SEARCH

SEARCH BY CITATION

Keywords:

  • anaemia;
  • β-blockers;
  • bleeding;
  • cirrhosis;
  • gastritis;
  • gastrointestinal hemorrhage;
  • portal hypertension

Abstract

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Portal hypertensive gastropathy (PHG) occurs as a complication of cirrhotic or non-cirrhotic portal hypertension. Although the pathogenesis of PHG is not completely understood, evidence suggests that the key factor for the development of PHG is portal hypertension. PHG is clinically important because it may cause acute (and even) massive or insidious, blood loss. The diagnosis of PHG is (only) made endoscopically; it is most often characterized by an abnormality of the gastric mucosa described as a mosaic-like pattern resembling ‘snake-skin’, with or without red spots and the endoscopic pattern is key its diagnosis. Unfortunately, standardization of the endoscopic diagnostic criteria for PHG is poor and consensus is generally lacking, resulting in a wide range of reported prevalence. Pharmacological therapies, presumably reducing portal pressure and gastric blood flow, have been used to treat acute bleeding; propanolol, a non-selective β-blocker (24–480 mg/day), has been used most frequently. Endoscopic treatment for PHG bleeding plays a small, if any, role in the treatment of PHG. TIPS and shunt surgery have not been extensively analysed as a treatment for acute or chronic PHG bleeding, but they appear to lessen the severity of PHG. Secondary prophylaxis of PHG bleeding with non-selective β-blockers is recommended. There is not enough evidence to support the use of β-blockers in primary prophylaxis of PHG bleeding, even in cases of severe PHG (however, non-selective β-blockers are recommended if varices are present). Further studies are needed to clarify the role of PHG in suspected chronic gastrointestinal bleeding.


Introduction and definition

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Portal hypertensive gastropathy (PHG) occurs as a complication of cirrhotic or non-cirrhotic portal hypertension. PHG is clinically important because it may cause acute (and even) massive, or insidious, blood loss. It is characterized by an endoscopic abnormality of the gastric mucosa that is classically described as a mosaic-like pattern that resembles the skin of a snake, with or without red spots (1). In the past, these lesions were described as inflammatory or erosive, and were thought to be a form of gastritis in patients with cirrhosis and portal hypertension. However, in 1985, it was demonstrated that the pathological change is characterized by vascular ectasia rather than mucosal inflammation (2) and further suggested that these mucosal lesions should be named congestive gastropathy rather than gastritis. Now, the preferred term is portal hypertensive gastropathy. The endoscopic findings are not specific for PHG and can be seen in other diseases. In addition, similar changes can be seen in the small bowel and colonic mucosa of patients with portal hypertension and these have been named portal hypertensive enteropathy (3, 4) and portal hypertensive colopathy (5, 6) respectively.

Epidemiology

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

The overall prevalence of PHG ranges from 20 to 98% of patients with known cirrhosis, depending on author (1, 7–14). This discrepancy is perhaps related to several factors: patient selection, absence of uniform diagnostic criteria and classification, and differences in inter- and intra-observer interpretation of endoscopic lesions (2). Several studies (7–11) have shown a higher prevalence in patients with high Child–Pugh (CP) scores, patients with oesophageal varices (EVs) or with a history of EV treatment (sclerotherapy or ligation). However, these studies have been inconclusive and have generally failed to identify variables that predict for the presence of PHG. In general, most patients with PHG have a mild form of the disease. In a previous review, mild PHG was reported to be present in 60% of patients, while severe PHG was present in up to 46% of cases (1).

In one study of a large cohort of cirrhotic patients carefully studied in the Hepatitis Antiviral Long-term Treatment against Cirrhosis (HALT-C) trial, it was reported that the prevalence of PHG was 37% (12). All 1050 patients had advanced fibrosis or cirrhosis documented on liver biopsy. Mild PHG was reported in 34%, while 3% had severe PHG. In this study, non-African American race, low albumin, elevated bilirubin and low platelets were independently associated with the presence of PHG. Interestingly, EVs were identified in 40% of patients with PHG but 17% of patients without PHG. Further, patients with PHG were significantly more likely to have medium or large EV (odds ratio 6.4; 95% confidence interval 3.47–11.79) and it was more prevalent in patients with more fibrosis on liver biopsy (26% in patients with Ishak score of 3 and 51% in patients with Ishak score of 6). These data suggest that PHG could be associated with more severe portal hypertension. In another study (9), the prevalence of PHG was 21.6% in 222 patients with cirrhosis, without a history of prior bleeding who at endoscopy did not have medium or large sized EV and who did not undergo any kind of endoscopic therapy and were not taking propanolol; 90% of the cases of PHG were mild.

Even though it is not clear what factors play the most important role in the development of PHG, there is a general agreement that – based on longitudinal studies – the prevalence of PHG increases with EV obliteration (1).

Pathophysiology

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

The pathogenesis of PHG is not completely understood and is likely to be complicated. However, evidence suggests that the key factor for the development of PHG is portal hypertension. This is based on the premise that portal hypertension is necessary for PHG to develop. In addition, some small studies have shown improvement of PHG after transjugular intrahepatic portosystemic shunt (TIPS) and shunt surgery (15, 16) suggesting an association between PHG and the severity of portal hypertension. The degree of portal hypertension necessary for development of PHG is controversial, and many studies have failed to show a linear correlation with the severity of portal hypertension (10, 11, 13, 17). Interestingly, two studies (2, 18) have shown a significant difference in hepatic venous pressure gradient between patients without PHG and patients with severe PHG but not between patients without PHG and mild PHG or between patients with mild and severe PHG.

