disposition and first-pass metabolism of ethanol in humans: is it gastric or hepatic and does it...

Disposition and first-pass metabolism of ethanol in humans: Is it gastric or hepatic and does it depend on gender?

Objective: To assess the extent and site of the first-pass metabolism of ethanol and to examine whether first-pass metabolism and disposition of ethanol are dependent on gender. Methods: After a standardized lunch, healthy subjects (six women and six men) received on two separate occasions a 60-minute intravenous infusion of ethanol (0.3 gm/kg) an d concomitantly an equimolar dose of d,-ethanol/kg either orally (over 20 minutes) or intraduodenally (infused over 30 minutes). Blood levels, urinary excretion of da- and d,-ethanol, and sedative effects were monitored for 6 hours. Disposi- tion and first-pass metabolism of ethanol were evaluated by applying an open two-compartment model with Michaelis-Menten elimination. Results: Comparison of the corresponding intravenous/oral versus intravenous/intraduodenal data of each individual revealed that total first-pass metabolism (gastric plus hepatic) was not pronounced in either males (9.1% + 4.0%; mean + SD) or females (8.4% f 3.1%) and that this first-pass metabolism was partly of gastric origin. Dose-corrected values for area under the blood concentration-time curve were on average 28% higher (p < 0.0001) in the women than in the men. Mean total blood ethanol disappearance rate was higher (p < 0.001) in women (3.92 -C 0.40 mmolfi * hr) than in men (3.19 + 0.48 mmol/L . hr). Renal clearance was gender-independent and negligible. A linear relationship (p < 0.001) could be found between the blood levels of ethanol and sedation index. Because the slope was steeper in women (1.04) than in men (0.42) a higher central nervous system sensitivity to the sedative effects of ethanol in women can be assumed. Conclusions: Under realistic life conditions (social drinking of moderate doses of ethanol after a light lunch) only a minor, gender-independent first-pass metabolism is observed that is partly of gastric origin. (Clin Pharmacol Ther 1996;59:503-13.)

Eva Ammon, MSc, Christian SchZfer, MD, Ute Hofmann, PhD, and Ulrich Klotz, PhD Stuttgart, Germany

The widely used “social” drug ethanol has been extensively studied for decades. Especially the pharmacokinetics of ethanol has received considerable attention because of its forensic importance and its nonlinearity which is due to Michaelis-Menten elimination (for review see Holford’). Likewise, there is considerable controversy concerning the extent and site of the presystemic elimination of ethanol. The different arguments for and against a gastric or he-

From the Dr. Margarete Fischer-Bosch Institute of Clinical Phar- macology and Robert Bosch Hospital.

Supported by the Robert Bosch Foundation, Stuttgart, Germany. Received for publication Sept. 6, 199.5; accepted Nov. 18, 1995. Reprint requests: Ulrich Klotz, PhD, Dr. Margarete Fischer-

Bosch Institut fur Klinische Pharmakologie, Auerbachstrasse 112, D-70376 Stuttgart, Germany.

Copyright 0 1996 by Mosby-Year Book, Inc. 0009-9236/96/$5.00 + 0 13/l/70779

patic contribution to its first-pass metabolism have been recently summarized.2~” In addition, there are some indications that this first-pass metabolism may be lower in women than in men, which could explain the higher blood ethanol levels observed in female subjects.4

Disposition and first-pass metabolism of ethanol seem to be affected both by several characteristics of the tested subjects (e.g., genetic constitution, age, gender, nutritional status, gastrointestinal motility, and liver function) and the mode of ethanol intake (e.g., dose, route and time of administration, volume and concentration of ethanol, and food consumption).‘-” These numerous confounding factors and the well-known considerable intraindividual and interindividual variability of the pharmacokinetics of ethanol435 contribute to the difficulties in generaliz- ing data obtained from a single trial. In particular, in

503

504 Amnmon et al. CLINICAL PHARMACOLOtiY & THERAPEUTICS

MAY 1996

all studies published so far the extent of first-pass metabolism was assessed by separate experiments with intravenous and oral dosing of ethanol on different occasions. The so-called stable isotope tech- nique6 can be applied in such studies to reduce the large day-to-day variability in the disposition of high clearance drugs, including ethanol. By concomitant administration of unlabeled (da) and deuterium (d,)-labeled ethanol through two different routes (e.g., intravenous as reference and oral), bioavailability and first-pass metabolism can be estimated more accurately because intraindividual variability will be absent.7 We have applied this approach to assess the first-pass metabolism of ethanol and to differentiate between hepatic and gastric first-pass metabolism. To accomplish this aim, healthy male and female subjects received, by way of three different routes (intravenous, oral, and intraduodenal), da- and d,-ethanol under realistic life conditions (e.g., moderate dosing with ethanol after a light meal at lunch time).

