- Email: [email protected]
Pathogenesis of portal hypertensive gastropathy: A clinical and experimental review
Portal hypertension Pathogenesis of portal hypertensive gastropathy: A clinical and experimental review Masayuki Ohta, MD, Shohei Yamaguchi, MD, Norikazu Gotoh, MD, and Morimasa Tomikawa, MD, Fukuoka, Japan
Portal hypertensive gastropathy (PHG) is recognized as a clinical entity in portal hypertension, but the pathogenesis of PHG is still unclear. Therefore, we reviewed the current state of knowledge concerning the portal hypertensive gastric mucosa and hypothesized the pathogenesis of PHG. Elevated portal pressure can induce changes of local hemodynamics, thus causing congestion in the upper stomach and gastric tissue damage. These changes may then activate cytokines and growth factors, such as tumor necrosis factor α, which are substances that activate endothelial constitutive nitric oxide synthase and endothelin 1 in the portal hypertensive gastric mucosa. Overexpressed nitric oxide synthase produces an excess of nitric oxide, which induces hyperdynamic circulation and peroxynitrite overproduction. The overproduction of peroxynitrite, together with endothelin overproduction may cause an increased susceptibility of gastric mucosa to damage. When combined with the characteristics of impaired mucosal defense and healing, these factors may together produce PHG in patients with portal hypertension. (Surgery 2002;131:S165-70.) From the Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
ESOPHAGEAL VARICES HAVE LONG been considered the major cause of upper gastrointestinal hemorrhage in patients with portal hypertension. However, gastric mucosal lesions are another frequent cause of upper gastrointestinal bleeding in these patients, accounting for 20% to 40% of all bleeding.1 Initially, mucosal bleeding is caused by erosive gastritis, namely inflammation of the gastric mucosa.2 McCormack et al3 demonstrated that the gastric mucosa, observed as hemorrhagic gastritis endoscopically in portal hypertensive patients, showed mucosal and submucosal ectasia but not necessarily any inflammatory changes. Since the term congestive gastropathy3 was introduced, these mucosal lesions were termed congestive gastropathy or portal hypertensive gastropathy (PHG).4 Now PHG is recognized as a clinical entity in portal hypertension. The macroscopic appearance of PHG is endoscopically characterized by either mosaic pattern or snake-skin pattern.4,5 In addition, PHG can be subdivided into 2 grades, mild and severe.6 Reprint requests: Masayuki Ohta, MD, Department of Surgery, Matsuyama Red Cross Hospital, 1 Bunkyo-cho, Matsuyama 7908524, Japan. Copyright © 2002 by Mosby, Inc. 0039-6060/2002/$35.00 + 0 11/0/119499 doi:10.1067/msy.2002.119499
Portal hypertensive gastric mucosa in animal models has an increased susceptibility to damage due to noxious factors such as alcohol and aspirin.7 Investigating the mechanism of this increased susceptibility in experimental studies has been thought to be important for elucidating the pathogenesis of PHG. Many studies have reported the hemodynamic data of PHG; however, such findings remain controversial. In addition, several articles have also suggested that vasoactive factors are related to the pathogenesis of portal hypertensive gastric mucosa, but this aspect is not confirmed. This article reviews the current state of knowledge concerning the hemodynamics, relative vasoactive factors, and mucosal defense system of portal hypertensive stomach; the pathogenesis of PHG is also hypothesized. HEMODYNAMICS AND PHG Portal hemodynamics. Since decompression shunt surgery is effective for the treatment of PHG,2 the elevated portal venous pressure must play an important role in the pathogenesis of the PHG. McCormack et al3 measured the wedged hepatic venous pressure in 18 portal hypertensive patients, and there was no significant difference in the pressure between the patients with and without PHG (17.3 ± 1.6 vs 16.0 ± 1.6 mm Hg). Two other SURGERY S165
S166 Ohta et al
Surgery January 2002
Fig 1. Hypothesized pathogenesis of PHG.
studies demonstrated no significant difference in the hepatic venous pressure gradient (HVPG) between portal hypertensive patients with and without PHG.5,8 The direct portal venous pressure also showed no substantial difference between the 2 groups.9 Iwao et al6 subdivided PHG into mild and severe, and thus demonstrated significant differences in HVPG between portal hypertensive patients with severe PHG and without PHG but not between those with severe PHG and with mild. Therefore, an elevated portal pressure may be an important but not a critical factor in the pathogenesis of PHG. Endoscopic injection sclerotherapy has been reported to result in a deterioration of PHG in several articles.10 PHG significantly worsens 6 to 9 months after the eradication of varices by sclerotherapy; however, a gradual improvement in PHG normally follows.10 Recurrent small veins requiring additional sclerotherapy appear more frequently in patients with PHG.10 When portal hypertensive patients with varices and PHG are treated by sclerotherapy, PHG has a high likelihood of bleeding. Nakayama11 et al demonstrated that a gastrorenal shunt may play a protective role in the development of PHG after sclerotherapy. These findings show that the development of PHG is related more to the local hemodynamics, namely, upper gastric congestion, than to the portal hemodynamics. We9 have described the portal-variceal pressure gradient (portal venous pressure minus esophageal variceal pressure) to be an index of upper gastric congestion. Although the portal venous pressure
and esophageal variceal pressure did not differ in cirrhotic patients with or without PHG before the respective treatments, the portal-variceal pressure gradient significantly increased in the PHG-positive group more than in the PHG-negative group. In two studies9,12 of Doppler ultrasonography, the portal venous blood flow is reported unchanged in cirrhotic patients with PHG compared with those without PHG. Iwao et al12 showed that the portal blood flow was similar in cirrhotic patients and normal controls. Therefore the portal blood flow is not considered to play an important role in the development of PHG. Gastric mucosal blood flow. Many experimental studies13,14 with animal models of portal hypertension have shown a consistent increase in the total and mucosal gastric blood flow using different techniques. However, several studies that have measured the gastric mucosal blood flow in cirrhotic patients with PHG by means of laser-Doppler flowmetry, using the same instruments, have reported conflicting results. Some studies show an increased gastric mucosal blood flow in PHG,9 while others show a decreased mucosal blood flow in PHG.15 It thus remains highly controversial as to whether the condition of PHG is active or passive congestion. Although various biases exist such as chronic anemic conditions in patients and some difficulty for the contact techniques of endoscopic probe in laser-Doppler flowmetry, it seems that these conflicting results may be caused mainly by the characteristic angioarchitecture of the gastric mucosa in portal hypertension. Hashizume et al16 demonstrat-
Surgery Volume 131, Number 1
ed dilated microvessels in the mucosa and submucosa in cirrhotic patients in a morphologic study. A recent histologic study investigated the diameter and thickness of the wall of mucosal capillaries in the stomachs of cirrhotic patients and healthy volunteers.17 Although the diameter of the fundal capillaries did not differ among cirrhotic patients and controls, the wall thickness of the capillaries in cirrhotic patients was significantly greater than that in the controls. In the experimental studies,18 gastric mucosal microvessels are also dilated in portal hypertension. However, electron microscopic studies demonstrate that endothelial cells of the gastric mucosal capillaries in the surface layer are enlarged and the capillary lumina are narrowed in portal hypertensive models.18,19 A report that oxygenation of the gastric mucosal surface is reduced in the animal models of portal hypertension20 probably supports the belief that a narrowed capillary lumina exists in the portal hypertensive gastric mucosa. We therefore assume that the superficial mucosal blood flow decreases while the total mucosal flow increases in portal hypertensive gastric mucosa. The controversial data on laserDoppler flowmetry in PHG regarding cirrhotic patients may thus be due to these small variations. VASOACTIVE FACTORS AND PORTAL HYPERTENSIVE GASTRIC MUCOSA Nitric oxide. Nitric oxide (NO), a potent vasodilator, is generated from the terminal guanidio-nitrogen atoms of L-arginine by an enzyme, NO synthase (NOS).21 Distinct cDNAs for NOS enzyme have been isolated for an inducible NOS and 2 constitutive NOSs, one synthesized by endothelial cells (endothelial constitutive NOS, ecNOS) and the other by neuronal cells.21 In the gastric mucosa, NO plays a major role in mucosal defense by modulating the mucosal circulation.22 The NO protects the gastric mucosa against injury by ethanol and endothelin-1(ET-1), whereas the inhibition of NOS increases gastric mucosal injury.23 However, excessive NO production has a cytotoxic potential, thus the increasing gastric mucosal injury.23,24 In portal hypertension, NO is overproduced. The plasma levels of the NO metabolite and cGMP increase in portal hypertensive models.25 The NOS mRNA expression and NOS activity are enhanced in portal hypertensive vessels.25,26 Overproduced NO mediates the systemic hyperdynamic circulation and development of collaterals.27 This increased activity of NOS is most likely caused by the activation of constitutive NOS isoform because the inhibition of inducible NOS activity by dexamethasone does not affect the
