partitions into the aqueous fraction. A high molarity alkaline solution is necessary to recover the 6-keto-opioids, presumably by the salt-out effect. Ammonium hydroxide was chosen for its volatile characteristic. Any base that is co-extracted into the hexane/chloroform is evaporated along with it and does not appear in the residue. A postpropionylation purification step was previously found to be necessary to recover r and ec-hydroxytriazolam from urine extracts (18). The recoveries of other benzodiazepine metabotites such as nordiazepam, temazepam, 2-hydroxyethyflurazepam,


oxazepam, and lorazepam also measured in the same method were not improved by the purification step. Guillot et al. (19) underlined the importance of removing propionylation by products to minimize maintenance of the mass spectrometer. They removed the excess propionic anhydride by reacting it with methanol. The formed propionic acid was evaporated with reaction solvent consisting of 10% pyridine in chloroform. In the method described by Smith et al. (7) the remaining propionic anhydride was hydrolyzed with dilute sulfuric acid. The formed propionic acid was removed by washing with ethyl acetate. The opioid derivatives were then recovered by alkalinizing the aqueous phase then extracting it with methylene chloride.

Broussard et al. (10) described the formation of the oximes of hydrocodone, hydromorphone, oxycodone, and oxymorphone with hydroxylamine HCl. They added the hydroxylamine reagent to the urine after enzymatic hydrolysis but prior to extraction. This approach was investigated here by adding 5 mg methoxyamine HCI to 1 mL of the hydrolyzed urine followed by further heating at 56~ for I h. The urine was then extracted, propionylated, and purified as described in the method. The resulting chromatogram contained many more extraneous peaks. The largest co-eluted with hydromorphone. The reaction of methoxyamine HCI with the 6-keto-opioids proceeds more quickly in extracted urine and produces fewer extraneous peaks due to endo- geneous urinary substances.

Hydrolysates obtained by heating with hydrochloric acid produce extracts with high backgrounds and numerous extraneous peaks. Acid hydrolysis has been reported to yield lower recoveries of codeine (20) and morphine-3-glucuronide (21) when compared to [3-glucuronidase hydrolysis. A purified ~-glucuronidase facilitates the hydrolysis of the opiate conjugates while preserving the integrity of heroin and 6-AM (22). Crude preparations of Helix pomatia [3-glucuronidase contain a deacetylating esterase that reduces these two opiates, when present, to morphine (23). Suitable enzyme preparations from E. coli (22,23), Helixpomatia (21), and limpets (9) have been used by others. Morphine-3- glucuronide is readily hydrolyzed by [~-glucuronidase (22). Morphine-6-glucuronide was shown to be hydrolyzed more slowly than morphine-3-glucuronide (24). The former required a one-day incubation at 50~ with 1500 U of E. coli [3-glucuronidase for complete hydrolysis. There was little difference using this enzyme preparation at 37 or 50~ Codeine-6-glucuronide appeared to be hydrolyzed even more slowly. In this work, an overnight (16 h) incubation at 37~ using 5000 U of E. coli l~-glucuronidase was used to improve codeine recovery. There was no practical advantage to extending hydrolysis time beyond 16 h. No hydrolysis studies for the 6-keto-opioids have been reported. This is probably due to the difficulty in obtaining the pure glucuronide powders. Further work is required to opti- mize hydrolysis conditions for all of the opioid glucuronides spe- cific for each source of [3-glucuronidase.

References

1. B.D. Paul, J.M. Mitchell, L.D. Mell, Jr., and J. Irving. Gas chromatography/electron impact mass fragmentometric determination of uri-

nan/6-acetylmorphine, a metabolite of heroin. J. Anal. ToxicoL 13: 2-7 (1989).

2. R. Wasels and E Belleville. Gas Chromatographic-mass spectro- metric procedures used for the identification and determination of morphine, codeine and 6-monoacetylmorphine. J. Chromatogr. 674: 225-234 (1994).

3. B.D. Paul, L.D. Mell, Jr., J.M. Mitchell, J. Irving, and A.J. Novak. Simultaneous identification and quantitation of codeine and morphine in urine by capillary gas chromatography and mass spec- troscopy. J. Anal ToxicoL 9" 222-226 (1985).

4. W. Huang, W. Andollo, and W.L. Hearn. A solid phase extraction technique for the isolation and identification of opiates in urine. J. AnaL ToxicoL 16:307-310 (1992).

5. B.H. Chen, E.H. Taylor, and A.A. Pappas. Comparison of derivatives for determination of codeine and morphine by gas chromatography/mass spedrometry. J. Anal. ToxicoL 14:12-17 (1990).

6. M. Kushnir and F.M. Urry. Comparison of four derivitizing reagents for 6-acetylmorphine GC/MS analysis in urine. Abstract. Clin. Chem. 43:S213 (1997).