MATERIAL AND METHODS Subj,,t,. Twelve drug-free, healthy subjects (age

range, 24 to 52 years; six men with a mean body weight of 72 + 9 [SD] kg and six women with a mean body weight of 65 -+ 10 kg; one light smoker in each group) participated in the study after each signed an informed consent form. The study was approved by the Ethical Committee of our hospital. Each subject had a medical examination and was carefully asked about former illnesses, usual alcohol and drug intake, and smoking habits. All subjects had normal endocrinologic, hepatic, and renal functions as assessed by routine laboratory parameters. Subjects (“occasional social drinkers,” 565 gm ethanol per week) refrained from ethanol for at least 36 hours before the study.

Study design. The study was a randomized cross- over design and consisted of two parts separated by a l-week interval. On both occasions, all subjects received at noon a standardized lunch (chicken with rice a la mandarin; 27 gm fat, 5 gm protein, 57 gm carbohydrates, 381 kcal) and a 60-minute intravenous infusion of da-ethanol (6.5 1 mmol corresponding to 0.3 gmlkg body weight; 95% ethanol diluted with a 10% glucose solution to an approximately 5% ethanol solution) was started at 12:30 PM. At this time, the equimolar dose of 2,2,2-d,-ethanol was administered either orally over 20 minutes (diluted to 250 ml with 10% glucose) or infused intraduodenally over 30 minutes (approximately 10% ethanol

solution in 10% glucose) through a gastroduodenal feeding tube (outer diameter, 2.8 mm; Freka, Fre- senius, Bad Homburg, Germany). The position of the infusion port in the lower duodenum (below the major duodenal papilla) was checked by transillumi- nation.

Blood samples (5 ml) were collected before dosing (to test for previous intake of ethanol), at 10, 20 (end of ethanol drinking), 30 (end of intraduodenal infusion), 45, 60 (end of intravenous infusion), 75, and 90 minutes, and at 2, 2.5, 3, 4, 5, and 6 hours from a vein in the contralateral arm of each subject. At the same time points, sedative effects of ethanol were assessed by a sedation index formed from four visual analogue scales (each 10 cm) and by measurement of the choice reaction time by the Leeds Psychometer (ZAK, Simbach/Inn, Germa- ny). Total urine was collected in two fractions (0 to 3 hours and 3 to 6 hours). No food was allowed during the study, and subjects remained in supine position.

Sample preparation and analysis. Concentrations of d,- and d,-ethanol in blood and urine were determined by a specific gas chromatography/mass spectrometry (GC/MS) method with use of d,- ethanol as the internal standard. Immediately after collection, samples were divided into aliquots (0.2 ml), mixed with internal standard (1 kmol), and stored at -20” C until analysis. In brief, after pre- cipitation of protein with perchloric acid (0.5 mol/L), extraction with dichloromethane (2 X 2 ml), and derivatization (1 hour at room temperature) with 3,5-dinitrobenzoyl chloride (10 mg) and pyri- dine (5 l~,l), samples (2 ~1) were assayed by GUMS in electron-impact selected ion monitoring (SIM) mode (70 eV) with use of the masses 240, 243, and 245 for the d,-, d,-, and d,-derivatives, respectively. We used a 5790B mass-selective detector (Hewlett- Packard Company, Palo Alto, Calif.) with a 5890A gas chromatograph, equipped with a capillary col- umn (HP-l, 25 m x 0.2 mm internal diameter; film thickness, 0.11 pm). With each series of biological samples, a standard curve (7 calibrators in the range of 0.25 to 12.5 mmol/L) was prepared and three different quality controls in duplicate were run.

Pharmacokinetic analysis. After intravenous administration, pharmacokinetic parameters of ethanol were calculated according to an open two-compartment model with Michaelis-Menten elimination from the central compartment by use of the TopFit programs with the weighting function of l/y. Blood ethanol disappearance rate was calculated from the

CLINICAI. PHARMACOLOGY & THEP.Al’EUTI(‘S \‘OLUME 59, NUMBEK 5 Amvnon et al. 505

“‘I,, , , , / / , , I I , , I, / , I / / , , I I I I, I ,

0 0 1 1 2 2 3 3 4 4 5 5 6 6

time [h] time [h]

Fig. 1. Blood concentration-time profiles of do-ethanol (squares) and d,-ethanol (circles) in Fig. 1. Blood concentration-time profiles of do-ethanol (squares) and d,-ethanol (circles) in one male subject (U.K.) after the concomitant oral administration of equimolar doses of one male subject (U.K.) after the concomitant oral administration of equimolar doses of do-ethanol (6.51 mmol corresponding to 0.3 gm ethanol/kg) and d,-ethanol. do-ethanol (6.51 mmol corresponding to 0.3 gm ethanol/kg) and d,-ethanol.

pseudolinear decline of the blood levels from 2 to 5 hours by linear regression. Area under the whole blood concentration-time curve (AUC) was calculated by applying the trapezoidal rule, with extrapolation from the last measured concentration to infinity by use of the final slope. Total first-pass metabolism was calculated from the differences between the AUC values from the concomitant intravenous and oral administrations. Likewise, hepatic first-pass metabolism was assessed from the concomitant intravenous and intraduodenal administrations and gastric first-pass metabolism from the differences between total and hepatic first-pass metabolism. The apparent volume of distribution (V,,) was calculated model-independently by the equation Dose * AUMC/(AUC)2, in which AUMC is the area under the first moment curve, and renal clearance was calculated from the ratio Ae/AUC over 6 hours, in which Ae is the amount of unchanged drug excreted into urine.

