Journal of Analytical Toxicology, Vol. 16, March/April 1992

Alcohol-Specific Cocaine Metabolites in Serum and Urine of Hospitalized Patients Alan H.B. Wu* and Taofiq A. Onigbinde

Department of Pathology and Laboratory Medicine, University of Texas Medical School, Houston, Texas 77030 K e n t G. J o h n s o n

Department of Pathology and Laboratory Medicine, Toxicology Laboratory, Hermann Hospital, Houston, Texas 77030 G a r y H. W i m b i s h

Department of Pharmacology, Texas College of Osteopathic Medicine, Fort Worth, Texas 76107

Abstract

I

The serum and urine from 44 consecutive patients that tested

positive for the cocaine metaboUte benzoylecgonine (BE) were examined for free cocaine, ecgonine methyl ester (EME), and other metabolites by gas chromatography/ion trap mass spectrometry (GC/MS). In 13 of these patients, unique ethanolrelated cocaine metabolites, cocaethylene and ecgonine ethyl ester (EEE), were detected in urine and serum. One was from a newborn baby whose mother's blood was positive for cocaine and negative for cocaethylene. In two other patients, isopropanol was also consumed with cocaine and ethanol. In one of these two, cocaisopropylene and ecgonine isopropyl ester (EPE) were identified in urine. The urine ethanol concentration in 7 of the 13 cocaethylene-positive patients ranged from 19 to 322 mg/dL. In the other six, ethanol was not detected in the urine. However, each of these latter patients had either prior serum results that were positive for ethanol or admitted to recent alcohol abuse. In the remaining 31 of 44 cocaine-positive patients, ethanol and the alcohol-specific cocaine metabolites were absent. The detection of alcoholrelated cocaine metabolites is fairly common in a cocainepositive patient population.

Introduction

The metabolism of cocaine in humans has been extensively studied. A significant fraction of a cocaine dose is converted to benzoylecgonine and ecgonine methyl ester as the major metabolites (1). Concentrations of BE exceeding 150 ng/mL in urine by GC/MS are considered positive according to the National Institute for Drug Abuse's guidelines tbr drug testing in the workplace (2). In addition to free cocaine, as many as 11 metabolites have been observed in human urine, depending on measurement conditions (3). Cocaine is often abused in combination with other recreational drugs. Simultaneous consumption of cocaine and heroin is common, and undoubtedly contributes to many cocaine-associated deaths. Tardiff et al. found cocaine with heroin and other opiates in 39% of cocaine-positive fatalities during an 11-month period in New York City (4). Cocaine is also frequently abused 9 Authorto whomcorrespondenceshould be addressed.

1 32

in conjunction with alcohol and can lead to significant cardiovascular (5) and hepatotoxic consequences (6). Fatal malignant hyperthermia has been described in one patient on recreational cocaine and ethanol (7). Studies have indicated that in the presence of ethanol, cocaine is partially converted to unique metabolites by the liver. Some of these metabolites include cocaethylene and ecgonine ethyl ester, 3'-hydroxy- and 4'-hydroxy-benzoylecgonine ethyl esters, and 4'-hydroxy-3'-methoxybenzoylecgonine ethyl ester (8,9). Recently, the cocaethylene metabolite has been shown to have high affinity for the dopamine transport system, which may attribute to the enhanced euphoria associated with the concomitant use of cocaine and ethanol (10). This report presents the clinical summaries and supporting laboratory data on hospitalized patients who have unique cocaine metabolites related to the combination of cthanol, isopropanol, and cocaine abuse.

