Metabolism of flurazepam by the small intestine The metabolism offiurazepam-5- I"C has been studied in man following catheterization of the portal and hepatic veins. Flurazepam was administered through a tube into the stomach in one patient and into the duodenum in two patients. Thin-layer chromatographs of portal vein blood showed that there was a rapid and early appearance of metabolites offiurazepam consistent with the metabolism of the flurazepam by the intestinal mucosa and at times when the concentrations in the hepatic vein and peripheral blood were very much lower than those in the portal vein. The major metabolites identified in portal vein blood were the mono- and didesethyl metabolites offiurazepam. Considerable hepatic uptake offlurazepam and its metabolites occurred, as evidenced by the lower concentrations of the parent compound and metabolites in the hepatic vein. Thus, "first-pass" metabolism offiurazepam following oral administration occurs in the small bowel mucosa of man as well as in the liver.
W. A. Mahon, M.D., T. Inaba, Ph.D., and R. M. Stone, M.D. Toronto, Ontario, Canada Division of Clinical Pharmacology, Departments of Medicine and Surgery, Toronto General Hospital, and Department of Pharmacology, University of Toronto
The metabolism of flurazepam after oral administration in man has been reported . 7 The investigators have shown that following oral administration little of the parent compound can be detected in the systemic circulation, whereas several metabolites can be identified. During repetitive dosing" the parent compound did not achieve concentrations above the sensitivity limit of the assay (3 ng/ml), whereas the hySupported in pari by a grant from Hoffmann-La Roche, Ltd. (Canada). Presented in part at the Fall Meeting of the American Society for Pharmacology and Experimental Therapeutics, New Orleans, Aug. 16 to 19, 1976. Received for publication Dec. 2, 1976. Accepted for publication March 11, 1977. Reprint requests to: Dr. W. A. Mahon, Room 3-200, College Wing, Toronto General Hospital, Toronto, Ontario, Canada M5G lL7.
228
droxyethyl metabolite was measurable shortly after a 30-mg oral dose but was not detectable after 24 hr. N-desalkyl flurazepam was a major circulating metabolite which reached steady state after 7 days. Animal studies have reported that the N-desalkyl and the N-hydroxyethyl metabolites were more active than flurazepam." Kaies and associates" have shown that the hypnotic effect of flurazepam is greater on the second and third nights of repeated administration than on the first night and have suggested that it was due to accumulated active metabolites rather than the parent compound. Material and methods
The subjects were three patients whose characteristics and investigations are shown in Table I. All were admitted for hemodynamic
Volume 22 Number 2
229
Flurazepam metabolism in small bowel
10000
INTRAGASTRIC
INTRAOUODENAL
1000
L 0::--
.../~......
----.....
/
I I
" ~. .. /...........
~
r!
dpm/ml
INTRADUODENAL
I
fr ,.,"
i
V i
100
o
50
100
150 200
0
15
PORTAL VEIN PERIPHERAL HEPATIC VEIN
30
45
60
0
15
30
45
60
MINUTES
Fig. 1. Total radioactivity (dpm/rnl) measured over time in three experiments. Sampies were taken from the portal vein, a peripheraI artery, and a hepatic vein as shown. In two patients (No. I, left, and No. 3, right) , fiurazepam-ö-t-C was administered into the duodenum in a dose of 39 /LCi (1.4 mg) and in one patient (No. 2, center), the drug was given into the stornach in a dose of 42 pO (1.5 mg).