7. M.L. Smith, R.O. Hughes, B. Levine, S. Dickerson, W.D. Darwin, and E.J. Cone. Forensic drug testing for opiates. VI. Urine testing for hydromorphone, hydrocodone, oxymorphone, and oxycodone with commercial opiate immunoassays and gas chromatography-mass spedrometry. J. Anal ToxicoL 19:18-26 (1995).

8. L.J. Bowie and P.B. Kirkpatrick. Simultaneous determination of monoacetylmorphine, morphine, codeine, and other opiates by GC/MS. ]. Anal ToxicoL 13:326-329 (1989).

9. J.W. Soper, D. Canfield, L. Busch, T. Roberts, and R. Crooks. A single step GC/MS procedure for quantitation of 6-monoacetylmorphine, coupled with simultaneous quantitation (6) and/or detection (8) of 14 additional opiate compounds, following derivatization with propionic anhydride. Abstract. American Academy of Forensic Science annual meeting, February 19-24, 1996, Nashville, TN.

10. L.A. Broussard, L.C. Presley, T. Pittman, R. Clouette, and G.H. Wimbish. Simultaneous identification and quantitation of codeine, morphine, hydrocodone and hydromorphone in urine as trimethylsilyl and oxime derivatives by gas chromatography-mass spectrometry. Clin. Chem. 43:1029-1032 (1997).

11. K.M. Monti, R.L. Foltz, and D.M. Chinn. Analysis of naltrexone and 6-~-naltrexol in plasma and urine by gas chromatography/negative ion chemical ionization mass spectrometry. J. Anal. Toxicol. 15: 136-140 (1991 ).

12. C.L. O'Neal and A. Poklis. Simultaneous determination of acetylcodeine, monoacetylmorphine and other opiates in urine by GC-MS. ]. Anal ToxicoL 21: 427-432 (1997).

13. W.H. Huang, D.E. Moody, R.L. Foltz, and S.L. Walsh. Determination of naltrexone and 6-1~-naltrexol in plasma by solid-phase extraction and gas chromatography-negative ion chemical ionization-mass spedrometry. ]. Anal Toxicol. 21:252-257 (1997).

14. Dj. Bortz and L.D. Bowers. Mass-to-charge ratio misassignments of molecular ions in ion-trap mass spectrometry. J. Anal. ToxicoL 22: 257-259 (1998).

15. B.L. Posey and S.N. Kimble. High performance liquid chromatographic study of codeine, norcodeine, and morphine as indicators of codeine ingestion. J. Anal ToxicoL 8:68-74 (1984).

16. M. Kushnir, C. Jones, R. Struempler, and F.M. Urry. Eliminating inter- ferences of related compounds in opiates GC/MS analysis. Abstract. Clin. Chem. 42:$218 (1996).

17. J.R Brendler, R.E. Clouette, G.H. Wimbish, and J.C. Garriott. The determination of codeine, morphine, hydrocodone and hydromorphone in blood by use of a dual derivatization technique. Abstract. American Academy of Forensic Science annual meeting, February 9-14, 1998, San Francisco, CA.

18. R. Meatherall. GC-MS confirmation of urinary benzodiazepine metabolites. ]. Anal ToxicoL 18:369-381 (1994).

19. J.G. Guillot, M. Lefebvre, and J.R Weber. Determination of heroin, 6-acetylmorphine and morphine in biological fluids using their pro- pionyl derivatives with ion trap GC-MS. J. Anal. Toxicol. 21: 127-133 (1997).

20. F.T. Delbeke and M. Debackere. Influence of hydrolysis procedures

185


on the urinary concentrations of codeine and morphine in relation to doping analysis. J. Pharm. Biomed. Anal. 11" 339-343 (1993).

21. A. Solans, R. de la Torre, and J. Segura. Deterimination of morphine and codeine in urine by gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 8:905-909 (1990).

22. M. Zezulak, J.J. Snyder, and S.g. Needleman. Simultaneous analysis of codeine, morphine and heroin after [3-glucuronidase hydrolysis. J. Forensic ScL 38:1275-1285 (1993).

23. J.H. Nichols, P.A. Chezick, and G.M. Lawson. Confirmation of urine opiates by GC/MS. Abstract. Clin. Chem. 38:1012 (1992).

24. R.W. Romberg and L. Lee. Comparison of the hydrolysis rates of morphine-3-glucuronide and morphine-6-glucuronide with acid and [3-glucuronidase. ]. Anal. ToxicoL 19:157-162 (1995).

Manuscript received August 25, 1998; revision received November 6, 1998.

186

GC-MS Confirmation of Codeine, Morphine, 6 … · 6-Acetylmorphine, Hydrocodone, Hydromorphone,...

Documents

Transcript of GC-MS Confirmation of Codeine, Morphine, 6 … · 6-Acetylmorphine, Hydrocodone, Hydromorphone,...