European Journal of

Clinical Pharmacology

Eur J Clin Pharmacol (1985) 29: 193-197

Springer-Verlag 1985

Absorption of Glibenclamide from Different Sites

of the Gastro-Intestinal Tract
D. Brockmeier 1, H.-G. Grigoleit 1, and H. Leonhardt 2
1Hoechst Aktiengesellschaft, Klinische Forschung, Frankfurt/Main and
2St~idtische Krankenanstalten, Medizinische Klinik, RoonstraBe 30, Dtiren, Federal Republic of Germany

Summary. In a study of eight volunteers and six patients, glibenclamide was placed at different sites of
the gastro-intestinal tract under visual control. The
dose was instilled once into the stomach and once into the duodenum of the eight volunteers in a randomized crossover design. The six patients underwent diagnostic colonoscopy, and the dose was
placed into the ascending colon if pathological findings were not present.
The area under the concentration-time curve,
completed by extrapolation, and the mean residence
time of the drug in the body were calculated. These
pharmacokinetic characteristics were examined using a Jonckheere test for ordered alternatives and a
Wilcoxon signed rank pair test.
The means of the areas under the curve were
477 _+131 ng. h ml- 1 for the stomach, 475 _+142 ng. h
m l - 1 for the duodenum and 486 + 301 ng. h ml- 1 for
the colon. The mean residence time changed from
2.67 __+0.35 h for the stomach to 2.42 ___0.48 h for the
duodenum and 3.55 + 0.68 h for the colon.
These results indicate that although glibenclamide is absorbed from all three sites of the gastro-intestinal tract to the same extent, the rates of absorption
are different. It is discussed whether these findings
really confirm the pH-partition hypothesis in drug
absorption. Since glibenclamide - a weak acid - has
a pK-value of about 6.5, these data seem to confirm
the pH-partition hypothesis of drug absorption.

Keywords: Glibenclamide, intestinal absorption;

small and large intestine, bioavailability

Oral administration is the most convenient and popular way of administering drugs. However, little is
known about local absorption characteristics along
the gastro-intestinal tract for most drugs. In contrast,
other sites such as the buccal and rectal mucosa have

been characterized as potent absorption sites for certain drugs which undergo extensive first-pass elimination. However, whether suppositories are exclusively absorbed from the rectum after dissolution is
still debatable, c.f. De Boer et al. (1982) and Kruis et
al. (1982).
In order to assess the absorption features at different sites in the gastro-intestinal tract it would be
necessary to retain the dose in position once administered. The effects of a prolonged or reduced gastric
emptying time induced by metoclopramide or propantheline on the rate and extent of bioavailability of
digoxin and paracetamol have been demonstrated
by Nimmo et al. (1973) and Manninen et al. (1973).
These findings seem to support the idea that certain drugs are not absorbed equally well all along the
gastro-intestinal tract, but that there are regions with
enhanced absorption facilities. One of the established ideas is that ionized drugs cannot pass the gastro-intestinal mucosa as easily as the unionized
form; another is that in normal subjects the pH
changes within the gastro-intestinal tract from acidic
to alkaline. The combination of both concepts suggests at least an absorption spectrum for the gastrointestinal tract if not an absorption window (Kiibler
1964) with 'all-or-nothing' characteristics depending
on the pK values of the drug considered (Hogben et
al. 1959).
Under normal experimental conditions neither
the gastric emptying time nor the transit time
through the intestine - and thus the pH of the environment - can be controlled or measured, nor can
the amount absorbed from the different physiological sites be estimated.
However, the drug can be placed initially at different sites and by this exclusion of certain sites the
further absorption can be studied. We have investigated the absorption of a number of drugs which
show significant differences in pH-dependent water
solubility, octanol/water partitioning and p K values
(Grigoleit and Brockmeier 1984, personal communi-

194

D. Brockmeier et al.: Gastro-Intestinal Glibenclamide Absorption

Table I. Clinical details of the subjects and patients, and the pharmacokinetic characteristics of glibenclamide
Subj.