Table I. Patient characteristics, laboratory investigation at time of study, and hemodynamic investigations
Patient Age No. (yr) 42 2
47
3
50
Diagnosis AIcohoIic cirrhosis Macronodu1ar cirrhosis Micronodular cirrhosis
Bilirubin, total! Alk. direct SGOT phos. (mg/dl) (IU/L) (IU/L)
Albumin/ globulin (gm/dl)
ProPortal Hepatic blood thrombin vein time (sec) presflow patient/ sure (ml! (mm Hg) minlmr) control
Portal bloodflow as % of total hepatic bloodflow
1.0/0.8
53
57
3.5/2.6
10.7/11.0
31
576
48
0.6/0.2
46
129
3.1/3.7
10.8/9.3
27
648
86
0.8/0.2
43
160
3.3/3.7
11/9.5
28
805
0
investigation of the portal circulation prior to surgical portal decompression for portal hypertension. The present study was added on to the diagnostic studies. At the time of the study the liver disease of the patients was stable, and they had not been receiving drugs nor taking alcohol during the previous 3 mo. Catheterization of the portal vein was performed as follows: The umbilical vein was exposed by a supraumbilical
midline incision, and following dilatation of the vein with a probe a catheter was introduced into the portal vein. The catheter was kept open by slow continuous infusions of heparinized saline until investigations were carried out 3 to 4 days later. On the day of the experiment, the hepatic vein was catheterized with a catheter introduced from the right arm; the left brachial artery was
230
Mahon, Inaba, and Stone
r-------
Clinical Pharmacology and Therapeutics
, - - - - - - - 30
5 MINUTE - - - - - - - . . ,
::L U1+1D PORTAL VEIN
HEPATIC VEIN
PORTAL VEIN
ooou.~/ 3
400
.pm
200
200
100
200
HIDF
•
'0
3
cpm
cpm
F
2
HEPATIC
e
10
---,
VEIN
PER'PHERAL
200~ 200~
cpm
100
cpm
1+10
2 F
3
•
MINUTE - -
10
100
3 5
1+10 F 10
3
2 10
!l
cm FROM ORIGIN
Fig. 2. Thin-layer radiochromatograms of sampIes of serum from the portal and hepatic veins and peripheral artery taken 5 and 30 min after intraduodenal administration (Experiment 1). For this and subsequent figures: F, flurazepam; 1 + 1D, mono- and didesethyl flurazepam; 2, N-hydroxyethyl flurazepam; 3, N-desalkyl flurazepam.
.-------15
, - - - - - - - - I HOUR
MINUTE - - - - - - - ,
HEPATIC
PORTAL VEIN
PORTAL
VEIN
GOO
400~
cpm 200
'+10
F
400
cpm
200
bL 4
6
e
VEIN
000
400
cpm 200
F
2
HEPATIC
1+10 F 400
cpm 200
1+10
1+10
246810
VEIN
10
'0
cm FROM ORIGIN
Fig. 3. Thin-layer radiochromatograms of sampies of serum from portal and hepatic veins taken 15 min and 1 hr after the intragastric administration (Experiment 2).
also catheterized. Total hepatic blood flow was measured with the use of indocyanine green.! Portal vein blood flow was measured by means of the intraduodenal instillation of xenon-I33. 8, 9 The xenon was introduced in solution through a tube placed under fluoroscopic control. Flurazepam (specific activity, 27.7 /LCi/mg) was dissolved in 95% ethanol, sterilized by passage through a Millipore filter, and diluted in sterile water to a final concentration of 30% ethanol. The dose of 1.4 to 1.5 mg of the material contained in a volume of 2 ml was injected into the duodenum in two patients and into the stornach in one patient. The material was flushed through the tube with 20 ml water. At intervals samples of blood were drawn simultaneously from the portal vein, the hepatic vein,
and the peripheral artery. Total radioactivity in serum was determined in Aquasol (New England Nuclear Corp.). Flurazepam and its metabolites were extracted with ethyl acetate and chromatographed on a silica gel thin-Iayer chromatography (TLC) plate (E. Merck, 0.25-mm thick). The plate was developed in ethyl acetate: ethanol: concentrated ammonia (v/v, 95:5:0.5) along with reference flurazepam and its metabolites according to the method of Schwartz and Postrna. 7 The silica gel was scraped off in 0.5-cm sections and counted in Aquasol. Results
Fig. I shows the results in three experiments. On the vertical axis the disintegrations per min-
Va/urne 22 Number 2
