Journal of Hepatology, 1992; 16:215-218
215
©1992 ElsevierScientificPublishers Ireland Ltd. All rights reserved. 0168-8278/92/$05.00 HEPAT01214 Rapid Publication
Blockade of ATP-sensitive K + channels by glibenclamide reduces portal pressure and hyperkinetic circulation in portal hypertensive rats Richard Moreau, Hirokazu Komeichi, St6phane Cailmail and Didier Lebrec Laboratoire d'Hbmodynamique Splam'hnique. Unit~de Recherches de Physiopathologie Hbpatique. I NSERM U-24. H6pital Beaujon, Clichy. France
Certain results of in vitro studies raise the possibility that blockade of ATP-sensitive K + channels by glibenclamide may induce vasoconstriction. Therefore, this substance might decrease portal pressure and hyperkinetic circulation in animals with portal hypertension. Thus, systemic and regional hemodynamics (radioactive microspheres) were measured before and 20 min after a bolus intravenous injection of glibenclamide (20 mg/kg) in conscious rats with portal vein stenosis. Blood pressure decreased significantly from 14.5_+1.5 to 12.2+_1.2 (mean _+SE). Cardiac index significantly decreased by 24%, portal tributary blood flow by 31%, and hepatic artery blood flow by 35%. Systemic vascular resistance significantly increased by 38%, portal territory vascular resistance and hepatic artery vascular resistance by 61%, each, and renal vascular resistance by 17%. Arterial pressure, heart rate, and renal blood flow were unchanged. Moreover, glibenclamide blunted the vasodilating action of diazoxide (an ATP-sensitive K + channel opener). These results show that in rats with extrahepatic portal hypertension the blockade of ATP-sensitive K + channels by glibenclamide reduces portal pressure and hyperkinetic circulation. Key words: Sulfonylurea; Systemic and splanchnic hemodynamics; Extrahepatic portal hypertension
In vitro studies have shown that arterial smooth muscle cells contain ATP-sensitive K + channels whose blockade by the sulfonylurea glibenclamide leads to vasoconstriction (1-4). As a result, glibenclamide might reduce portal pressure and the hyperkinetic circulation in portal hypertension. In these portal-hypertensive animals, however, systemic and regional hemodynamic responses to glibenclamide have not yet been investigated. Thus, the aim of the present study was to determine the effects of glibenclamide administration on systemic and regional hemodynamics in conscious rats with extrahepatic portal hypertension.
studied. All rats were allowed free access to food and water until 14-18 h before the study, when food was withdrawn. Portal hypertension was induced by ligation of the portal vein (5). Under ether anesthesia, the abdomen was incised and the portal vein exposed. A polyethylene catheter of 0.96 mm external diameter was passed along the portal vein, and 3-0 silk was used to ligate both the catheter and the portal vein. All rats were studied "at least 3 weeks after the operation. Protocols performed in this laboratory are approved by the French Agricultural Office in conformity with European legislation for research involving animals.
Protocols
Materials and Methods Animals
Twelve adult male Sprague-Dawley rats (Charles River Laboratoires, Saint-Aubin-L6s-Elbeuf, France) were
Six rats received glibenclamide alone (20 mg/kg i.v. bolus) and plasma glucose concentrations were measured before and after glibenclamide. Four additional experiments were performed. In the first set of experiments, two rats received a single dose of the ATP-sensitive K +
Correspondence to: Dr. R. Moreau, INSERM U-24, H6pital Beaujon,92118 Clichy,France.
216 channel opener, diazoxide (30 mg/kg i.v. bolus) (3,4). In the second set of experiments, two rats received glibenclamide and diazoxide (30 mg/kg i.v.) 15 min later. In the third set of experiments, one rat received the vehicle of glibenclamide only. In the fourth set of experiments, one rat received the vehicle of diazoxide only. H e m o d y namic measurements were performed under baseline conditions and 20 min after the onset of either drug or vehicle administration.
