Therapeutic options for portal hypertension

Raffaella Lionetti
MD
Research Fellow and Medical Assistant

Mario Angelico
MD
Director
Gastroenterology Unit
Tor Vergata University
Fatebenefratelli Hospital
Rome
Italy
E:[email protected]

Portal hypertension (PH) is a syndrome defined by a pathological increase in portal venous pressure above its physiological value, or an increase in the hepatic venous pressure gradient (HVPG) above 3–5mmHg. When the HVPG rises above the threshold of 10–12mmHg, complications of PH may occur, and therefore this cutoff defines patients with clinically significant PH.(1) Measurement of HVPG is the gold standard for the diagnosis and quantification of PH. The response of HVPG to medical therapy correlates with protection from complications of PH, namely gastrointestinal bleeding from ruptured gastro-oesophageal varices, or portal hypertensive gastropathy, ascites and metabolic disorders. Decreases in HVPG below 12mmHg, or at least 20% from baseline values, are the best predictive criteria demonstrating the efficacy of treatment.

Mortality from variceal bleeding is the leading cause of death in cirrhosis. Many studies have shown that the risk of bleeding is higher in patients with large oesophageal varices, yet a consistent proportion of bleeding (25% at two years) also occurs in patients with small varices.(2)

The main two factors contributing to the development of PH are increased vascular resistance and increased hepatic blood flow.

• The increase in vascular resistance to portal blood flow is localised to the hepatic sinusoids and the collateral vessels. Increased hepatic vascular resistance in cirrhosis not only is a mechanical consequence of the architectural disorder caused by liver disease, but is also caused by a “dynamic” component due to the active contraction of activated stellate cells and vascular smooth muscle cells. The dynamic component may be modified by endogenous factors and pharmacological agents. Hepatic vascular resistance is further increased because of an imbalance between vasodilatory and vasoconstrictor substances,(2) the latter being predominant. This is important because it provides the basis for using vasodilators.
• The second factor contributing to the pathogenesis of PH is the increase in portal blood flow due to a splanchnic arteriolar vasodilation. The increased portal inflow can be counteracted by using splanchnic vasoconstrictors, such as beta-blockers.

Treatment options and clinical scenarios
Nonselective beta-blockers, such as propranolol and nadolol, are the most commonly used drugs in the treatment of PH since they reduce portal pressure by decreasing portal and collateral blood flow.(3) This is achieved by a decrease in cardiac output, determined by the blockade of beta(1)- adrenoceptors in the heart; and by splanchnic vasoconstriction caused by the blockade of vasodilatory beta(2)-receptors in the splanchnic vasculature.(4) This explains why cardioselective beta-blockers are less effective in reducing portal pressure.

The decrease in portal pressure is accompanied by a reduction in the pressure of oesophageal varices.(4) Adequate protection from the risk of bleeding (ie, a HVPG reduction below 12mmHg or of at least 20%(5) is achieved in one-third to one-half of treated patients. Nonresponse to treatment is in part due to an increase in portal collateral resistance, which attenuates the reduction in portal pressure promoted by the decrease in portal flow.

The dosage of beta-blockers should be individualised. Adjustment of the dose is usually done by stepwise increases by carefully looking at clinical tolerance, heart rate and arterial blood pressure. In general, progressive doses are administered beginning at 20mg/12h and increasing every two days until the heart rate decreases by 25%, or up to the maximum tolerated dosage.(3) Attention should be paid to avoid reaching less than 55 beats/min or a systolic arterial pressure lower than 90mmHg.

beta-blockers are contraindicated in patients with asthma, chronic obstructive pulmonary disease, aortic stenosis, atrioventricular block, intermittent claudication and psychosis. Bradycardia and diabetes are relative contraindications.

The frequency of adverse effects during long-term follow-up is about 15%.(3) The most frequent are dyspnoea, bronchospasm, insomnia, fatigue, impotence and asthenia.

The medical treatment of PH includes at least three different scenarios:

• Primary prophylaxis (or prevention of first bleeding).
• Treatment of the acute bleeding episode.
• Prevention of rebleeding (secondary prophylaxis).

Primary prophylaxis
The aim of prophylactic therapy is to prevent variceal bleeding and bleeding-related deaths. The effectiveness of beta-blockers, such as propranolol and nadolol, in the prevention of the first variceal bleeding has been proved. On average, beta-blockers reduce by 50% the chance of haemorrhage in cirrhotic patients at risk.(6) However, 30–40% of patients taking beta-blockers do not achieve an appropriate haemodynamic response to treatment, mostly due to the onset of side-effects, or to increasing hepatic resistances.(7) A meta-analysis shows not only a significant benefit of beta-blockers in relation to bleeding (OR 0.54 and 95% CI 0.39–0.74), but also a marked trend in reduced mortality close to statistical significance (0.75 and 95% CI 0.57–1.06).(7) The beneficial effect of propranolol is limited to the period of administration, so that once treatment is initiated it should be maintained indefinitely.

