Intestinal
Abs orption of Amiodarone in Man
Albrecht Pfeiffer, MD, Monique Rongier,
Nicole TA,
Vidon, PhD, and Jean-Jacques
Martine
Bovet,
Bernier,
TA,
MD
jejunal absorption rate of amiodarone and the influence of lipids on it were studied in human volunteers using the intestinal perfusion technique. A nutrient solution (Realmentyl#{174}, Sopharga Laboratories, France) with 300 mg of the drug was infused for 120 minutes at the ligament of Treitz. The segment tested was 25 cm long. Two caloric loads
The
3.3 Kcal/min (solution A) and 1.3 Kcal/min (solution B), A caloric load 2.5 times higher than B, were administered. Minor interindividual differences in amiodarone absorption rate were observed (20.2 to 3 1.7%) with solution A. Amiodarone absorption correlated with lipid absorption significantly. Since the maximal plasma concentrations of the drug and the area under the curve (AUC/24 hours) did not correlate with the amount of amiodarone absorbed, the wide fluctuations of amiodarone pharmacokinetics must mainly be due to amiodarone tissue distribution and metabolic pathway. of
the
nutrient
containing
total
solution, lipid
and
A
miodarone, a benzofurane derivative introduced into clinical practice in 1962 for the treatment of angina pectoris,’ is now extensively used against a wide variety of serious arrhythmias including supraventricular and ventricular tachycardias.2’3 From a pharmacokinetic standpoint, this amphiphilic substance remains to be a truly challenging drug. In man, after oral administration, an interindividual range between 3 and 10 hours is being observed for maximal plasma concentrations.48 Bioavailability ranges from 33%7 to 86%.69 Long half1ife,102 kinetics of tissue distribution,10 high distribution volume10 and complex metabolic pathway13 contribute to the explanation of the wide fluctuations of amiodarone pharmacokinetics. Further factors concerning drug absorption are not yet fully elucidated and should be taken into consideration, i.e., highly different interindividual drug absorption rates from jejunum and their possible dependence on simultaneously ingested nutrients or incomplete drug absorption possibly due to low intestinal absorption rates. The aim of this study was to evaluate the jejunal absorption rate of amiodarone using the intestinal
From INSERM U.290, H#{244}pital Saint-Lazare, 75010 Paris, France. Dr. Pfeiffer is supported by the DFG (Deutsche Forschungsgemeinschaft). Address for correspondence: A. Pfeiffer, MD, c/o Hospital MUnchenBogenhausen, Englschalkinger Str. 77, 0-8000 Munchen 81, FRG.
J ClIn Pharmacol
1990;30:615-620
intubation technique.’4 To investigate the nutrients on the drug absorption, amiodarone given in a nutrient solution at two dilutions.
MATERIAL
AND
effect
of was
METHODS
Subjects Ten healthy male volunteers with a mean age of 25.1 (23-26) years, weight of 66.7 ± 5.3 kg (mean ± SD) and height of 177 ± 10 cm participated in the study. After explanation of the protocol, all subjects gave written informed consent. The protocol has been approved by the ethical committee of Saint-Lazare Hospital and was carried out in accordance with the updated Helsinki declaration. The subjects were advised not to take any medication during the 8 days previous to the study and none but amiodarone during the day of the test. All subjects were free of gastrointestinal, renal, hepatic, and cardiac disease. Prior to the study, they received a physical examination which included blood pressure and electrocardiogram.
Perfusion
General
Techniques
and
Blood
Procedure. Amiodarone-HC1 the jejunum was evaluated in ten ing’4 a nutrient solution containing The effect of lipid concentration
Samples absorption in subjects by infusthe drug. on the absorption
615
PFEIFFER
of amiodarone two dilutions
was evaluated of the nutrient
in four solution.
