BIOPHARMACEUTICS AND DRUG DISPOSITION, VOL. 13, 445-460 (1992)

PHARMACOKINETICS AND METABOLISM OF CODEINE IN HUMANS TOM B. VREE*t and CORRIEN P. W. G. M. VERWEY-VAN WISSEN* *Department of Clinical Pharmacy, and tlnstitute of Anaesthesiology, Academic Hospital Nijmegen Sint Radboud, Geert Grooteplein Zuid 8, Nijmegen, The Netherlands

ABSTRACT Codeine (30 mg phosphate) was metabolized by eight human volunteers to the following six metabolites: codeine-6-glucuronide 81 .O+ 9.3 per cent, norcodeine 2.16? 1.44 per cent, morphine 0- 562 0-39 per cent, morphine-3-glucuronide 2-10? 1-24 per cent, morphine-6-glucuronide 0-80k 0.63 per cent, and normorphine 2.44f 2-42 per cent. Two out of eight volunteers were unable to O-dealkylate codeine into morphine and lack therefore the cytochrome P450 IID6 isoenzyme. The half-life of codeine was 1 a47 f 0.32 h , that of codeine-6-glucuronide 2.75 f 0.79 h, and that of morphine-3-glucuronide 1a71 ? 0-51 h. The systemic clearance of codeine was 2280? 840 ml min-I, the renal clearance of codeine was 93.8 k 29-8 ml rnin-', and that of codeine-6-glucuronide was 122f 39.2 ml min-'. The plasma AUC of codeine-6-glucuronide is approximately 10 times higher than that of codeine. Protein binding of codeine and codeine-6-glucuronide in vivo was 56- 1 ? 2-5 per cent and 34.0f3.6 per cent, respectively. The in vitro protein binding of norcodeine was 2 3 - 5 k 2 .9 per cent; of morphine, 46.5? 2.4 per cent; of normorphine, 23.5 k3.5 per cent; of morphine-3-glucuronide, 27 -0k0.8 per cent; and of morphine-6-glucuronide, 36.7k3.8 per cent. KEY WORDS Codeine Norcodeine Codeinedglucuronide Morphine Morphine-3-glucuronide Morphine-6-glucuronide Renal excretion protein binding humans

INTRODUCTION The morphinomimetic drug codeine is used as an analgesic and antitussive agent. Codeine is metabolized by conjugation with glucuronic acid to codeine-6-glucuronide,and to a minor extent via N-demethylation to norcodeine and via O-demethylation to morphine.IJ The possible metabolites generated by these three metabolic pathways are shown in Figure 1. For the correlation of observed analgesic or antitussive effect and the corresponding plasma concentration of the active agent, most if not all analytical methods paid attention to codeine, norcodeine, and r n ~ r p h i n e . ~Formation -~ of the latter agent may contribute to development of dependence associated with chronic Addressee for correspondence: Dr T. B. Vree, Department of Clinical Pharmacy, Academic Hospital Nijmegen Sint Radboud, Geert Grooteplein Zuid 8, 6525 GA Nijmegen, The Netherlands.

0142-2782/92/060445- 16$8.OO 01992 by John Wiley & Sons, Ltd.

Received 22 October 1991 Revised 6 January 1992

446

TOM B. VREE AND CORRIEN P. W. G . M. VERWEY-VAN WISSEN

codeine administration. It is becoming increasingly obvious that glucuronidation of morphine at the 6-position does not reduce the analgesic effect; morphine-6-glucuronide has a stronger analgesic effect than morphine itself,1°-18 and its plasma concentration is higher than that of morphine.I2 In analogy to morphine, codeine-6-glucuronide should possess an activity similar to codeine itself; analogous to morphine, the plasma concentration of codeine-6-glucuronide is much higher than that of codeine itself .19,mRecently the HPLC analysis of codeine with its metabolites and conjugates was developed.20 The aim of this investigation was to study the pharmacokinetics, metabolism, and renal excretion of codeine with its metabolites norcodeine, morphine, codeine-6-glucuronide, morphine 3-glucuronide, and morphine 6-glucuronide in humans.