Materials and Methods

Subject samples. All subjects were patients examined in the emergency room of Hermann Hospital, Houston, Texas. Blood samples collected at admission into gray- (sodium fluoride) or red- (no preservative) top tubes were assayed in this study. Random urine samples were collected without preservatives within 24 h after admission and assayed. All samples were stored either frozen or at 4~ Table I summarizes the age, sex, race, and clinical diagnoses for the patients positive for ethanol-specific cocaine metabolites. All were under the age of 40, and all were either black or Hispanic. Examination of medical records indicated that many of these patients inhaled cocaine in the form of "crack." Reagents. All standards for cocaine and metabolites were obtained from Radian Laboratories. The internal standard, difluorobenzoylecgonine, was from ElSohly Laboratories. All derivatizing agents were from Sigma Chemical Co. Solvents used for extractions were spectrophotometric grade. Instrumentation. Urine samples were initially screened for BE by using the EMIT assay on the ETS enzyme immunoassay analyzer (Syva). All samples testing positive for BE (300-ng/mL cutoff limit) were confirmed by the Finnigan ITS40 GC/ion trap MS analyzer via electron impact ionization (EI).

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

Journal of Analytical Toxicology, Vol. 16, March/April 1992

Extraction procedures. A serum or urine sample was spiked with 90 ~tL of 300 ng/mL of difluorobenzoylecgonine and adjusted to pH 6.0 with 2 mL of 0.1 mol/L phosphate buffer. Samples were applied onto solid-phase extraction columns (Bond Elut Certify II, Analytichem Int.), which had been prepared by passing 2 mL of methanol and 0. l mol/L phosphate buffer prior to sample addition. Columns were rinsed with 3 mL water, 3 mL 0.1 mol/L HC1, and 9 mL methanol. Cocaine and metabolites were eluted with 2 mL methylene chloride-isopropyl alcohol (80:20) in 2% ammonium hydroxide. Extracts were evaporated to dryness under nitrogen at room temperature and derivatized with 50 ~ each of pentafluoropropionic acid and pentafluoro- 1propanol for 25 rain at 70~ The derivatives were evaporated under nitrogen, reconstituted with 300 ~.L ethyl acetate, and transferred to autosampler vials. An alternate derivative was prepared on some samples using either 1-iodopropane (1 l) or 2-iodopropane (Sigma). After extraction with solid-phase columns, dried extracts were solubilized with 50 ktL of l: 10:0.05 0.1 mol/L tetramethylphenyl ammonium hydroxide, n-dimethyl acetamide, and 25% tetramethyl ammonium hydroxide in methanol, and derivatized with 50 I.tL iodopropane after a 15-min incubation at 60~ The derivatives were evaporated under dry nitrogen, reconstituted with 300 t.tL ethyl acetate, and transferred to autosampler vials. A volume of 1 I,d_.of sample was injected onto a DB-5 capillary GC column (J&W Scientific). Quantitative analyses were performed for free cocaine, BE, EME, and cocaethylene based on standard curves. The limits of detection, quantitation, and linearity for the BE assay are 9, 25, and 5000 ng/mL, respectively (12). The assay was linear to 500 ng/mL for cocaine, cocaethylene, and EME. The LOD and LOQ for cocaine and the other metabolites were not determined, but are assumed to be similar to that of BE. Because standards were not available for cocaisopropylene and the ecgonine ethyl and isopropyl esters, no quantitative results are given for these metabolites. Alcohol measurements. Ethanol concentrations in serum and urine were measured on a DuPont aca IV analyzer using the alcohol dehydrogenase method (13). The LOD for this assay is l

mg/dL. An LOQ of 10 mg/dL was used for both serum and urine ethanol. Samples not assayed immediately were capped and stored at 4~ Isopropanol and acetone concentrations were measured from urine at the Mayo Medical Laboratories, Rochester, MN, using the headspace gas chromatography assay.

Z~

m*

i

~

--

C~IM

1

TOT

tWx

=o

m

/

l.g.