Statistical analysis. Statistical analysis to evaluate differences between genders was performed with the two-sided, unpaired Mann-Whitney U test. In each group, the two intravenous experiments were compared by applying the two-sided paired Wilcoxon signed-rank test. A p value of less than 0.05 was

considered to be significant. All values are reported as mean 2 SD.

RESULTS With the described GC/MS assay, d,- and d,-

ethanol blood levels down to 0.05 mmol/L (limit of quantification) could be measured. The within-day reproducibility of the method was assessed by ex- tracting and analyzing up to eight identical samples in 1 day. The concentrations added were 0.25 mmol/L, 5 mmol/L, and 10 mmol/L for d,-ethanol and 0.25 mmol/L and 10 mmol/L for d,-ethanol. The coefficients of variation were 10.62% (mean, 0.23 mmol/L; IZ = 7) 2.96% (mean, 5.35 mmol/L; n = 8) and 1.36% (mean, 11.53 mmol/L, n = 8) for d,,- ethanol and 5.28% (mean, 0.23 mmol/L; n = 8) and 1.09% (mean, 10.71 mmol/L; n = 8) for d,-ethanol, respectively. The between-day reproducibility was determined by measuring aliquots of three different quality controls (0.5 mmol/L, 2.5 mmol/L, and 8.75 mmol/L) on each day. The coefficients of variation were 7.25% (mean, 0.54 mmol/L; n = 32) 6.42% (mean, 2.55 mmol/L, n = 32) and 8.50% (mean, 8.95 mmol/L; II = 32) for do-ethanol and 13.43% (mean, 0.55 mmol/L; n = 32) 6.32% (mean, 2.55 mmol/L; y2 = 32) and 6.21% (mean, 8.94 mmol/L;

506 Amman et al. CLINICAL PHAFUdACOLOGY &THERAPEUTICS

MAY1996

8 12 F.S.

16 10

14

0 0 0 1 2 3 4 5 6 0 1 2 3 4 5 6

12

GE

F'I

SO.

0 1 2 3 4 5 6 0 1 2 3 4 5 6

he M time %I

0 0 1 2 3 4 5 6

Fig. 2, A. Individual blood concentration-time profiles of ethanol in six male subjects after a 60-minute intravenous infusion of 0.3 gm/kg do-ethanol (triangles) with concomitant administration of an equimolar dose of d,-ethanol in form of an oral solution (open circles) or a 30-minute intraduodenal infusion (solid circles). Solid and open symbols refer to the two trials, respectively.

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 59, NUMBER 5

0 B

Ammon et al. 507

I%

14 AG AZ

I I

12 10

5 8 p 6 g 6

4 4

2 2

0 0 1 2 3 4

0 5 6 0 1 2 3 4 5 6

18 16

16

16

14

EA MS

2

0 0 1 2 3 4 5 6 0 1 2 3 4 5 6

18

10 16

S.J. 14 S.S. 8

12

6 10

I - 8

0 4 5 UJ 6

2 4

2

0 0 0 1 2 3 4 5 6 0 1 2 3 4 5 6

Fig. 2, B. Individual blood concentration-time profiles of ethanol in six female subjects after a 60-minute intravenous infusion of 0.3 gm/kg do-ethanol (ftiangles) with concomitant administration of an equimolar dose of d,-ethanol in form of an oral solution (open circles) or a 30-minute intraduodenal infusion (solid circles). Solid and open symbols refer to the two trials, respectively.

508 Awwuon et al. CLINICAL PHARh4ACOLOGY & THERAPk.UTICS

MAY 1996

Table I. Individual pharmacokinetic data of ethanol in healthy volunteers

Part I (iv + po) AUC(O-m) BEDR V

Dose (mmollL hr) (mmol/L . hr) (mmoir. hr) aw Km (mmoliL)

Volunteer (mmol) iv PO iv PO Total K.7 WW iv iv

Men U.K. B.B. G.E. F.S. S.O. T.J.