1
2
3
4
5
6
7
8

Sex

Female
Female
Male
Male
Female
Male
Female
Male

Age
[years]
27
29
18
24
32
27
25
29

Weight
[kg]
62
51
80
60
60
68
64
82

Height
[cm]
169
160
186
172
170
175
164
189

Cmax
[ng- ml-1]

Tmax
[h]

Male
Female
Female
Female
Female
Female

33
26
44
32
44
44

74
58
58
59
78
54

180
168
162
167
168
162

MTtot~]
[h]

178
253
119
182
196
126
174
204

181
298
162
186
206
175
145
90

1.25
2.00
1.50
2.00
2.00
2.50
1.25
1.25

0.75
0.75
0.75
0.75
2.50
0.50
2.00
2.50

391
577
336
381
680
345
507
596

390
645
399
464
747
365
376
413

2.40
2.58
2.67
2.59
3.37
2.99
2.31
2.46

2.04
1.89
2.13
2.13
2.90
2.37
2.64
3.26

477
131

475
142

2.67
0.35

2.42
0.48

Mean
_+SD
9
10
11
12
13
14

AUDC
[rig. h. ml-1]

85
142
79
42
84
179

2.50
2.50
2.00
6.00
1.50
2.50

329
565
419
269
263
1052

3.69
3.24
3.75
4.26
2.35
4.03

486
301

3.55
0.68

Mean
_+SD
Subjects 1-8: Application in stomach and duodenum (S = stomach, D = duodenum)
Subjects 9-14: Application in colon (C = colon)

cation). We report here on the absorption of

glibenclamide after placing the drug in the stomach,
d u o d e n u m and ascending colon u n d e r visual control.

Materials and Methods

Subjects
Fourteen subjects (5 males and 9 females) participated in the experiments. In eight volunteers, glibenclamide was instilled into the stomach a n d into the duo d e n u m in an o p e n crossover r a n d o m i z e d design
with a wash-out period o f at least 7 days; in six patients the dose was administered into the ascending
colon. D e m o g r a p h i c data are listed in Table 1.
Before the study, each subject was informed
a b o u t the investigation and the study procedures,
and gave written consent to participate. Medical history, physical examination, and laboratory tests
(haematology, biochemical status, urinalysis, electrocardiogramm, chest X-ray) were carried out before
study.
Some m i n o r deviations f r o m the normal range of
the above mentioned variables were judged by the
investigators as being of no significance with respect
to the safety o f the subjects and to the aim of the
experiment.

The 6 patients in w h o m the drug was placed in

the ascending colon underwent diagnostic colonoscopy. The inclusion criterion for these and the eight
volunteers was that no pathological findings could
be detected at endoscopy.
For two patients (9, 14) the diagnosis was carcinophobia, three patients h a d irritable bowel s y n d r o m e
(10, 11, 13) and one (12) was examined to exclude inf l a m m a t o r y disease of the colon, which was, however, not present. Because any evidence of disease
might influence the absorption capability only patients were included in w h o m this could be ruled out
with a high probability.

Drug Administration
The subjects were not allowed to smoke nor to ingest
any alcohol or caffeine-containing beverages or
food. With the exception o f the drug used in the trial,
the subjects also abstained f r o m taking drugs, including those sold over-the-counter, for at least 24 h
before and during the study.
The volunteers reported to the clinic after fasting
overnight. For introduction of the gastroscope into
the stomach and d u o d e n u m , the m o u t h and p h a r y n x
were anaesthetized using a local anaesthetic spray
containing 10% lidocaine. The gastroscope was introduced with the subjects lying on their left side.

D. Brockmeier et al.: Gastro-Intestinal Glibenclamide Absorption

When the desired site in the upper gastro-intestinal

tract was reached a catheter was inserted via the gastroscope. A suspension was prepared freshly by placing one 1.75 mg Semi-Euglucon-N tablet, Hoechst
AG, Frankfurt/Main, FRG, in a syringe, filling the
syringe with 4 ml of a 0.9% NaCl-solution, and shaking until the tablet had disintegrated completely. The
suspension was then injected through the catheter.
Saline solution was then instilled (3 ml) to ensure
that all the drug was washed out from the syringe
and catheter, followed by 5 ml of air. The volunteers
had to remain in the recumbent position for a further
4 h. The patients were prepared for colonoscopy following the standard gastroenterological procedures.
The drug was administered through the endoscope
using the same procedure as described above. No
food was allowed for two hours. Fluid intake was not
permitted for three hours after administration of the
drug; thereafter beverages were allowed ad libitum.