&

NorrnorphlnMqlucuronIde

Norrnorphlne-3glucuronlde

Norrnorphlne

I

Morphlnedglucuronlde

Codeine

Codelnedglucuronlde

Norcodelno

NortnoQhlne

Figure 1. Molecular structures of codeine and its possible metabolites

CODEINE PHARMACOKINETICS AND METABOLISM

447

MATERIALS AND METHODS Drugs

Codeine and codeine phosphate were obtained from the hospital pharmacy. Norcodeine, normorphine, morphine-3-glucuronide,and morphine-6-glucuronide were obtained from Sigma (St Louis, MO, USA). Morphine was obtained from Diosynth (Oss, The Netherlands). Codeine-6-glucuronide was identified in human plasma and urine by deglucuronidation.

Subjects Eight human volunteers (four males, four females; seven Caucasians and one Japanese) with ages ranging from 20 to 50 years, took an oral dose of 30 mg codeine phosphate (22 mg codeine; 10mg tablets obtained from the Hospital Pharmacy). The volunteers did not use any concomitant medication. The study had the approval of the hospital Ethical committee and informed consent was obtained from the volunteers.

Drug analysis

A Spectroflow 400 high performance liquid chromatograph (Kratos, Rotterdam, The Netherlands) equipped with an electrochemicaldetector (Model 5100 A, ESA, Kratos, Rotterdam, The Netherlands) was used. A stainless-steel column (250 x 4.6 mm I.D.; Chrompack, Middelburg, The Netherlands)packed with CP-tm-Spher C8, particle size 8 pm (Chrompack, Middelburg, The Netherlands) was used. The solvent consisted of 0.25 g heptanesulphonic acid (HSA), 0.68 g KH2P04, 5 - 5 ml H3P0425 per cent in water adjusted to a total volume of 425 ml, then 20 ml methanol and 60 ml acetonitrile were added, and the solvent was degassed with a gentle stream of helium. The flow was 2*0mlrnin-' at a pressure of 9.1 MPa (150 bar). The temperature of the column was 40 "C. The injection volume was 50 pl. The capacity factors of codeine and its metabolites are morphine-3-glucuronide, 2.2; morphine-6-glucuronide, 3 6; normorphine, 6 - 0 ; morphine, 6.3; codeine-6-glucuronide, 9.9; norcodeine, 17 -4; and codeine, 19-2. For sample injection a Marathon autosampler (Interscience, Breda, The Netherlands) was used. The analytical cell (Model 5010, ESA) consists of two units, which can be regulated separately. The potentials of detector cell 1 and detector cell 2 were + 0-65V and + 0.85 V for the measurement of morphine-3-glucuronide, codeine-6-glucuronide, norcodeine, and codeine. For the measurement of morphine-6-glucuronide, normorphine, and morphine the potentials of the cells were +0.25 V and +0.35 V, respectively. Both cells were connected to a twochannel Chromjet integrator (Spectra Physics, Eindhoven, The Netherlands).

448

TOM B. VREE A N D CORRIEN P. W. G . M. VERWEY-VAN WISSEN

Sampling Blood samples were collected at regular time intervals after administration for 24 h, centrifuged at 2600 g and the plasma separated and stored at - 20 "C prior to analysis. Urine was collected on spontaneous voiding for 48 h, and an aliquot stored at - 20 "C prior to analysis.