Acute variceal bleeding
This is one of the most dramatic events in the life of cirrhotic patients and is associated with high mortality at the first episode.(8) Treatment involves several steps, including resuscitation and replacement of the blood loss, and specific treatment of the bleeding. Drug therapy consists of the treatment of associated complications, such as bacterial infections and encephalopathy, and in the use of vasoactive drugs in order to stop bleeding.

Treating associated complications
Norfloxacin should be administered orally at the dosage of 400mg twice daily for seven days, through a nasogastric tube or intravenously.(9) Lactulose should be given orally or by enema.

Stopping the bleeding
Vasoactive drugs effective in stopping acute bleeding include vasopressin, terlipressin, somatostatin and its analogues.

Vasopressin is no longer used due to adverse effects. Its action can be ameliorated with the co-administration of nitroglycerin, which results in a more marked reduction in HVPG and in better tolerance.

Terlipressin is a synthetic analogue of vasopressin with prolonged effects. Placebo-controlled studies have proved the efficacy of terlipressin (2mg every four hours) in the treatment of acute bleeding from ruptured oesophagastric varices and its capacity to reduce mortality. The most common adverse effects are arterial hypotension and bradycardia. Once haemostasis is obtained, terlipressin can be administered for five days in smaller doses (1mg every four hours) to prevent early rebleeding. In a recent study, terlipressin was found to be as effective as endoscopic sclerotherapy in controlling the bleeding and in the prevention of early rebleeding, but with a lower rate of complications. Terlipressin has been used with success in the treatment of bleeding cirrhotics while they were being brought to the hospital from their home by ambulance.(10)

Somatostatin also decreases portal flow and pressure at pharmacological dosages. Its effectiveness in reducing HVPG is much greater following bolus injection than after continuous infusion, the latter showing only a transient effect. It is used in continuous intravenous infusion at doses of 250–500µg/h after one to three boluses of 250µg.

Octreotide is a synthetic analogue of somatostatin, effective in reducing complications of variceal bleeding after emergency sclerotherapy. A recent meta-analysis (of heterogeneous studies) suggested its efficacy as firstline treatment both given as monotherapy(11) or in association with endoscopic treatment.

Prevention of rebleeding
Rebleeding may occur in approximately two-thirds of patients who have already bled and represents a major clinical concern in the first six months after the initial haemorrhagic episode. Both treatment with beta-blockers(5) and endoscopic sclerotherapy have been demonstrated to significantly reduce the proportion of patients rebleeding compared with those who have not been treated. Yet drug therapy is usually preferred because there are fewer side-effects.

The combination of beta-blockers with nitrates has been shown to increase the protective efficacy of b-blockers against bleeding, although no improvement in mortality has so far been reported.(12) This combination has also been found to be superior compared with sclerotherapy or band ligation alone in a population of patients with compensated cirrhosis.(13) On the other hand, the combination of band ligation with drug treatment has been reported to be better than sclerotherapy alone, possibly being the best currently available choice of treatment.

When these approaches are ineffective, the patient should be considered for a transjugular intrahepatic portosystemic shunt,(14) or be submitted to a distal splenorenal shunt or to other types of portocaval shunt surgery.

References

1. Garcia-Tsao G, Groszmann R, Fisher R, et al. Hepatology 1985;5:419-24.
2. Cales P, Desmorat H, Vinel JP, et al. Gut 1990;31:1298-302.
3. Lebrec D, Nouel O, Corbic M, Benhamou J-P. Lancet 1980;ii:180-2.
4. Bosch J, Masti R, Kravetz D, et al. Hepatology 1984;4:1200-5.
5. Groszmann R, Bosch J, Grace N, et al. Gastroenterology 1990;99:1401-7.
6. D’Amico G, Pagliaro L, Bosch J. Hepatology 1995;22:332-54.
7. Pagliaro L, D’Amico G, Sorensen TIA, et al. Ann Intern Med 1992;117:59-70.
8. Smith JL, Graham DY. Gastroenterology 1982;82:968-73.
9. Bernard B, Grange JD, Khac EN. Hepatology 1999;29:1655-61.
10. Levacher S, Letoumelin P, Pateron D. Lancet 1995;346:865-8.
11. Corley DA, Cello JP, Adkisson W. Gastroenterology 2001;120:161-9.
12. Villanueva C, Balanzò J, Novella MT. N Engl J Med 1996;334:1624-9.
13. Villanueva C, Minana J, Ortiz J. N Engl J Med 2001;345:647-55.
14. Papatheodoridis GV, Goulis J, Leandro G. Hepatology 1999;30:612-22.