Ohta et al S167
hyperdynamic circulation of portal hypertension.28 However, recent reports have demonstrated that a cirrhotic liver is impaired in the intrahepatic vasodilatory response to NO agonist and has a reduced production of NO.29 In the portal hypertensive gastric mucosa of the animals, the ecNOS mRNA and protein levels and constitutive NOS enzyme activity all increase more than the controls.14 However, the inducible NOS enzyme activity remains unchanged. The NOS activity and NO metabolites in the gastric mucosa of PHG also increase in cirrhotic patients.30 Overproduced NO increases the gastric mucosal blood flow in the portal hypertensive models.14,31 Gastric mucosal necrosis induced by ethanol in portal hypertension increases after the administration of high-dose N-nitro-L-arginine methyl ester (NOS inhibitor) but decreases after low-dose administration.14 Excessive NO production by an overexpression of ecNOS plays an important role in the increased susceptibility to damage. An excess amount of NO can produce peroxynitrite, which initiates membrane lipid peroxidation and thus results in cell injury.32 Tarnawski33 et al demonstrated the overproduction of peroxynitrite and enhanced lipid peroxidation in the portal hypertensive gastric mucosa. As a result, the overproduction of NO and peroxynitrite may be the underlying mechanism for the increased susceptibility of portal hypertensive gastric mucosa to damage, which is probably related to the pathogenesis of PHG. Tumor necrosis factor α (TNF-α), which is a multifunctional cytokine, is also overproduced in the portal hypertensive gastric mucosa.34 This cytokine can increase vascular permeability and cause both structural and metabolic changes in vascular endothelial cells.35 We demonstrated that TNF-α might regulate ecNOS expression in the portal hypertensive stomach, since anti-TNF-α treatment reversed an overexpression of ecNOS mRNA and protein and its enzyme activity.34 Endothelin. ET-1 is a 21-amino acid peptide initially noted for its powerful vasoconstrictor properties and also has been recently recognized to have pleiotropic biologic effects, depending on its receptors.36 Specific endothelin receptors are divided into at least 2 types: endothelin A receptor (ETAR) and endothelin B receptor (ETBR). The ETAR is localized in vascular smooth muscle and mediates vasoconstriction.36 The ETBR is predominantly expressed in endothelial cells and mediates the production of NO and prostacyclin.36 The submucosal injection of ET-1 induces gastric ulceration by decreasing the gastric mucosal blood flow, ischemia, and increased secretion of acid.37 ET-1
S168 Ohta et al
also enhances microvascular permeability of the stomach through the activation of ETAR.38 In cirrhotic patients, the plasma level of ET-1 has been shown to increase.39 Furthermore, ET-1 plays an important role in increased intrahepatic vascular resistance in portal hypertensive animals.40 The expression of ETAR and ETBR mRNAs increases in the vasculature of portal hypertension.41 The enhanced ETBR expression may be related to the maintenance of elevated portal pressure in the animals.41 The expression of ET-1, ETAR, and ETBR mRNAs is also enhanced in the portal hypertensive esophagus.42 In portal hypertensive gastric mucosa of the animals, ET-1 mRNA and protein levels increase in comparison with the controls.43 Although ETAR antagonist does not affect the gastric mucosal blood flow, it significantly reduces the gastric mucosal injury induced by ethanol in the portal hypertensive animals.43 In addition, ETAR/ETBR antagonist reverses hyperpermeability of portal hypertensive gastric mucosa.44 As a result, the overproduction of ET-1 in the portal hypertensive stomach can impair mucosal microcirculation through the ETAR and may thus also be related to the pathogenesis of PHG. Exogenous ET-1 increases the ecNOS mRNA expression and NO production through the ETBR, and ET-1 enhances pulmonary ecNOS level in the portal hypertensive models.45 The overproduction of ET-1 may modulate ecNOS regulation in the portal hypertensive gastric mucosa through the ETBR. Exogenous TNF-α elevates the plasma levels of ET-1,46 and TNF-α regulates the expression of the ET-1 gene in cultured endothelial cells.47 Therefore, TNF-α may activate not only the ecNOS gene but also the ET-1 gene in the portal hypertensive stomach. MUCOSAL DEFENSE AND HEALING OF THE PORTAL HYPERTENSIVE STOMACH Although vasoactive factors may be related to the pathogenesis of PHG, an impaired mucosal defensive system also plays an important role in the increased susceptibility of portal hypertensive gastric mucosa to damage. The gastric mucosal gel layer thickness significantly decreases in portal hypertensive animals.20 The mucosal gel layer thickness correlates positively with the surface epithelial cell pH.20 The contents of mucosal glycoprotein, hexosamine, in the portal hypertensive gastric mucosa have been shown to decrease in clinical and experimental studies.48 The oxygenation of the gastric mucosal surface decreases in the portal hypertensive gastric mucosa, whereas the
Surgery January 2002
total gastric mucosal blood flow increases in portal hypertension.20 These blunted mucosal defenses may cause an impaired adaptive cytoprotection against damage in the gastric mucosa of portal hypertensive rats.49 In the portal hypertensive gastric mucosa, angiogenesis is impaired after ethanol-induced injury.18 Vascular endothelial growth factor (VEGF) is an endothelial-specific angiogenetic factor. The elevation in mRNA expression of VEGF and its receptors and the phosphorylation of the receptors after ethanol-induced injury in the portal hypertensive gastric mucosa are significantly reduced compared with the sham-operated controls.50 Adrenomedullin, a potent vasodilatory peptide, is implicated in the hemodynamics in cirrhotic patients.51 The expression of adrenomedullin mRNA and protein in the portal hypertensive gastric mucosa after ethanolinduced injury significantly decreases in comparison with the controls.52 Extracellular signal-regulated kinase 2 (ERK2) is included in the mitogen-activated protein kinase cascade. The phosphorylation and activity of ERK2 do not increase in the portal hypertensive gastric mucosa following ethanol injury vs baseline values, whereas they significantly increase in the controls.53 These findings may help to explain the mechanism for impaired angiogenesis and healing after gastric mucosal injury due to portal hypertension. HYPOTHESIS OF PATHOGENESIS OF PHG Regarding the pathogenesis of PHG, in portal hypertension, elevated portal pressure can induce changes in the local hemodynamics, thus causing congestion in the upper stomach and gastric tissue damage, thereby activating cytokines and growth factors, such as TNF-α (Fig 1). These substances activate ecNOS and probably also ET-1. An overexpressed ecNOS produces an excessive amount of NO, which may induce systemic hyperdynamic circulation, increased gastric mucosal blood flow, and peroxynitrite overproduction. The overproduction of peroxynitrite, together with ET-1 overproduction, thus causes an increased susceptibility of gastric mucosa to undergo damage. When combined with the characteristics of an impaired mucosal defense and healing, these factors may together produce PHG in portal hypertension. An overproduction of NO and ET-1 are probably related to the pathogenesis of PHG. However, the long-term administration of NO synthase inhibitor can normalize the gastric blood flow on the 4-day but not on the 8-day regimen.54 The long-term inhibition of NO may activate other vasodilatory factors, such as prostaglandins, which may regulate
Ohta et al S169
Surgery Volume 131, Number 1 gastric mucosal blood flow.31 The interaction of NO and prostaglandin in the hemodynamics of portal hypertension has been suggested.55 However, the results regarding contents of endogenous prostaglandins including the gastric mucosa in cirrhotic patients remain controversial.56,57 It also remains controversial whether cyclooxygenase-2 plays an important role in the hemodynamics of portal hypertension.58,59 Further study of the gastric hemodynamics and prostaglandins under the longterm inhibition of NO synthase is thus called for.