ute per milliliter (dpm/ml) is shown on a logarithmie seale and time is shown on the horizontal axis. In the first experiment (Fig. I, left) sampling was earried out over 30 min from the hepatie vein and over 200 min from portal and peripheral sampling sites. It ean be seen that there was a sharp decline in radioaetivity measured in portal blood. The differenee in the radioactivity between the portal and the hepatie sampling sites indieates uptake of radioactivity by the liver. The second experiment was earried out over 60 min in the patient following intragastric fiurazepam (Fig. I, center). A difference in portal and hepatie coneentrations of radioactivity indieates uptake by the liver. The lowest eurve, that of the peripheral blood, shows the lowest eoneentration of radioaetivity, the difference probably being due to dilution of the radioaetive drugs in the systemic eireulation or perhaps also uptake by other tissues. The third experiment (Fig. I, right) was done after the intraduodenal instillation of fiurazepam. Sampies were taken at 1 min as weil as at 5,15, etc., min up to 60 min. Even at Imin signifieant radioaetivity was detected; a peak was reached at 15 min and declined thereafter. Fig. 2 shows thin-Iayer radiochromatograms from samples of blood taken from the first patient at 5 min and 30 min. At 5 min, 20% ofthe radioactivity in the portal vein sampie was fiurazepam, the remainder being almost equal coneentrations of mono- and didesethyl fiurazepam and N-desalkyl flurazepam, as weil as a low concentration of N-hydroxyethyl fiurazepam. In hepatic vein blood at the same time there were only very low concentrations of all the above compounds. At 30 min, 10% of the radioactivity in the portal vein sampie was flurazepam and the remainder was mainly mono- and didesethyl and N-desalkyl metabolites; in the hepatic and peripheral blood sampies, the absolute concentration of radioactivity was notably lower than that in the portal blood; the relative concentration of flurazepam in both was about 10% of the total radioactivity. Fig. 3 shows thin-layer radiochromatograms
Flurazepam metabolism in small bowel
5 MINUTE
I MINUTE
F 600
1+10
F
600
3
400
400
cpm 200
231
cpm
2
200
1+10
10
cm FROM ORIGIN
Fig. 4. Thin-layer radiochromatograms of sampIes of portal vein blood taken 1 and 5 min after the intraduodenal administration (Experiment 3).
from sampies of blood taken at 15 min and I hr from the second patient. At 15 min, low concentrations of flurazepam and the mono- and didesethyl metabolite were found. At 1 hr, 15% of the radioaetivity was flurazepam, the remainder being mono- and didesethyl metabolites and the N-desalkyl metabolite. The hepatic sampie at 1 hr contained low concentrations of fiurazepam and higher concentrations of metabolites. Fig. 4 shows thin-Iayer radiochromatograms of sampies of blood taken from the portal vein in the third patient. At 1 min, both mono- and didesethyl metabolites were identified, but the dominant substance was fiurazepam. At 5 min, the mono- and didesethyl metabolites and N-desalkyl metabolite were detected and were of concentrations comparable to that of the parent compound; N-hydroxyethyl fiurazepam was also present. Blood taken at other sampling times showed similar patterns. Early sampies of portal blood contained radioactivity that was largely flurazepam, while in later sampies the radioactivity represented the mono- and didesethyl metabolites and the N-desalkyl metabolites. The metabolites in later sampies of portal blood formed up to 85% of the radioactivity. Hepatic and peripheral blood contained radioactivity in much lower concentrations than that in the portal blood and very little fiurazepam. Discussion
These studies reveal that shortly after the introduction of flurazepam-Svv'C into the small
232
Mahon, Inaba, and Stone
bowel, flurazepam and metabolites are detectable in portal blood. When the drug was introduced into the stomach the findings were similar, but there was a delay in appearance of radioactivity in the portal vein blood which was probably due to the time required for gastric emptying. That the metabolites of flurazepam are the consequence of metabolism in small bowel mucosa is suggested by the very early appearance of the metabolites in portal blood at times when the hepatic and peripheral blood contain either no flurazepam or none of its metabolites. It is also evident that there is considerable hepatic uptake of flurazepam and metabolites, since only very low levels of radioactivity were found in the hepatic vein and peripheral blood. Our studies do not permit evaluation of the relative contribution to