Sampling and Assay

Blood samples (6 ml) were collected before and 0.25,
0.5, 0.75, 1.0, 1.25, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, and
6 h after administration. Serum 2 ml was separated
from the coagulated blood and frozen at - 20 C until assayed for glibenclamide. The concentration of
glibenclamide was determined using radioimmunoassay (Heptner et al. 1984).

195

~Glibenclamide
2OO- TCncentratin [ng/ml]

1OO

O
Time [h]~"
Fig. 1. The concentration-time profiles of glibenclamide for the
various sites of application are shown. The median concentration
curves are depicted. In 8 subjects, a suspension containing 1.75 nag
glibenclamide was instilled into the stomach (O) and duodenum
( 0 ) in a cross over design. The same preparation was instilled into
the colon () in a group of 6 patients. In each case the dose was
placed under endoscopic control

Changes in the mean absorption time due to the

placement of the dose at different sites along the gastro-intestinal tract will therefore be reflected in the
total mean time.
Mean times were estimated by a method published elsewhere (Brockmeier 1984), using the same
extrapolation which served for the estimation of the
total area under the curve (AUDC).

Statistics
PharmacokineticAnalyses
The amount absorbed was estimated by the total
area under the concentration-time curve according
to Dost's law of corresponding areas (Dost 1958,
1964). Since total clearance, which relates area to
amount, was not known for each individual, the
areas served as a relative measurement. Areas were
calculated by the linear trapezoidal rule for the entire
concentration-time data (AUD) and completed by
extrapolation. The extrapolation (AUCext0 was calculated from a single exponential adjusted to the terminal loglinear phase of the individual concentration-time profiles. The sum of AUD and AUCextr
gave the total area under the curve (AUDC).
The rate of absorption was estimated by the total
mean residence time of the drug in the body (MTtotal)
(von Hattingberg and Brockmeier 1982). After oral
administration of a drug, the total mean residence
time is the sum of the mean absorption time
(MT~us. appI) - which will depend on where the drug is
placed in the gastro-intestinal tract - and the mean
time in the body (MTvss) (von Hattingberg et al.
1980).

Considering the physico-chemical properties of glibenclamide, i. e. pH-dependent dissociation and solubility, octanol/water partitioning (Hajdu et al. 1969),
the surface of absorption, and the agitation of the
suspension along the gastro-intestinal tract, it was
expected that the rate of absorption would be lowest
in the colon, higher in the stomach, and highest in
the duodenum, and that the same might be valid with
respect to the extent of absorption. We therefore
used a Jonckheere test (Jonckheere 1954; Pahnke et
al. 1983, personal communication) for the ordered
alternatives, ignoring the fact that the values of the
eight volunteers with administration into the stomach and duodenum are correlated. However, a test
which also takes this fact into account is - to our
knowledge - not available.
In addition, the rate and extent of absorption after instilling the dose into the stomach and duodenum was compared by a one-sided Wilcoxon signed
rank pair test, using an analogical hypothesis to that
mentioned above. Due to the repeated testing, a
Bonferroni correction was applied, setting the level
of significance at p = 0.025.

D. Brockmeier et al.: Gastro-Intestinal Glibenclamide Absorption

196
Area Under the Curve
(AUDC) h.Lgh / m l ]

1000

800
P

600
l

stomach

duodenum

colon

Fig.2. The area under the data completed by extrapolation

(AUDC, ordinate Q) - as a measure of amount absorbed - does
not depend on the site of application in the gastro-intestinal (abscissa). However, larger interindividual differences are seen with
the application in the colon. Paired values are connected by a line

4-

Mean Residence Time

(MT) [hi

stomach

duodenum

colon

Fig.3. The mean residence time (MT, ordinate S ) is depicted for

the 3 different sites of application (abscissa) of glibenclamide. The
differences are significant using a Jonckheere test on ordered
alternatives. For further details refer to the text