Whether primary parenchymal liver disease is required for the development of PHG is controversial. For example, patients with portal hypertension because of portal vein thrombosis alone may develop PHG. However, other factors clearly are associated with the presence and severity of PHG. These include prior treatment of EV, aetiology of portal hypertension (cirrhotic vs. non-cirrhotic), severity of primary liver disease, H. pylori infection (8, 10, 12–14, 18–23). Interestingly, a history of endoscopic treatment for EV has been shown an increase in incidence and severity of PHG (24, 25), but such changes are usually transitory and reversible (26, 27).

Abnormalities in the gastric microcirculation seem to be responsible for the congestion seen in PHG (28). However, some controversy remains as to whether this congestion is active or passive. Some data suggest that gastric mucosal blood flow might be decreased, but the total blood flow (blood flow in mucosa and submucosa of the stomach) may in fact be increased in PHG (15, 17, 24, 29) suggesting that PHG develops because of congestion caused by blockade of gastric blood drainage rather than by hyperaemia. On the other hand, this point is controversial as other studies have shown conflicting results (30). The mucosa in PHG appears to be highly susceptible to injury, as it has been shown to exhibit impaired healing and mucosal defense, predisposing it to bleeding (31, 32). The abnormalities in the regulation of the gastric microcirculation producing mucosal surface hypoxia (33, 34) and additional alterations in epithelial cell integrity, probably mediated through local factors such as overproduction of nitric oxide (NO), oxygen free radicals, endothelin-1, tumour necrosis factor α and prostaglandins (35–39) have been implicated. However, a clear understanding of the molecular basis for this phenomenon is still lacking.

Diagnosis and classification

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

The diagnosis of PHG is endoscopic. Since the initial description by McCormack et al. (2), the spectrum of lesions that have been considered to be consistent with PHG is extensive and includes the following: a classic, ‘mosaic-like pattern’, which consists of multiple erythematous areas, rectangular or diamond-shaped, outlined by a delicate white or yellowish network; red point lesions, cherry red spots, ‘scarlatina’ type rash, brown spots, and petechiae (40–43). In a classification proposed by the NIEC (New Italian Endoscopic Club for the Study and Treatment of Oesophageal Varices) (44), only two type of lesions were proposed as diagnostic for PHG: ‘mosaic-like pattern’ and red marks, the latter described as red lesions of variable diameter, flat or slightly protruding into the lumen of the stomach. The most common locations for PHG are the body and fundus of the stomach (40, 41), although clinicians often note findings throughout the stomach. While endoscopy has been considered to be the gold standard diagnostic tool, over the last several years, less invasive methods such as capsule endoscopy have been used to diagnose PHG (45), with good visualization of the mucosa. However, its accuracy as a diagnostic method has not been validated. Other imaging methods like dynamic CT scan and upper gastrointestinal (GI) series with barium have only been described in small series (46, 47).

Even though the histopathology of this entity is well described as vascular ectasia of the mucosal and submucosal veins and capillaries, biopsies are rarely done to confirm this diagnosis. This is often because of concerns about potential bleeding – although it should be emphasized that there is no literature indicating that there is a clearly increased risk of bleeding after biopsy of PHG.

Beyond diagnosis, classification of PHG remains highly controversial, with a lack of general consensus (Table 1). The NIEC (44) recommends a two-category classification system for PHG including mild and severe. Mild PHG is diagnosed when only ‘mosaic-like pattern’ (MLP) lesions are found. MLP can be mild, moderate or severe (Fig. 1). Severe PHG is diagnosed when red marks, with or without MLP, are found (Fig. 1). Other authors (48) have proposed grading PHG as mild, moderate and severe. These two classifications are comparable with regard to the limitations in specificity and sensitivity, but the two-category classification system had significantly better intra- and interobserver agreement and reproducibility (49). The NIEC two-category classification system is consistent with the initial classification proposed by McCormack (2) as mild and severe. It is also simple and separates two categories clearly according to the risk of bleeding (3.5–31% in mild PHG and 38–62% in severe PHG). Additionally, Baveno III (50) recommended a classification of mild and severe according to the last NIEC consensus.

Table 1.   Classification of portal hypertensive gastropathy
ClassificationNIEC (44)McCormack et al. (2)Tanoue et al. (48)
CategoryTwo-category systemTwo-category systemThree-category system
  1. NIEC, New Italian Endoscopic Club for the Study and Treatment of Oesophageal Varices.