475.4 29.86 25.78 1.70 1.46 3.16 0.403 113.00 25.70 436.3 18.77 16.75 1.99 1.88 3.87 0.475 136.00 19.80 423.3 30.14 28.42 1.58 1.67 3.25 0.400 97.70 14.70 520.9 27.70 23.97 1.31 1.22 2.53 0.450 142.00 46.00 410.2 24.77 23.53 1.37 1.45 2.82 0.437 115.00 15.70 566.5 29.52 27.69 1.53 1.53 3.06 0.393 116.00 9.20

Mean 472.1 26.79 24.36 1.58 1.54 3.12 0.426 119.95 21.85 5 SD + 61.3 2 4.41* t 4.20 ? 0.25 L 0.22* 2 0.45* + 0.033* ? 16.29t 2 13.05

Women S.J. A.G. S.S. E.A. A.Z. M.S.

403.7 31.22 29.26 2.00 1.72 3.72 0.397 86.80 24.50 442.8 33.59 32.04 1.89 1.74 3.63 0.418 112.00 59.60 481.9 32.58 30.34 1.77 2.00 3.77 0.349 80.70 1.65 306.1 23.44 21.20 1.91 1.81 3.72 0.394 71.60 4.23 468.9 31.93 27.69 1.63 1.85 3.48 0.336 100.00 8.51 442.8 30.56 27.60 1.86 2.06 3.92 0.350 86.80 4.63

Mean 424.4 30.55 28.02 1.84 1.86 3.71 0.374 89.65 17.19 ? SD ? 63.9 k 3.64* +- 3.74 2 0.13 2 0.14* ? 0.15” k 0.033* -e 14.34t 2 22.34

iv, Intravenous administration; po, oral administration; id, intraduodenal administration; AUC(O-r-), area under the whole blood concentration-time curve; BEDR, blood ethanol disappearance rate; V,,, apparent steady-state volume of distribution; V,,,, maximal metabolic rate; K,, Michaelis-Menten constant; FPM, first-pass metabolism.

*p < 0.05; tp < 0.01; $p = 0.015.

IZ = 32) for da-ethanol, respectively. In addition, the accuracy over the entire concentration range measured was assessed by the blind addition of various amounts of d,- and da-ethanol to blank blood. There was a close correlation (r2 > 0.99) and good agreement between the spiked concentrations and the actually measured concentrations (y = 0.97x + 2.05 for do-ethanol and y = 1.02x - 1.93 for da-ethanol).

A pilot study with concomitant oral administration of equimolar doses of d,- and d,-ethanol veri- fied that no isotope effect exists. This can be seen in Fig. 1, inasmuch as both curves were identical.

The individual blood concentration-time profiles of the six female and six male subjects after the administration of da- and da-ethanol by different routes are presented in Fig. 2. From the two intravenous experiments it is obvious that day-to-day variability was relatively small, which can be also seen by comparison of the calculated pharmacokinetic parameters (except for apparent Michaelis- Menten constant [K,]) summarized in Table I. According to minimized Akaike, Schwarz, and

lmbimbo criteria, best fits (r2 2 0.99) of the intravenous curves were obtained if a two- compartment open model with linear input rate and Michaelis-Menten elimination (weighting function l/y) was applied. At least 92% of the calculated AUC was based on the measured blood levels because the extrapolated AUC was always 3%.

Peak concentrations between 8 and 18 mmol/L were reached at 60 minutes (end of intravenous infusion) and about 40 minutes after the start of the 30-minute intraduodenal infusion. After oral dosing, peak concentration values were lower (6 to 8 mmol/L) and appeared after 30 to 90 minutes. All dose-corrected AUC values in women exceeded (p < 0.0001) those in men (Table I), which could be due to the smaller (p = 0.033) distribution volume (0.36 -C 0.04 versus 0.40 + 0.04 L/kg). Comparison of the corresponding intravenous and oral data from each individual, a total first-pass metabolism of 9.1% + 4.0% and 8.4% -+ 3.1% could be calculated in men and women, respectively. In both groups, the larger part of this low first-pass metabolism was

C:LINICAI. PHARMACOLOGY & THERAPEL’TI(‘S VOI.L~hl~ 59, NUMBER 5 Awmton et al. 509

Part II (iv + id)

AUC(O-x) (mmollL . hr) iv id

BEDR V n,ux aw Km (mmollL . hr) (mmoliL . hr) (mmollL) FPMrvra, FPMgasMc

iv id Total Vss V-&) iV iv (So) i%,)

33.09 35.06 1.54 1.59 3.13 0.382 121.00 41.00 23.34 22.67 2.05 2.04 4.09 0.394 110.00 7.96 31.77 29.50 1.87 1.47 3.34 0.317 113.00 13.30 30.92 29.49 1.75 1.66 3.41 0.336 107.00 11.20 23.11 20.65 1.27 1.15 2.42 0.419 142.00 17.50 29.61 30.41 1.62 1.53 3.15 0.387 101.00 4.19

28.64 27.96 1.68 1.57 3.26 0.373 115.67 15.86 2 4.35 i- 5.34 5 0.27$ -c 0.291 i- 0.54$ -t 0.038 -c 14.501 ? 13.13

27.80 26.81 2.42 2.36 4.78 0.384 108.00 54.00 40.59 41.73 1.91 1.86 3.77 0.346 82.30 8.60 36.84 33.26 2.33 2.28 4.61 0.280 93.70 7.23 27.59 27.59 2.09 1.89 3.98 0.381 67.00 9.38 31.93 34.53 1.75 1.81 3.56 0.357 101.00 9.98 31.85 30.49 2.08 2.04 4.12 0.329 88.40 6.07

32.77 32.40 2.10 2.04 4.14 0.346 90.07 15.88 25.11 2 5.48 t- 0.25$ F 0.23$ ? 0.4f.Q t- 0.039 k 14.49t 2 18.73

contributed by the gastric site (6.1% + 5.3% and 5.9% t 4.8%, respectively).