subjects
using
Solutions. Two dilutions (A and B) of a nutrient solution (Realmentyl#{174}, Sopharga Laboratories, France) containing the same concentration of amiodaroneHC1 (0.52 ± 0.01 mg/mL) were prepared. Realmentyl contained 18% proteins, 27% lipids, and 55% carbohydrates. The lipid fraction was mainly composed of unsaturated long-chain fatty acids with oleic and linoleic acids making about 60% of lipid calories (corn and soybean oils with added linoleic acid plus 35% lipids as egg yolk powder). Solution A contained a 2.5 times higher lipid concentration than solution B. A and B contained a caloric load of 3.3 kcal/5 mL and 1.3 kcal/5 mL, respectively. Polyethylene glycol 4000 (PEG) was added (5 g/L) as nonabsorbable marker of fluid movements. The osmolality of the solutions was adjusted to 300 mOsmol/kg by adding NaCl. A and B containing amiodarone-HC1 were warmed to 37#{176}Cunder permanent stirring. Their pH was 6.8 and 6.7, respectively. In study 1, A was infused in six subjects (A-F) while in study 2 A and B were infused in four subjects (G-K) in a random order with an interval of 1 month. Experimental Design. The experiments were carried out over a 2-day period: On day 1, a double lumen tube with a mercury bag attached to the distal tip was passed under radiological control. After an overnight fast and assessment of the infusion point just below the ligament of Treitz, the jejunal infusion was performed at a flow rate of 5 mL/min for 120 minutes. Intestinal fluid was aspirated continuously at a rate of I mL/min with a peristaltic pump 25 cm below the infusion point. After an equilibration period of 60 minutes, four consecutive 15-minute samples were obtained. The total amiodarone amount administered was 283 ± 4 mg. It has been calculated as the difference of the infused amiodarone amount (312 ± 6 mg) minus the amiodarone amount in the samples.
ET
AL
tions were expressed after hydrolysis as fatty acids. Fatty acids were extracted by the technique of Blankenhorn et al’6 and titrated by the Dole Technique.17 Biliary phospholipids were not measured by this technique. Lipase activity was estimated by Marchis-Mouren’s method.18 Bile salts were measured by a fluorimetric method;19’2#{176} amiodarone in intestinal fluid and blood samples were determined by the high performance liquid chromatography (HPLC) method described previously, known to be a highly sensitive HPLC method (approximately 5 tg/L).8
Mathematical
Results are expressed in mean ± SD. The absorption rates of amiodarone and lipids were calculated according to the rules of the perfusion technique.14 From the mean percent absorption rate of amiodarone in subjects A-F (study 1), a profile for the mean amount of amiodarone absorbed over the whole length of the intestine was constructed as a function of time. This model, previously described in detail,21 assumes that a constant fraction of the drug entering each centimeter of intestine is absorbed and that the infused solution progresses in the small bowel with a velocity of I cm /min 22 The area under the plasma concentration-time curve (AUC/24 hours) was calculated using the trapezoidal rule. All plasma concentrations and calculated AUC/24 hour values were matched for 300 mg amiodarone. For statistical analysis, the paired and unpaired Student’s t tests were used as appropriate. For correlation, a linear regression coefficient with the least squares method of analysis was applied.
RESULTS Blood Individual
Samples. Blood samples were collected into heparinized tubes before and 15, 30, 45, 60, 90, 120, 150 minutes and 3, 3.5, 4, 6, 8, and 24 hours after the start of the drug infusion. Jejunal and plasma samples were stored at -20#{176}C until analysis. Analytical
Procedure
In all intestinal samples, Hyden’s turbidimetric
616
#{149} J Clin Pharmacol
PEG 4000 method.’5
1990;30:615-620
was measured Lipid concentra-
by
Samples
values of maximal plasma concentration calculated AUG and time to reach maximal plasma concentration (tmax) obtained with solution A (3.3 kcal/min) and B (1.3 kcal/min) are demonstrated in Table I. Maximal plasma concentration varied between 32 g/L and 666 tg/L. Area under the plasma concentration time curve ranged from 23.2 mg-min-l to 624.0 mg-min’l1. Maximal plasma concentration and AUG were significantly correlated (P < .001). Time to reach maximal plasma concentration with a mean value of 4.7 hours varied between 2.0 hours and 8.0 hours. (Cmax),
Blood
Analysis
INTESTINAL
0 01 it
NC”.J c’i
_4
-l
00
01
CI
V C
ui
-
i
cj o
I
(0
I
‘
.
c’
C’)
N
01
N IL)
,-
,-
N
E
0
0
N
0
0)
concentration (370 and AUG values ± 116 mg’min.l1) during the infusion
and
B (G’-K’)
±
g/L
129
(242 ± 7 were not of solution
respectively.
