Treatment of portal hypertension includes pharmacologic management aimed at decreasing portal pressure and endoscopic therapy aimed at obliterating esophageal varices. When medical treatment fails, shunt procedures can be used to decompress high portal pressure. The goal of treatment is to interrupt the process that leads to the development of varices, refractory ascites, and hydrothorax.
De Franchis R. Revising consensus in portal hypertension: report of the Baveno V consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol
Garcia-Tsao G, Bosch J. Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med
Nonselective β-blockers (propranolol, nadolol, and timolol) are the cornerstones of long-term management in patients with portal hypertension (Table 47–2). β-Blockers work by decreasing cardiac output by blocking β1-receptors and by vasoconstricting splanchnic vessels via blockade of β2-receptors, thereby leaving unopposed α-adrenergic activity. Unfortunately, not all patients respond to maximally tolerated doses of β-blockers. In patients who fail to show a reduction in HPVG, there is often an increase in portocollateral resistance. The long-acting nitrate isosorbide mononitrate, when used in conjunction with β-blockers, has been shown to counteract this increase in portocollateral resistance. Oral nitrates may cause systemic hypotension, which limits their clinical usefulness. Carvedilol, a nonselective β-blocker with intrinsic anti-alpha-1 adrenergic activity, produces a greater reduction in the HVPG than nonselective β-blockers given as monotherapy and has been shown to be well tolerated in clinical trials.
Table 47–2. Drugs Used in the Management of Portal Hypertension. ||Download (.pdf)
Table 47–2. Drugs Used in the Management of Portal Hypertension.
|Drug||Class of Drug||Starting Dose||Maximum Dose|
|Propranolol||Nonselective β-blocker||40 mg twice daily||640 mg/day|
|Nadolol||Nonselective β-blocker||40 mg daily||320 mg/day|
|Timolol||Nonselective β-blocker||10 mg daily||40 mg/day|
|Isosorbide mononitrate||Long-acting nitrate||20 mg daily||240 mg/day|
|Spironolactone||Aldosterone antagonist||25 mg daily||400 mg/day|
|Furosemide||Loop diuretic||40 mg daily||80 mg/day|
|Octreotide||Splanchnic vasoconstrictor||50 mcg bolus, followed by 50 mcg/h||50 mcg/h|
|Norfloxacin||Quinolone antibiotic||400 mg twice daily||—|
The use of β-blockers in patients with cirrhosis is limited by their side-effect profile, which includes hypotension, fatigue, lethargy, depression, and dyspnea in patients with associated pulmonary disease. Due to concomitant diseases such as reactive airway disease, congestive heart failure, bradycardia, and heart block, 15–20% of patients are unable to take β-blockers. In patients who are candidates for β-blocker therapy, dose titration should be adjusted to patient tolerance.
Vasopressin and its analog, terlipressin, and somatostatin and its analogs, octreotide and vapreotide, have been the agents used for the management of acute variceal bleeding. These vasoconstrictive drugs act by decreasing splanchnic blood flow, resulting in a lowering of portal pressure, and by decreasing splanchnic hyperemia. Vasopressin has been demonstrated to decrease HVPG by as much as 23%. Its use is limited, however, due to significant complications, including myocardial infarction, hypertension, hyponatremia, peripheral vascular ischemia, bradycardia, and fluid retention. These side effects may be ameliorated with the concurrent use of nitroglycerin, but caution is recommended with its use, due to the risk of hypotension.
Terlipressin has a milder side-effect profile than vasopressin, and it does not appear to alter renal excretion of sodium. Its intravenous half-life is 4 hours, and its use results in a 21% decrease in HVPG. Randomized controlled trials comparing terlipressin with a placebo or no pharmacologic treatment in patients with acute variceal hemorrhage have demonstrated a significant survival benefit for terlipressin. It is safer and more effective than either vasopressin or vasopressin plus nitroglycerin.
Somatostatin has been shown to decrease portal pressure in patients with portal hypertension, but to a lesser extent than vasopressin. It works by inhibiting vasodilatory peptides from the gastrointestinal tract that have been shown to contribute to the maintenance of portal hypertension. Due to its short half-life (2 minutes), somatostatin is used as a continuous infusion after an initial bolus to treat acute variceal hemorrhage.
The somatostatin analogs, namely octreotide, lanreotide, and vapreotide, have similar pharmacologic properties to somatostatin.