the metabolism of flurazepam by the small bowel and by the liver, but it is clear that metabolism occurred before the "first pass" of the drug through the liver. Our search of the literature indicates for the first time that in man the small intestine metabolizes benzodiazepine. Patients with varying degrees of liver disease who were being investigated prior to portal vein surgery were studied. It is not known if the metabolic processes in such patients differ from those in normal man. Questions which arise from such tracer studies are whether the metabolic proces ses observed in the small bowel might be saturated by therapeutic doses of the drug and whether under these circumstances perhaps more of the parent compound would reach the liver. Saturation may not have occurred with the use of the small (1.4 to 1. 5 mg) doses of flurazepam, but it can be speculated that it may occur with the usual hypnotic dose of flurazepam (15 to 30 mg). In elaborate studies Hoensch and associates" demonstrated that the various enzymes capable of drug-metabolizing activity were three to ten times as active in the rat epithelial cells of the upper small bowel villus as in the crypt cells. The activity of the enzymes decreased progressively from proximal to distal small bowel. It has been reported by Dencker and coworkers" that in man after oral administration
Clinical Pharmacology and Therapeutics
of imipramine, desmethylimipramine concentrations in portal blood were always higher than in peripheral blood. They suggested that the difference in desmethylimipramine concentrations was due to an enterohepatic circulation of desmethylimipramine. In view of the rapid appearance of desmethylimipramine and the observations in our laboratory that demethylation of imipramine can occur in the rat small intestine, * we would suggest that dealkylation of imipramine mayaIso occur in the intestinal mucosal cells of man and is the mechanism responsible for the concentration of desmethylimipramine in portal blood. A general issue of significance from these studies is the relative importance of metabolism of drugs by the small bowel and the liver. We wish to acknowledge Dr. S. 1. Weyman, former Medical Director of Hoffmann-La Roche, Ltd. (Canada), for advice and support. We express our thanks to Dr. W. E. Scott, Hoffmann-La Roche Inc., Nutley, N. J., who made available flurazepam-5 14C and the metabolites of flurazepam used in these studies, and Misses Judith Rennie and Maytak Pang and Mrs. Diane Ellis, R.N., for their excellent technical assistance. *Inaba. T.: Unpublished observations,
References 1. Bradley, S. E., Ingelfinger, F. J., and Bradley, G. P.: The estimation of hepatic blood flow in man, 1. Clin. luvest. 24:890-897, 1945. 2. Dencker, H., Dencker, S. J., Greer, A., and Nagy, A.: Intestinal absorption, demethylation, and enterohepatic circulation of imipramine, CLIN. PHARMACOL. THER. 19:584-586, 1976. 3. Hoensch, H., Woo, C. H., Raffin, S. B., and Schmid, R.: Oxidative metabolism of foreign compounds in rat small intestine: Cellular 10calization and dependence on dietary iron, Gastroenterology 70: 1063-1070, 1976. 4. Kaies, A., Bixler, E. 0., Scharf, M., and Kaies, J. D.: Sieep laboratory studies of flurazepam: A model for evaluating hypnotic drugs, CLIN. PHARMACOL. THER. 19:576-583, 1976. 5. Kaplan, S. A., deSilva, 1. A. F., Alexander, J. K., Strojny, N., Weinfelf, R. E., Puglisi, C. V., and Weissman, L.: Blood level profile in man following chronic oral administration of flurazepam hydrochloride, J. Pharm. Sei. 62:1932-1935, 1972. 6. RandalI, L. 0., and Kappell, B.: Pharmacologi-
Vo/ume 22 Number 2
cal activity of some benzodiazepines and their metabolites, in Garattini, S., Mussini, E., and Randali , L. 0., editors: The benzodiazepines, New York, 1973, Raven Press, pp. 27-51. 7. Schwartz, M. A., and Postma, E.: Metabolism of flurazepam, a benzodiazepine in man and dog, J. Pharm. Sei. 59: 1800-1805, 1970. 8. Shizgal, H. M., and Goldstein, M. S.: Measure-
Flurazepam metabolism in small bowel
233
ment of portal and total hepatic blood flow by intestinal xenon technique, Surgery 72:83-90, 1972. 9. Stone, R. M., tenHove, W., Effros, R., and Leevy, C. M.: Portal vein blood flow, its estimation and significance, Gastroentero1ogy 62: 186, 1972.