15.
16. 17.
18.
REFERENCES 1. Terés J, Bordas JM, Bru C, Diaz F, Brugeura M, Rodés J. Upper gastrointestinal bleeding in cirrhosis: Clinical and endoscopic correlations. Gut 1976;17:37-40. 2. Sarfeh IJ, Juler GL, Stemmer EA, Mason GR. Results of surgical management of hemorrhagic gastritis in patients with gastroesophageal varices. Surg Gynecol Obstet 1982; 155:167-70. 3. McCormack TT, Sims J, Eyre-Brook I. Gastric lesions in portal hypertension; inflammatory or congestive? Gut 1985; 26:1226-32. 4. Kotzampassi K, Eleftheriadis E, Aleras H. The ‘mosaic-like’ pattern of portal hypertensive gastric mucosa after variceal eradication by sclerotherapy. J Gastroenterol Hepatol 1990;5:659-63. 5. Lin WJ, Lee FY, Lin HC, Tsai YT, Lee SD, Lai KH, et al. Snake skin pattern gastropathy in cirrhotic patients. J Gastroenterol Hepatol 1991;6:145-9. 6. Iwao T, Toyonaga A, Sumino M, Takagi K, Oho K, Nishizono M, et al. Portal hypertensive gastropathy in patients with cirrhosis. Gastroenterology 1992;102:2060-5. 7. Sarfeh IJ, Tarnawski A, Hajduczek A, Stachura J, Bui HX, Krause WJ. The portal hypertensive gastric mucosa: histologic, ultrastructural, and functional analysis after aspirin induced damage. Surgery 1988;104:79-85. 8. Quintero E, Pique JM, Bombi JA, Bordas JM, Sentis J, Elena M, et al. Gastric mucosal vascular ectasias causing bleeding in cirrhosis. A distinct entity associated with hypergastrinemia and low serum levels of pepsinogen I. Gastroenterology 1987;93:1054-61. 9. Ohta M, Hashizume M, Higashi H, Ueno K, Tomikawa M, Kishihara F, et al. Portal and gastric hemodynamics in cirrhotic patients with portal-hypertensive gastropathy. Hepatology 1994;20:1432-6. 10. Tanoue K, Hashizume M, Wada H, Ohta M, Kitano S, Sugimachi K. Effects of endoscopic injection sclerotherapy on portal hypertensive gastropathy: A prospective study. Gastrointest Endosc 1992;38:582-5. 11. Nakayama M, Iwao T, Oho K, Toyonaga A, Tanikawa K. Role of extravariceal collateral channels in the development of portal-hypertensive gastropathy before and after sclerotherapy. J Gastroenterol 1998;33:142-6. 12. Iwao T, Toyonaga A, Ikegami M, Sumino M, Oho K, Shigemori H, et al. Portal vein hemodynamics in cirrhotic patients with portal hypertensive gastropathy: An echoDoppler study. Hepatogastroenterology 1994;41:230-4. 13. Kitano S, Koyanagi N, Sugimachi K, Kobayashi M, Inokuchi K. Mucosal blood flow and modified vascular responses to norepinephrine in the stomach of rats with liver cirrhosis. Eur Surg Res 1982;14:221-30. 14. Ohta M, Tanoue K, Tarnawski AS, Pai R, Itani R, Sander FC, et al. Overexpressed nitric oxide synthase in portal-hyper-
19.
20.
21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
tensive stomach of rat: A key to increased susceptibility to damage? Gastroenterology 1997;112:1920-30. Iwao T, Toyonaga A, Ikegami M, Oho K, Sumino M, Harada H, et al. Reduced gastric mucosal blood flow in patients with portal-hypertensive gastropathy. Hepatology 1993; 18:36-40. Hashizume M, Tanaka K, Inokuchi K. Morphology of gastric microcirculation in cirrhosis. Hepatology 1983;3:1008-12. Misra V, Misra SP, Dwivedi M. Thickened gastric mucosal capillary wall: A histological marker for portal hypertension. Pathology 1998;30:10-3. Ichikawa Y, Tarnawski A, Sarfeh IJ, Ishikawa T, Shimada H. Distorted microangioarchitecture and impaired angiogenesis in gastric mucosa of portal hypertensive rats. Gastroenterology 1994;106:702-8. Tarnawski AS, Sarfeh IJ, Stachura J, Hajduczek A, Bui HX, Dabros W, et al. Microvascular abnormalities of the portal hypertensive gastric mucosa. Hepatology 1988;8:1488-94. Tomikawa M, Akiba Y, Kaunitz JD, Kawanaka H, Sugimachi K, Sarfeh IJ, et al. New insights into impairment of mucosal defense in portal hypertensive gastric mucosa. J Gastrointest Surg 2000;4:458-63. Sessa WC. The nitric oxide synthase family of proteins. J Vasc Res 1994;31:131-43. Whittle BJR. Neuronal and endothelium-derived mediators in the modulation of the gastric microcirculation: Integrity in the balance. Br J Pharmacol 1993;110:3-17. Lopez-Belmonte J, Whittle BJR, Moncada S. The actions of nitric oxide donors in the prevention or induction of injury to the rat gastric mucosa. Br J Pharmacol 1993;108:73-8. Ferraz JGP, Tigley A, Wallace JL. Paradoxical effects of Larginine on gastric mucosal integrity. Eur J Pharmacol 1994;260:107-11. Cahill PA, Foster C, Redmond EM, Gingalewski C, Wu Y, Sitzmann JV. Enhanced nitric oxide synthase activity in portal hypertensive rabbits. Hepatology 1995;22:598-606. Morales-Ruiz M, Jiménez W, Pérez-Sala D, Ros J, Leivas A, Lamas S, et al. Increased nitric oxide synthase expression in arterial vessels of cirrhotic rats with ascites. Hepatology 1996;24:1481-6. Lee FY, Colombato LA, Albillos A, Groszmann RJ. Administration of N-nitro-L-arginine ameliorates portal-systemic shunting in portal hypertensive rats. Gastroenterology 1993;105:1464-70. Fernández M, Garcia-Pagán JC, Casadevall M, Bernadich C, Piera C, Whittle BJR, et al. Evidence against a role for inducible nitric oxide synthase in the hyperdynamic circulation of portal hypertensive rats. Gastroenterology 1995;108:1487-95. Rockey DC, Chung JJ. Reduced nitric oxide production by endothelial cells in cirrhotic rat liver: Endothelial dysfunction in portal hypertension. Gastroenterology 1998; 114: 344-51. El-Newihi HM, Kanji VK, Mihas AA. Activity of gastric mucosal nitric oxide synthase in portal hypertensive gastropathy. Am J Gastroenterol 1996;91:535-8. Casadavall M, Panés J, Piqué JM, Marroni N, Bosch J, Whittle BJR. Involvement of nitric oxide and prostaglandins in gastric mucosal hyperemia of portal-hypertensive anesthetized rats. Hepatology 1993;18:628-34. Radi R, Beckman JS, Bush KM, Freemen BA. Peroxynitrite induced membrane lipid peroxydation: The cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 1991;288:481-7. Tarnawski AS, Tomikawa M, Kawanaka H, Sugimachi K, Sarfeh IJ. Increased oxygen free radicals, nitrotyrosine and