Sample preparation Plasma. Plasma (0.6 ml) was subjected to an extraction with the Baker-10 extraction system (Baker Chemicals, cat. no. 70180, Deventer, The Netherlands) fitted with 1 ml disposable extraction columns packed with reversed phase Octadecylsilane (C18) bonded to silica gel (cat. no. 7020-01). The extraction column was conditioned with 2 x 1 ml of methanol, 2 x 1 ml of water and 1 ml 500 mM diammoniumsulphate of pH 9.3. Plasma (0.6 ml) diluted with an additional 0.5 ml of 500 mM diammoniumsulphate buffer (PH 9-3), was placed on the column. The column was washed with 2 x l m l of 5mM diammoniumsulphate(PH 9.3). The sample was eluted with 400 p1 of the mixture 25%. 0.01 M KH2P0,+ 10 per cent acetonitrile+ 10 per cent Urine. Urine (0-2 ml) and 0.5 ml of 500 mM diammoniumsulphate (pH 9.3) were placed on the extraction column. The urine was processed in a similar way to plasma but with the following differences: the column was washed four times with 1 ml of 5 mM diammoniumsulphate (pH 9- 3) and the sample was eluted with 1 ml of 0.01 M KH2P04+ 10 per cent acetonitrile+ 10 per cent H3P0425 per cent. Deconjugation Deglucuronidation was carried out with 100 p1urine, 50 pl of B-glucuronidase (Escherichia coli, type VII-A, Sigma cat. no. G-7646, 20,000Uml-l) and 150 ~ 1 0 . 0 2M KH2P04buffer of pH 6.8. The reaction was completed in 17 h at 37 "C. Thereafter 0.7 ml500 mM diammoniumsulphate was added and the sample extracted.

Calibration curves glucuronide The increase in the concentrations of codeine in urine after deconjugation represented the concentration of the conjugate. A calibration curve was constructed with the help of the following formula: [ Codeine-6-gluc] = d [ codeine I .Mcodeine-6-g,uc/Mcodeine

where d [codeine] is the difference between the concentrations of codeine before

CODEINE PHARMACOKINETICS AND METABOLISM

449

and after deconjugation and M is the relative molecular mass. Thereafter calibration curves for codeine-6-glucuronide were constructed by spiking urine with known concentrations of the compound. Calibration curves of plasma codeine-6-glucuronide were obtained by spiking plasma with known (measured after deglucuronidation) concentration of the glucuronides in urine.20

Recovery and reproducibility The calibration curves were prepared by adding a variable quantity of stock solution to blank plasma/urine. The correlation coefficients were 0.999 or more for morphine-6-glucuronide,morphine-3-glucuronide,normorphine, morphine, codeine, norcodeine, and codeine-6-glucuronide.20 The recoveries from the extraction procedure were, respectively: morphine 92 per cent; morphine-3-glucuronide, morphine-6-glucuronide, and codeine-6-glucuronide 100 per cent; codeine and norcodeine 85 per cent. The reproducibility was approximately 95 per cent for the concentration ranges of all the above mentioned compounds.20

Quantification limit The quantification limit for codeine, codeine-6-glucuronide,and norcodeine is 5 ng ml- in plasma and 25 ng ml- in urine. The quantification limit for morphine is 5 ng xn- in plasma and 20 ng ml- in urine, for morphine-3-glucuronide

long in plasma and 70ngnl-I in urine; and for morphine-6-glucuronide in plasma and 50 ng ml- in urine (signal to noise ratio of 3).

5 ng ml-

Protein binding Protein binding (in vivo) of codeine and codeine-6-glucuronidewas measured in volunteer plasma samples by means of Amicon micropartition system MPS-1 (Amicon, Danvers, Ma 01923, USA; Oosterhout, The Netherlands). An aliquot of 1 ml was placed in the filter, centrifuged for 30 min at 40oO g. Of the filtrate 600 pl was extracted as plasma and injected onto the column. The average protein binding (kSD) in each volunteer was calculated from six plasma samples obtained at different time intervals over a concentration range of 0-1-1 pg ml- l . No drug binding to the membranes was observed. For the in vifroprotein binding measurement, samples of bank pooled human plasma were spiked with known concentrations of codeine and of its metabolites and treated as mentioned above.

Renal clearance The apparent renal clearance values of the parent drug and its conjugates was calculated by two methods:

450

TOM B. VREE AND CORRIEN P. W. G . M. VERWEY-VAN WISSEN

1. The average renal clearance rate in each urine sample, by dividing the rate of excretion by the total plasma concentration at midpoint of the measured time-interval. 2. The total amount excreted in the urine was divided by the AUC,. The correlation coefficient between these two methods was 0 954.