Mild‘Mosaic-like pattern’ Fine speckling or ‘scarlatina’ type of rashGrade 1:
 Mild: diffusely pink areola Mild reddening
 Moderate: flat red spot in centre of pink areola Superficial reddening Congestive mucosa
 Severe: diffusely red areola ‘Snake-skin’ pattern 
ModerateNot consideredNot consideredGrade 2: Severe redness and a fine reticular pattern separating areas of raised mucosa
Severe‘Red marks’: Cherry red spots, confluent or notGrade 3: Grade 2 plus point bleeding
 Red lesions of variable diameter, flat or slightly protruding. Discrete or confluent Diffuse hemorrhage
image

Figure 1.  Mild and severe portal hypertensive gastropathy (PHG). In (a) and (b) are shown representative images of mild PHG. This form of PHG is characterized by a ‘snake-skin’ appearance, without stigmata of bleeding or substantial erythema or oedema. In (c) and (d) are shown representative images of severe PHG. This form of PHG is characterized by extensive red lesions, which may be discrete or protruding from the mucosa. There is often evidence of irregularity in the mucosa, indicative of inflammation. Cherry red spots may be discrete or confluent. Subtle oozing may also be present [as in (d)].

Download figure to PowerPoint

Portal hypertensive gastropathy has also been proposed as a marker for the diagnosis of portal hypertension, with conflicting results. Its sensitivity, positive predictive value (PPV) and negative predictive value (NPV) vary considerably among studies, but its specificity has been reported to be above 95% (14, 43). The endoscopic finding of both PHG and EV might improve the accuracy of EV alone in the diagnosis of cirrhosis (51).

As mentioned above, PHG is an endoscopic diagnosis and unfortunately this is problematic on several levels. First, other disorders may look like PHG endoscopically and it is important to recognize that endoscopic findings described as classic for PHG are not 100% specific for this disorder. The other critical issue is that there is significant intra- and inter-observer disagreement about endoscopic findings typical of PHG such as the so-called ‘mosaic-like pattern’ (40). This makes assigning a patient with an endoscopic lesion that appears to be PHG as having portal hypertension and cirrhosis (although we recommend that in patients with an endoscopic lesion that is consistent with PHG, portal hypertension and cirrhosis should be considered).

The endoscopic differential diagnosis includes several disorders. An important diagnostic consideration is gastric antral vascular ectasia (GAVE) or watermelon stomach, which is characterized endoscopically by linear red stripes, separated by normal mucosa, giving the appearance of a watermelon, most commonly seen in the gastric antrum or proximal stomach (Table 2). GAVE can be seen in patients with cirrhosis, but also in autoimmune diseases and connective tissue diseases including atrophic gastritis, scleroderma, sclerodactily, pernicious anaemia. Histopathogically, in patients with cirrhosis, in comparison to PHG, GAVE appears to be characterized by more extensive ectasia, thrombi and spindle cell proliferation (52). Although GAVE has been associated with cirrhosis and some have proposed an aetiological link to portal hypertension, the lesion responds poorly to TIPS or β-blockers, making it less likely that GAVE is a portal hypertension-mediated lesion (16, 53).

Table 2.   Comparison between portal hypertensive gastropathy and gastric antral vascular ectasia
 PHGGAVE
  1. GAVE, gastric antral vascular ectasia; PHG, portal hypertensive gastropathy.

Predominant locationBody and fundusAntrum
Classic endoscopic appearanceMosaic like pattern or red marksLinear red stripes separated by normal mucosa
HistologyMild to moderate dilation of veins and capillaries of gastric mucosa and submucosa. No changes in vessel wallsMarked dilation of capillaries and venules in gastric mucosa and submucosa with areas of intimal thickening and thrombi
Portal hypertensionAlways presentCan be seen without portal hypertension
Associated conditionsOnly seen in conditions that cause portal hypertensionAutoimmune and connective tissue diseases (scleroderma, pernicious anemia, hypothyroidsm), liver cirrhosis
Endoscopic therapyNoYes, argon plasma coagulation preferred
Response to β-blockers and TIPSYesNo

Several other entities should be considered in the differential diagnosis of PHG; these include simple acute gastritis [caused by non-steroidal anti-inflammatory drug (NSAIDS) or H. pylori], which may have a mosaic like pattern endoscopically and which is characterized by prominent inflammatory cell infiltrate with minor vascular dilation and only affects the mucosa (42). Other uncommon diseases that lead to an endoscopic lesion resembling PHG. Such processes have been described as case reports and include polycythaemia (54), gastric purpura (55) and Osler Weber Rendu disease.

Natural history of portal hypertensive gastropathy

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Studies describing the natural history of PHG have had relatively long follow-up (a median follow up of 4 years) (7–10, 13). The end points that are appropriate in the follow up of PHG are: incidence, progression/regression, acute GI bleeding, chronic GI bleeding and survival – and these will be covered individually below.

Incidence of portal hypertensive gastropathy

In patients with cirrhosis undergoing endoscopy, typically to screen for the presence of varices, the incidence of PHG fluctuates between 30 and 45% at the end of follow up period. In one study (9) describing the cumulative incidence, PHG was identified at a level of 3% at 1 year, 10% at 2 years and 24% at 3 years. Less than 10% of new PHG cases were severe. In some studies, the presence of EV, concomitant treatment of EV (EVL and/or sclerotherapy) and more advanced disease have been associated with the development of PHG (7–10, 13, 19, 26). One study reported that in patients with cirrhosis and EVs, patients treated with sclerotherapy seemed to have an increased risk of subsequently developing PHG (13). It has also been suggested that PHG is more common in patients with gastro-oesophageal varices than in patients with EV alone (42 vs. 11%; P<0.01) (13).