Octreotide, which is used widely in the United States, is thought to act through the inhibition of the vasodilator glucagon. The intravenous half-life of octreotide, like that of somatostatin, is short; therefore, it is given as a bolus followed by a continuous infusion. Although intravenous octreotide may decrease portal pressure and bleeding from esophageal varices, its use has not been shown to improve overall survival. Similarly, vapreotide in combination with endoscopic therapy has been shown to decrease bleeding from esophageal varices more effectively than endoscopic therapy alone, but without a benefit in survival.
Together with a low-sodium diet, continuous spironolactone (100 mg/day) treatment results in a modest decrease in portal pressure. The efficacy and safety of losartan and irbesartan, angiotensin II receptor blockers, in lowering portal pressure has been established in Child Pugh class A patients but the risk of systemic hypotension and renal failure precludes their use in patients with decompensated cirrhosis. Randomized controlled trials with established clinical endpoints are needed to establish clinical efficacy.
Calés P, Masliah C, Bernard B, et al. Early administration of vapreotide for variceal bleeding in patients with cirrhosis. N Engl J Med.
Groszmann RJ, Garcia-Tsao G, Bosch J, et al. Portal Hypertension Collaborative Group. Beta-blockers to prevent gastroesophageal varices in patients with cirrhosis. N Engl J Med.
Endoscopic Variceal Ligation (EVL) and Sclerotherapy
EVL and sclerotherapy are local treatments for varices and do not alter the pathophysiologic processes that lead to their development. As such, even with successful EVL, varices eventually recur. EVL consists of the placement of rubber bands on variceal columns that lead to localized mucosal and submucosal necrosis and replacement of the varix by scar tissue. Band ligation carries the risk of causing esophageal ulcerations that have the potential to bleed.
Sclerotherapy involves injecting a sclerosing agent (ethanolamine oleate, sodium tetradecyl sulfate, sodium morrhuate) into or adjacent to a varix. Sclerotherapy leads to localized thrombosis and inflammation with subsequent scar formation. Both local and systemic complications of sclerotherapy occur in 20–40% of patients. Side effects include bacteremia, fever, chest pain, mediastinitis, acute esophageal ulceration (with bleeding potential), and esophageal strictures. Because of greater efficacy and fewer side effects, EVL has replaced sclerotherapy as the endoscopic treatment of choice.
Garcia-Pagán JC, Bosch J. Endoscopic band ligation in the treatment of portal hypertension. Nat Clin Pract Gastroenterol Hepatol.
The transjugular intrahepatic portosystemic shunt (TIPS) was first introduced in the 1980s, as an alternative to surgically performed shunts. TIPS should be reserved for patients who fail pharmacologic and endoscopic management of both acute and recurrent variceal bleeding.
Placement of a TIPS bypasses the fibrosed liver and allows for unhindered blood flow between the portal and hepatic veins. A catheter is inserted into the right internal jugular vein and advanced to the hepatic venous system (usually the right hepatic vein). A needle is then used to cannulate the liver, creating a tract to the portal vein. The transhepatic tract is dilated, and a flexible metal stent is placed, resulting in a shunt between the hepatic and portal veins. Placement of TIPS is successful in greater than 90% of cases and is effective in acutely decreasing portal pressure.
Early complications of TIPS include fever, infection, renal dysfunction, intrahepatic or intraperitoneal hemorrhage, and liver failure. Furthermore, hyperbilirubinemia, secondary to hemolytic anemia, may persist until the TIPS stent is re-epithelialized. Long-term complications from TIPS placement include hepatic encephalopathy and stent occlusion. Over 20% of patients with TIPS will either develop or have worsening hepatic encephalopathy and should be prophylactically treated with lactulose, neomycin, or rifaximin. Risk factors for developing hepatic encephalopathy include older age, larger stent diameter, and prior episodes of hepatic encephalopathy. Seventy percent of patients who undergo TIPS with non-coated stents will develop stent occlusion within 1 year and 90% by 2 years. Angioplasty or additional stent placement is successful in treating stent occlusion and decreases the reocclusion rate to 10% at 2 years. However, these complication rates are significantly reduced with the use of coated stents which have replaced non-coated stents as standard therapy. Patients treated with TIPS should have Doppler ultrasound surveillance at regular intervals to confirm stent patency, as TIPS dysfunction can result in variceal bleeding. However, angiographic assessment and measurements of portal pressure are more accurate in establishing TIPS dysfunction.
