34 Portal Hypertension, Varices, Gastropathy, and Gastric Antral Vascular Ectasia
Ibrahim Mostafa Ibrahim, Mostafa Ibrahim, and Nancy N. Fanous
The term “portal hypertension” was first introduced by Gilbert and Carnot in 1902. Substantial progress has been achieved in understanding the pathophysiology of portal hypertension over the past decades. This progress has led to the development of new therapeutic approaches such as pharmacological and endoscopic therapies, as well as surgical and radiological shunting procedures. Portal hypertension remains one of the most serious sequelae of chronic liver disease. Complications of portal hypertension such as gastrointestinal hemorrhage, hepatic encephalopathy, hepatorenal syndrome, and ascites continue to be the cause of significant morbidity and mortality in many countries.
34.2 Portal Hypertension: What Do We Need to Know?
34.2.1 Pathophysiology of Portal Hypertension
Portal hypertension (PHT) is a clinical syndrome defined by pathologic increase of portal venous pressure gradient (PVPG) between the portal vein (PV) and the inferior vena cava (IVC) to greater than 5 mm Hg. The hepatic venous pressure gradient (HVPG) accurately reflects the portal pressure gradient in the most common causes of cirrhosis, 1 and summarizes in a single measurement the interplay between two factors: hepatic resistance to portal flow and increased portal venous blood flow. 2
34.2.2 Noncirrhotic Portal Hypertension
The diseases leading to noncirrhotic portal hypertension (NCPH) are primarily vascular in nature and classified anatomically on the basis of site of resistance to blood flow as prehepatic, hepatic, and posthepatic; hepatic causes are further subdivided into presinusoidal, sinusoidal, and postsinusoidal (▶Table 34.1). 2 , 3 , 4
In NCPH, HVPG is normal or only mildly elevated and is significantly lower than PV pressure. 1 , 2 Schistosomiasis is one of the most common causes of NCPH worldwide. Two disease entities in NCPH, noncirrhotic portal fibrosis/idiopathic PHT (NCPF/IPH) and extrahepatic PV obstruction (EHPVO) are distinct diseases presenting with features of PHT but without evidence of significant parenchymal dysfunction. 3 , 4 , 5 Doppler ultrasound (US) is the first-line radiologic investigation in both disorders. Management in both NCPF/IPH and EHPVO is focused on management of acute variceal bleeding (AVB). 3 , 6
34.2.3 Cirrhotic Portal Hypertension: Natural History, Risk Stratification, and Individualizing Care
PHT is the main driving factor in the natural history of cirrhosis. HVPG measurement is the gold-standard method to assess the presence of clinically significant portal hypertension (CSPH), which is defined as HVPG greater than or equal to 10 mm Hg. Patients without CSPH have no gastroesophageal varices, and have a low 5-year risk of developing them. Ascites and gastroesophageal varices are the most frequent manifestations of CSPH. 6 , 7
For patients with compensated cirrhosis, the alternative term ‘‘compensated advanced chronic liver disease (cACLD)’’ has been proposed by Baveno VI to better reflect the ongoing progression of severe fibrosis to cirrhosis. In asymptomatic patients with known causes of chronic liver disease (CLD), liver stiffness by transient elastography (TE) is sufficient to suspect cACLD. Values less than 10 kPa in the absence of other clinical signs can exclude cACLD; values 10 to 15 kPa are suggestive of cACLD; values greater than 15 kPa are highly suggestive of cACLD. In patients with virus-related cACLD, noninvasive tests as TE greater than or equal to 20 to 25 kPa, alone or combined to platelets and spleen size are sufficient to diagnose CSPH. Confirmation of cACLD can be done by liver biopsy, upper gastrointestinal (GI) endoscopy, and HVPG. 7
Three different risk stages have been proposed based on 1-year mortality data: low-, intermediate-, and high-risk cirrhosis. Each category of risk is presented with the clinical features, HVPG value, the main outcome to prevent, the main pathophysiologic factor related with that category of risk. The 1-year mortality in these stages is less than or equal to 1%, 1 to 20%, and greater than or equal to 20%, respectively (▶Table 34.2). 8
34.3 Diagnosis of Portal Hypertension
34.3.1 Hepatic Venous Pressure Gradient
PHT is present if the HVPG is greater than or equal to 6 mm Hg; it typically becomes clinically significant with HVPG greater than or equal to 10 mm Hg, at which point varices may develop. Once the HVPG is greater than or equal to 12 mm Hg, patients are at risk for variceal bleeding and the development of ascites. 9