The significance of finding PHG as the only sign of portal hypertension (no evidence of esophageal or gastric varices) is unclear. Current AASLD guidelines (56) recommend a 1–3-year follow-up upper endoscopy after a negative endoscopy for EV depending on presence of hepatic decompensation. Available data suggest that the finding of asymptomatic PHG alone on first endoscopy should not be an indication for more frequent endoscopic follow-up or for initiation of therapy to treat portal hypertension.

Progression/regression of portal hypertensive gastropathy

In most of cases (30–60%), PHG remains stable. Interestingly, however, PHG may also fluctuate, which suggests that it is a dynamic entity. It has been reported to progress from mild to severe in up to 30% of the cases and it regress or disappear in up to 20% of cases (7–9). EV ligation (EVL) is associated with faster progression of PHG (26, 57), but this worsening is usually transient and PHG can regress in up to 44% of patients after EVL (26). Another study found that only high CP scores at entry were associated with progression of PHG (8).

Acute gastrointestinal bleeding

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Portal hypertensive gastropathy is not the most common cause of significant upper GI bleeding in patients with portal hypertension, but bleeding is the most important complication of this disease. The incidence of acute upper GI bleeding from PHG varies widely (2–12%) (7–9). This is likely a result of the fact that it is often difficult to make a definitive diagnosis of GI bleeding in the setting of PHG, that there is inaccuracy in the classification of severe PHG (2), and that different studies have evaluated varied patient populations. Most cases (90–95%) of acute bleeding from PHG appear to occur in patients with severe PHG. In one study (58) of 1496 patients with upper GI bleeding, PHG was the cause of bleeding in 12 patients (0.8%). In patients with chronic liver disease, PHG was associated with 4% of all the cases of acute bleeding and 8% of the cases of non-variceal bleeding. The incidence of upper GI bleeding from PHG did not appear to correlate with the presence of EVs, size of EVs or prior history of EVs treatment (7).

Chronic gastrointestinal bleeding

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

It is difficult to estimate this incidence of chronic bleeding because of the lack of an appropriate definition of chronic bleeding. The incidence of suspected chronic bleeding from PHG also has been reported to range from 3 to 26% and appears to occur with equal frequency in patients with mild or severe PHG (7–9). Many studies have utilized the definition of ‘a decrease of 2 g/dl or more of haemoglobin between two time periods 6 months apart to catalogue chronic bleeding, provided the patient had not acutely bled in the meantime and was not taking NSAIDs’ (59). Others have taken that the simple presence of anaemia in a patient with cirrhosis to be evidence of chronic bleeding from PHG, while other studies have incorporated a positive fecal occult blood test (FOBT) as a diagnostic marker of chronic bleeding (60). Notwithstanding, definitions of chronic bleeding are generally poorly integrated and because they rely mostly on changes in haemoglobin, are unreliable because changes in hemoglobin are often non-specific, particularly in patients with chronic liver disease (because these patients may also have concomitant kidney disease, alcoholism, hypersplenism, bone marrow suppression, etc.). At this point, no studies have objectively quantitated chronic or occult blood loss from PHG or in patients with cirrhosis. Interestingly, studies that included patients with prior history of EV treatment (before or after the diagnosis of PHG) (7, 8, 26) had a higher incidence of suspected chronic GI bleeding.

Survival

There are no data on the impact of this entity in the overall survival in patients with portal hypertension. It is reasonable to presume that survival can be affected when a complication such as acute or chronic GI bleeding occurs. However, in studies reporting outcomes in patients with PHG (7–9), acute bleeding from PHG did not appear to represent a significant cause of mortality (<1%). This low rate of mortality is not enough to make any conclusions about the importance of PHG in survival of patients with cirrhosis.

Treatment and prevention

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Acute GI bleeding and perhaps chronic GI bleeding are the most important complications of PHG, occurring primarily in the setting of severe PHG. Only small, uncontrolled studies have evaluated the treatment of acute PHG bleeding (Table 3). A number of pharmacologic therapies have been used in an effort to treat acute bleeding, their effects being predicated on reduction of portal pressure and presumably gastric blood flow; propanolol, a non-selective β-blocker (24–480 mg/day), has been used most frequently. In one study (61), 14 patients with severe PHG and acute GI bleeding from PHG were treated with propanolol; bleeding stopped in 93% of patients within 3 days. Octreotide, a somatostatin analogue (100 μg bolus followed by an infusion of 25 μg/h for 48 h), may be effective in the treatment of acute PHG bleeding; in a randomized trial of 68 patients with of acute bleeding from PHG that compared octreotide, vasopressin and omeprazole (62), octreotide controlled bleeding in 100% of patients. Vasopressin and omeprazole alone were not as effective (they controlled bleeding in 64 and 59% of patients, respectively), but when they were used together they were 88% effective. Terlipressin, a vasopressin analog, may also be effective for treatment of acute bleeding caused by PHG and appears to have similar efficacy as octreotide (60, 63). Studies comparing β-blockers and octreotide are not available.

Table 3.   Treatment and prevention of portal hypertensive gastropathy
  1. PHG, portal hypertensive gastropathy; TIPS, transjugular intrahepatic portosystemic shunt.