S170 Ohta et al
34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
enhanced lipid peroxydation in portal hypertensive gastric mucosa: A key to increased susceptibility to injury. Gastroenterology 1999;116:A1282. Ohta M, Tarnawski AS, Itani R, Pai R, Tomikawa M, Sugimachi K, et al. Tumor necrosis factor α regulates nitric oxide synthase expression in portal hypertensive gastric mucosa of rats. Hepatology 1998;27:906-13. Stephens KE, Ishizaka A, Larryk JW, Raffin TA. Tumor necrosis factor causes increased pulmonary permeability and edema. Comparison to septic acute lung injury. Am Rev Respir Dis 1988;137:1364-70. Levin ER. Mechanisms of disease: Endothelin. N Engl J Med 1995;333:356-63. Lazaratos S, Kashimura H, Nakahara A, Fukutomi H, Osuga T, Urushidani T, et al. Gastric ulcer induced by submucosal injection of ET-1: Role of potent vasoconstriction and intraluminal acid. Am J Physiol 1993;265:G491-8. Filep JG, Sirois MG, Foldes-Filep E, Rousseau A, Plante GE, Fournier A, et al. Enhancement by endothelin-1 of microvascular permeability via the activation of ETA receptors. Br J Pharmacol 1993;109:880-6. Isobe H, Satoh M, Sasaki H, Nawata H. Increased plasma endothelin-1 levels in patients with cirrhosis and esophageal varices. J Clin Gastroenterol 1993;17:227-30. Kamath PS, Tyce GM, Miller VM, Edwards BS, Rorie DK. Endothelin-1 modulates intrahepatic resistance in a rat model of noncirrhotic portal hypertension. Hepatology 1999;30:401-7. Cahill PA, Hou M, Hendrickson R, Wang Y, Zhang S, Redmond EM, et al. Increased expression of endothelin receptors in the vasculature of portal hypertensive rats: Role in splanchnic hemodynamics. Hepatology 1998;28:396-403. Ohta M, Pai R, Kawanaka H, Ma T, Sugimachi K, Sarfeh IJ, et al. Expression of endothelin-1, and endothelin A and B receptors in portal hypertensive esophagus of rats. J Physiol Pharmacol 2000;51:57-67. Ohta M, Nguyen TH, Tarnawski AS, Pai R, Kratzberg YP, Sugimachi K, et al. Overexpression of endothelin-1 mRNA and protein in portal hypertensive gastric mucosa of rats: A key to increased susceptibility to damage? Surgery 1997; 122:936-42. Migoh S, Hashizume M, Tsugawa K, Tanoue K, Sugimachi K. Role of endothelin-1 in congestive gastropathy in portal hypertensive rats. J Gastroenterol Hepatol 2000;15:142-7. Zhang M, Luo B, Chen SJ, Abrams GA, Fallon MB. Endothelin-1 stimulation of endothelial nitric oxide synthase in the pathogenesis of hepatopulmonary syndrome. Am J Physiol 1999;277:G944-52. Mitaka C, Hirata Y, Ichikawa K, Yokoyama K, Emori T, Kanno K, et al. Effects of TNF-α on hemodynamic changes and circulating endothelium-derived vasoactive factors in dogs. Am J Physiol 1994;267:H1530-6. Marsden PA, Brenner BM. Transcriptional regulation of the
Surgery January 2002
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
endothelin-1 gene by TNF-α. Am J Physiol 1992;262:C85461. Tanoue K, Tarnawski AS, Kishihara F, Ohta M, Hashizume M, Sugimachi K, et al. Effect of teprenone on portal hypertensive gastric mucosa. Digestion 1996;57:35-40. Ninomiya K, Kitano S, Yoshida T, Bandoh T, Baatar D, Tsuboi S. Impaired adaptive cytoprotection to ethanolinduced damage in gastric mucosa of portal hypertensive rats. Dig Dis Sci 1999;44:1254-60. Kawanaka H, Tomikawa M, Jones MK, Pai R, Szabo IL, Sugimachi K, et al. Reduced expression of VEGF and its receptors in portal hypertensive gastric mucosa following alcohol injury—a key to impaired angiogenesis? Gastroenterology 2000;118:A1049. Genesca J, Gonzalez A, Catalan R, Segura R, Martinez M, Esteban R, et al. Adrenomedullin, a vasodilator peptide implicated in hemodynamic alterations of liver cirrhosis. Relationship to nitric oxide. Dig Dis Sci 1999;44:372-6. Tomikawa M, Wang H, Jones MK, Sugimachi K, Sarfeh IJ, Tarnawski AS. Reduced adrenomedullin expression in gastric mucosa of portal hypertensive rats after ethanolinduced injury. Ann Surg 1999;230:38-44. Kawanaka H, Tomikawa M, Jones MK, Pai R, Szabo IL, Sugimachi K, et al. Portal hypertensive gastric mucosa has reduced activation of MAP kinase (ERK2) in response to alcohol injury: A key to impaired healing? FASEB J 2001;15:574-6. Garcia-Pagán JC, Fernández M, Bernadich C, Pizcueta P, Piqué JM, Bosch J, et al. Effects of continued NO inhibition on portal hypertensive syndrome after portal vein stenosis in rat. Am J Physiol 1994;267:G984-90. Fernández M, Garcia-Pagán JC, Casadevall M, Mourelle MI, Piqué JM, Bosch J, et al. Acute and chronic cyclooxygenase blockade in portal-hypertensive rats: Influence on nitric oxide biosynthesis. Gastroenterology 1996;110:1529-35. Arakawa T, Sato H, Fukuda T, Nakamura H, Kobayashi K. Endogenous prostaglandin E2 in gastric mucosa of patients with alcoholic cirrhosis and portal hypertension. Gastroenterology 1987;93:135-40. Ohta M, Kishihara F, Hashizume M, Kawanaka H, Tomikawa M, Higashi H, et al. Increased prostacyclin content in gastric mucosa of cirrhotic patients with portal hypertensive gastropathy. Prostag Leukotr Ess 1995;53:41-5. Hou MC, Cahill PA, Zhang S, Wang YN, Hendrickson RJ, Redmond EM, et al. Enhanced cyclooxygenase-1 expression within the superior mesenteric artery of portal hypertensive rats: Role in the hyperdynamic circulation. Hepatology 1998;27:20-7. Tsugawa K, Hashizume M, Migou S, Kishihara F, Kawanaka H, Tomikawa M, et al. A selective cyclo-oxygenase-2 inhibitor, NS-398, may improve portal hypertension without inducing gastric mucosal injury. J Gastroenterol Hepatol 1999;14:642-51.