I

Table I. Clinical Summaries for CocaethylenePositive Patients fD.ZSx t U

Case 1 2 3 4 5* 6** 7 8t 9tt 10t 11 12 13

M

Age/race/sex Admittingdiagnosis 36BM 23BM 26HM 31BM 20BF 21HM 26BF 33 B F 1 day B F 38 H M 33BF 28BF 32B F

chest pain, rule out myocardial infarction chest pain, rule out myocardial infarction lacerations violent behavior Dihistine overdose isopropanol and ethanol abuse abdominalcramps, pregnancyof 14 weeks full-term labor and delivery full-term neonate isopropanol and ethanol abuse trauma due to assault suicidal ideation with substance abuse trauma due to domestic altercation

9150 cc of Dihistine, total dose: 300 mg codeine, 900 mg pseudoephedrine, 60 mg chlorpheniramine, and 5.9 g alcohol. *" Isopropanol concentration at admission = 245 mg/dL (toxic: >50 mg/dL); acetone = 56 mg/dL (toxic >50 mg/dL); and methanol = 0 mg/dL. t Udne from the neonate of this mother was unavailable. l't Serum from the mother of this infant was positive for cocaine and negative for ethanol and ethanol-specific metabolites. Unne from the mother was unavailable for analysis. Isopropanol concentration at admission = 127 mg/dL; acetone = 35.5 mg/dL; and methanol = 0 mg/dL.

I

'

Ji

J

t,.

,=_

=

_.

i

/

"="=='=""

==="

ILs.

101

9

. . ,

. . . .

u

. . . .

,

. . . .

i

. . . .

,

" ' ' 1

. . . .

,

. . . .

I

. . . .

9 fl

aB

WB

r

4g:

S:B

61:

ON:

, ' "

Figure 1. Total ion gas chromatogram and mass spectra of cocaine and metabolites extracted from urine of patients with (a) abuse with cocaine alone, (b) abuse with both cocaine and ethanol (Patient 1), and (c) abuse with cocaine,ethanol,and isopropanol(Patient10) The scan numbersand retention times are given on the abscissa, and the relativetotal ion intensity is given on the ordinate.

133

Journal of Analytical Toxicology, Vol. 16, March/April 1992

Results

Figure 1 shows the total ion gas chromatograms of extracted urine samples taken from patients who were abusing cocaine alone, cocaine in combination with ethanol, and cocaine with ethanol and isopropanoI (Patient I0). In Figure Ia, only ecgonine methyl ester, benzoylecgonine, and free cocaine are present. In Figure I b, ecgonine ethyl ester and cocaethylene are also present. In Figure lc, peaks determined to be ecgonine isopropyl ester and cocaisopropylene are present, in addition to free cocaine and the other four metabolites. The gas chromatographic retention times for the cocaine metabolites are consistent within the series. Increasing the length of the alkyl esters decreases volatility, increases their partition coefficient, and increases chromatographic retention times. The corresponding El mass spectra for cocaine and metabolites are also shown in Figure 1. In Figure 2, the chemical structure and major fragmentation ions for cocaine are shown along with the structures and proposed biochemical pathways for these cocaine metabolites. In all of these compounds the El mass spectra show a major ion at m/z 82 due to the fragmentation of the bicyclic ring system to methylpyrrole (3). For BE, the molecular (M § ion for the pentafluoro-l-propyl derivative is at m/z 421. The M § ion for free cocaine is at m/z 303, while for cocaethylene and cocaisopropylene the M § ions are at m/z 317 and m/z 331, respectively. The M § ion for the pentafluoropropionyl derivative of ecgonine methyl ester is at m/z 345, whereas the molecular ions of the ecgonine ethyl and isopropyl esters are m/z 359 and 373, respectively. The mass spectra of the ecgonine alkyl esters differ from free cocaine and the alkylated cocaine metabolites by the absence of m/z 105, since none of these esters have the benzoyl ion in their molecular structure. However, both cocaine and ecgonine methyl ester have ions at m/z 182, as shown in Figure 2, because the benzoate ion is lost from free cocaine. Cocaethylene and ecgonine ethyl ester also have corresponding ions at m/z 196, while cocaisopropylene and ecgonine