Samples
Amiodarone
absorption
rates
of three solution
are
shown
in Table
subjects (H’-K’) B (1.3 kcal/min),
II.
receiving the drug
absorption rates were between 20.2 and 31.7%. For the six subjects A-F (study I), a mean amiodarone absorption rate of 26.3 ± 1.9% was observed. Using this value, the profile of the computed mean amount of amiodarone absorbed over the total length of the intestijie was calculated as a function of time (Figure 1). As demonstrated by the shaded area, there was quantity after
the
of amiodarone end of the
in the intestine drug infusion.
to
rn
c 0
o C., -
G-K)
still a high be absorbed
r-4
IL) ‘.0
-
g/L and 282 different
116
With exception amiodarone with
,.0
4-
plasma ±
Intestinal
I
o
Maximal and 341 rng.min.t1 significantly
N
00 UI
AMIODARONE
A (subjects
.,j
(0 (0
0
OF
ABSORPTION
0
0 -4
,-,
Lipid absorption rates and corresponding lipase and bile salts flow rates are presented in Table III. Highest percent absorption rates of lipids were found when solution B (1.3 kcal/min) was infused. Lipid
Cci
E
_.j
(50
m . ,_.
_
absorption was lower under (3.3 kcal/min) corresponding pase and bile salts flow rates.
OLC)
UI -
N
‘5 E E’
#{149}
,
N
‘-
IC)
CV)
.2 -
0
(5 0..
N
infusion of solution A to low intraluminal liConsidering all intes-
tinal samples of each investigated cant (P < .001) correlation was amiodarone and lipid absorption
subject, a signifiobserved between rates expressed in
N
‘5
percent
‘‘
EX ‘5
UL1J
#{176}
(0
0
ci #{149} I-
amount
Correlation tion
Between 3 demonstrates curve of subjects
drug rate
amount of 26.3%
Figure
(0
N
of the
infused
(Figure
2).
Intestinal and Blood Samples. the mean plasma concentraA-F and the mean absorbed
0)0
ci
4-.
calculated over time.
according While
there
to the absorption was a parallel
ci ci UI
o
o
#{231}
0
fl
(0
0)
N
course of plasma concentration and the amount amiodarone absorbed during the 2-hour infusion nod, no correlation could be detected between and AUC/24 (Tables I and
‘‘
o UI
ci
h II).
and
the
absorbed
drug
of peCmax
amount
OLC)
‘.0
0)
(ON
N
-l
DISCUSSION The
‘5
intestinal
intubation
technique14
was
intro-
(‘JLC)
N
N
C’)
N
I
-J
E
#{176}
3
.
Q E’-’E C
o
CARDIOVASCULAR
Abs orption of Amiodarone in Man
Albrecht Pfeiffer, MD, Monique Rongier,
Nicole TA,
Vidon, PhD, and Jean-Jacques
Martine
Bovet,
Bernier,
TA,
MD
jejunal absorption rate of amiodarone and the influence of lipids on it were studied in human volunteers using the intestinal perfusion technique. A nutrient solution (Realmentyl#{174}, Sopharga Laboratories, France) with 300 mg of the drug was infused for 120 minutes at the ligament of Treitz. The segment tested was 25 cm long. Two caloric loads
The
3.3 Kcal/min (solution A) and 1.3 Kcal/min (solution B), A caloric load 2.5 times higher than B, were administered. Minor interindividual differences in amiodarone absorption rate were observed (20.2 to 3 1.7%) with solution A. Amiodarone absorption correlated with lipid absorption significantly. Since the maximal plasma concentrations of the drug and the area under the curve (AUC/24 hours) did not correlate with the amount of amiodarone absorbed, the wide fluctuations of amiodarone pharmacokinetics must mainly be due to amiodarone tissue distribution and metabolic pathway. of
the
nutrient
containing
total
solution, lipid
and
A
miodarone, a benzofurane derivative introduced into clinical practice in 1962 for the treatment of angina pectoris,’ is now extensively used against a wide variety of serious arrhythmias including supraventricular and ventricular tachycardias.2’3 From a pharmacokinetic standpoint, this amphiphilic substance remains to be a truly challenging drug. In man, after oral administration, an interindividual range between 3 and 10 hours is being observed for maximal plasma concentrations.48 Bioavailability ranges from 33%7 to 86%.69 Long half1ife,102 kinetics of tissue distribution,10 high distribution volume10 and complex metabolic pathway13 contribute to the explanation of the wide fluctuations of amiodarone pharmacokinetics. Further factors concerning drug absorption are not yet fully elucidated and should be taken into consideration, i.e., highly different interindividual drug absorption rates from jejunum and their possible dependence on simultaneously ingested nutrients or incomplete drug absorption possibly due to low intestinal absorption rates. The aim of this study was to evaluate the jejunal absorption rate of amiodarone using the intestinal
From INSERM U.290, H#{244}pital Saint-Lazare, 75010 Paris, France. Dr. Pfeiffer is supported by the DFG (Deutsche Forschungsgemeinschaft). Address for correspondence: A. Pfeiffer, MD, c/o Hospital MUnchenBogenhausen, Englschalkinger Str. 77, 0-8000 Munchen 81, FRG.