Acute bleeding Octreotide
 Terlipresson or vasopressin
 Propanolol
 TIPS
Chronic bleeding Not enough evidence for recommendations. Propanolol or TIPS could be considered
Prevention of first bleeding Not enough evidence for recommendations. Propanolol could be considered in severe PHG
Prevention of recurrent bleeding Propanolol

Endoscopic treatment for PHG bleeding plays a small, if any, role in the treatment of PHG – because bleeding is usually diffuse. Argon plasma coagulation, sclerotherapy, and possibly even coagulation therapy with the heater probe may be considered with focal bleeding – but there are no robust data that have addressed endoscopic treatment.

Anti-oxidants have also been used to treat PHG. In a small experimental study in rats (39), oral administration of vitamin E led to complete reversal of susceptibility of portal hypertensive gastric mucosa to alcohol injury. The effect of vitamin E appeared to be mediated through the restoration of (normal) ERK-2 signalling, which appeared to play a pivotal role in healing after gastric mucosal injury. Thalidomide (64) and prednisolone (65) have been used to treat acute PHG bleeding, and it has been found to be successful in isolated case reports.

Transjugular intrahepatic portosystemic shunt and shunt surgery have not been extensively analyzed as a treatment for acute or chronic PHG bleeding, but they appear to lessen the severity of PHG (15, 16). They should be considered in certain specific circumstances. However, TIPS and shunt surgery are both invasive and associated with substantial morbidity and mortality, and should be considered only as a last resort, and only in consultation with an expert in the management of portal hypertension.

Several small trials have addressed the issue of primary and secondary prevention for PHG related bleeding. In one study (61), 22 patients with EV without a prior history of bleeding from PHG were randomized to 6 weeks of placebo or 6 weeks of propanolol therapy (160 mg/day). PHG, assessed endoscopically, remained unchanged in the majority of patients although patients taking propanolol improved significantly (40%) compared with placebo (40 vs. 14%, respectively). Acute GI bleeding occurred in three patients (two in the group taking placebo and one in the patients taking propanolol). Another study (60) randomized 54 patients with cirrhosis and a recent history of acute (defined as the presence of haematemesis or melena) or chronic bleeding (defined as occult blood loss with transfusion requirements of 3 or more units of blood in 3 months and/or continuous iron therapy for longer than 50% of follow-up) from severe PHG to receive propanolol or no treatment. The rate of patients free of re-bleeding was significantly higher in the propanolol group compared with the non-treatment group (65 vs. 38% respectively at 12 months and 52 vs. 7% respectively at 30 months). Acute PHG bleeding was significantly more frequent in patients in the no treatment arm, but this difference was not seen in chronic PHG bleeding. In another study (62), 77 patients with a prior history of EV bleeding were randomized to EV ligation alone (40 patients) and a combination of EV ligation and propanolol (37 patients). PHG developed less frequently in the propanolol group (31 vs. 67%).

Based on the natural history of PHG, the risk of bleeding from mild PHG is low; thus, these patients generally do not require primary prophylaxis. In patients with mild PHG and varices, propanolol should be used (because varices should be treated with β-blockers). In patients with severe PHG who do not have an indication for prophylaxis of varices, prophylaxis with a non-selective β-blocker, should be considered, although the evidence to support this approach is equivocal. Although propanolol may even cause significant regression of severe PHG (60, 61), and presumably decreases the risk of bleeding, there is no evidence that this regression will have an impact on the risk of a first episode of PHG bleeding.

In summary, in patients with acute PHG bleeding, the duration of treatment with octreotide has been poorly defined, but it seems appropriate to treat for 48–72 h. After resolution of acute bleeding (or if the patient is haemodynamically stable), propanolol should be started as tolerated for secondary prophylaxis (66). Most of these patients have concomitant EV and propanolol will also prevent bleeding episodes from EV (67). Primary prophylaxis with propanolol for PHG bleeding should be considered in patients who have mild or severe PHG who will undergo EV ligation, including the risk of progression of PHG.

In terms of management of chronic bleeding, there is little definitive data with which to make recommendations. These patients should be treated with iron supplementation, orally, or even parentally if needed (newer IV iron formulations appear to be safe and effective). One study addressed the use of propanolol to treat chronic GI bleeding (60). In this study, there was no significant difference in the proportion of patients free of chronic GI bleeding between the group that received propanolol and the group that did not receive treatment (72 vs. 55% at 12 months; 63% vs. 40% at 30 months; P=0.2 respectively). As for acute bleeding from PHG, TIPS may be considered in well-compensated patients. Additionally, in those with decompensated cirrhosis, liver transplantation should be considered as part of management. Further studies are needed to justify the use of propanolol in suspected chronic GI bleeding from PHG, even though this is a common practice in many places.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

Portal hypertensive gastropathy is a common finding in patients with cirrhosis, described as a mosaic-like pattern at the time of endoscopy, and typically identified in the fundus and body of the stomach. Even though the pathophysiology of PHG is not completely understood, available evidence suggests that portal hypertension is necessary for its development. The role of other factors like previous or ongoing EV treatment, the presence of gastric varices, the severity of liver disease, and the severity of portal hypertension remain controversial. The endoscopic diagnosis of PHG has poor inter and intra-observer agreement, but the two-category classification system (mild and severe) is preferred because of its simplicity. PHG appears to fluctuate with time, and its complications include primarily acute or (suspected) chronic bleeding, either of which are usually seen in patients with severe PHG. Acute PHG bleeding can be treated with octreotide or propanolol. Secondary prophylaxis of PHG bleeding with non-selective β-blockers is recommended. There is not enough evidence to support the use of β-blockers in primary prophylaxis of PHG bleeding, even in cases of severe PHG (however, non-selective β-blockers are recommended if varices are present). Further studies are needed to clarify the role of PHG in suspected chronic GI bleeding.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References

The authors, certify that we have no financial arrangements (e.g., consultancies, stock ownership, equity interests, patent-licensing arrangements, research support, major honoraria, etc.) with a company whose product figures prominently in this manuscript or with a company making a competing product.