34.3.2 Noninvasive Tests
Transabdominal Ultrasound with Doppler Imaging
It may support a diagnosis of portal hypertension, but lack sensitivity. 10
TE values higher than 15 kPa could be used to diagnose cACLD. Values greater than or equal to 20 to 25 kPa are sufficient to rule in CSPH; and less than 20 kPa associated with platelet count greater than 150,000 could safely avoid screening of esophageal varices by endoscopy. 7
The cutoff values for patients with hepatitis C virus (HCV) infection and cirrhosis range from 11 to 17 kPa. The sensitivity and specificity are approximately 70 to 80% for F2 to F4 fibrosis. 13 , 14 Diagnostic accuracy is similar in patients with advanced-stage nonalcoholic fatty liver disease (NAFLD), with an area under the receiver operating characteristic (AUROC) curve of 0.94, sensitivity of 94%, and specificity of 95%. 15 In patients with autoimmune liver diseases, TE is very sensitive and specific for predicting advanced fibrosis in patients with primary biliary cholangitis and primary sclerosing cholangitis 16 however, it is less reliable than in autoimmune hepatitis due to significant hepatic inflammation that can overestimate stiffness. 17 There have been several meta-analyses of transient elastography testing, with a summary AUROC curve for diagnosing cirrhosis ranging from 0.90 to 0.95. 13 , 14 , 18 A meta-analysis of 40 studies of patients with CLD found a sensitivity of 83% and specificity of 89% for cirrhosis; however, for stage 2 fibrosis, the sensitivity was only 79% and specificity was 78%. 19
Evaluation for the underlying cause: For causes of cirrhotic and noncirrhotic PHT.
34.4 Treatment of Portal Hypertension
Therapies are aimed at achieving one of the following results:
Treatment of the Underlying Cause
In the era of new antiviral drugs, cirrhosis should be regarded as a disease whose mortality risk can be significantly reduced by a specific tailored approach. 7
Prevention and Management of the Other Complications of Portal Hypertension
Other complications of portal hypertension include spontaneous bacterial peritonitis (SBP), portal hypertensive gastropathy, hepatic hydrothorax, hepatopulmonary syndrome, portopulmonary hypertension.
Decreasing Portal Hypertension and Direct Treatment of Varices
Natural History of Varices
Varices are present in 50% of patients with cirrhosis and these form at a rate of 5 to 15% per year. Variceal bleeding occurs in one-third of patients with varices, and causes 70% of all upper gastrointestinal bleeding (UGIB) episodes in cirrhotic patients. Standardization of supportive care and new therapeutic options reduced bleeding-related mortality from about 50 to 15%—20% in the last three decades. 20 , 21 , 22
Screening and Surveillance
In patients with cirrhosis, the risk of varices is very low in patients with platelet count greater than or equal to 150,000 and the liver stiffness less than 20 kPa on TE, those patients may be followed up with annual platelet count and TE. 7 Endoscopic screening for varices should be performed if platelet count is less than 150,000 and TE is greater than 20 kPa: every 2 to 3 years in patients with compensated cirrhosis and no varices; 1 to 2 years in small Upper Gastrointestinal Tract Disease varices; and yearly or at the time of first decompensation in decompensated cirrhosis. 20
Esophageal varices (EV) are long columns of dilated veins, usually occurring within the lower third of the esophagus, immediately above the gastroesophageal junction (GEJ). EV are endoscopically graded according to the size (▶Table 34.3;▶Fig. 34.1) 23 ; however, the American Association of Study of Liver Disease (AASLD) recommends the classification into small and large esophageal varices based on a cutoff of 5 mm. In practice, the recommendations for medium-sized varices in the three-size classification are the same as for large varices in the two-size classification. 24
Gastric varices are supplied by the short gastric veins and drain into the deep intrinsic veins of the lower esophagus. These are classified according to site by the Sarin classification (▶Fig. 34.2, ▶Table 34.4). 25 Gastric varices account for 10 to 30% of variceal haemorrhage and can occur in up to 20% of patients with portal hypertension. 26 In our experience, the location of gastric varices according to the Sarin classification has no impact on management of acute gastric variceal bleeding (Video 34.1).