Absolute contraindications to TIPS placement include right heart failure and polycystic liver disease. Relative contraindications include systemic infection, portal vein thrombosis, biliary obstruction, and severe hepatic encephalopathy. Although TIPS can be successfully placed in most patients, placement is associated with a 30-day mortality rate of 3–15% and a procedure-related mortality rate of 2–5%.
Boyer TD, Haskal ZJ; American Association for the Study of Liver Diseases. The role of transjugular intrahepatic portosystemic shunt in the management of portal hypertension. Hepatology.
Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene coated stents for TIPS: results of a randomized study. Gastroenterology.
Variceal hemorrhage is not an indication for liver transplantation. However in appropriate candidates as determined by the Model for End-Stage Liver Disease (MELD) scoring system, liver transplantation can achieve up to 95% control of variceal bleeding. Liver transplantation continues to be limited by organ availability.
Early treatment with β-blockers prior to the development of complications of portal hypertension may halt or delay the progression of portal hypertension. Unfortunately, the results from a large multicenter randomized controlled trial have not supported the use of nonselective β-blockers in the preprimary prophylaxis of portal hypertension. However, data from this trial show that measurements of the HVPG predict the development of esophageal varices and hepatic decompensation.
Primary Prophylaxis for Variceal Hemorrhage
Because of the high morbidity and mortality in patients with cirrhosis associated with acute variceal bleeding, pharmacologic therapy aimed at preventing initial variceal bleeding is paramount (Figure 47–1). Nonselective β-blockers are the only established pharmacologic therapy for primary prophylaxis of variceal bleeding. A meta-analysis of randomized controlled trials has demonstrated their efficacy in decreasing the rates of a first variceal bleed from 24% to 15%. These studies involved primarily patients with Child-Pugh class A and B cirrhosis.
Management algorithm for primary prophylaxis of variceal hemorrhage. EGD, esophagogastroduodenoscopy; EVL, endoscopic variceal ligation.
Current practice guidelines on the management of portal hypertension recommend that patients with small varices who are at increased risk of bleeding (red wale sign on endoscopy, hepatic decompensation) should be treated prophylactically with β-blockers. Patients with small varices who are not at high risk may be started on β-blockers, but their long-term efficacy has not been established. All patients with medium-to large-sized varices who are judged to be compliant and are without contraindications to or intolerance of nonselective β-blockers should be treated with these drugs. Therapy must be continued for a lifetime, as the risk of bleeding recurs if treatment is discontinued.
Although combining a long-acting nitrate with a nonselective β-blocker increases the number of hemodynamic responders, it has not resulted in incremental clinical benefit (ie, by preventing initial variceal hemorrhage or leading to improved survival). Similarly, combining spironolactone with β-blockers showed no benefit in decreasing rates of bleeding episodes or survival compared with treatment using β-blockers alone. Current data do not support the use of combination therapy or monotherapy with nitrates for the prevention of a first variceal bleed. A recent study suggests that carvedilol may be superior to endoscopic variceal ligation for primary prophylaxis but more data are needed to establish its clinical role.
Grace ND, Garcia-Pagan JC, Angelico M, et al. Primary prophylaxis for variceal bleeding. In: de Franchis R (editor). Portal Hypertension IV: Proceedings from the 4th Baveno International Consensus Workshop. Blackwell Science, 2006:168–200.
Esophageal Variceal Ligation
Nonselective β-blockers or esophageal variceal ligation are considered to be first-line therapies for the prevention of initial variceal hemorrhage. However, 15% of patients have contraindications, and an additional 15–20% are intolerant of β-blocker therapy. EVL is an excellent alternative for these patients.
For patients who undergo esophageal variceal ligation, it is recommended that the procedure be performed biweekly until the varices are obliterated. As new varices may develop after eradication treatment, it is important to continue surveillance for varices every 6 months. Endoscopic ultrasound may be used to confirm eradication of esophageal varices, but it is likely of limited utility in clinical practice.
Recent meta-analyses have shown a benefit for EVL over β-blocker therapy in preventing first variceal hemorrhage, but EVL is associated with more serious side effects and offers no survival advantage. Randomized controlled trials of EVL have been limited by a short duration of follow-up. The studies with the largest number of patients and the longest follow-up have failed to show a difference between EVL and nonselective β-blockers, either for the prevention of initial variceal bleeding or for survival. Combination therapy using β-blockers plus EVL compared with EVL alone has not been shown to decrease the risk of first variceal hemorrhage and is associated with increased side effects.