References

  1. Top of page
  2. Abstract
  3. Introduction and definition
  4. Epidemiology
  5. Pathophysiology
  6. Diagnosis and classification
  7. Natural history of portal hypertensive gastropathy
  8. Acute gastrointestinal bleeding
  9. Chronic gastrointestinal bleeding
  10. Treatment and prevention
  11. Conclusions
  12. Acknowledgements
  13. References
  • 1
    Thuluvath P, Yoo H. Portal hypertensive gastropathy. Am J Gastroenterol 2002; 97: 29738.
    Direct Link:
  • 2
    McCormack T, Sims J, Eyre-Brook I, et al. Gastric lesions in portal hypertension: inflammatory gastritis or congestive gastropathy? Gut 1985; 26: 122632.
  • 3
    Menchen L, Ripoli C, Marin-Jimenez I, et al. Prevalence of portal hypertensive duodenopathy in cirrhosis: clinical and hemodynamic features. Eur J Gastroenterol Hepatol 2006; 18: 64953.
  • 4
    Higaki N, Matsui H, Imaoka H, et al. Characteristic endoscopic features of portal hypertensive enteropathy. J Gastroenterol 2008; 43: 32731.
  • 5
    Bini E, Lascarides C, Micale P, Weinshel E. Mucosal abnormalities of the colon in patients with portal hypertension: an endoscopic study. Gastrointest Endosc 2000; 52: 5116.
  • 6
    Ito K, Shiraki K, Sakai T, Yoshimura H, Nakano T. Portal hypertensive colopathy in patients with cirrhosis. World J Gastroenterol 2005; 11: 312730.
  • 7
    D'Amico G, Montanalbo L, The Liver Study Group of V et al. Cervello hospital. natural history of congestive gastropathy in cirrhosis. Gastroenterology 1990; 99: 155864.
  • 8
    Primignani M, Carpinelli L, the New Italian Endoscopic Club for the study and treatment of esophageal varices (NIEC). et al. Natural history of portal hypertensive gastropathy in patients with liver cirrhosis. Gastroenterology 2000; 119: 1817.
  • 9
    Merli M, Nicolini G, Angeloni S, et al. The natural history of portal hypertensive gastropathy in patients with liver cirrhosis and mild portal hypertension. Am J Gastroenterol 2004; 99: 195965.
    Direct Link:
  • 10
    Iwao T, Toyonaga A, Oho K, et al. Portal-hypertensive gastropathy develops less in patients with cirrhosis and fundal varices. J Hepatol 1997; 26: 123541.
  • 11
    Iwao T, Toyonaga A, Sumino M, et al. Portal hypertensive gastropathy in patients with cirrhosis. Gastroenterology 1992; 102: 20605.
  • 12
    Fontana R, Sanyal A, the HALT-C trial Group et al. Portal Hypertensive gastropathy in chronic Hepatitis C patients with bridging fibrosis and compensated cirrhosis: results from the HALT-C Trial. Am J Gastroenterol 2006; 101: 98392.
  • 13
    Sarin S, Sreenivas D, Lahoti D, Saraya A. Factors influencing development of portal hypertensive gastropathy in patients with portal hypertension. Gastroenterology 1992; 102: 9949.
  • 14
    Sarin S, Misra S, Singal A, Thorat V, Broor S. Evaluation of the incidence and significance of the “Mosaic Pattern” in patients with cirrhosis, noncirrhotic portal fibrosis, and extrahepatic obstruction. Am J Gastroenterol 1988; 83: 12359.
  • 15
    Mezawa S, Homma H, Ohta H, et al. Effect of transjugular intrahepatic portosystemic shunt formation on portal hypertensive gastropathy and gastric circulation. Am J Gastroenterol 2001; 96: 11559.
    Direct Link:
  • 16
    Kamath P, Lacerda M, Ahlquist D, et al. Gastric Mucosal Responses to intrahepatic portosystemic shunting in patients with cirrhosis. Gastroenterol 2000; 118: 90511.
  • 17
    Ohta M, Yamaguchi S, Tomikawa M. Pathogenesis of portal hypertensive gastropathy: A clinical and experimental review. Surgery 2002; 131: S165170.
  • 18
    Merkel C, Schillipiti M, Bighin R, et al. Portal hypertension and portal hypertensive gastropathy in patients with liver cirrhosis: a hemodynamic study. Dig Liver Dis 2003; 35: 26974.
  • 19
    Bayraktar Y, Balkanci F, Uzunalimoglu B, et al. Is portal hypertension due to liver cirrhosis a major factor in the development of portal hypertensive gastropathy? Am J Gastroenterol 1996; 91: 5548.
  • 20
    Sarwar S, Khan A, Alam A, et al. Effect on band ligation in portal hypertensive gastropathy and development of fundal varices. J Ayub Med Coll Abbottabad 2006; 18: 325..
  • 21
    Marques Chaves D, Sakai P, Mucenic M, et al. Comparative study of portal hypertensive gastropathy in schistosomiasis and hepatic cirrhosis. Endoscopy 2002; 34: 199202.