J ClIn Pharmacol
1990;30:615-620
intubation technique.’4 To investigate the nutrients on the drug absorption, amiodarone given in a nutrient solution at two dilutions.
MATERIAL
AND
effect
of was
METHODS
Subjects Ten healthy male volunteers with a mean age of 25.1 (23-26) years, weight of 66.7 ± 5.3 kg (mean ± SD) and height of 177 ± 10 cm participated in the study. After explanation of the protocol, all subjects gave written informed consent. The protocol has been approved by the ethical committee of Saint-Lazare Hospital and was carried out in accordance with the updated Helsinki declaration. The subjects were advised not to take any medication during the 8 days previous to the study and none but amiodarone during the day of the test. All subjects were free of gastrointestinal, renal, hepatic, and cardiac disease. Prior to the study, they received a physical examination which included blood pressure and electrocardiogram.
Perfusion
General
Techniques
and
Blood
Procedure. Amiodarone-HC1 the jejunum was evaluated in ten ing’4 a nutrient solution containing The effect of lipid concentration
Samples absorption in subjects by infusthe drug. on the absorption
615
PFEIFFER
of amiodarone two dilutions
was evaluated of the nutrient
in four solution.
subjects
using
Solutions. Two dilutions (A and B) of a nutrient solution (Realmentyl#{174}, Sopharga Laboratories, France) containing the same concentration of amiodaroneHC1 (0.52 ± 0.01 mg/mL) were prepared. Realmentyl contained 18% proteins, 27% lipids, and 55% carbohydrates. The lipid fraction was mainly composed of unsaturated long-chain fatty acids with oleic and linoleic acids making about 60% of lipid calories (corn and soybean oils with added linoleic acid plus 35% lipids as egg yolk powder). Solution A contained a 2.5 times higher lipid concentration than solution B. A and B contained a caloric load of 3.3 kcal/5 mL and 1.3 kcal/5 mL, respectively. Polyethylene glycol 4000 (PEG) was added (5 g/L) as nonabsorbable marker of fluid movements. The osmolality of the solutions was adjusted to 300 mOsmol/kg by adding NaCl. A and B containing amiodarone-HC1 were warmed to 37#{176}Cunder permanent stirring. Their pH was 6.8 and 6.7, respectively. In study 1, A was infused in six subjects (A-F) while in study 2 A and B were infused in four subjects (G-K) in a random order with an interval of 1 month. Experimental Design. The experiments were carried out over a 2-day period: On day 1, a double lumen tube with a mercury bag attached to the distal tip was passed under radiological control. After an overnight fast and assessment of the infusion point just below the ligament of Treitz, the jejunal infusion was performed at a flow rate of 5 mL/min for 120 minutes. Intestinal fluid was aspirated continuously at a rate of I mL/min with a peristaltic pump 25 cm below the infusion point. After an equilibration period of 60 minutes, four consecutive 15-minute samples were obtained. The total amiodarone amount administered was 283 ± 4 mg. It has been calculated as the difference of the infused amiodarone amount (312 ± 6 mg) minus the amiodarone amount in the samples.
ET
AL
tions were expressed after hydrolysis as fatty acids. Fatty acids were extracted by the technique of Blankenhorn et al’6 and titrated by the Dole Technique.17 Biliary phospholipids were not measured by this technique. Lipase activity was estimated by Marchis-Mouren’s method.18 Bile salts were measured by a fluorimetric method;19’2#{176} amiodarone in intestinal fluid and blood samples were determined by the high performance liquid chromatography (HPLC) method described previously, known to be a highly sensitive HPLC method (approximately 5 tg/L).8
Mathematical
Results are expressed in mean ± SD. The absorption rates of amiodarone and lipids were calculated according to the rules of the perfusion technique.14 From the mean percent absorption rate of amiodarone in subjects A-F (study 1), a profile for the mean amount of amiodarone absorbed over the whole length of the intestine was constructed as a function of time. This model, previously described in detail,21 assumes that a constant fraction of the drug entering each centimeter of intestine is absorbed and that the infused solution progresses in the small bowel with a velocity of I cm /min 22 The area under the plasma concentration-time curve (AUC/24 hours) was calculated using the trapezoidal rule. All plasma concentrations and calculated AUC/24 hour values were matched for 300 mg amiodarone. For statistical analysis, the paired and unpaired Student’s t tests were used as appropriate. For correlation, a linear regression coefficient with the least squares method of analysis was applied.