R. MOREAU et al.
Drugs Glibenclamide and diazoxide were purchased from Sigma Chemical (St. Louis, MO, U.S.A.). Thirty mg of glibenclamide were dissolved in 1 ml 0.1 N N a O H + 4 ml 5% dextrose, and 30 mg of diazoxide in 300 pl N,Ndimethyl-formamide.
Statistical analysis Values are expressed as m e a n s + S . E . C o m p a r i s o n s were performed using a Student's t-test for paired data. p < 0.05 was considered as significant.
Hemod)'tlamic m e a s l l r e m e n l s Four hours before h e m o d y n a m i c measurements, catheters were inserted under light diethyl ether anesthesia. Arterial pressure and heart rate were measured using a catheter inserted into a femoral artery. The left ventricle was cannulated via the right carotid artery. Cardiac and regional blood flows were measured by the radioactive microsphere method and the reference sample method as previously described (5). For the first h e m o d y n a m i c measurements, a precounted aliquot of approximately 60000, 16+1 l~m diameter, l'*lCe-labeled microspheres (spec. act. 10 mCi/g; New England Nuclear, Boston, MA, U.S.A.), suspended in Ficoll 70 (10% Pharmacia Fine Chemicals AB, Uppsala, Sweden) and Tween 80 (0.01%) and ultrasonically agitated, was injected into the ventricular catheter and flushed with 1 ml of isotonic saline over 45 s. During microsphere injection, a reference blood sample was drawn from the catheter in the femoral artery into a motor-driven syringe at 0.8 ml/min for 1 min. For the second set of h e m o d y n a m i c measurements, an injection of l~3Sn-labeled microspheres was given and the same technique was used. The animal was then killed with an overdose of pentobarbital sodium. Individual organs were dissected and placed in individual tubes for counting with a g a m m a - c o u n t e r (Computer G a m m a G 4 0 0 0 ; K o n t r o n , Montigny-Le-Bretonneux, France) at energy settings of 70-210 and 280-1000 keV for 113Sn and 141Ce channel was corrected using xX3Sn and 1'*lCe standards. Adequate microsphere mixing was assumed with a difference < 1 0 % between left and right kidney. Regional blood flows were calculated by the following formula: organ blood flow ( m l . m i n - 1 . I00 g - 1)= [organ radioactivity (cpm)/radioactivity injected (cpm)] × CI ( m l . m i n - 1. 100 g - 1). Portal tributary blood flow was calculated as the sum of stomach, intestine, colon, spleen, and mesenteric-pancreas flows. Cardiac index, stroke volume index, and systemic and regional vascular resistance were calculated as previously described (5).
Results Glibenclamide significantly decreased cardiac index, stroke volume index, portal pressure, portal tributary blood flow, and hepatic artery blood flow, and significantly increased systemic vascular resistance, portal territory vascular resistance, hepatic artery vascular resistance, and renal vascular resistance (Table 1, Fig. 1). Plasma glucose concentrations were 0.91+0.11 and 0.88+0.09 mmol/1, before and after glibenclamide administration, respectively. TABLE 1 Systemic, splanchnic and renal hemodynamic responses to glibenclamide (20 mg/kg i.v.) in rats with portal vein stenosis Baseline After glibenclamide Heart rate 399+14~ 392+17 (beats. min - t) Mean arterial pressure 106+2 109+3 (mmHg) Cardiac index 33.2 + 1.9 24.9 + 1.2b (ml-min- t. 100 g- l) Stroke volume index 86+8 65+3 b (/d.min-I-100g -l) Systemic vascular resistance 258+12 355+16b [ 1 0 3 x (dyn.s.cm -5. 100 g)] Portal pressure 14.5+1.5 12.2+1.2b (mmHg) Portal tributary blood flow 7.2+0.6 4.9+0.5 b (ml.min-l.100g -1) Portal territory vascular 1057+99 1649+130b resistance [103 × (dyn.s.cm -5- 100 g)] Hepatic artery blood flow 1.3+0.2 0.8+0.I b (ml.min- I. 100 g- 1) Hepatic artery vascular 71 +10 112+13 b resistance [105 x (dyn.s-cm -5-100 g)] Hepatocollateral vascular 165+18 206+23 resistance [ 1 0 3 × (dyn.s.cm -5. 100 g)] Renal blood flow 3.0+0.3 2.7+0.2 (ml.min- 1. 100 g- 1) Renal vascular resistance 29 + 3 34__+3b [105 x (dyn.s.cm -5. 100 g)] Mean + SE. bSignificantly different from baseline value (p < 0.05).