Between 2 and 5 hours, d,- and da-ethanol concentrations declined in a superimposable and pseudolinear fashion, and a higher (p < 0.001; n = 12) total blood ethanol disappearance rate was observed in women (3.92 ? 0.40 mmol/L) than in men (3.19 2 0.48 mmol/L). In contrast, men exhibited higher (p < 0.0001) maximal metabolic rate (V,,) values (118 t 15 mmol/L * hr) than women (90 + 14 mmol/L * hr). Minor amounts of unchanged ethanol could be recov- ered in the urine (2% to 4% of dose), and the low renal clearance averaged 6.7 ml/min in men and 8.5 mlimin in women.

In most cases peak ethanol blood levels were reached after 1 hour, whereas maximal sedative effects occurred at variable time points (0.33 to 6 hours). On average, the maximal prolongation of choice reaction time was 20%. Sedation index in- creased maximally from basal mean levels of 5 to an average of 28. Between all individual sedation indexes and total ethanol blood levels a significant @ < 0.0001) linear relationship was found (Fig. 3). The slope of this response curve was steeper in women than in men (1.04 versus 0.42) indicating a higher sensitivity of the female central nervous system to the sedative effects of ethanol.

DISCUSSION

13.66 13.66 10.76 7.89

5.71 0.00 13.47 8.84 5.01 0.00 6.20 6.20

9.13 6.10 ? 3.98 t 5.34

6.28 2.72 4.61 4.61 6.88 0.00 9.56 9.56

13.28 13.28 9.69 5.42

8.38 5.93 2 3.10 2 4.79

Because there is a considerable controversy concerning the extent, site, and gender-related differences of the first-pass metabolism of etha- no1,2,3 we have tried to approach this problem by a new study design. So far all studies have compared AUC values (calculated by the trapezoidal rule) from separate experiments (after intravenous and oral administrations on different days). Because it has been stated that intrasubject variation of ethanol metabolism is greater than inter- subject variation” (e.g., within-subject variation of blood ethanol disappearance rate and apparent volume of distribution (V) was 58% and 27%, respectively), we tried to eliminate this confounding factor by applying the so-called stable isotope technique with concomitant intravenous and oral (or intraduodenal) administrations. Because no isotope effect was observed (see superimposable curves of d,- and d,-ethanol in Figs. 1 and 2) it can be concluded that the body handled both agents in an identical manner.

Concerning the dose-dependent (nonlinear) disposition of ethanol, different partly overlap- ping phases of the time profiles have been despribed’>h,“‘ml” and could be also observed in our female and male individuals (Fig. 2):

CLINICAL PHAIWL~COLOGY & THERWEUTICS MAY 1996 5 10 Awunon et al.

IO 20 30

EtOHtO, [mmol/l]

Fig. 3. Relationship between all individual total blood alcohol concentrations (EtOH,,, = sum of d,- and d,-ethanol) and sedation indexes formed from four visual analog scales (corrected for prestudy values) for all subjects of both trials. Solid line (y = 0.42x + 5.4; r = 0.34; p < 0.0001) represents male (squares) and broken line (y = 1.04x + 3.5; r = 0.65; p < 0.0001) female (triangles) subjects.

l An initial steeper decline in most of the intravenous and intraduodenal blood ethanol curves that represents mainly rapid distribution from the blood (central compartment) to peripheral tissues (peripheral compartment)

l A pseudolinear phase between about 2 and 5 hours that persisted until ethanol concentrations decreased to approximately 2.8 mmol/L (13 mg/dl)

l A curvilinear phase representing classic linear disposition below the above ethanol concentrations

Such “threshold” levels (9 to 20 mgidl) have been already noted by several other groups.10,12,13 Because of the complex pharmacokinetics of ethanol, different models for evaluation of blood level time profiles have been applied. However, there is good evidence that an open two-compartment model with Michaelis-Menten elimination from the central compartment describes best the disposition of ethanol.‘0,13 In our subjects optimal fits of curves and the most reliable data were generated by use of this model.