W. A. Mahon, M.D., T. Inaba, Ph.D., and R. M. Stone, M.D. Toronto, Ontario, Canada Division of Clinical Pharmacology, Departments of Medicine and Surgery, Toronto General Hospital, and Department of Pharmacology, University of Toronto
The metabolism of flurazepam after oral administration in man has been reported . 7 The investigators have shown that following oral administration little of the parent compound can be detected in the systemic circulation, whereas several metabolites can be identified. During repetitive dosing" the parent compound did not achieve concentrations above the sensitivity limit of the assay (3 ng/ml), whereas the hySupported in pari by a grant from Hoffmann-La Roche, Ltd. (Canada). Presented in part at the Fall Meeting of the American Society for Pharmacology and Experimental Therapeutics, New Orleans, Aug. 16 to 19, 1976. Received for publication Dec. 2, 1976. Accepted for publication March 11, 1977. Reprint requests to: Dr. W. A. Mahon, Room 3-200, College Wing, Toronto General Hospital, Toronto, Ontario, Canada M5G lL7.
228
droxyethyl metabolite was measurable shortly after a 30-mg oral dose but was not detectable after 24 hr. N-desalkyl flurazepam was a major circulating metabolite which reached steady state after 7 days. Animal studies have reported that the N-desalkyl and the N-hydroxyethyl metabolites were more active than flurazepam." Kaies and associates" have shown that the hypnotic effect of flurazepam is greater on the second and third nights of repeated administration than on the first night and have suggested that it was due to accumulated active metabolites rather than the parent compound. Material and methods
The subjects were three patients whose characteristics and investigations are shown in Table I. All were admitted for hemodynamic
Volume 22 Number 2
229
Flurazepam metabolism in small bowel
10000
INTRAGASTRIC
INTRAOUODENAL
1000
L 0::--
.../~......
----.....
/
I I
" ~. .. /...........
~
r!
dpm/ml
INTRADUODENAL
I
fr ,.,"
i
V i
100
o
50
100
150 200
0
15
PORTAL VEIN PERIPHERAL HEPATIC VEIN
30
45
60
0
15
30
45
60
MINUTES
Fig. 1. Total radioactivity (dpm/rnl) measured over time in three experiments. Sampies were taken from the portal vein, a peripheraI artery, and a hepatic vein as shown. In two patients (No. I, left, and No. 3, right) , fiurazepam-ö-t-C was administered into the duodenum in a dose of 39 /LCi (1.4 mg) and in one patient (No. 2, center), the drug was given into the stornach in a dose of 42 pO (1.5 mg).