HEMODYNAMIC RESPONSES TO GLIBEN CLAMIDE IN PORTAL HYPERTENSION
Portal pressure
E E:
14
Portal tributary blood flow
~
8
E before
Arterial pressure 130
16.5_4.5mmHg; portal tributary blood flow was 8.4+0.2 and 3.7+0.1 ml.min- t. 100 g- '; portal territory vascular resistance was 901 +133 and 2319+l [(dyn.s. c m - S - 1 0 0 g ) x l 0 3 ] ; hepatic artery blood flow was 0.63+__0.05 and 0.63+0.13 m l . m i n - l . 1 0 0 g - 1 ; hepatic artery vascular resistance was 139___62 and 165+44 [(dyn-s'cm -s'100 g) xl0S]; hepatocollateral vascular resistance was 148+17 and 356+86 [(dyn.s.cm -5. 100 g) x103]; renal blood flow 3.48+0.01 and 1.58 + 0.94 ml. min- 1. 100 g - 1; and renal vascular resistance was 25 _ 3 and 100 _ 64 [(dyn. s.cm - 5. 100 g) x 105]. Finally, neither the vehicle of glibenclamide nor the vehicle of diazoxide changed systemic and regional hemodynamics.
110
Discussion
6.5
3.0
after
before
Cardiac index
after
,p
o o r-
217
o~
"I,-
30
E E
E
E 20
before
after
90
before
after
Fig. I. Hemodynamic effects of glibenclamide in conscious rats with portal vein stenosis.
The effects of diazoxide alone were as follows: Mean arterial pressure was 106+3 and 88+1 mmHg (before and after diazoxide, respectively); heart rate was 382___5 and 446 + 5 beats-min - 1; stroke volume index was 97 + 5 and ll2+__6Fll.min-l-100g-1; cardiac index was 36.8 + 3.4 and 50.9 ___13.0 ml. min- 1.100 g- 1; systemic vascular resistance was 233+28 and 147+36 [(dyn's' cm -5- 100 g)x 103]; portal pressure was 13.0+ 1.0 and 14.5 ___0.5 mmHg; portal tributary blood flow was 6.9 ___0.4 and 7.3 ___0.8 ml. min - 1.100 g- 1; portal territory vascular resistance was 1080_ 115 and 818 _ 84 [(dyns . c m - 5 " 1 0 0 g ) x l 0 3 ] ; hepatic artery blood flow was 1.01 +__0.42 and 2.30+0.70 m l - m i n - l . 1 0 0 g - l ; hepatic artery vascular resistance was 103 _ 45 and 3 4 _ 10 [(dyns.cm -5.100 g) x105]; hepatocollateral vascular resistance was 149+__2 and 161+12 [(dyn.s.cm-5. 100 g) × 103]; renal blood flow 2.59+0.04 and 2.88+0.45 ml-min -1. 100 g-1; and renal vascular resistance was 33 ___2 and 25 ___4 [(dyn" s'cm-5. 100 g)x 105]. The effects of glibenclamide plus diazoxide were as follows: Mean arterial pressure was 110+14 and 123+8 mmHg (before and after combined therapy, respectively); heart rate was 355 + 4 and 358___10 beats. m i n - l ; stroke volume index was 96+1 and 66+6~dm i n - ] ' 1 0 0 g - 1 ; cardiac index was 34.0+0.9 and 23.6 + 2.5 ml-min- 1. 100 g - 1; systemic vascular resistance was 259+27 and 423+71 [(dyn-s'cm -5' 100 g) ×103]; portal pressure was 15.5+1.5 and