  • 22
    Quintero E, Pique J, Bombi J, et al. Gastric mucosal vascular ectasias causing bleeding in cirrhosis. A distinct entity associated with hypergastrenemia and low serum levels of pepsinogen I. Gastronterology 1987; 93: 105461.
  • 23
    Pan WD, Xun RY, Chen YM. Correlations of portal hypertensive gastropathy of hepatitis B cirrhosis with other factors. Hepatobiliary Pancreatic Dis Int 2002; 1: 52731.
  • 24
    Yuksel O, Koklu S, Arhan M, et al. Effects of esophageal varices eradication on portal hypertensive gastropathy and fundal varices: a retrospective and comparative study. Dig Dis Sci 2006; 51: 2730.
  • 25
    Yoshikawa I, Murata I, Nakano S, Otsuki M. Effects of variceal ligation on portal hypertensive gastropathy and gastric mucosal blood flow. Am J Gastroenterol 1998; 93: 714.
    Direct Link:
  • 26
    Sarin S, Shahi H, Jain M, et al. The natural history of portal hypertensive gastropathy: influence of variceal eradication. Am J Gastroenterol 2000; 95: 288893.
  • 27
    Hou M, Lin C, Chen C, et al. Changes in portal hypertensive gastropathy after endoscopic variceal sclerotherapy or ligation: an endoscopic observation. Gastrointest Endosc 1995; 42: 13944.
  • 28
    Hashizume M, Tanaka K, Inokuchi K. Morphology of Gastric Microcirculation in Cirrhosis. Hepatology 1983; 3: 100812.
  • 29
    Gupta R, Sawant P, Parameshwar RV, et al. Gastric mucosal blood flow and hepatic perfusion index in patients with portal hypertensive gastropathy. J Gastroenterol Hepatol 1998; 13: 9216.
  • 30
    Ohta M, Hashizume M, Higashi H, et al. Portal and gastric mucosal hemodynamics in cirrhotic patients with portal-hypertensive gastropathy. Hepatology 1994; 20: 14326.
  • 31
    Perini R, Camara P, Ferraz J. Pathogenesis of portal hypertensive gastropathy: translating basic research into clinical practice. Nat Clin Pract Gastroenterol Hepatol 2009; 6: 1508.
  • 32
    Ferraz J, Wallace J. Underlying mechanisms of portal hypertensive gastropathy. J Clin Gastroenterol 1997; 25: S738.
  • 33
    Tarnawski A, Srafeh IJ, Stachura J, et al. Microvascular abnormalities of the portal hypertensive gastric mucosa. Hepatology 1998; 8: 148894.
  • 34
    Albillos A, Colombato LA, Enriquez R, et al. Sequence of morphological and hemodynamic changes of gastric microvessels in portal hypertension. Gastroenterology 1992; 102: 206670.
  • 35
    Ferraz JGP, Tigley A, Wallace JL. Paradoxical effects of l-arginine on gastric mucosal integrity. Eur J Pharmacol 1994; 260: 10711.
  • 36
    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: 1427.
  • 37
    Lopez-Talavera J, Merril W, Groszmann R. Tumor necrosis factor alpha: a major contributor to the hyperdynamic circulation in pre-hepatic portal hypertensive rats. Gastroenterology 1995; 108: 7617.
  • 38
    Payen JL, Cales P, Pienkowski P, et al. Weakness of mucosal barrier in portal hypertensive gastropathy of alcoholic cirrhosis. Effects of propanolol and enprostil. J Hepatol 1995; 23: 68996.
  • 39
    Kawanaka H, Tomikawa M, Jones M, et al. Defective mitogen-activated protein kinase (ERK2) signaling in gastric mucosa of portal hypertensive rats: potential therapeutic implications. Hepatology 2001; 34: 9909.
  • 40
    Cales P, Zabotto B, Meskens C, et al. Gastroesophageal endoscopic features in cirrhosis. Gastroenterology 1990; 98: 15662.
  • 41
    Vigeneri S, Termini R, Piraino A, et al. The stomach in liver cirrhosis. Gastroenterology 1991; 101: 4728.
  • 42
    Triger D, Hosking S. The gastric mucosa in portal hypertension. J Hepatol 1989; 8: 26772.
  • 43
    Papazian A, Braillon A, Dupas JL, Sevenet F, Capron JP. Portal hypertensive gastropathy: an endoscopic study. Gut 1986; 27: 1199203.
  • 44
    Spina GP, Arcidiacono R, Bosch J, et al. Gastric endoscopic features in portal hypertension: final report of a consensus conference, Milan, Italy, September 19, 1992. J Hepatol 1994; 21: 4617.
  • 45
    Canlas K, Dobozi B, Lin S, et al. Using capsule endoscopy to Identify GI Tract Lesions in cirrhotic patients with portal hypertension and chronic anemia. J Clin Gastroenterol 2008; 42: 8448.
  • 46
    Ishihara K, Ishida R, Saito T, et al. computed tomography features of portal hypertensive gastropathy. J Comput Assist Tomogr 2004; 28: 8325.
  • 47