There is no role for endoscopic sclerotherapy, TIPS, surgical shunts, or liver transplantation in the primary prophylaxis of variceal hemorrhage.
dela Peña J, Brullet E, Sanchez-Hernández R, et al. Variceal ligation plus nadolol
compared with ligation for prophylaxis of variceal rebleeding: a multicenter trial. Hepatology.
Garcia-Tsao G, Sanyal AJ, Grace ND, et al; Practice Guidelines Committee of the American Association for the Study of Liver Diseases; Practice Parameters Committee of the American College of Gastroenterology. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology.
Gonzalez R, Zamora J, Gomez-Camarero J, et al. Meta-analysis: combination endoscopic and drug therapy to prevent variceal rebleeding in cirrhosis. Ann Intern Med
Lo GH, Lai KH, Cheng JS, et al. Endoscopic variceal ligation plus nadolol
compared with ligation alone for the prevention of variceal bleeding: a prospective, randomized trial. Hepatology.
Acute Variceal Hemorrhage
Patients with acute variceal bleeding may present with hematemesis, melena, or hematochezia. The initial management includes immediate resuscitation, administration of prophylactic antibiotics and vasoactive agents, and diagnostic endoscopy (Figure 47–2). It is important to consider elective intubation in patients with active bleeding to protect the airway from aspiration, especially if there is concomitant hepatic encephalopathy.
Management algorithm for acute variceal hemorrhage. EGD, esophagogastroduodenoscopy; EVL, endoscopic variceal ligation; PHG, portal hypertensive gastropathy; TIPS, transjugular intrahepatic portosystemic shunt.
Resuscitative measures should be aimed at replacing blood volume to a goal of a hematocrit of 25%, thereby avoiding increases in portal pressure and potential exacerbation of variceal bleeding associated with aggressive transfusion. Correcting thrombocytopenia and coagulopathy have not been shown to improve survival from an acute variceal bleed. Nonetheless, transfusing platelets in order to increase levels to greater than 50,000 and transfusing fresh frozen plasma to reverse coagulopathy may be beneficial in the acute setting. The excessive use of saline should be avoided in resuscitation, as it can worsen or precipitate the formation of ascites and volume overload.
Infection in the setting of acute variceal bleeding has been associated with early rebleeding and a high mortality rate. Patients with bleeding from varices are at high risk of developing infection, including spontaneous bacterial peritonitis. Short-term antibiotics should be administered to all patients with cirrhosis and acute variceal bleeding. Norfloxacin (400 mg twice daily for 7 days) or quinolone antibiotics with a similar spectrum of activity (eg, levofloxacin, ciprofloxacin) are the preferred agents. In a recent study, intravenous ceftriaxone (1 g/day) was found to be superior to norfloxacin in patients with severely decompensated cirrhosis.
As soon as variceal hemorrhage is suspected, administration of vasoactive agents should be initiated. These agents include vasopressin, vasopressin plus nitroglycerin, terlipressin, and somatostatin or its analogs. Use of vasopressin is limited by its serious side effects, which can be partially counteracted by the addition of nitroglycerin. The vasopressin analog terlipressin has fewer side effects than vasopressin and is the only pharmacologic agent that has a survival benefit in patients with acute variceal hemorrhage.
Octreotide has been shown to have equal efficacy to sclerotherapy in controlling bleeding. It is safe with very few side effects and can be used continuously for up to 5 days. However, its use has been associated with tachyphylaxis, and its vasoactive effects may be transient. When combined with endoscopic therapy, octreotide improves management of acute bleeding compared with EVL alone, but without a survival benefit.
β-Blockers should not be administered during an active bleed, due to the physiologic blunting of heart rate in the setting of hypotension.
Endoscopic evaluation should occur as soon as possible and within the first 12 hours of a patient presenting with an acute variceal bleed. In addition to confirming the source of bleeding, endoscopic treatments are successful in controlling hemorrhage in 90% of cases. EVL is the procedure of choice. The combination of EVL and a somatostatin analog has been shown to be more effective than either EVL or a pharmacologic agent, given as monotherapy, and is the treatment of choice.
Balloon tamponade is 80% effective in immediate control of hemorrhage from esophagogastric varices. However, serious complications occur when the balloon is left inflated for more than 24 hours. The complications, which occur in 20% of cases, include aspiration, esophageal ulceration and necrosis, and esophageal perforation. Balloon tamponade is best utilized as a short-term bridge to TIPS or liver transplantation.