The present study performed in conscious rats with extrahepatic portal hypertension shows that glibenclamide induced decreased cardiac output but caused no changes in arterial pressure, and thus induced systemic vasoconstriction. Our results show that the decrease in cardiac output was due to a reduced stroke volume. Since it has been shown that glibenclamide decreases cardiac contractility in isolated perfused rat hearts (6), this mechanism might account for the reduced stroke volume found in this study. The results show that glibenclamide induced an increase in portal territory vascular resistance which caused reduced portal tributary blood flow. This, in turn, induced a portal hypotensive effect. On the other hand, since hepatocollateral vascular resistance tended to rise following glibenclamide administration, this effect might have limited the reduction of portal pressure. In this study, glibenclamide also elicited hepatic artery vasoconstriction which induced a fall in hepatic artery blood flow. Since the normal response to decreased portal tributary blood flow is an adenosine-mediated increase in hepatic artery blood flow (7), these results suggest that glibenclamide might inhibit the vasodilating action of adenosine on the hepatic artery. Finally, in the present study, glibenclamide induced slight renal hypoperfusion and vasoconstriction. It should be emphasized, however, that decreases in renal blood flow did not reach statistical significance and renal vasoconstriction was less marked than vasoconstriction in other territories. Glibenclamide has been shown to block vascular ATPsensitive K + channels (1). Moreover, in the present study glibenclamide suppressed the systemic and regional vaso-
218
R. MOREAU et al.
dilation induced by the ATP-sensitive K ÷ channel opener, diazoxide. As a result, the blockade of ATPsensitive K ÷ channels seems to be the cause of the glibenclamide vasoconstrictor effect. Since glibenclamide elicited increased vascular tone in the systemic, splanchnic and renal vascular beds in portal hypertensive rats,
this suggests that under baseline conditions ATP-sensitive K + channels are opened and responsible for a vasodilator tone in these territories. In conclusion, the results of the present study showed that in portal hypertensive rats, glibenclamide lowers portal hypertension and hyperkinetic circulation.
References
relaxation response to BRL 34915 (cromakalim), minoxidil sulfate and diazoxide in vascular smooth muscle, J Pharmacol Exp Ther 1989; 248: 149-56. 5 Lee SS, Girod C, Valla D, Geoffroy P, Lebrec D. Effects of pentobarbital anesthesia on splanchnic hemodynamics of normal and portal-hypertensive rats. Am J Physiol 1985; 249: G528-32. 6 Mitani A, Kinoshita K, Fukamachi K, el al. Effects of glibenclamide and nicorandil on cardiac function during ischemia and reperfusion in isolated perfused hearts. Am J Physiol 1991; 261: H1864-71. 7 Ezzat WR, Lautt WW. Hepatic arterial pressure-flow autoregulation is adenosine mediated. Am J Physiol 1987; 252: H836-45.