Results

Figure I shows the median values of the concentration-time data for the eight volunteers with application into the stomach and duodenum respectively,
and for the six patients with the dose placed in the
ascending colon. The concentration-time profiles
differ clearly in shape depending on the sites of application. The individual curves correspond in principle with those depicted, i. e. they differ more in the
maximum amplitude and less in shape.
This is also reflected in the interindividual differences of the AUDC and MT values:the means of the
A U D C values (Fig. 2 and Table 1) do not differ significantly, indicating that on average the same
amount was absorbed at each site of application.
However, the individual values differ much more after application in the ascending colon than in the stom-

ach and duodenum. In contrast, the interindividual

differences between the total mean times are not as
pronounced as with the AUDC values (Fig.3 and
Table 1), since the profiles of the individual curves
are similar. The mean of the MTtotal-values is smallest
for the duodenum. It is slightly larger for the stomach but clearly higher for the colon, indicating that
absorption is fastest from duodenum and slowest
from the colon. For the majority (15 out of 22 curves)
extrapolation amounted to less than 10% of total the
area and less than 20% in the remaining cases. For
three subjects (11, 12, 14) in whom the dose was instilled into the colon, a reasonable adjustment of the
terminal slope was not possible; in these cases we
used the mean terminal rate constant in extrapolation.
However, with respect to the area under the
curve, the null hypothesis of 'no group differences'
was not rejected by the Jonckheere test, regardless of
whether the A U D (without extrapolation) or the
A U D C values were used in the test. Concerning the
total mean time (MTtotal), the alternative hypothesis
of 'descending order' of the rate of absorption from
duodenum to colon must be accepted, again regardless of whether the values with or without extrapolation were used in the test. When comparing the area
under the curve and the total mean time after instilling glibenclamide into the stomach and duodenum
respectively, the one-sided Wilcoxon signed rank
pair test did not reject the null hypothesis of no differences in the mean - whether the values with or
without extrapolation were used.
Discussion

We have studied the absorption of glibenclamide

from different sites of the gastro-intestinal tract. The
drug was placed at three selected sites under direct
visual control. The mean amount of drug absorbed
was the same, regardless of the site at which the dose
was initially placed. We cannot claim that with
placement of the dose in the stomach the drug is actually or even exclusively absorbed from the stomach. The S-shaped initial rise in the blood level
curves after instilling the drug into the stomach for
several individuals indicates that a transport mechanism may be involved with this site of application.
This may be a process of dissolution, since glibenclamide is insoluble (according to the nomenclature
of USP XX) below a pH of 7.0 (Hajdfl et al. 1969), or
of protracted transport to the site of absorption within the intestinal tract. However, neither the amount
absorbed nor the rate of absorption - as measured by
the mean residence time - is significantly different
from application in the duodenom.

D. Brockmeier et al. : Gastro-Intestinal Glibenclamide Absorption

197

In contrast to the amount, the rate of absorption

from the colon is significantly lower when compared
with the two other sites.
Considering the pK-value of the weak acid glibenclamide, these findings seem to confirm the pHpartition hypothesis. However, solubility rises with
increasing dissociation (Hajdu et al. 1969). Moreover, most theoretical considerations proceed on the
assumption that the pH in the stomach is generally
far below 7.4. This assumption is, however, not valid,
as has been demonstrated by Kuna (1964) on 1556
samples in 312 subjects and by Bowman et al. (1967)
and Meldrum et al. (1972).
We have learned from similar studies on piretanide, furosemide, and nomifensine (Meyer et al.
1982; Grigoleit and Brockmeier 1984, personal communication) that the pH-partition hypothesis is - if at
all - only of predictive value if solubility and octanol/water partitioning remain reasonably constant
or sufficiently high over the entire physiological pH
range, which is rarely the case. Otherwise these factors may be more important - as the surface of
absorption is in general - than dissociation in the
ionized form.
Blume et al. (1984) stated that glibenclamide is
poorly absorbed from the lower part of the gastro-intestinal tract. This statement is based only on the pH
partition hypothesis using an incorrect pKa value
of 5.3 instead of 6.3 to 6.8 depending on the evaluation method (Hajd6 et al. 1969). Contrary to their
statement, the same amounts are absorbed from the
colon as from the upper part of the gastro-intestinal
tract, but the rate of absorption is clearly slower. Because of the slower absorption from the colon, it cannot be ruled out that the extent of absorption may also be affected in some cases by the time the drug
spends in contact with the absorbing surface.
However, the main fraction of an orally administered dose of glibenclamide will be absorbed in the
upper part of the gastro-intestinal tract, and even
that portion which passes the small intestine unabsorbed has a clear chance of being made available to
the body via the large intestine.