The predictors of variceal bleeding include presence of decompensated cirrhosis (Child–Turcotte–Pugh [CTP] class B or C), size of varices, and presence of high-risk stigmata upon endoscopy (red wale marks/cherry red spots). 27 The 1-year rate of recurrent variceal hemorrhage is approximately 60%. 28 The 6-week mortality with each episode of variceal hemorrhage is approximately 15 to 20%. 29
Treatment of the etiologic agent is expected to improve both liver structure and function, and this could reflect on reduction of PHT and development of varices. Treatment with nonselective β-blockers (NSBB) is not recommended in those patients. 20
34.4.1 Primary Prophylaxis
Primary prophylaxis refers to the prevention of a first variceal hemorrhage in a patient with varices. Guidelines recommend usage of one of two approaches: pharmacologic prophylaxis using NSBB, or endoscopic prophylaxis using endoscopic variceal ligation (EVL) . Both NSBB and EVL are superior to no treatment for the prevention of a first variceal hemorrhage in patients with medium- and large-sized varices and patients with small varices who have red signs or are Child’s score B or C (Video 34.2). 7
According to Baveno VI, traditional NSBB (propranolol, nadolol) and carvedilol are valid first-line treatments for primary prophylaxis of variceal hemorrhage. Carvedilol proved to be more effective than traditional NSBB in reducing HVPG, but has not been adequately compared head-to-head to traditional NSBB in clinical trials. 7
Nadolol and propranolol at a starting dose of 20 to 40 mg/d, respectively are used in patients with good tolerability and no contraindication toβ-blockers. A meta-analysis that included seven trials with 797 patients found that patients treated with β-blockers had improved outcomes compared with controls. 30 Carvedilol has been recommended in a dose of 12.5 mg twice daily for patients with Child A cirrhosis, and 6.25 mg twice daily for patients with Child B or C cirrhosis. 31 Other drugs such as simvastatin, clonidine, molsidomine, metoclopramide, pentoxifylline, verapamil, and losartan are being evaluated. 32 The effect of NSBB can be monitored through either starting at a low dose then increasing to a maximum tolerated dose as needed to achieve a resting heart rate of about 55 to 60 beats/min 33 ; or a decrease in HVPG of at least 10% from baseline or to less than or equal to 12 mm Hg after chronic treatment with NSBB is clinically relevant and is associated with a significant reduction in risk of variceal bleeding and decompensation. Similarly, acute HVPG response to intravenous propranolol may be used to identify responders to β-blockers. 7 HVPG measurements may be useful in clinical trials; however, it is not feasible for routine practice.
EVL is recommended for patients with medium or large varices who are intolerant of or have contraindications to β-blockers, or if a goal reduction in HVPG cannot be achieved. Effectiveness of EVL versus NSBB for primary prophylaxis has been studied and several meta-analyses have been published. The overall data suggest that EVL is as effective as NSBB with somewhat less hemorrhage but no changes in overall mortality. 34 , 35 , 36 , 37 , 38
NSBB do not prevent development or progression from small to large varices and have significant side effects. On the other hand, EVL should be performed by expert endoscopists to avoid complications including banding-induced ulcerations and bleeding (Video 34.3). Also, patients require routine endoscopic surveillance post-EVL due to the probability of variceal recurrence. 39 The frequency of endoscopic evaluation depends on multiple factors such as whether the patient has varices, size of varices and risk signs, and if the patient had compensated or decompensated liver disease (▶Fig. 34.3). In general, patients require three to four sessions for eradication of varices. Combination therapy was not more effective than EVL alone in preventing hemorrhage or death. 40 Therefore, decision to choose either option should be individualized based on local resources and level of experience.
Prophylactic endoscopic sclerotherapy (ES) for esophageal varices is not recommended as it carries higher rates of complications without substantial benefit. 41 , 42 Though it has been studied for this use in patients with large gastric varices, 7 the Baveno consensus guidelines did not recommend cyanoacrylate injection for primary prophylaxis of gastric varices. For the time being, patients with gastric varices should continue to receive NSBB for primary prophylaxis. Surgical shunts and transjugular intrahepatic portosystemic shunt (TIPS) have been proposed for primary prophylaxis; however, the available data do not support their use.