1 Standen NB, Quayle JM, Davies NW, Brayden JE, Huang Y, Nelson MT. Hyperpolarizing vasodilators activate ATP-sensitive K + channels in arterial smooth muscle. Science 1989: 245: 177-80. 2 Quast U, Cook NS. Moving together: K" channel openers and ATPsensitive K" channels. Trends Pharmacol Sci 1989; 10: 431-5. 3 Quast U, Cook NS. In vitro and in vivo comparison of two channel openers, diazoxide and cromakalim, and their inhibition by glibenclamide. J Pharmacol Exp Ther 1989: 250: 261-71. 4 Winquist RJ, Heaney LA, Wallace AA, et al. Glyburide blocks the
215
©1992 ElsevierScientificPublishers Ireland Ltd. All rights reserved. 0168-8278/92/$05.00 HEPAT01214 Rapid Publication
Blockade of ATP-sensitive K + channels by glibenclamide reduces portal pressure and hyperkinetic circulation in portal hypertensive rats Richard Moreau, Hirokazu Komeichi, St6phane Cailmail and Didier Lebrec Laboratoire d'Hbmodynamique Splam'hnique. Unit~de Recherches de Physiopathologie Hbpatique. I NSERM U-24. H6pital Beaujon, Clichy. France
Certain results of in vitro studies raise the possibility that blockade of ATP-sensitive K + channels by glibenclamide may induce vasoconstriction. Therefore, this substance might decrease portal pressure and hyperkinetic circulation in animals with portal hypertension. Thus, systemic and regional hemodynamics (radioactive microspheres) were measured before and 20 min after a bolus intravenous injection of glibenclamide (20 mg/kg) in conscious rats with portal vein stenosis. Blood pressure decreased significantly from 14.5_+1.5 to 12.2+_1.2 (mean _+SE). Cardiac index significantly decreased by 24%, portal tributary blood flow by 31%, and hepatic artery blood flow by 35%. Systemic vascular resistance significantly increased by 38%, portal territory vascular resistance and hepatic artery vascular resistance by 61%, each, and renal vascular resistance by 17%. Arterial pressure, heart rate, and renal blood flow were unchanged. Moreover, glibenclamide blunted the vasodilating action of diazoxide (an ATP-sensitive K + channel opener). These results show that in rats with extrahepatic portal hypertension the blockade of ATP-sensitive K + channels by glibenclamide reduces portal pressure and hyperkinetic circulation. Key words: Sulfonylurea; Systemic and splanchnic hemodynamics; Extrahepatic portal hypertension
In vitro studies have shown that arterial smooth muscle cells contain ATP-sensitive K + channels whose blockade by the sulfonylurea glibenclamide leads to vasoconstriction (1-4). As a result, glibenclamide might reduce portal pressure and the hyperkinetic circulation in portal hypertension. In these portal-hypertensive animals, however, systemic and regional hemodynamic responses to glibenclamide have not yet been investigated. Thus, the aim of the present study was to determine the effects of glibenclamide administration on systemic and regional hemodynamics in conscious rats with extrahepatic portal hypertension.
studied. All rats were allowed free access to food and water until 14-18 h before the study, when food was withdrawn. Portal hypertension was induced by ligation of the portal vein (5). Under ether anesthesia, the abdomen was incised and the portal vein exposed. A polyethylene catheter of 0.96 mm external diameter was passed along the portal vein, and 3-0 silk was used to ligate both the catheter and the portal vein. All rats were studied "at least 3 weeks after the operation. Protocols performed in this laboratory are approved by the French Agricultural Office in conformity with European legislation for research involving animals.
Protocols
Materials and Methods Animals
Twelve adult male Sprague-Dawley rats (Charles River Laboratoires, Saint-Aubin-L6s-Elbeuf, France) were
Six rats received glibenclamide alone (20 mg/kg i.v. bolus) and plasma glucose concentrations were measured before and after glibenclamide. Four additional experiments were performed. In the first set of experiments, two rats received a single dose of the ATP-sensitive K +
Correspondence to: Dr. R. Moreau, INSERM U-24, H6pital Beaujon,92118 Clichy,France.
216 channel opener, diazoxide (30 mg/kg i.v. bolus) (3,4). In the second set of experiments, two rats received glibenclamide and diazoxide (30 mg/kg i.v.) 15 min later. In the third set of experiments, one rat received the vehicle of glibenclamide only. In the fourth set of experiments, one rat received the vehicle of diazoxide only. H e m o d y namic measurements were performed under baseline conditions and 20 min after the onset of either drug or vehicle administration.