The so-called back-extrapolation method is very often applied, especially in forensic medicine. This method is based on the blood ethanol disappearance rate (the term k,, is often used) calculated from the pseudolinear decline (as seen here between 2 and 5 hours). In addition, by extrapolating the slope to

zero time, the concentration extrapolated to time zero (Co) is derived and used to calculate V (D/Co).3,14 Because the slope varies widely (e.g., from 9 to 36 mg/dl * hr)15 and depends on the selec- tion of points used for the linear regression analysis, this method appears questionable. In our study, total blood ethanol disappearance rate (sum of do and d3) was higher in female than in male subjects, a phenomenon described earlier.16-18 The observed difference is probably not primarily caused by genetic factors because those differences were also seen in male and female siblings.” Probably the higher free and total testosterone levels observed in men can account for the slower blood ethanol disappearance rate, inasmuch as this hormone has an inhibitory effect on alcohol dehydrogenase.” Our values (Table I) are in good agreement with Wid- mark’s classic @slope” (10 to 25 mg/dl * hr; mean, 16 mg/dl [3.5 mmol/L * hr]) or the means given by Jones et al. in normal subjects after oral administration (14 mg/dl * hr)” or intravenous infusion (17 mg/dl * hr).”

With the extrapolation method, V values between 0.43 and 0.59 L/kg have been reported.‘,” As pointed out by Rangno et al.,” this V is calculated by use of an incorrect Co value and consequently will overestimate V by about 17%. In addition, the

CLISICAL PHARMACOLOGY & THERAPEUTICS VOLUME 59, NUMBER 5 Amman et al. 5 11

method is valid only after intravenous administration of drugs whose disposition can be described by the one-compartment model.*’ Therefore our slightly lower V value of 0.4 L/kg (calculated model- independently) appears more realistic. Independent of the calculation of V, it has been shown earlier that the V of ethanol is smaller in women than in men because of the lower water content in the body of women than men. ” We could confirm this find- ing, which is an important factor contributing to higher AUC values in women than men, if identical doses (based on total body weight) of ethanol are administered (see Table I and Leung*l). Such gender-related differences in AUC could also be explained by differences in the rate or capacity to metabolize ethanol. Although we could not find a difference between sexes in the highly variable apparent K, values, V,, was lower in women than in men, which implies that females exhibit a lower capacity to metabolize ethanol. This could be attrib- utable to the well-known fact that liver size is smaller in female subjects.22

It is generally accepted that elimination of ethanol occurs primarily by metabolism, and it is assumed that the renal route can be neg1ected.i This could be substantiated by the low recovery (2% to 4% of administered dose) of unchanged ethanol in urine in this study. In addition, renal clearance averaged only 6.7 and 8.5 ml/min in men and women, respectively.

If ethanol is administered orally its bioavailability depends on several factors, such as dose, rate of input, food intake, liver function, and volume and ethanol concentration of ingested solution.1,‘5,1”,23-26 Changes in bioavailability cannot be explained by differences in absorption that is rapid and complete but are due to alteration in first-pass metabolism.’ This first-pass metabolism is more pronounced with low doses, with subjects in the fed state, with long-term ethanol consumption, and with intact gastric mucosa.27,28 In addition, it has been claimed that first-pass metabolism is higher in women than in men.4 Our study in healthy volunteers was designed to simulate realistic life conditions, for example, ethanol was ingested in moderate doses (“social drinking”) after lunch. Both factors fa- vor first-pass metabolism. Nevertheless, first-pass metabolism was of minor extent and almost negligible (Table I). The observed range (4.6% to 13.7%, no gender-related differences) fits very well to a recent study, in which a mean value of 4.2% was obtained for a white European population.2” Bioavailability has also approached unity in many other studies.1,10Y12,24

By comparing the results of two experiments (intravenous/oral versus intravenousiintraduodenal), we tried to separate total first-pass metabolism into gastric and hepatic components (Table I). It could be shown that a major contribution to the small total first-pass metabolism was provided by the gastric site. This observation is qualitatively similar to earlier find- ings of Caballeria et a1.,28 who showed significantly higher blood alcohol concentrations and negligible first-pass metabolism when alcohol was infused intraduodenally in healthy men than when it was taken orally by the same subjects. Moreover, these authors demonstrated that patients with subtotal gastrectomies exhibited oral blood alcohol concentrations approxi- mating those obtained after intravenous infusion, with no evidence of a significant first-pass metabolism.28

Because a mean first-pass metabolism of less than 10% was observed in our individuals, it was not surprising that no difference in the first-pass metabolism was obvious between the two genders. In similar studies in men and women, oral bioavailability has been independent of gender and was close to lOO%.*i In contrast, Frezza et a1.4 observed a lower first-pass metabolism in women than in men that was attributed to lower gastric alcohol dehydrogenase activity in women than in men. The obvious divergent results on gender-related differences in alcohol metabolism obtained by various laboratory groups might be partly caused by variabilities in genotypes of the tested subjects or might be caused by different study designs and dosing regimens.