Portal hypertensive gastropathy (PHG) is recognized as a clinical entity in portal hypertension, but the pathogenesis of PHG is still unclear. Therefore, we reviewed the current state of knowledge concerning the portal hypertensive gastric mucosa and hypothesized the pathogenesis of PHG. Elevated portal pressure can induce changes of local hemodynamics, thus causing congestion in the upper stomach and gastric tissue damage. These changes may then activate cytokines and growth factors, such as tumor necrosis factor α, which are substances that activate endothelial constitutive nitric oxide synthase and endothelin 1 in the portal hypertensive gastric mucosa. Overexpressed nitric oxide synthase produces an excess of nitric oxide, which induces hyperdynamic circulation and peroxynitrite overproduction. The overproduction of peroxynitrite, together with endothelin overproduction may cause an increased susceptibility of gastric mucosa to damage. When combined with the characteristics of impaired mucosal defense and healing, these factors may together produce PHG in patients with portal hypertension. (Surgery 2002;131:S165-70.) From the Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
ESOPHAGEAL VARICES HAVE LONG been considered the major cause of upper gastrointestinal hemorrhage in patients with portal hypertension. However, gastric mucosal lesions are another frequent cause of upper gastrointestinal bleeding in these patients, accounting for 20% to 40% of all bleeding.1 Initially, mucosal bleeding is caused by erosive gastritis, namely inflammation of the gastric mucosa.2 McCormack et al3 demonstrated that the gastric mucosa, observed as hemorrhagic gastritis endoscopically in portal hypertensive patients, showed mucosal and submucosal ectasia but not necessarily any inflammatory changes. Since the term congestive gastropathy3 was introduced, these mucosal lesions were termed congestive gastropathy or portal hypertensive gastropathy (PHG).4 Now PHG is recognized as a clinical entity in portal hypertension. The macroscopic appearance of PHG is endoscopically characterized by either mosaic pattern or snake-skin pattern.4,5 In addition, PHG can be subdivided into 2 grades, mild and severe.6 Reprint requests: Masayuki Ohta, MD, Department of Surgery, Matsuyama Red Cross Hospital, 1 Bunkyo-cho, Matsuyama 7908524, Japan. Copyright © 2002 by Mosby, Inc. 0039-6060/2002/$35.00 + 0 11/0/119499 doi:10.1067/msy.2002.119499
Portal hypertensive gastric mucosa in animal models has an increased susceptibility to damage due to noxious factors such as alcohol and aspirin.7 Investigating the mechanism of this increased susceptibility in experimental studies has been thought to be important for elucidating the pathogenesis of PHG. Many studies have reported the hemodynamic data of PHG; however, such findings remain controversial. In addition, several articles have also suggested that vasoactive factors are related to the pathogenesis of portal hypertensive gastric mucosa, but this aspect is not confirmed. This article reviews the current state of knowledge concerning the hemodynamics, relative vasoactive factors, and mucosal defense system of portal hypertensive stomach; the pathogenesis of PHG is also hypothesized. HEMODYNAMICS AND PHG Portal hemodynamics. Since decompression shunt surgery is effective for the treatment of PHG,2 the elevated portal venous pressure must play an important role in the pathogenesis of the PHG. McCormack et al3 measured the wedged hepatic venous pressure in 18 portal hypertensive patients, and there was no significant difference in the pressure between the patients with and without PHG (17.3 ± 1.6 vs 16.0 ± 1.6 mm Hg). Two other SURGERY S165
S166 Ohta et al
Surgery January 2002
Fig 1. Hypothesized pathogenesis of PHG.
studies demonstrated no significant difference in the hepatic venous pressure gradient (HVPG) between portal hypertensive patients with and without PHG.5,8 The direct portal venous pressure also showed no substantial difference between the 2 groups.9 Iwao et al6 subdivided PHG into mild and severe, and thus demonstrated significant differences in HVPG between portal hypertensive patients with severe PHG and without PHG but not between those with severe PHG and with mild. Therefore, an elevated portal pressure may be an important but not a critical factor in the pathogenesis of PHG. Endoscopic injection sclerotherapy has been reported to result in a deterioration of PHG in several articles.10 PHG significantly worsens 6 to 9 months after the eradication of varices by sclerotherapy; however, a gradual improvement in PHG normally follows.10 Recurrent small veins requiring additional sclerotherapy appear more frequently in patients with PHG.10 When portal hypertensive patients with varices and PHG are treated by sclerotherapy, PHG has a high likelihood of bleeding. Nakayama11 et al demonstrated that a gastrorenal shunt may play a protective role in the development of PHG after sclerotherapy. These findings show that the development of PHG is related more to the local hemodynamics, namely, upper gastric congestion, than to the portal hemodynamics. We9 have described the portal-variceal pressure gradient (portal venous pressure minus esophageal variceal pressure) to be an index of upper gastric congestion. Although the portal venous pressure
and esophageal variceal pressure did not differ in cirrhotic patients with or without PHG before the respective treatments, the portal-variceal pressure gradient significantly increased in the PHG-positive group more than in the PHG-negative group. In two studies9,12 of Doppler ultrasonography, the portal venous blood flow is reported unchanged in cirrhotic patients with PHG compared with those without PHG. Iwao et al12 showed that the portal blood flow was similar in cirrhotic patients and normal controls. Therefore the portal blood flow is not considered to play an important role in the development of PHG. Gastric mucosal blood flow. Many experimental studies13,14 with animal models of portal hypertension have shown a consistent increase in the total and mucosal gastric blood flow using different techniques. However, several studies that have measured the gastric mucosal blood flow in cirrhotic patients with PHG by means of laser-Doppler flowmetry, using the same instruments, have reported conflicting results. Some studies show an increased gastric mucosal blood flow in PHG,9 while others show a decreased mucosal blood flow in PHG.15 It thus remains highly controversial as to whether the condition of PHG is active or passive congestion. Although various biases exist such as chronic anemic conditions in patients and some difficulty for the contact techniques of endoscopic probe in laser-Doppler flowmetry, it seems that these conflicting results may be caused mainly by the characteristic angioarchitecture of the gastric mucosa in portal hypertension. Hashizume et al16 demonstrat-
Surgery Volume 131, Number 1
ed dilated microvessels in the mucosa and submucosa in cirrhotic patients in a morphologic study. A recent histologic study investigated the diameter and thickness of the wall of mucosal capillaries in the stomachs of cirrhotic patients and healthy volunteers.17 Although the diameter of the fundal capillaries did not differ among cirrhotic patients and controls, the wall thickness of the capillaries in cirrhotic patients was significantly greater than that in the controls. In the experimental studies,18 gastric mucosal microvessels are also dilated in portal hypertension. However, electron microscopic studies demonstrate that endothelial cells of the gastric mucosal capillaries in the surface layer are enlarged and the capillary lumina are narrowed in portal hypertensive models.18,19 A report that oxygenation of the gastric mucosal surface is reduced in the animal models of portal hypertension20 probably supports the belief that a narrowed capillary lumina exists in the portal hypertensive gastric mucosa. We therefore assume that the superficial mucosal blood flow decreases while the total mucosal flow increases in portal hypertensive gastric mucosa. The controversial data on laserDoppler flowmetry in PHG regarding cirrhotic patients may thus be due to these small variations. VASOACTIVE FACTORS AND PORTAL HYPERTENSIVE GASTRIC MUCOSA Nitric oxide. Nitric oxide (NO), a potent vasodilator, is generated from the terminal guanidio-nitrogen atoms of L-arginine by an enzyme, NO synthase (NOS).21 Distinct cDNAs for NOS enzyme have been isolated for an inducible NOS and 2 constitutive NOSs, one synthesized by endothelial cells (endothelial constitutive NOS, ecNOS) and the other by neuronal cells.21 In the gastric mucosa, NO plays a major role in mucosal defense by modulating the mucosal circulation.22 The NO protects the gastric mucosa against injury by ethanol and endothelin-1(ET-1), whereas the inhibition of NOS increases gastric mucosal injury.23 However, excessive NO production has a cytotoxic potential, thus the increasing gastric mucosal injury.23,24 In portal hypertension, NO is overproduced. The plasma levels of the NO metabolite and cGMP increase in portal hypertensive models.25 The NOS mRNA expression and NOS activity are enhanced in portal hypertensive vessels.25,26 Overproduced NO mediates the systemic hyperdynamic circulation and development of collaterals.27 This increased activity of NOS is most likely caused by the activation of constitutive NOS isoform because the inhibition of inducible NOS activity by dexamethasone does not affect the