R. MOREAU et al.
Drugs Glibenclamide and diazoxide were purchased from Sigma Chemical (St. Louis, MO, U.S.A.). Thirty mg of glibenclamide were dissolved in 1 ml 0.1 N N a O H + 4 ml 5% dextrose, and 30 mg of diazoxide in 300 pl N,Ndimethyl-formamide.
Statistical analysis Values are expressed as m e a n s + S . E . C o m p a r i s o n s were performed using a Student's t-test for paired data. p < 0.05 was considered as significant.
Hemod)'tlamic m e a s l l r e m e n l s Four hours before h e m o d y n a m i c measurements, catheters were inserted under light diethyl ether anesthesia. Arterial pressure and heart rate were measured using a catheter inserted into a femoral artery. The left ventricle was cannulated via the right carotid artery. Cardiac and regional blood flows were measured by the radioactive microsphere method and the reference sample method as previously described (5). For the first h e m o d y n a m i c measurements, a precounted aliquot of approximately 60000, 16+1 l~m diameter, l'*lCe-labeled microspheres (spec. act. 10 mCi/g; New England Nuclear, Boston, MA, U.S.A.), suspended in Ficoll 70 (10% Pharmacia Fine Chemicals AB, Uppsala, Sweden) and Tween 80 (0.01%) and ultrasonically agitated, was injected into the ventricular catheter and flushed with 1 ml of isotonic saline over 45 s. During microsphere injection, a reference blood sample was drawn from the catheter in the femoral artery into a motor-driven syringe at 0.8 ml/min for 1 min. For the second set of h e m o d y n a m i c measurements, an injection of l~3Sn-labeled microspheres was given and the same technique was used. The animal was then killed with an overdose of pentobarbital sodium. Individual organs were dissected and placed in individual tubes for counting with a g a m m a - c o u n t e r (Computer G a m m a G 4 0 0 0 ; K o n t r o n , Montigny-Le-Bretonneux, France) at energy settings of 70-210 and 280-1000 keV for 113Sn and 141Ce channel was corrected using xX3Sn and 1'*lCe standards. Adequate microsphere mixing was assumed with a difference < 1 0 % between left and right kidney. Regional blood flows were calculated by the following formula: organ blood flow ( m l . m i n - 1 . I00 g - 1)= [organ radioactivity (cpm)/radioactivity injected (cpm)] × CI ( m l . m i n - 1. 100 g - 1). Portal tributary blood flow was calculated as the sum of stomach, intestine, colon, spleen, and mesenteric-pancreas flows. Cardiac index, stroke volume index, and systemic and regional vascular resistance were calculated as previously described (5).
Results Glibenclamide significantly decreased cardiac index, stroke volume index, portal pressure, portal tributary blood flow, and hepatic artery blood flow, and significantly increased systemic vascular resistance, portal territory vascular resistance, hepatic artery vascular resistance, and renal vascular resistance (Table 1, Fig. 1). Plasma glucose concentrations were 0.91+0.11 and 0.88+0.09 mmol/1, before and after glibenclamide administration, respectively. TABLE 1 Systemic, splanchnic and renal hemodynamic responses to glibenclamide (20 mg/kg i.v.) in rats with portal vein stenosis Baseline After glibenclamide Heart rate 399+14~ 392+17 (beats. min - t) Mean arterial pressure 106+2 109+3 (mmHg) Cardiac index 33.2 + 1.9 24.9 + 1.2b (ml-min- t. 100 g- l) Stroke volume index 86+8 65+3 b (/d.min-I-100g -l) Systemic vascular resistance 258+12 355+16b [ 1 0 3 x (dyn.s.cm -5. 100 g)] Portal pressure 14.5+1.5 12.2+1.2b (mmHg) Portal tributary blood flow 7.2+0.6 4.9+0.5 b (ml.min-l.100g -1) Portal territory vascular 1057+99 1649+130b resistance [103 × (dyn.s.cm -5- 100 g)] Hepatic artery blood flow 1.3+0.2 0.8+0.I b (ml.min- I. 100 g- 1) Hepatic artery vascular 71 +10 112+13 b resistance [105 x (dyn.s-cm -5-100 g)] Hepatocollateral vascular 165+18 206+23 resistance [ 1 0 3 × (dyn.s.cm -5. 100 g)] Renal blood flow 3.0+0.3 2.7+0.2 (ml.min- 1. 100 g- 1) Renal vascular resistance 29 + 3 34__+3b [105 x (dyn.s.cm -5. 100 g)] Mean + SE. bSignificantly different from baseline value (p < 0.05).