Because disposition of ethanol is nonlinear (dose and concentration dependent), calculations of first- pass metabolism on the basis of comparison of AUC values can be a problem because AUC and bioavailability are dependent on the input rate of ethanol (e.g., rate of infusion or ingestion).30,3i In such cases the so-called integrated form of the Michaelis- Menten equation should be preferred.23T31’32 How- ever, use of this equation requires Co (extrapolated with great error), K, (highly variable), and V,,, from intravenous experiments (often population means from the literature are taken) and thus is hardly applicable. In addition, as pointed out by Metzler and Tong, “. . . it is not possible to estimate V max and K, of a Michaelis-Menten type pharmacokinetic model with any precision from a single- dose experiment.“33 One pragmatic solution to this mathematical problem is to adjust the input rates through the different administration routes in such a way that similar rates of ethanol delivery to the liver are accomplished, resulting in concen-

512 Amman et al. CLINICAL PHAKhlACOLOGY & ‘I’HEKiWEUTI(‘S

hL4Y 19%

trations and time profiles that are very similar after intravenous and oral (or intraduodenal) administration; thus, the AUC method should re- main valid.30 Because we have achieved this by the selected duration of infusions (60 minutes for intravenous and 30 minutes for intraduodenal) and ingestion (over 20 minutes) of ethanol, the obtained results are valid at least for the administered doses given after a meal at lunch time to healthy female and male volunteers. In contrast to all other studies performed thus far, our results are not confounded by random artifacts or variability caused by the large day-to-day variability of the high clearance compound ethanol.

In our study, we also tried to measure, at least qualitatively, the pharmacodynamic response to ethanol in terms of sedation as assessed by choice reaction time and sedation index. There was some interindividual variability in the temporal relations between blood ethanol concentrations and sedative effects, confirming earlier observations.34 However, the well-known ethanol-impaired vigilance35 could be related to ethanol blood levels. In both women and men we found significant (p < 0.0001) linear correlations between the sedation indexes and ethanol blood levels. Because the slope was much steeper in female (1.04) than in male (0.42) subjects it could be concluded that the central nervous system of women is more sensitive to the sedative effects of ethanol.

In conclusion, for moderate oral doses of ethanol (social drinking) taken after a meal, only a minor (4.6% to 13.7%) first-pass metabolism exists. This first-pass metabolism is partly of gastric origin but not dependent on gender. The higher AUC values seen in women are caused mainly by the smaller V and may be partly due to lower V,,, values, indicating that females may have a lower metabolic capacity for ethanol.

We are indebted to Mrs. A. Henker and H. Kohler for editorial assistance. We thank Mrs. E. Schneider, RN, and the staff of the endoscopy unit for assistance in conducting the study. We are grateful to Drs. M. Eichelbaum, G. Engel, and K. KivistG for helpful discussions.

References 1. Holford NHG. Clinical pharmacokinetics of ethanol.

Clin Pharmacokinet 1987;13:273-92. 2. Gentry RT, Baraona E, Lieber CS. Agonist: gastric

first pass metabolism of alcohol. J Lab Clin Med 1994;123:21-6.

3. Levitt MD. Antagonist: the case against first-pass metabolism of ethanol in the stomach. J Lab Clin Med 1994;123:28-31.

4. Frezza M, Di Padova C, Pozzato G, Terpin M, Bara- ona E, Lieber CS. High blood ethanol levels in women: the role of decreased gastric alcohol dehydrogenase activity and first pass metabolism. N Engl J Med 1990;322:95-9.

5. Kopun M, Propping P. The kinetics of ethanol absorption and elimination in twins and supplementary repetitive experiments in singleton subjects. Eur J Clin Pharmacol 1977;11:337-44.

6. Eichelbaum M. The application of drugs labelled with stable isotopes in pharmacokinetics and clinical pharmacokinetics. In: Breimer DD, Speiser P, eds. Topics in pharmaceutical sciences. Amsterdam: Elsevier Sci- ence, 1985:237-48.

7. Jones AW, Jonsson fi. Between-subject and within- subject variations in the pharmacokinetics of ethanol. Br J Clin Pharmacol 1994;37:427-31.

8. Heinzel G, Woloszczak R, Thomann P. TopFit ver- sion 2.0: pharmacokinetic and pharmacodynamic data analysis system for the PC. Stuttgart: G Fischer, 1993.

9. Wagner JG. Intrasubject variation in elimination half- lives of drugs which are appreciably metabolized. J Pharmacokinet Biopharm 1973;1:165-73.

10. Rangno RE, Kreeft JH, Sitar DS. Ethanol dose- dependent elimination: Michaelis-Menten vs classical kinetic analysis. Br J Clin Pharmacol 1981;12:667-73.

11. Jones AW, Hahn RG, Stalberg HP. Pharmacokinetics of ethanol in plasma and whole blood: estimation of total body water by the dilution principle. Eur J Clin Pharmacol 1992;42:445-8.