RESULTS Blood Individual
Samples. Blood samples were collected into heparinized tubes before and 15, 30, 45, 60, 90, 120, 150 minutes and 3, 3.5, 4, 6, 8, and 24 hours after the start of the drug infusion. Jejunal and plasma samples were stored at -20#{176}C until analysis. Analytical
Procedure
In all intestinal samples, Hyden’s turbidimetric
616
#{149} J Clin Pharmacol
PEG 4000 method.’5
1990;30:615-620
was measured Lipid concentra-
by
Samples
values of maximal plasma concentration calculated AUG and time to reach maximal plasma concentration (tmax) obtained with solution A (3.3 kcal/min) and B (1.3 kcal/min) are demonstrated in Table I. Maximal plasma concentration varied between 32 g/L and 666 tg/L. Area under the plasma concentration time curve ranged from 23.2 mg-min-l to 624.0 mg-min’l1. Maximal plasma concentration and AUG were significantly correlated (P < .001). Time to reach maximal plasma concentration with a mean value of 4.7 hours varied between 2.0 hours and 8.0 hours. (Cmax),
Blood
Analysis
INTESTINAL
0 01 it
NC”.J c’i
_4
-l
00
01
CI
V C
ui
-
i
cj o
I
(0
I
‘
.
c’
C’)
N
01
N IL)
,-
,-
N
E
0
0
N
0
0)
concentration (370 and AUG values ± 116 mg’min.l1) during the infusion
and
B (G’-K’)
±
g/L
129
(242 ± 7 were not of solution
respectively.
Samples
Amiodarone
absorption
rates
of three solution
are
shown
in Table
subjects (H’-K’) B (1.3 kcal/min),
II.
receiving the drug
absorption rates were between 20.2 and 31.7%. For the six subjects A-F (study I), a mean amiodarone absorption rate of 26.3 ± 1.9% was observed. Using this value, the profile of the computed mean amount of amiodarone absorbed over the total length of the intestijie was calculated as a function of time (Figure 1). As demonstrated by the shaded area, there was quantity after
the
of amiodarone end of the
in the intestine drug infusion.
to
rn
c 0
o C., -
G-K)
still a high be absorbed
r-4
IL) ‘.0
-
g/L and 282 different
116
With exception amiodarone with
,.0
4-
plasma ±
Intestinal
I
o
Maximal and 341 rng.min.t1 significantly
N
00 UI
AMIODARONE
A (subjects
.,j
(0 (0
0
OF
ABSORPTION
0
0 -4
,-,
Lipid absorption rates and corresponding lipase and bile salts flow rates are presented in Table III. Highest percent absorption rates of lipids were found when solution B (1.3 kcal/min) was infused. Lipid
Cci
E
_.j
(50
m . ,_.
_
absorption was lower under (3.3 kcal/min) corresponding pase and bile salts flow rates.
OLC)
UI -
N
‘5 E E’
#{149}
,
N
‘-
IC)
CV)
.2 -
0
(5 0..
N
infusion of solution A to low intraluminal liConsidering all intes-
tinal samples of each investigated cant (P < .001) correlation was amiodarone and lipid absorption
subject, a signifiobserved between rates expressed in
N
‘5
percent
‘‘
EX ‘5
UL1J
#{176}
(0
0
ci #{149} I-
amount
Correlation tion
Between 3 demonstrates curve of subjects
drug rate
amount of 26.3%
Figure
(0
N
of the
infused
(Figure
2).
Intestinal and Blood Samples. the mean plasma concentraA-F and the mean absorbed
0)0
ci
4-.
calculated over time.
according While
there
to the absorption was a parallel
ci ci UI
o
o
#{231}
0
fl
(0
0)
N
course of plasma concentration and the amount amiodarone absorbed during the 2-hour infusion nod, no correlation could be detected between and AUC/24 (Tables I and
‘‘
o UI
ci
h II).
and
the
absorbed
drug
of peCmax
amount
OLC)
‘.0
0)
(ON
N
-l
DISCUSSION The
‘5
intestinal
intubation
technique14
was
intro-
(‘JLC)
N
N
C’)
N
I
-J
E
#{176}
3
.
Q E’-’E C
o
CARDIOVASCULAR