HEMODYNAMIC RESPONSES TO GLIBEN CLAMIDE IN PORTAL HYPERTENSION
Portal pressure
E E:
14
Portal tributary blood flow
~
8
E before
Arterial pressure 130
16.5_4.5mmHg; portal tributary blood flow was 8.4+0.2 and 3.7+0.1 ml.min- t. 100 g- '; portal territory vascular resistance was 901 +133 and 2319+l [(dyn.s. c m - S - 1 0 0 g ) x l 0 3 ] ; hepatic artery blood flow was 0.63+__0.05 and 0.63+0.13 m l . m i n - l . 1 0 0 g - 1 ; hepatic artery vascular resistance was 139___62 and 165+44 [(dyn-s'cm -s'100 g) xl0S]; hepatocollateral vascular resistance was 148+17 and 356+86 [(dyn.s.cm -5. 100 g) x103]; renal blood flow 3.48+0.01 and 1.58 + 0.94 ml. min- 1. 100 g - 1; and renal vascular resistance was 25 _ 3 and 100 _ 64 [(dyn. s.cm - 5. 100 g) x 105]. Finally, neither the vehicle of glibenclamide nor the vehicle of diazoxide changed systemic and regional hemodynamics.
110
Discussion
6.5
3.0
after
before
Cardiac index
after
,p
o o r-
217
o~
"I,-
30
E E
E
E 20
before
after
90
before
after
Fig. I. Hemodynamic effects of glibenclamide in conscious rats with portal vein stenosis.
The effects of diazoxide alone were as follows: Mean arterial pressure was 106+3 and 88+1 mmHg (before and after diazoxide, respectively); heart rate was 382___5 and 446 + 5 beats-min - 1; stroke volume index was 97 + 5 and ll2+__6Fll.min-l-100g-1; cardiac index was 36.8 + 3.4 and 50.9 ___13.0 ml. min- 1.100 g- 1; systemic vascular resistance was 233+28 and 147+36 [(dyn's' cm -5- 100 g)x 103]; portal pressure was 13.0+ 1.0 and 14.5 ___0.5 mmHg; portal tributary blood flow was 6.9 ___0.4 and 7.3 ___0.8 ml. min - 1.100 g- 1; portal territory vascular resistance was 1080_ 115 and 818 _ 84 [(dyns . c m - 5 " 1 0 0 g ) x l 0 3 ] ; hepatic artery blood flow was 1.01 +__0.42 and 2.30+0.70 m l - m i n - l . 1 0 0 g - l ; hepatic artery vascular resistance was 103 _ 45 and 3 4 _ 10 [(dyns.cm -5.100 g) x105]; hepatocollateral vascular resistance was 149+__2 and 161+12 [(dyn.s.cm-5. 100 g) × 103]; renal blood flow 2.59+0.04 and 2.88+0.45 ml-min -1. 100 g-1; and renal vascular resistance was 33 ___2 and 25 ___4 [(dyn" s'cm-5. 100 g)x 105]. The effects of glibenclamide plus diazoxide were as follows: Mean arterial pressure was 110+14 and 123+8 mmHg (before and after combined therapy, respectively); heart rate was 355 + 4 and 358___10 beats. m i n - l ; stroke volume index was 96+1 and 66+6~dm i n - ] ' 1 0 0 g - 1 ; cardiac index was 34.0+0.9 and 23.6 + 2.5 ml-min- 1. 100 g - 1; systemic vascular resistance was 259+27 and 423+71 [(dyn-s'cm -5' 100 g) ×103]; portal pressure was 15.5+1.5 and
The present study performed in conscious rats with extrahepatic portal hypertension shows that glibenclamide induced decreased cardiac output but caused no changes in arterial pressure, and thus induced systemic vasoconstriction. Our results show that the decrease in cardiac output was due to a reduced stroke volume. Since it has been shown that glibenclamide decreases cardiac contractility in isolated perfused rat hearts (6), this mechanism might account for the reduced stroke volume found in this