Bowman WC, Rand M J, West GB (1967) Textbook of pharmacology. Blackwell, London

Brockmeier D (1984) In vitro - in vivo correlation, a time scaling
problem? III. Evaluation of mean times. Arzneimittelforsch/
Drug-Res 34 [II]: 1604-1607
De Boer AG, Moolenaar F, De Leede LGJ, Breimer DD (1982)
Rectal drug administration: Clinical pharmacokinetic considerations. Clin Pharmacokinet 7: 285- 311
Dost FH (1958) Ober ein einfaches statistisches Dosis-UmsatzGesetz. Klin Wochenschr 34:655-657
Dost FH (1964) Ein neues Verfahren zur Ermittlung des Umsatzes
yon Stoffen exogener und endogener Herkunft im Blut ohne
Kenntnis der Distribution. Antibiot Chemother 12: 149-158
Hajdfi P, Kohler KF, Schmidt FH, Springier H (1969) Physikalisch-chemische und analytische Untersuchungen an HB 419.
Arzneimittelforsch/Drug Res 19:1381-1386
Heptner W, Badian M, Baudner S, Hellstern C, Irmisch R, Rupp
W, Weimer K, Wissmann H (1984) A radioimmunoassay for determination of glibenclamide and other sulfonureas. Pharm Res
1 [5]: 215-220
Hogben CAM, Tocco DJ, Brodie BB, Schanker LS (1959) On the
mechanism of intestinal absorption of drugs. J Pharrnacol Exp
Ther 125:275-282
Jonckheere AR (1954) A distribution-free k-sample test against ordered alternatives. Biometrika 41:133-145
Kruis W, Bull U, Eisenburg J, Paumgartner G (1982) Retrograde
colonic spread of sulphasalazine enemas. Scand J Gastroenterol 17:933-938
Kfibler W (1965) Qualitative Blutspiegeluntersuchungen fiber die
Kinetik der Dtinndarmresorption. Gastroenterologia 104
[Suppl]: 231-235
Kuna S (1964) The pH of gastric juice in the normal resting stomach. Arch Int Pharmacodyn 152:79-97
Manninen V, Melin J, Apalathi A, Karesoja M (1973) Altered absorption of digoxin in patients given propanthelie and metoclopramide. Lancet 24:398-400
Meldrum SJ, Watson BW, Riddle HC (1972) pH profile of gut as
measured by radiotelemetry capsule. Br Med J 2:104-106
Meyer BH, Mfiller FO, Grigoleit H-G, Heptner W (1982) Endoscopic placement in volunteers to investigate site of absorption
of piretanide: Preliminary report. IRCS Med Sci 10:14
Nimmo J, Heading RC, Tothill P, Prescott LF (1973) Pharmacological modification of gastric emptying. Effects of propantheline and metoclopramide on paracetamol absorption. Br Med J
10:587-589
von Hattingberg HM, Brockmeier D, Voegele D (1980) A method
for in vivo - in vitro correlation using the additivity of mean
times in biopharmaceutical models. In: Rietbrock N, Woodcock BG, Neuhaus G (eds) Methods in clinical pharmacology.
Vieweg, Braunschweig, Wiesbaden, pp 85-93
von Hattingberg HM, Brockmeier D (1982) A concenpt for the assessment of bioavailability in complex systems in terms of
amounts and rates. In: Bozler G, van Rossum JM (eds) Titisee
Conference. Pharmacokinetic during drug development: Data
analysis and evaluation techniques (Oct. 1978). Gustav Fischer,
Stuttgart, pp 315-323

Acknowledgements. The authors are indepted to Dr.

W. Heptner for measuring glibenclamide, to Miss
G. Lef~vre for her skillful assistance in the evaluation
of the pharmacokinetic characteristics, and to Mr.
A.Reeves for his helpful suggestions about the
manuscript.

References
Blume H, All SL, Siewert M (1984) Zur pharmazeutischen Qualit~itvon glibenclamidhaltigen Fertigarzneimitteln. Phann Z 129
[17]: 983-989

Received: 24. 5.1985

accepted: June 10, 1985
Dr. D. Brockmeier
Hoechst AG
Klinische Forschung
Abteilung ffir Biometrie und Dokumentation
D-6230 Frankfurt/Main 80
Federal Republic of Germany