Table I. Patient characteristics, laboratory investigation at time of study, and hemodynamic investigations
Patient Age No. (yr) 42 2
47
3
50
Diagnosis AIcohoIic cirrhosis Macronodu1ar cirrhosis Micronodular cirrhosis
Bilirubin, total! Alk. direct SGOT phos. (mg/dl) (IU/L) (IU/L)
Albumin/ globulin (gm/dl)
ProPortal Hepatic blood thrombin vein time (sec) presflow patient/ sure (ml! (mm Hg) minlmr) control
Portal bloodflow as % of total hepatic bloodflow
1.0/0.8
53
57
3.5/2.6
10.7/11.0
31
576
48
0.6/0.2
46
129
3.1/3.7
10.8/9.3
27
648
86
0.8/0.2
43
160
3.3/3.7
11/9.5
28
805
0
investigation of the portal circulation prior to surgical portal decompression for portal hypertension. The present study was added on to the diagnostic studies. At the time of the study the liver disease of the patients was stable, and they had not been receiving drugs nor taking alcohol during the previous 3 mo. Catheterization of the portal vein was performed as follows: The umbilical vein was exposed by a supraumbilical
midline incision, and following dilatation of the vein with a probe a catheter was introduced into the portal vein. The catheter was kept open by slow continuous infusions of heparinized saline until investigations were carried out 3 to 4 days later. On the day of the experiment, the hepatic vein was catheterized with a catheter introduced from the right arm; the left brachial artery was
230
Mahon, Inaba, and Stone
r-------
Clinical Pharmacology and Therapeutics
, - - - - - - - 30
5 MINUTE - - - - - - - . . ,
::L U1+1D PORTAL VEIN
HEPATIC VEIN
PORTAL VEIN
ooou.~/ 3
400
.pm
200
200
100
200
HIDF
•
'0
3
cpm
cpm
F
2
HEPATIC
e
10
---,
VEIN
PER'PHERAL
200~ 200~
cpm
100
cpm
1+10
2 F
3
•
MINUTE - -
10
100
3 5
1+10 F 10
3
2 10
!l
cm FROM ORIGIN
Fig. 2. Thin-layer radiochromatograms of sampIes of serum from the portal and hepatic veins and peripheral artery taken 5 and 30 min after intraduodenal administration (Experiment 1). For this and subsequent figures: F, flurazepam; 1 + 1D, mono- and didesethyl flurazepam; 2, N-hydroxyethyl flurazepam; 3, N-desalkyl flurazepam.
.-------15
, - - - - - - - - I HOUR
MINUTE - - - - - - - ,
HEPATIC
PORTAL VEIN
PORTAL
VEIN
GOO
400~
cpm 200
'+10
F
400
cpm
200
bL 4
6
e
VEIN
000
400
cpm 200
F
2
HEPATIC
1+10 F 400
cpm 200
1+10
1+10
246810
VEIN
10
'0
cm FROM ORIGIN
Fig. 3. Thin-layer radiochromatograms of sampies of serum from portal and hepatic veins taken 15 min and 1 hr after the intragastric administration (Experiment 2).
also catheterized. Total hepatic blood flow was measured with the use of indocyanine green.! Portal vein blood flow was measured by means of the intraduodenal instillation of xenon-I33. 8, 9 The xenon was introduced in solution through a tube placed under fluoroscopic control. Flurazepam (specific activity, 27.7 /LCi/mg) was dissolved in 95% ethanol, sterilized by passage through a Millipore filter, and diluted in sterile water to a final concentration of 30% ethanol. The dose of 1.4 to 1.5 mg of the material contained in a volume of 2 ml was injected into the duodenum in two patients and into the stornach in one patient. The material was flushed through the tube with 20 ml water. At intervals samples of blood were drawn simultaneously from the portal vein, the hepatic vein,
and the peripheral artery. Total radioactivity in serum was determined in Aquasol (New England Nuclear Corp.). Flurazepam and its metabolites were extracted with ethyl acetate and chromatographed on a silica gel thin-Iayer chromatography (TLC) plate (E. Merck, 0.25-mm thick). The plate was developed in ethyl acetate: ethanol: concentrated ammonia (v/v, 95:5:0.5) along with reference flurazepam and its metabolites according to the method of Schwartz and Postrna. 7 The silica gel was scraped off in 0.5-cm sections and counted in Aquasol. Results
Fig. I shows the results in three experiments. On the vertical axis the disintegrations per min-
Va/urne 22 Number 2