isopropyl ester have ions at m/z 210. These incremental increases of 14 atomic mass units reflect the addition of methylene groups to the respective molecular ions. To verify the structure of cocaisopropylene, the chromatogram and mass spectrum from Patient 10 were compared against benzoylecgonine derivatized with both l- and 2-iodopropane. These derivatives produce coca- 1-propylene and coca-2-propylene (cocaisopropylene), respectively. The resulting gas chromatogram for the 2-iodopropyl derivative showed a retention time that was identical to the cocapropylene identified from Patient 10. The retention time for the l-iodopropyl derivative was 24 s longer. Coca-l-propylene should have a longer retention time than coca2-propylene because straight-chain alkyl groups are less volatile than the corresponding branched alkyl isomers. The mass spectra for these two derivatives also confirmed that the cocaproplene identified in Patient 10 is the 2-iodopropyl derivative of BE. The spectra in Figures lc (urine sample) and 3b (coca-2-propylene) are nearly identical. They both have the m/z 168 ion, which is absent in the coca-l-propylene mass spectrum (Figure 3a). The 168 ion is also present in ecgonine isopropyl ester, formed from the loss of both benzoate and 2-propyl groups from the coca-2-propylene metabolite. Evidently, the corresponding 1-propyl group is not lost in the coca- 1-propylene metabolite. Table II tabulates the concentrations of serum and urine

p,

INT

*,

218

94

~3

o

N

C-OCH

,a.

...........................

3

i

!

c.

!

~

o 3

II

CH~

3:31

I~

IS8

,i,i,],[,r,l,l,l,l,l,l,i

1811 (MW. 3311

I'l'l,l,l'l,l,l,lrl,l,l,l,l'1,1rl,l,lrl,l,lr],]ll,I,

,i,f,

150

25a

31B

3~

488

Q

(MW - 373) 18~82

,3

~e~':=~ - -o-........."~' ..

O

.

!i~ a i

,.~_v

~oc.

i

c,

o

, 3

,,

L-~'_._ ' ~az; !,~ 'k~._V ;

(uw- r

(Mw- aoa)

(Mw- 3 ~ IN1

~~

t N

M

I

C-(~H

I

I

3

o t

331

Figure 2. Chemical structures, molecular weights, major fragmentation

patterns, and proposed in vivo biochemical pathways for cocaine in the presence of ethanol and isopropanol. The pentafluoropropionic derivative of the ecgonine esters and the pentafluoro-l-propanol derivative of BE are shown.

134

,i,i,

l,i ,i ,i, i,i ,i ,f,i,

110

I.$8

T L

i,i ,i ,i, i,i ,i ,i, i,l,]

288

;50

,i,i,

'ann

i,i "['1'1'1'1'[

3S8

' I' I'r'l

~

']'1'

4S8

Figure 3. Mass spectrum of the (a) 1-iodopropyl and (b) 2-iodopropyl derivatives of benzoylecgonine. Relative total ion intensity is plotted against the mass/charge ratio.

Journal of Analytical Toxicology, Vol. 16, March/April 1992

ethanol, free cocaine, BE, ecogonine methyl and ethyl esters, and cocaethylene. Samples not collected in tubes with sodium fluoride may be falsely high for ecgonine methyl and ethyl esters. Isenschmid et al. showed that samples are most stable when they are adjusted to pH 5.0, preserved with NaF, and stored below 4~ (14). Although free cocaine can be enzymatically and chemically hydrolyzed to ecgonine methyl ester and benzoylecgonine, respectively, Hearn demonstrated that cocaethylene is produced by the liver (10).