Transjugular Intrahepatic Portosystemic Shunt (TIPS)
TIPS is used to manage variceal bleeding when endoscopic and pharmacologic treatments fail. In 90% of cases of uncontrolled bleeding, TIPS is successful in stopping bleeding. Unfortunately, mortality remains high (27–50%), probably because of the severity of illness, and rebleeding can occur in the short and long term. Predictors of early mortality include creatinine greater than 1.7 mg/dL, bilirubin greater than 5 mg/dL, a Child-Pugh score greater than 12, an APACHE II score greater than 18, and a Patch prognostic index greater than 18. TIPS is the rescue therapy of choice after two failed attempts of combined endoscopic and pharmacologic management.
The use of polytetrafluoroethylene-coated stents results in significantly less stent dysfunction and lower clinical relapse rates. Patients receiving coated stents, compared with those receiving uncoated stents, had lower rates of recurrent bleeding and hepatic encephalopathy after 2 years of follow-up. Recent studies have shown that the early use of TIPS with coated stents in patients at high risk for treatment failure (ie, Child Pugh class C patients or Child Pugh class B patients with active bleeding at endoscopy) is associated with a decrease in treatment failure and in mortality.
Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene coated stents for TIPS: results of a randomized study. Gastroenterology.
Garcia-Pagan JC, Caca K, Bureau C, et al. Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med.
Prevention of Recurrent Variceal Hemorrhage
Once the acute bleed has been controlled, secondary prevention of rebleeding is paramount. First-line therapy for prevention of recurrent variceal hemorrhage is the combination of pharmacologic therapy and EVL. For patients who fail to respond to medical therapy, TIPS, surgical shunts or liver transplantation for appropriate candidates are the options (Figure 47–3A).
A. The role of hepatic venous pressure gradient measurements in preventing recurrent variceal hemorrhage. EVL, endoscopic variceal ligation; TIPS, transjugular intrahepatic portosystemic shunt.
Pharmacologic treatment for prevention of recurring variceal hemorrhage is initiated by administration of a nonselective β-blocker, titrated to tolerance. β-Blockers have been shown to decrease rates of rebleeding from 63% to 42% and to decrease mortality from 27% to 20%. The hemodynamic goal of β-blocker therapy is to decrease the HVPG by greater than 20% or to a level less than 12 mm Hg (Figure 47–3B). Achieving this goal has been shown to reduce rebleeding rates to less than 15%.
B. Prevention of recurrent variceal hemorrhage. EVL, endoscopic variceal ligation; HVPG, hepatic venous pressure gradient; TIPS, transjugular intrahepatic portosystemic shunt.
In patients who have an incomplete response to β-blocker monotherapy, one can consider the addition of isosorbide mononitrate. With combination therapy, the rebleeding rate appears to be lower, but at a cost of significantly greater side effects and no survival benefit.
If patients are unable to tolerate β-blockers, EVL is the preferred endoscopic treatment to prevent recurrent variceal hemorrhage. Although EVL is more effective than sclerotherapy at preventing rebleeding, the risk of rebleeding after EVL remains high, reaching 30–50% at 2 years. The combination of a nonselective β-blocker and EVL has been shown to result in a decreased rate of variceal rebleeding, a lower transfusion requirement, a lower recurrence of esophageal varices, and a trend toward improved survival. However, there are slightly increased side effects with combination therapy. Current guidelines recommend combination therapy as the treatment of choice for prevention of variceal rebleeding,.
EVL sessions should be performed every 7–14 days until complete variceal obliteration is achieved. Once varices are eradicated, continued surveillance every 6–12 months is required to screen for recurrent varices.
In patients who do not respond to pharmacologic or endoscopic therapy, alone or in combination, decompression of the portal system by use of TIPS with coated stents is the next option. Liver transplantation provides the only definitive treatment for rebleeding as it directly addresses the underlying liver pathology that leads to portal hypertension and the formation of varices. However, variceal bleeding is not by itself an indication for liver transplantation, which should be reserved for appropriate candidates.
Numerous meta-analyses have demonstrated that TIPS is more effective than endoscopic therapy in preventing rebleeding from varices. The drawback of TIPS is an increase in hepatic encephalopathy and the lack of a survival benefit. Endoscopic and pharmacologic therapy remains the first-line therapy for prevention of rebleeding, with TIPS reserved as a rescue therapy.