study. The results show that glibenclamide induced an increase in portal territory vascular resistance which caused reduced portal tributary blood flow. This, in turn, induced a portal hypotensive effect. On the other hand, since hepatocollateral vascular resistance tended to rise following glibenclamide administration, this effect might have limited the reduction of portal pressure. In this study, glibenclamide also elicited hepatic artery vasoconstriction which induced a fall in hepatic artery blood flow. Since the normal response to decreased portal tributary blood flow is an adenosine-mediated increase in hepatic artery blood flow (7), these results suggest that glibenclamide might inhibit the vasodilating action of adenosine on the hepatic artery. Finally, in the present study, glibenclamide induced slight renal hypoperfusion and vasoconstriction. It should be emphasized, however, that decreases in renal blood flow did not reach statistical significance and renal vasoconstriction was less marked than vasoconstriction in other territories. Glibenclamide has been shown to block vascular ATPsensitive K + channels (1). Moreover, in the present study glibenclamide suppressed the systemic and regional vaso-
218
R. MOREAU et al.
dilation induced by the ATP-sensitive K ÷ channel opener, diazoxide. As a result, the blockade of ATPsensitive K ÷ channels seems to be the cause of the glibenclamide vasoconstrictor effect. Since glibenclamide elicited increased vascular tone in the systemic, splanchnic and renal vascular beds in portal hypertensive rats,
this suggests that under baseline conditions ATP-sensitive K + channels are opened and responsible for a vasodilator tone in these territories. In conclusion, the results of the present study showed that in portal hypertensive rats, glibenclamide lowers portal hypertension and hyperkinetic circulation.
References
relaxation response to BRL 34915 (cromakalim), minoxidil sulfate and diazoxide in vascular smooth muscle, J Pharmacol Exp Ther 1989; 248: 149-56. 5 Lee SS, Girod C, Valla D, Geoffroy P, Lebrec D. Effects of pentobarbital anesthesia on splanchnic hemodynamics of normal and portal-hypertensive rats. Am J Physiol 1985; 249: G528-32. 6 Mitani A, Kinoshita K, Fukamachi K, el al. Effects of glibenclamide and nicorandil on cardiac function during ischemia and reperfusion in isolated perfused hearts. Am J Physiol 1991; 261: H1864-71. 7 Ezzat WR, Lautt WW. Hepatic arterial pressure-flow autoregulation is adenosine mediated. Am J Physiol 1987; 252: H836-45.
1 Standen NB, Quayle JM, Davies NW, Brayden JE, Huang Y, Nelson MT. Hyperpolarizing vasodilators activate ATP-sensitive K + channels in arterial smooth muscle. Science 1989: 245: 177-80. 2 Quast U, Cook NS. Moving together: K" channel openers and ATPsensitive K" channels. Trends Pharmacol Sci 1989; 10: 431-5. 3 Quast U, Cook NS. In vitro and in vivo comparison of two channel openers, diazoxide and cromakalim, and their inhibition by glibenclamide. J Pharmacol Exp Ther 1989: 250: 261-71. 4 Winquist RJ, Heaney LA, Wallace AA, et al. Glyburide blocks the