Discussion It is apparent that ethanol-specific metabolites can be detected in the absence of ethanol in blood or urine. No ethanol was detected in the urine of 6 of the 13 cocaethylene-positive patients. Patient 5 overdosed on Dihistine 2t/2 h prior to admission. Ethanol was also ingested, as this medication is present as a 5% alcohol elixir. In Patients 6, 8, and 11, serum alcohol concentrations of 38, 32, and 217 mg/dL, respectively, were reported 12-18 h before the urine collection. Patient 7 and the mother of Patient 9 reported an alcohol history of six beers per day were reported in the medical record. From these data, it appears that the rate of clearance for cocaethylene and ecgonine ethyl ester may be slower than that of ethanol itself. This study also shows that isopropyl-specific cocaine metabolites can be detected when isopropanol is ingested with cocaine. Toxic concentrations of isopropanol and acetone were recovered in the urine o f Patients 6 and 10. However, isopropanol specific cocaine metabolites were only detected in the latter patient, despite the fact that the isopropanol concentration was lower than

in the former. It may be p o s s i b l e that in P a t i e n t 6 cocaine and isopropanol were not simultaneously present, or were present in insufficient concentrations in the blood to allow for the formation of isopropyl-specific metabolites. It might also be possible that not all individuals are able to produce these metabolites. The simple presence of a high concentration of isopropanol may be insufficient to spontaneously convert cocaine to these derivatives, since incubation of the urine from Patient 6 for 24 h at room temperature was not sufficient to produce these metabolites. Further studies are needed to determine the pathophysiologic significance of these metabolites. Both of these patients were discharged following uneventful recoveries. The data from Patient 9 demonstrates that ethanol-specific cocaine metabolites can be present in newborns of cocaineaddicted mothers. The BE concentration from the serum of the mother was 388 ng/mL. This sample, collected 2 h before delivery, was negative for the presence of cocaethylene. However, because this sample was not collected in NaF, all of the cocaethylene may have hydrolyzed to benzoylecgonine. Urine from the mother was not available for analysis. Both ethanol and cocaine can readily pass through the placenta from mother to newborn, and they contribute to developmental problems in the neonate (15,16). It is also likely that the ethanol-specific metabolites are also able to enter fetal blood from maternal blood. Whether or not the fetal liver can convert cocaine and ethanol to these metabolites remains unknown. Most animal studies and case reports describe the toxic effects of cocaine alone. There have been only a few studies involving the combination of cocaine with other drugs, such as alcohol. Part of the reason for this is because many clinical laboratories confirm cocaine using selected ion monitoring GC/MS techniques targeted towards benzoylecgonine alone. Thus, these lab-

Table II. Laboratory Findings for Cocaethylene-Positive Patients* Case no. 1 2

Sample type serum urine serum-NaF

urine 3 4 5

6 7

8 9 10t't 11 12

plasma-EDTA urine serum-NaF urine serum-NaF urine serum urine urine serum-NaF urine urine serum-NaF urine serum-NaF urine serum-NaF

urine 13

serum-NaF urine

Relative time (h)t

Free cocaine

BE

EME

Cocaethylene

EEE

Alcohol (mg/dL)

0 0 0 0 0 +1 0 -0.5 0 +2 0 +12 0 0 +12 0 0 +2 0 +18 0 +1 0 +2

52 1050 132 642 < 25 35,300 < 25 97 65 4020 < 25 127 57,000 < 25 900 26 < 25 540 < 25 1000 < 25 38,200 < 25 17,500

335 9480 1170 12,400 544 9050 87 1170 1860 20,600 < 25 607 4000 1202 97,300 1930 490 31,000 300 20,000 400 206,000 2200 160,000

100 3700 95 9550 252 29,200 21 874 38 38,600 < 25 544 37,400 96 12,700 537 84 5000 < 25 1300 < 25 150,000 < 25 82,500

Alcohol-specific cocaine metabolites in serum and urine of hospitalized patients.

The serum and urine from 44 consecutive patients that tested positive for the cocaine metabolite benzoylecgonine (BE) were examined for free cocaine, ...
417KB Sizes 0 Downloads 0 Views