12. Wilkinson PIS, Sedman AJ, Sakmar E, Kay DR, Wag- ner JG. Pharmacokinetics of ethanol after oral administration in the fasting state. J Pharmacokinet Bio- pharm 1977;5:207-24.

13. Wilkinson PK, Sedman AJ, Sakmar E, Earhart RH, Weidler DJ, Wagner JG. Blood ethanol concentrations during and following constant-rate intravenous infusion of alcohol. Clin Pharmacol Ther 1976;19:213- 23.

14. Al-Lanqawi Y, Moreland TA, McEwen J, Halliday F, Durnin CJ, Stevenson IH. Ethanol kinetics: extent of error in back extrapolation procedures. Br J Clin Pharmacol 1992;34:316-21.

15. Jones AW. Disappearance rate of ethanol from the blood of human subjects: implications in forensic tox- icology. J Forensic Sci 1993;38:104-18.

16. Smith GD, Shaw LI, Maini PK, Ward RJ, Peters TJ, Murray JD. Mathematical modelling of ethanol metabolism in normal subjects and chronic alcohol mis- users. Alcohol Alcohol 1993;28:25-32.

17. Cole-Harding S, Wilson JR. Ethanol metabolism in men and women. J Stud Alcohol 1987;48:380-7.

18. Mishra L, Sharma S, Potter JJ, Mezey E. More rapid

CLINICAL PHARiMACOLOGY & THERAPEUTICS VOLUrMF 5Y, NUMBER 5 Awwnon et al. 513

elimination of alcohol in women as compared to their male siblings. Alcohol Clin Exp Res 1989;13:752-4.

19. Mezey E, Oesterling JE, Potter JJ. Influence of male hormones on rates of ethanol elimination in man. Hepatology 1988;8:742-4.

20. Klotz U. Pathophysiological and disease-induced changes in drug distribution volume: pharmacokinetic implications. Clin Pharmacokinet 1976;1:204-18.

21. Marshall AW, Kingstone D, Boss M, Morgan MV. Ethanol elimination in males and females: relationship to menstrual cycle and body composition. Hepa- tology 1983;3:701-6.

22. Leung NWY, Farrant P, Peters TJ. Liver volume measurements by ultrasonography in normal subjects and alcoholic patients. J Hepatol 1986;2:157-86.

23. Roine RP, Gentry RT, Lim RT, Baraona E, Lieber CS. Effect of concentration of ingested ethanol on blood alcohol levels. Alcohol Clin Exp Res 1991;15: 734-8.

24. Wagner JG. Lack of first-pass metabolism of ethanol at blood concentrations in the social drinking range. Life Sci 1986;39:407-14.

25. Wedel M, Pieters JE, Pikaar NA, Ockhuizen T. Ap- plication of a three-compartment model to a study of the effects of sex, alcohol dose and concentration, exercise and food consumption on the pharmacokinetics of ethanol in healthy volunteers. Alcohol Alco- hol 1991;26:329-36.

26. Whitfield JB, Martin NG. Alcohol consumption and alcohol pharmacokinetics: interactions within the normal population. Alcohol Clin Exp Res 1994;18:238-43.

27. DiPadova C, Worner TM, Julkunen RJK, Lieber CS. Effects of fasting and chronic alcohol consumption on

the first-pass metabolism of ethanol. Gastroenterol- ogy 1987;92:1169-73.

28. Caballeria J, Frezza M, Hernandez-Mufioz R, et al. Gastric origin of the first-pass metabolism of ethanol in humans: effect of gastrectomy. Gastroenterology 1989;97:1205-9.

29. Derr RF. Simulation studies on ethanol metabolism in different human populations with a physiological pharmacokinetic model. J Pharm Sci 1993;82:677- 82.

30. Smith T, De Master EG, Furne JK, Springfield J, Levitt MD. First-pass gastric mucosal metabolism of ethanol is negligible in the rat. J Clin Invest 1992;89: 1801-6.

31. Levitt MD, Levitt DG. The critical role of the rate of ethanol absorption in the interpretation of studies purporting to demonstrate gastric metabolism of ethanol. J Pharmacol Exp Ther 1994;269:297- 304.

32. Wagner JG, Wilkinson PK, Ganes DA. Parameter V, and K, for elimination of alcohol in young male subjects following low doses of alcohol. Alcohol Al- cohol 1989;24:555-64.

33. Metzler CM, Tong DDM. Computational problems of compartmental models with Michaelis-Menten-type elimination. J Pharm Sci 1981;70:733-7.

34. Radow R, Hurst PM. Temporal relations between blood alcohol concentration and alcohol effect: an experiment with human subjects. Psychopharmacol- ogy (Berl) 1985;85:260-6.

35. Koelega HS. Alcohol and vigilance performance: a review. Psychopharmacology (Berl) 1995;118:233- 49.

disposition and first-pass metabolism of ethanol in humans: is it gastric or hepatic and does it...

Documents