-

Pharmacokinetic and statistical analysis Regression lines, standard deviations, and Student’s t-tests were calculated according to standard statistical procedures AUC, was calculated using the trapezoidal rule (with extrapolation Ct/P to t = 03). Total body clearance C1= dose/AUC, (assuming F = 1). Mean Residence Times (MRT = MRTDisposition -k MRTAbsorption with MRTAbsorption < < MRTDisposition) Of codeine and the conjugates were calculated as AUMC.AUC- l . Intrinsic of codeine-6-glucuronide) was calculated as MRT( = MRTDisposition MRTMetabolite - MRTParent.22Steady state volume of distribution Vss was calculated as dose. AUMC.AUC-2. Peak plasma concentrations (Cmm)were read from the plasma concentration-time curve. The tl/,values were calculated from the slope of the terminal portion of the semilogarithmic plasma concentration-time curve by linear regression analysis.

RESULTS Figure 2 shows the plasma concentration-time curves and renal excretion ratetime profiles of codeine and its metabolites in a volunteer after an oral dose of 30 mg codeine phosphate ( = 22 mg codeine base). Codeine was rapidly conjugated with glucuronic acid into codeine-6-glucuronide. Codeine, in this low dose, was only measurable in plasma for 5 h, while codeine-6-glucuronide was detectable in plasma for at least 15 h. The AUC, of codeine-6-glucuronide constituted 91-6 per cent of the sum of AUC, values of codeine and its metabolites measurable in plasma (Table 3). This volunteer O-dealkylates codeine into morphine, which in turn is instantaneously and predominantly glucuronidated at the 3 position. The half-lives of codeine and morphine-3-glucuronide were almost similar, 1- 8 h, while that of codeine-6-glucuronide was longer, 3 h. The renal excretion rate-time curves show that codeine-6-glucuronide was the main metabolite (77.3 per cent) and that codeine was excreted only as a small portion of the dose (3-9 per cent). Morphine was formed and excreted accounting for 0.31 per cent of the dose, its morphine-3-glucuronide metabolite for a 10 times higher fraction, 2 - 7 per cent, and its morphine-6-glucuronide metabolite for 0.55 per cent. Norcodeine was also formed and accounted 1.5 per cent of the administered dose. The observed half-life of codeine-6-glucuronide was 3 h, and longer than the observed half-life of codeine (1 - 8 h).

CODEINE PHARMACOKINETICS AND METABOLISM

45 1

- plasma conc. ug/ml - renal excr. rate ug/min

-- ...... 1 ,..: i - I

100

22 mg Codeine

P.O.

i..,

--

i

I

i..,

!*i

i

- i

L..?

'"1 "'"I

t

'0' I

:I-

i

-I

----

I

I

"""i .....I.,

I

Figure 2. Plasma concentration-time curves and renal excretion rate-time profiles of codeine (C), its metabolites norcodeine (norC), morphine (M), and the glucuronides codeine-6-glucuronide (C-6-gluc), morphine-3-glucuronide(M-3-gluc), morphine-6-glucuronide(M-6-gluc) in a volunteer after an oral dose of 30 mg codeine phosphate (22 mg codeine)

Figure 3 shows data similar to that in Figure 2, but this volunteer is unable to 0-dealkylate codeine into morphine. In plasma and urine only codeine, codeine-6-glucuronideand norcodeine was detected. This lack of 0-dealkylation slightly increases the amount and percentage of codeine and norcodeine that was excreted in urine.

452

TOM B. VREE AND CORRIEN P. W. G. M. VERWEY-VAN WISSEN

.

plasma conc. ug/ml

renal excr. rate ug/min 101 9..

!""i i ! i

i j i

22 mg Codeine p.0.

'....! i

i...)

i

t......b

i !.....b

i......)

I(

i

...a

,.i

..., b.......) I.....

C-6-gluc 82.3 %

...........

i b.....,

i

?....b

I._.