In our experience, we do prophylactic ES for gastric varices by glue only in two situations: (1) management of acute esophageal variceal bleeding if associated with gastric varices of any type; (2) during prophylactic EVL for esophageal varices in a patient with associated gastric varices. Theoretically, obliteration of esophageal varices may lead to increased pressure in gastric varices thus increasing the possibility of gastric hemorrhage.
34.4.2 Management of Acute Variceal Bleeding
Management of AVB necessitates the work of a multidisciplinary team, and works simultaneous on four axes: initial resuscitation, treatment of acute bleeding, secondary prophylaxis, and management of treatment failure.
Risk Stratification and Resuscitation
Evaluation of hemodynamic status and intravascular volume monitoring is crucial through careful history taking, vital signs with measuring orthostatic hemodynamic changes, and laboratory testing (complete blood count [CBC], coagulation parameters, blood grouping and cross-matching of 2–4 blood units, liver and kidney function) is initially performed.
Risk stratification into low- and high-risk patients on admission according to validated prognostic scales should be done. High-risk patients can be identified for early intervention, thus reducing morbidity and mortality. Low-risk patients can be discharged safely. The Glasgow–Blatchford score (GBS) should be used in every patient on initial presentation. 43 The Rockall score can also be used, but requires knowledge of endoscopic findings to be fully completed. 44
Airway protection up to endotracheal intubation may be considered in patients with massive bleeding and/or hepatic encephalopathy. 45 Routine use of nasogastric tube (NGT) in patients with AVB is controversial. It is unclear whether NGT placement aggravates hemorrhage from varices or Mallory–Weiss tears. Although studies have failed to demonstrate a benefit with regard to clinical outcomes 46 ; however, in our experience a NGT placement and washing with cold saline together with the start of pharmacologic therapy can help decompress the stomach, help prevent aspiration, and assist in clearing the field before endoscopy.
Circulatory support is done through two large-bore peripheral intravenous lines or a central line. Replacement is done by packed red blood cells (RBCs) following the restrictive transfusion strategy aiming for a hemoglobin level of 7 to 8 g/L. 6 A study by Villanueva et al compared the efficacy and safety of a restrictive transfusion strategy (transfusion when the hemoglobin level fell below 7 g/L) with those of a liberal transfusion strategy (transfusion when the hemoglobin level fell below 9 g/L) in a total of 921 patients with severe acute upper GI bleeding. The probability of survival at 6 weeks was higher in the restrictive-strategy group than in the liberal-strategy group (95 vs. 91%). Further bleeding occurred in 10% of the patients in the restrictive-strategy group as compared with 16% of the patients in the liberal-strategy group (p = 0.01), and adverse events occurred in 40% as compared with 48% (p = 0.02). The probability of survival was significantly higher in the subgroup of patients with cirrhosis and Child–Pugh class A or B disease, but not in those with cirrhosis and Child–Pugh class C disease. Within the first 5 days, the portal-pressure gradient increased significantly in patients assigned to the liberal strategy, but not in those assigned to the restrictive strategy. 47
Coagulopathy should be corrected as needed with fresh frozen plasma and platelet transfusion. The role of recombinant factor VIIa cannot yet be recommended for routine clinical use in patients with variceal hemorrhage. 48
Antibiotic prophylaxis is recommended in AVB (▶Table 34.5). A systematic review including eight placebo-controlled trials in 864 patients found the antibiotics were associated with a significant reduction in mortality and bacterial infections as bacteremia, pneumonia, SBP, and urinary tract infections. 49 Intravenous ceftriaxone (1 g/d for 7 days) proved superior to norfloxacin in a randomized-controlled trial (alternatives include trimethoprim–sulfamethoxazole [one double-strength tablet twice daily]or ciprofloxacin [500 mg orally every 12 hours]). 50 Antibiotics may also reduce the risk of recurrent bleeding in hospitalized patients who bled from esophageal varices. 51 Guidelines recommend that short-term (maximum 7 days) antibiotic prophylaxis should be instituted in any patient with cirrhosis and GI hemorrhage. In patients with advanced cirrhosis, intravenous ceftriaxone may be preferable, particularly in centers with a high prevalence of quinolone-resistant organisms. 52
Hepatic encephalopathy (HE) prevention in patients with cirrhosis and upper GI bleeding may be achieved by either lactulose or rifaximin. 7
Thiamine should be given in alcoholic subjects monitored for withdrawal symptoms. 53