ute per milliliter (dpm/ml) is shown on a logarithmie seale and time is shown on the horizontal axis. In the first experiment (Fig. I, left) sampling was earried out over 30 min from the hepatie vein and over 200 min from portal and peripheral sampling sites. It ean be seen that there was a sharp decline in radioaetivity measured in portal blood. The differenee in the radioactivity between the portal and the hepatie sampling sites indieates uptake of radioactivity by the liver. The second experiment was earried out over 60 min in the patient following intragastric fiurazepam (Fig. I, center). A difference in portal and hepatie coneentrations of radioactivity indieates uptake by the liver. The lowest eurve, that of the peripheral blood, shows the lowest eoneentration of radioaetivity, the difference probably being due to dilution of the radioaetive drugs in the systemic eireulation or perhaps also uptake by other tissues. The third experiment (Fig. I, right) was done after the intraduodenal instillation of fiurazepam. Sampies were taken at 1 min as weil as at 5,15, etc., min up to 60 min. Even at Imin signifieant radioaetivity was detected; a peak was reached at 15 min and declined thereafter. Fig. 2 shows thin-Iayer radiochromatograms from samples of blood taken from the first patient at 5 min and 30 min. At 5 min, 20% ofthe radioactivity in the portal vein sampie was fiurazepam, the remainder being almost equal coneentrations of mono- and didesethyl fiurazepam and N-desalkyl flurazepam, as weil as a low concentration of N-hydroxyethyl fiurazepam. In hepatic vein blood at the same time there were only very low concentrations of all the above compounds. At 30 min, 10% of the radioactivity in the portal vein sampie was flurazepam and the remainder was mainly mono- and didesethyl and N-desalkyl metabolites; in the hepatic and peripheral blood sampies, the absolute concentration of radioactivity was notably lower than that in the portal blood; the relative concentration of flurazepam in both was about 10% of the total radioactivity. Fig. 3 shows thin-layer radiochromatograms
Flurazepam metabolism in small bowel
5 MINUTE
I MINUTE
F 600
1+10
F
600
3
400
400
cpm 200
231
cpm
2
200
1+10
10
cm FROM ORIGIN
Fig. 4. Thin-layer radiochromatograms of sampIes of portal vein blood taken 1 and 5 min after the intraduodenal administration (Experiment 3).
from sampies of blood taken at 15 min and I hr from the second patient. At 15 min, low concentrations of flurazepam and the mono- and didesethyl metabolite were found. At 1 hr, 15% of the radioaetivity was flurazepam, the remainder being mono- and didesethyl metabolites and the N-desalkyl metabolite. The hepatic sampie at 1 hr contained low concentrations of fiurazepam and higher concentrations of metabolites. Fig. 4 shows thin-Iayer radiochromatograms of sampies of blood taken from the portal vein in the third patient. At 1 min, both mono- and didesethyl metabolites were identified, but the dominant substance was fiurazepam. At 5 min, the mono- and didesethyl metabolites and N-desalkyl metabolite were detected and were of concentrations comparable to that of the parent compound; N-hydroxyethyl fiurazepam was also present. Blood taken at other sampling times showed similar patterns. Early sampies of portal blood contained radioactivity that was largely flurazepam, while in later sampies the radioactivity represented the mono- and didesethyl metabolites and the N-desalkyl metabolites. The metabolites in later sampies of portal blood formed up to 85% of the radioactivity. Hepatic and peripheral blood contained radioactivity in much lower concentrations than that in the portal blood and very little fiurazepam. Discussion
These studies reveal that shortly after the introduction of flurazepam-Svv'C into the small
232
Mahon, Inaba, and Stone
bowel, flurazepam and metabolites are detectable in portal blood. When the drug was introduced into the stomach the findings were similar, but there was a delay in appearance of radioactivity in the portal vein blood which was probably due to the time required for gastric emptying. That the metabolites of flurazepam are the consequence of metabolism in small bowel mucosa is suggested by the very early appearance of the metabolites in portal blood at times when the hepatic and peripheral blood contain either no flurazepam or none of its metabolites. It is also evident that there is considerable hepatic uptake of flurazepam and metabolites, since only very low levels of radioactivity were found in the hepatic vein and peripheral blood. Our studies do not permit evaluation of the relative contribution to