0.

0

\C

C-6-gluc T1/2 4 h

T1/2 1.8 h

1

I

0

!!

I

10

I 15

I

20

I

25

1

30 h

Figure 3. Plasma concentration-time curves and renal excretion rate-time profiles of codeine (C), its metabolite norcodeine (norC), and the glucuronide codeine-6-glucuronide (Cd-gluc), in a volunteer after an oral dose of 30 mg codeine phosphate (22 mg codeine). This volunteer is unable to 0dealkylate codeine into morphine and lacks the cytochrome P450 IID6 i s ~ e n z y m e ' ~ ~ ~ '

Table 1 summarizes the pharmacokinetic data of the eight volunteers. Table 2 shows the AUC values in pmol 1- h and in percentage of the sum of AUC, values. It is clear that codeine-6-glucuronide is the principal compound (89 f4 per cent) that was present in the body after the oral dose of 30mg codeinephosphate. Table 3 shows the percentage of the dose excreted in the urine. The

*Below detection limit.

2365 297

514 260

2350

1.86 3.38 2.42 1.52

1.0 2.0 1.3

82.0 830 60.0 6.8

450

2321

3.23 5-22 4.01 1.99

1.8 3.0 1.8

*

41.5 1040 54- 1

676

3786

2.98 4-53 4.35 1.55

1.3 3.0 2.7

*

34.0 781 65.8

1673 416

311

2.53

1*70 1659

4.14 6.67

*

*

1.8 4-0

*

*

70-0 75 1

7 m 83

3.21 4-91

*

*

1.0 2-9

*

*

74.8 693

Subject number, gender and body weight (kg) 3 4 5 6 f f m m 80 75 68 90

2- 10 2.83 2.72 0.73

1-1 1.7 1.5

*

72.9 1518 32.2

2 f 62

3012

2.85 3.44 2.56 0-59

1.6 2.0 1.6

Half-life (h) Codeine Codeine-6-gluc Morphine-3-gluc

MRT (h) Codeine Codeined-gluc Morphine-3-gluc Codeined-gluc CL (mlmin-I) Codeine vss (1) Codeine

40.8 1271 37.7 6-6

1 f 60

C,,(ng ml-9 Codeine Codeined-gluc Morphine-3-gluc Morphined-gluc

Parameter

Table 1. Pharmacokinetic variables of codeine and its metabolites in humans

189

1115

2.83 5.27 3.61 2.40

1.4 3.4 1.4

*

118.8 863 71.3

8 m 70

389 (157)

2280 (840)

2.90 (0.70) 4-53 (1.26) 3.27 (0.82) 1-63 (0.70)

1.47 (0-32) 2-75 (0.79) 1.71 (0.51)

66.9 (27.9) 968 *O (290) 53.5 (15.6)

Mean (SD)

P VI W

*Not detected.

3-9 93.8 2 3

AUC (To total) Codeine Codeined-gluc Morphine-3-gluc

-

0.409 9.82 0.237

1 f

AUC,_,(pmoll-' h) Codeine Codeined-gluc Morphine-3-gluc

Parameter

5.9 92.0 2.1

0.520 8.16 0.188

2 f

7.2 88.0 4.8

4.4 91 - 9 3.7

0.530 11-08 0.447 4.2 87.6 8.2

0.326 6.75 0.635

Subject number and gender 4 5 f m

0.525 6.42 0.353

3 f

*

11.4 88.6

*

5-90

0.762

m

6

*

7.7 92.3

*

0.736 8.82

7 m

5.5

9.9 84.6

1* 102 9.36 0 * 609

8 m

Table 2. AUC values &moll-' h and per cent total AUC) of codeine and metabolites in humans

7.37 (3.05) 89.81 (3.09) 4-43 (2.28)

Mean (SD)

m

CODEINE PHARMACOKINETICS AND METABOLISM

455

average percentage of the dose excreted reaches 91 - 9 k 4 - 8per cent of the dose, which suggests that the oral bioavailability of codeine is 100 per cent (F=1). The renal clearance values of codeine and codeine-6-glucuronide were similar, being approximately 100mlmin-l as shown in Table 3. The protein binding experiments show that codeine and its metabolite morphine had low protein binding values of approximately50 per cent, while that of the glucuronides varied around 30 per cent, and those of norcodeine and normorphine were approximately 20 per cent (Table 4).