Ohta et al S167
hyperdynamic circulation of portal hypertension.28 However, recent reports have demonstrated that a cirrhotic liver is impaired in the intrahepatic vasodilatory response to NO agonist and has a reduced production of NO.29 In the portal hypertensive gastric mucosa of the animals, the ecNOS mRNA and protein levels and constitutive NOS enzyme activity all increase more than the controls.14 However, the inducible NOS enzyme activity remains unchanged. The NOS activity and NO metabolites in the gastric mucosa of PHG also increase in cirrhotic patients.30 Overproduced NO increases the gastric mucosal blood flow in the portal hypertensive models.14,31 Gastric mucosal necrosis induced by ethanol in portal hypertension increases after the administration of high-dose N-nitro-L-arginine methyl ester (NOS inhibitor) but decreases after low-dose administration.14 Excessive NO production by an overexpression of ecNOS plays an important role in the increased susceptibility to damage. An excess amount of NO can produce peroxynitrite, which initiates membrane lipid peroxidation and thus results in cell injury.32 Tarnawski33 et al demonstrated the overproduction of peroxynitrite and enhanced lipid peroxidation in the portal hypertensive gastric mucosa. As a result, the overproduction of NO and peroxynitrite may be the underlying mechanism for the increased susceptibility of portal hypertensive gastric mucosa to damage, which is probably related to the pathogenesis of PHG. Tumor necrosis factor α (TNF-α), which is a multifunctional cytokine, is also overproduced in the portal hypertensive gastric mucosa.34 This cytokine can increase vascular permeability and cause both structural and metabolic changes in vascular endothelial cells.35 We demonstrated that TNF-α might regulate ecNOS expression in the portal hypertensive stomach, since anti-TNF-α treatment reversed an overexpression of ecNOS mRNA and protein and its enzyme activity.34 Endothelin. ET-1 is a 21-amino acid peptide initially noted for its powerful vasoconstrictor properties and also has been recently recognized to have pleiotropic biologic effects, depending on its receptors.36 Specific endothelin receptors are divided into at least 2 types: endothelin A receptor (ETAR) and endothelin B receptor (ETBR). The ETAR is localized in vascular smooth muscle and mediates vasoconstriction.36 The ETBR is predominantly expressed in endothelial cells and mediates the production of NO and prostacyclin.36 The submucosal injection of ET-1 induces gastric ulceration by decreasing the gastric mucosal blood flow, ischemia, and increased secretion of acid.37 ET-1
S168 Ohta et al
also enhances microvascular permeability of the stomach through the activation of ETAR.38 In cirrhotic patients, the plasma level of ET-1 has been shown to increase.39 Furthermore, ET-1 plays an important role in increased intrahepatic vascular resistance in portal hypertensive animals.40 The expression of ETAR and ETBR mRNAs increases in the vasculature of portal hypertension.41 The enhanced ETBR expression may be related to the maintenance of elevated portal pressure in the animals.41 The expression of ET-1, ETAR, and ETBR mRNAs is also enhanced in the portal hypertensive esophagus.42 In portal hypertensive gastric mucosa of the animals, ET-1 mRNA and protein levels increase in comparison with the controls.43 Although ETAR antagonist does not affect the gastric mucosal blood flow, it significantly reduces the gastric mucosal injury induced by ethanol in the portal hypertensive animals.43 In addition, ETAR/ETBR antagonist reverses hyperpermeability of portal hypertensive gastric mucosa.44 As a result, the overproduction of ET-1 in the portal hypertensive stomach can impair mucosal microcirculation through the ETAR and may thus also be related to the pathogenesis of PHG. Exogenous ET-1 increases the ecNOS mRNA expression and NO production through the ETBR, and ET-1 enhances pulmonary ecNOS level in the portal hypertensive models.45 The overproduction of ET-1 may modulate ecNOS regulation in the portal hypertensive gastric mucosa through the ETBR. Exogenous TNF-α elevates the plasma levels of ET-1,46 and TNF-α regulates the expression of the ET-1 gene in cultured endothelial cells.47 Therefore, TNF-α may activate not only the ecNOS gene but also the ET-1 gene in the portal hypertensive stomach. MUCOSAL DEFENSE AND HEALING OF THE PORTAL HYPERTENSIVE STOMACH Although vasoactive factors may be related to the pathogenesis of PHG, an impaired mucosal defensive system also plays an important role in the increased susceptibility of portal hypertensive gastric mucosa to damage. The gastric mucosal gel layer thickness significantly decreases in portal hypertensive animals.20 The mucosal gel layer thickness correlates positively with the surface epithelial cell pH.20 The contents of mucosal glycoprotein, hexosamine, in the portal hypertensive gastric mucosa have been shown to decrease in clinical and experimental studies.48 The oxygenation of the gastric mucosal surface decreases in the portal hypertensive gastric mucosa, whereas the
Surgery January 2002
total gastric mucosal blood flow increases in portal hypertension.20 These blunted mucosal defenses may cause an impaired adaptive cytoprotection against damage in the gastric mucosa of portal hypertensive rats.49 In the portal hypertensive gastric mucosa, angiogenesis is impaired after ethanol-induced injury.18 Vascular endothelial growth factor (VEGF) is an endothelial-specific angiogenetic factor. The elevation in mRNA expression of VEGF and its receptors and the phosphorylation of the receptors after ethanol-induced injury in the portal hypertensive gastric mucosa are significantly reduced compared with the sham-operated controls.50 Adrenomedullin, a potent vasodilatory peptide, is implicated in the hemodynamics in cirrhotic patients.51 The expression of adrenomedullin mRNA and protein in the portal hypertensive gastric mucosa after ethanolinduced injury significantly decreases in comparison with the controls.52 Extracellular signal-regulated kinase 2 (ERK2) is included in the mitogen-activated protein kinase cascade. The phosphorylation and activity of ERK2 do not increase in the portal hypertensive gastric mucosa following ethanol injury vs baseline values, whereas they significantly increase in the controls.53 These findings may help to explain the mechanism for impaired angiogenesis and healing after gastric mucosal injury due to portal hypertension. HYPOTHESIS OF PATHOGENESIS OF PHG Regarding the pathogenesis of PHG, in portal hypertension, elevated portal pressure can induce changes in the local hemodynamics, thus causing congestion in the upper stomach and gastric tissue damage, thereby activating cytokines and growth factors, such as TNF-α (Fig 1). These substances activate ecNOS and probably also ET-1. An overexpressed ecNOS produces an excessive amount of NO, which may induce systemic hyperdynamic circulation, increased gastric mucosal blood flow, and peroxynitrite overproduction. The overproduction of peroxynitrite, together with ET-1 overproduction, thus causes an increased susceptibility of gastric mucosa to undergo damage. When combined with the characteristics of an impaired mucosal defense and healing, these factors may together produce PHG in portal hypertension. An overproduction of NO and ET-1 are probably related to the pathogenesis of PHG. However, the long-term administration of NO synthase inhibitor can normalize the gastric blood flow on the 4-day but not on the 8-day regimen.54 The long-term inhibition of NO may activate other vasodilatory factors, such as prostaglandins, which may regulate
Ohta et al S169
Surgery Volume 131, Number 1 gastric mucosal blood flow.31 The interaction of NO and prostaglandin in the hemodynamics of portal hypertension has been suggested.55 However, the results regarding contents of endogenous prostaglandins including the gastric mucosa in cirrhotic patients remain controversial.56,57 It also remains controversial whether cyclooxygenase-2 plays an important role in the hemodynamics of portal hypertension.58,59 Further study of the gastric hemodynamics and prostaglandins under the longterm inhibition of NO synthase is thus called for.