the metabolism of flurazepam by the small bowel and by the liver, but it is clear that metabolism occurred before the "first pass" of the drug through the liver. Our search of the literature indicates for the first time that in man the small intestine metabolizes benzodiazepine. Patients with varying degrees of liver disease who were being investigated prior to portal vein surgery were studied. It is not known if the metabolic processes in such patients differ from those in normal man. Questions which arise from such tracer studies are whether the metabolic proces ses observed in the small bowel might be saturated by therapeutic doses of the drug and whether under these circumstances perhaps more of the parent compound would reach the liver. Saturation may not have occurred with the use of the small (1.4 to 1. 5 mg) doses of flurazepam, but it can be speculated that it may occur with the usual hypnotic dose of flurazepam (15 to 30 mg). In elaborate studies Hoensch and associates" demonstrated that the various enzymes capable of drug-metabolizing activity were three to ten times as active in the rat epithelial cells of the upper small bowel villus as in the crypt cells. The activity of the enzymes decreased progressively from proximal to distal small bowel. It has been reported by Dencker and coworkers" that in man after oral administration
Clinical Pharmacology and Therapeutics
of imipramine, desmethylimipramine concentrations in portal blood were always higher than in peripheral blood. They suggested that the difference in desmethylimipramine concentrations was due to an enterohepatic circulation of desmethylimipramine. In view of the rapid appearance of desmethylimipramine and the observations in our laboratory that demethylation of imipramine can occur in the rat small intestine, * we would suggest that dealkylation of imipramine mayaIso occur in the intestinal mucosal cells of man and is the mechanism responsible for the concentration of desmethylimipramine in portal blood. A general issue of significance from these studies is the relative importance of metabolism of drugs by the small bowel and the liver. We wish to acknowledge Dr. S. 1. Weyman, former Medical Director of Hoffmann-La Roche, Ltd. (Canada), for advice and support. We express our thanks to Dr. W. E. Scott, Hoffmann-La Roche Inc., Nutley, N. J., who made available flurazepam-5 14C and the metabolites of flurazepam used in these studies, and Misses Judith Rennie and Maytak Pang and Mrs. Diane Ellis, R.N., for their excellent technical assistance. *Inaba. T.: Unpublished observations,
References 1. Bradley, S. E., Ingelfinger, F. J., and Bradley, G. P.: The estimation of hepatic blood flow in man, 1. Clin. luvest. 24:890-897, 1945. 2. Dencker, H., Dencker, S. J., Greer, A., and Nagy, A.: Intestinal absorption, demethylation, and enterohepatic circulation of imipramine, CLIN. PHARMACOL. THER. 19:584-586, 1976. 3. Hoensch, H., Woo, C. H., Raffin, S. B., and Schmid, R.: Oxidative metabolism of foreign compounds in rat small intestine: Cellular 10calization and dependence on dietary iron, Gastroenterology 70: 1063-1070, 1976. 4. Kaies, A., Bixler, E. 0., Scharf, M., and Kaies, J. D.: Sieep laboratory studies of flurazepam: A model for evaluating hypnotic drugs, CLIN. PHARMACOL. THER. 19:576-583, 1976. 5. Kaplan, S. A., deSilva, 1. A. F., Alexander, J. K., Strojny, N., Weinfelf, R. E., Puglisi, C. V., and Weissman, L.: Blood level profile in man following chronic oral administration of flurazepam hydrochloride, J. Pharm. Sei. 62:1932-1935, 1972. 6. RandalI, L. 0., and Kappell, B.: Pharmacologi-
Vo/ume 22 Number 2
cal activity of some benzodiazepines and their metabolites, in Garattini, S., Mussini, E., and Randali , L. 0., editors: The benzodiazepines, New York, 1973, Raven Press, pp. 27-51. 7. Schwartz, M. A., and Postma, E.: Metabolism of flurazepam, a benzodiazepine in man and dog, J. Pharm. Sei. 59: 1800-1805, 1970. 8. Shizgal, H. M., and Goldstein, M. S.: Measure-
Flurazepam metabolism in small bowel
233
ment of portal and total hepatic blood flow by intestinal xenon technique, Surgery 72:83-90, 1972. 9. Stone, R. M., tenHove, W., Effros, R., and Leevy, C. M.: Portal vein blood flow, its estimation and significance, Gastroentero1ogy 62: 186, 1972.