    Chang D, Levine M, Ginsberg G, Rubesin S, Laufer I. Portal hypertensive gastropathy: radiographic findings in eight patients. Am J Roent 2000; 175: 160912.
  • 48
    Tanoue K, Hashizume M, Wada H, et al. Effects of endoscopic injection sclerotherapy on portal hypertensive gastropathy: a prospective study. Gastrointest Endosc 1992; 38: 5825.
  • 49
    Yoo H, Eustace J, Verma S, et al. Accuracy and reliability of the endoscopic classification of portal hypertensive gastropathy. Gastrointest Endosc 2002; 56: 67580.
  • 50
    De Franchis. Updating consensus in portal hypertension: report of the Baveno II Consensus workshop on definitions, methodology and therapeutic strategies in portal hypertension. J Hepatol 2000; 33: 84652.
  • 51
    Oberti F, Burtin P, Maiga M, et al. Gastroesophageal endoscopic signs of cirrhosis: independent diagnostic accuracy, interassociation, and relationship to etiology and hepatic dysfunction. Gastrointest Endosc 1998; 48: 14857.
  • 52
    Payen JL, Cales P, Voigt JJ, et al. Severe portal hypertensive gastropathy and antral vascular ectasia are distinct entities in patients with cirrhosis. Gastroenterology 1995; 108: 13844.
  • 53
    Spahr L, Villenueve J-P, Dufresne M-P, et al. Gastric antral vascular ectasia in cirrhotic patients: absence of relation with portal hypertension. Gut 1999; 44: 73942.
  • 54
    Misra S, Dwivedi M, Misra V, Barthwal R. Portal-hypertensive–gastropathy-like changes in a patient with secondary polycytemia: reversal of endoscopic and histopathologic changes with phlebotomy. Gastrointest Endosc 2002; 56: 9246.
  • 55
    Hsu , et al. Gastric purpura. An unusual endoscopic feature in gastrointestinal hemorrhage. Gastrointest Endosc 1997; 45: 52729.
  • 56
    Garcia-Tsao G, Sanyal A, Grace N, Carey W, and the Practice Guidelines Committee of the American Association for the Study of Liver Diseases, the Practice Parameters Committee of the American College of Gastroenterology. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology 2007; 46: 92238.
  • 57
    Lo GH, Cheng JS, Hsu PI, et al. The effects of endoscopic variceal ligation and propanolol on portal hypertensive gastropathy: a prospective controlled trial. Gastrointest Endosc 2001; 53: 57984.
  • 58
    Gostout C, Viggiano T, Balm R. Acute gastrointestinal bleeding from portal hypertensive gastropathy: prevalence and clinical features. Am J Gastroenterol 1993; 88: 20303.
  • 59
    De Franchis R. Portal hypertension II. Proceedings of the Second Baveno International Consensus Workshop on Definitions, Methodology, and Therapeutic Strategies. Oxford: Blackwell Science, 1996; 3055.
  • 60
    Perez-Ayuso R, Pique J, Bosch J, et al. Propanolol in prevention of recurrent bleeding from severe portal hypertensive gastropathy in cirrhosis. Lancet 1991; 337: 14314.
  • 61
    Hosking S, Kennedy H, Seddon I, Triger D. The role of propanolol in congestive gastropathy of portal hypertension. Hepatology 1987; 7: 43741.
  • 62
    Zhou Y, Qiao L, Wu J, Hu H, Xu C. Control of bleeding in portal hypertensive gastropathy: comparison of the efficacy of octreotide, vasopressin, and omeprazole in the control of acute bleeding in patients with portal hypertensive gastropathy. A controlled study. J Gastroenterol Hepatol 2002; 17: 9739.
  • 63
    Ripoll C, Garcia-Tsao G. Treatment of gastropathy and gastric antral vascular ectasia in patients with portal hypertension. Curr Treat Options Gastroenterol 2007; 10: 48394.
  • 64
    Bruha R, Marecek Z, Spicak J, et al. Double-blind randomized, comparative multicenter study of the effect of terlipressin in the treatment of acute esophageal variceal and/or hypertensive gastropathy bleeding. Hepatogastroenterology 2002; 49: 11616.
  • 65
    Karajeh M, Hurlstone D, Stephenson T, Ray-Chaudhuri D, Gleeson D. Refractory bleeding from portal hypertensive gastropathy: a further novel role for thalidomide therapy? Eur J Gastroenterol Hepatol 2006; 18: 5458.
  • 66
    De Franchis Evolving Concepts in portal hypertension. report of the baveno IV concensus workshop on methodology of diagnosis and therapy of portal hypertension. J Hepatol 2005; 43: 16776.
  • 67
    Cremers MI, Oliveira AP, Alves AL, Freitas J. Portal hypertensive gastropathy: treatment with corticosteroids. Endoscopy 2002; 34: 177.