15.
16. 17.
18.
REFERENCES 1. Terés J, Bordas JM, Bru C, Diaz F, Brugeura M, Rodés J. Upper gastrointestinal bleeding in cirrhosis: Clinical and endoscopic correlations. Gut 1976;17:37-40. 2. Sarfeh IJ, Juler GL, Stemmer EA, Mason GR. Results of surgical management of hemorrhagic gastritis in patients with gastroesophageal varices. Surg Gynecol Obstet 1982; 155:167-70. 3. McCormack TT, Sims J, Eyre-Brook I. Gastric lesions in portal hypertension; inflammatory or congestive? Gut 1985; 26:1226-32. 4. Kotzampassi K, Eleftheriadis E, Aleras H. The ‘mosaic-like’ pattern of portal hypertensive gastric mucosa after variceal eradication by sclerotherapy. J Gastroenterol Hepatol 1990;5:659-63. 5. Lin WJ, Lee FY, Lin HC, Tsai YT, Lee SD, Lai KH, et al. Snake skin pattern gastropathy in cirrhotic patients. J Gastroenterol Hepatol 1991;6:145-9. 6. Iwao T, Toyonaga A, Sumino M, Takagi K, Oho K, Nishizono M, et al. Portal hypertensive gastropathy in patients with cirrhosis. Gastroenterology 1992;102:2060-5. 7. Sarfeh IJ, Tarnawski A, Hajduczek A, Stachura J, Bui HX, Krause WJ. The portal hypertensive gastric mucosa: histologic, ultrastructural, and functional analysis after aspirin induced damage. Surgery 1988;104:79-85. 8. Quintero E, Pique JM, Bombi JA, Bordas JM, Sentis J, Elena M, et al. Gastric mucosal vascular ectasias causing bleeding in cirrhosis. A distinct entity associated with hypergastrinemia and low serum levels of pepsinogen I. Gastroenterology 1987;93:1054-61. 9. Ohta M, Hashizume M, Higashi H, Ueno K, Tomikawa M, Kishihara F, et al. Portal and gastric hemodynamics in cirrhotic patients with portal-hypertensive gastropathy. Hepatology 1994;20:1432-6. 10. Tanoue K, Hashizume M, Wada H, Ohta M, Kitano S, Sugimachi K. Effects of endoscopic injection sclerotherapy on portal hypertensive gastropathy: A prospective study. Gastrointest Endosc 1992;38:582-5. 11. Nakayama M, Iwao T, Oho K, Toyonaga A, Tanikawa K. Role of extravariceal collateral channels in the development of portal-hypertensive gastropathy before and after sclerotherapy. J Gastroenterol 1998;33:142-6. 12. Iwao T, Toyonaga A, Ikegami M, Sumino M, Oho K, Shigemori H, et al. Portal vein hemodynamics in cirrhotic patients with portal hypertensive gastropathy: An echoDoppler study. Hepatogastroenterology 1994;41:230-4. 13. Kitano S, Koyanagi N, Sugimachi K, Kobayashi M, Inokuchi K. Mucosal blood flow and modified vascular responses to norepinephrine in the stomach of rats with liver cirrhosis. Eur Surg Res 1982;14:221-30. 14. Ohta M, Tanoue K, Tarnawski AS, Pai R, Itani R, Sander FC, et al. Overexpressed nitric oxide synthase in portal-hyper-
19.
20.
21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
tensive stomach of rat: A key to increased susceptibility to damage? Gastroenterology 1997;112:1920-30. Iwao T, Toyonaga A, Ikegami M, Oho K, Sumino M, Harada H, et al. Reduced gastric mucosal blood flow in patients with portal-hypertensive gastropathy. Hepatology 1993; 18:36-40. Hashizume M, Tanaka K, Inokuchi K. Morphology of gastric microcirculation in cirrhosis. Hepatology 1983;3:1008-12. Misra V, Misra SP, Dwivedi M. Thickened gastric mucosal capillary wall: A histological marker for portal hypertension. Pathology 1998;30:10-3. Ichikawa Y, Tarnawski A, Sarfeh IJ, Ishikawa T, Shimada H. Distorted microangioarchitecture and impaired angiogenesis in gastric mucosa of portal hypertensive rats. Gastroenterology 1994;106:702-8. Tarnawski AS, Sarfeh IJ, Stachura J, Hajduczek A, Bui HX, Dabros W, et al. Microvascular abnormalities of the portal hypertensive gastric mucosa. Hepatology 1988;8:1488-94. Tomikawa M, Akiba Y, Kaunitz JD, Kawanaka H, Sugimachi K, Sarfeh IJ, et al. New insights into impairment of mucosal defense in portal hypertensive gastric mucosa. J Gastrointest Surg 2000;4:458-63. Sessa WC. The nitric oxide synthase family of proteins. J Vasc Res 1994;31:131-43. Whittle BJR. Neuronal and endothelium-derived mediators in the modulation of the gastric microcirculation: Integrity in the balance. Br J Pharmacol 1993;110:3-17. Lopez-Belmonte J, Whittle BJR, Moncada S. The actions of nitric oxide donors in the prevention or induction of injury to the rat gastric mucosa. Br J Pharmacol 1993;108:73-8. Ferraz JGP, Tigley A, Wallace JL. Paradoxical effects of Larginine on gastric mucosal integrity. Eur J Pharmacol 1994;260:107-11. Cahill PA, Foster C, Redmond EM, Gingalewski C, Wu Y, Sitzmann JV. Enhanced nitric oxide synthase activity in portal hypertensive rabbits. Hepatology 1995;22:598-606. Morales-Ruiz M, Jiménez W, Pérez-Sala D, Ros J, Leivas A, Lamas S, et al. Increased nitric oxide synthase expression in arterial vessels of cirrhotic rats with ascites. Hepatology 1996;24:1481-6. Lee FY, Colombato LA, Albillos A, Groszmann RJ. Administration of N-nitro-L-arginine ameliorates portal-systemic shunting in portal hypertensive rats. Gastroenterology 1993;105:1464-70. Fernández M, Garcia-Pagán JC, Casadevall M, Bernadich C, Piera C, Whittle BJR, et al. Evidence against a role for inducible nitric oxide synthase in the hyperdynamic circulation of portal hypertensive rats. Gastroenterology 1995;108:1487-95. Rockey DC, Chung JJ. Reduced nitric oxide production by endothelial cells in cirrhotic rat liver: Endothelial dysfunction in portal hypertension. Gastroenterology 1998; 114: 344-51. El-Newihi HM, Kanji VK, Mihas AA. Activity of gastric mucosal nitric oxide synthase in portal hypertensive gastropathy. Am J Gastroenterol 1996;91:535-8. Casadavall M, Panés J, Piqué JM, Marroni N, Bosch J, Whittle BJR. Involvement of nitric oxide and prostaglandins in gastric mucosal hyperemia of portal-hypertensive anesthetized rats. Hepatology 1993;18:628-34. Radi R, Beckman JS, Bush KM, Freemen BA. Peroxynitrite induced membrane lipid peroxydation: The cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 1991;288:481-7. Tarnawski AS, Tomikawa M, Kawanaka H, Sugimachi K, Sarfeh IJ. Increased oxygen free radicals, nitrotyrosine and