DISCUSSION

The metabolism of codeine involves N-demethylation to norcodeine, 3-0demethylation to morphine, and glucuronide conjugation at the 6-position. Also norcodeine can be glucuronidated at the 6-position and morphine at the 6 and 3 positions, leading to the possible nine metabolites as depicted in Figure 1. Substitution at the 6-OH position does not lead to a reduction of analgesic activity. 12J7 Therefore a pharmacokinetic analysis must deal with preferably all the metabolites given in Figure 1. A series of authors deal with one or two codeine m e t a b o l i t e ~ , ~ ~while ~ ~ ~ codeine-6-glucuronide ~~3-2~ was included in the N e l ~ o n P, ~e r s s ~ n and , ~ ~Yue.2*30.31 studies of Chen,19*28

Half-life The low dose of 30 mg codeine phosphate (22 mg codeine base) resulted in a very short half-life of 1- 8 h for codeine, which differed from the half-life of codeine-6-glucuronide. In this aspect the group of volunteers differed from the human subject reported by Chen et al. ,19,28 who showed identical half-lives for the two compounds. The (maximal) plasma concentrations of codeine in our group of volunteers were slightly lower than those reported by Chen.19128When a dose of 60 mg codeine phosphate was taken orally, the codeine elimination was biphasic, with half-lives of 1 8 and 3 * 5 h, and then the half-lives of codeine and its 6-glucuronide metabolite were similar (Vree, unpublished results). This The is in agreement with data reported by Guay et al.32and Chen et a1.19*28 intrinsic MRT of codeine-6-glucuronide is shorter than that of codeine, 1.63k0-70 vs 2.90f0.70h (p=0.035). This indicates that the rate of glucuronidation of codeine governs the kinetics of codeine, and contributes to the apparent t% of 3 h of codeine-6-glucuronide.

Metabolism Two of the eight volunteers were unable to O-demethylate codeine and to form morphine with its glucuronide metabolites. This lack of 0demethylation capacity is correlated with the lack of Odemethylation capacity of

98.8 111 42.1

CL ml min-I Codeine Codeine-6-gluc Morphine-3-gluc

2 f

86.5 160 65.4

101 - 0

0.1

1.0 0.3

0.03

3.7 95.0 0.9

*Not detected; subject number 8 Japanese.

0-8 0.2 0.3 93.9

*

3.3 88.6 0.7

1 f

Excreted 070 dose Codeine Codeine-6-gluc Norcodeine Morphine Morphine-3-gluc Morphined-gluc Normorphine Total

Parameter

90-0 85.8 78.9

1.5

0.3 2.7 0.6 2.3 88.6

1.5

0.7 2.1 1.1 2.6 93.4 67.1 158 71 - 9

3.9 77.3

2.9 82.5

154 142 35.2

4.1 78.2 1.2 0.9 1.8 1.9 2.6 90.7

Subject number and gender 5 3 4 f f m

*

72.2 169

89.8

* * *

*

4.5 81.3 4.0

6 m

*

116 114

93.6

* * *

*

7-0 82.3 4.3

7 m

65.6 82-5 84.1

5.9 62.9 3.2 0.9 4.2 0.7 6.8 84.6

8 m

93.8 (29.8) 122 (39.2) 62.9 (20.0)

4.41 (1.38) 81-01 (9.31) 2.16 (1.44) 0.56 (0.39) 2.10 (1.24) 0.80 (0.63) 2.44 (2.42) 91 -95 (4.81)

Mean (SD)

Table 3. Percentage of the dose excreted in the urine and renal clearance of codeine and its metabolites in eight human volunteers

€

z

m

v)

v)

I

w m m