S170 Ohta et al
34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
enhanced lipid peroxydation in portal hypertensive gastric mucosa: A key to increased susceptibility to injury. Gastroenterology 1999;116:A1282. Ohta M, Tarnawski AS, Itani R, Pai R, Tomikawa M, Sugimachi K, et al. Tumor necrosis factor α regulates nitric oxide synthase expression in portal hypertensive gastric mucosa of rats. Hepatology 1998;27:906-13. Stephens KE, Ishizaka A, Larryk JW, Raffin TA. Tumor necrosis factor causes increased pulmonary permeability and edema. Comparison to septic acute lung injury. Am Rev Respir Dis 1988;137:1364-70. Levin ER. Mechanisms of disease: Endothelin. N Engl J Med 1995;333:356-63. Lazaratos S, Kashimura H, Nakahara A, Fukutomi H, Osuga T, Urushidani T, et al. Gastric ulcer induced by submucosal injection of ET-1: Role of potent vasoconstriction and intraluminal acid. Am J Physiol 1993;265:G491-8. Filep JG, Sirois MG, Foldes-Filep E, Rousseau A, Plante GE, Fournier A, et al. Enhancement by endothelin-1 of microvascular permeability via the activation of ETA receptors. Br J Pharmacol 1993;109:880-6. Isobe H, Satoh M, Sasaki H, Nawata H. Increased plasma endothelin-1 levels in patients with cirrhosis and esophageal varices. J Clin Gastroenterol 1993;17:227-30. Kamath PS, Tyce GM, Miller VM, Edwards BS, Rorie DK. Endothelin-1 modulates intrahepatic resistance in a rat model of noncirrhotic portal hypertension. Hepatology 1999;30:401-7. Cahill PA, Hou M, Hendrickson R, Wang Y, Zhang S, Redmond EM, et al. Increased expression of endothelin receptors in the vasculature of portal hypertensive rats: Role in splanchnic hemodynamics. Hepatology 1998;28:396-403. Ohta M, Pai R, Kawanaka H, Ma T, Sugimachi K, Sarfeh IJ, et al. Expression of endothelin-1, and endothelin A and B receptors in portal hypertensive esophagus of rats. J Physiol Pharmacol 2000;51:57-67. Ohta M, Nguyen TH, Tarnawski AS, Pai R, Kratzberg YP, Sugimachi K, et al. Overexpression of endothelin-1 mRNA and protein in portal hypertensive gastric mucosa of rats: A key to increased susceptibility to damage? Surgery 1997; 122:936-42. Migoh S, Hashizume M, Tsugawa K, Tanoue K, Sugimachi K. Role of endothelin-1 in congestive gastropathy in portal hypertensive rats. J Gastroenterol Hepatol 2000;15:142-7. Zhang M, Luo B, Chen SJ, Abrams GA, Fallon MB. Endothelin-1 stimulation of endothelial nitric oxide synthase in the pathogenesis of hepatopulmonary syndrome. Am J Physiol 1999;277:G944-52. Mitaka C, Hirata Y, Ichikawa K, Yokoyama K, Emori T, Kanno K, et al. Effects of TNF-α on hemodynamic changes and circulating endothelium-derived vasoactive factors in dogs. Am J Physiol 1994;267:H1530-6. Marsden PA, Brenner BM. Transcriptional regulation of the
Surgery January 2002
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
endothelin-1 gene by TNF-α. Am J Physiol 1992;262:C85461. Tanoue K, Tarnawski AS, Kishihara F, Ohta M, Hashizume M, Sugimachi K, et al. Effect of teprenone on portal hypertensive gastric mucosa. Digestion 1996;57:35-40. Ninomiya K, Kitano S, Yoshida T, Bandoh T, Baatar D, Tsuboi S. Impaired adaptive cytoprotection to ethanolinduced damage in gastric mucosa of portal hypertensive rats. Dig Dis Sci 1999;44:1254-60. Kawanaka H, Tomikawa M, Jones MK, Pai R, Szabo IL, Sugimachi K, et al. Reduced expression of VEGF and its receptors in portal hypertensive gastric mucosa following alcohol injury—a key to impaired angiogenesis? Gastroenterology 2000;118:A1049. Genesca J, Gonzalez A, Catalan R, Segura R, Martinez M, Esteban R, et al. Adrenomedullin, a vasodilator peptide implicated in hemodynamic alterations of liver cirrhosis. Relationship to nitric oxide. Dig Dis Sci 1999;44:372-6. Tomikawa M, Wang H, Jones MK, Sugimachi K, Sarfeh IJ, Tarnawski AS. Reduced adrenomedullin expression in gastric mucosa of portal hypertensive rats after ethanolinduced injury. Ann Surg 1999;230:38-44. Kawanaka H, Tomikawa M, Jones MK, Pai R, Szabo IL, Sugimachi K, et al. Portal hypertensive gastric mucosa has reduced activation of MAP kinase (ERK2) in response to alcohol injury: A key to impaired healing? FASEB J 2001;15:574-6. Garcia-Pagán JC, Fernández M, Bernadich C, Pizcueta P, Piqué JM, Bosch J, et al. Effects of continued NO inhibition on portal hypertensive syndrome after portal vein stenosis in rat. Am J Physiol 1994;267:G984-90. Fernández M, Garcia-Pagán JC, Casadevall M, Mourelle MI, Piqué JM, Bosch J, et al. Acute and chronic cyclooxygenase blockade in portal-hypertensive rats: Influence on nitric oxide biosynthesis. Gastroenterology 1996;110:1529-35. Arakawa T, Sato H, Fukuda T, Nakamura H, Kobayashi K. Endogenous prostaglandin E2 in gastric mucosa of patients with alcoholic cirrhosis and portal hypertension. Gastroenterology 1987;93:135-40. Ohta M, Kishihara F, Hashizume M, Kawanaka H, Tomikawa M, Higashi H, et al. Increased prostacyclin content in gastric mucosa of cirrhotic patients with portal hypertensive gastropathy. Prostag Leukotr Ess 1995;53:41-5. Hou MC, Cahill PA, Zhang S, Wang YN, Hendrickson RJ, Redmond EM, et al. Enhanced cyclooxygenase-1 expression within the superior mesenteric artery of portal hypertensive rats: Role in the hyperdynamic circulation. Hepatology 1998;27:20-7. Tsugawa K, Hashizume M, Migou S, Kishihara F, Kawanaka H, Tomikawa M, et al. A selective cyclo-oxygenase-2 inhibitor, NS-398, may improve portal hypertension without inducing gastric mucosal injury. J Gastroenterol Hepatol 1999;14:642-51.