IMAJ VOL 15 February 2013 REVIEWS

Evaluation and Treatment of Esophageal Varices in the

Cirrhotic Patient
Eyal Ashkenazi MD1, Yulia Kovalev MD1 and Eli Zuckerman MD2
1
Liver Unit, Carmel Medical Center, Haifa
2
Liver Unit, Carmel Medical Center affiliated with Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel

later into the liver parenchyma, causing fibrosis and sinusoidal

Abstract: Portal hypertension is the leading cause of morbidity and mor- hypertension.
tality in liver cirrhosis. Complications of portal hypertension To assess portal pressure hemodynamically, the most accu-
in cirrhotic patients include esophageal and gastric varices, rate way is to measure the portal pressure directly. In practice,
portal hypertensive gastropathy, ascites, hepatorenal syn- however, direct portal pressure measurement is not recom-
drome, hepatopulmonary syndrome and portopulmonary mended because of potential complications. Instead, wedge
hypertension. The hepatic venous pressure gradient should hepatic pressure can be measured by placing a catheter in
be at least 10 mmHg for esophageal varices to appear, and the hepatic vein, occluding it with a balloon, and measuring
more than 12 mmHg for acute esophageal variceal bleeding. the post-occlusion pressure. This wedge pressure accurately
This article reviews the pathophysiology responsible for portal
reflects the intrasinusoidal portal pressure, especially in cir-
hypertension and its complications, and the treatments used
rhotic patients. Reducing the free hepatic pressure from
for esophageal varices in the setting of primary and secondary
the wedge hepatic pressure results in a gradient, termed the
prophylaxis and during active bleeding.
hepatic venous pressure gradient. Based on measurements of
IMAJ 2013; 15: 109115
the wedge and free hepatic pressures, and the HVPG pres-
Key words: portal hypertension, esophageal variceal bleeding, primary
2

prophylaxis, secondary prophylaxis, carvedilol, simvastatin, sure, intrahepatic PHT can be divided into pre-sinusoidal,
transjugular intrahepatic portosystemic shunt (TIPS) sinusoidal or post-sinusoidal PHT, as mentioned earlier. This
classification is based on different hemodynamics for each
sub-classification:
Pre-sinusoidal PHT: Normal wedge pressure and free
hepatic venous pressure (normal HVPG)

P hemodynamically by pathologically elevated portal pres-

ortal hypertension is a common clinical syndrome defined
HVPG = hepatic venous pressure gradient

sure. Elevated pressure indicates a pathologic process in the

Table 1. Classification of portal hypertension
portal system. Portal hypertension can be classified anatomi-
cally as pre-hepatic, hepatic or post-hepatic, depending on
1. Pre-hepatic
the site of the pathology. The hepatic pathology can be further Portal vein thrombosis (independent of cause), splenic vein
divided into pre-sinusoidal, sinusoidal or post-sinusoidal por- tahrombosis, cavernous transformationof the portal vein, splenic
arteriovenous fistula, idiopathic tropical splenomegaly
tal hypertension, according to different etiologies and different
2. Intrahepatic
sites within the liver that can be affected by these etiologies a. Pre-sinusoidal
[Table 1]. The most common cause of PHT in the western 1

Schistosomiasis, chronic viral hepatitis (HBV and HBC), cirrhosis

biliaris primaria, myeloproliferative diseases, focal nodular
world is liver cirrhosis, accounting for 90% of cases. The most hyperplasia, idiopathic portal hypertension, sarcoidosis,
frequent cause of pre-hepatic PHT in the western world is tuberculosis, Wilsons disease, hemochromatosis, amyloidosis,
remaining storage diseases, polycystic liver disease, infiltration of
portal vein thrombosis [1]. The most common cause of post- liver hilus (independent of cause), benign and malignant neoplasms
hepatic PHT is Budd-Chiari syndrome, which results from b. Sinusoidal
obstruction of the hepatic veins or the inferior vena cava in Liver cirrhosis (independent of cause), acute viral and alcoholic
hepatitis, acute fatty liver of pregnancy
the area of these veins [2]. The etiology for portal hypertension c. Post-sinusoidal
differs in various parts of the world. For example, in Africa the Venous occlusion disease, alcoholic hyaline sclerosis of central veins
3. Extrahepatic
most common cause of PHT is schistosomiasis, which begins Hepatic vein thrombosis (Budd-Chiari disease), inflammatory/
as a pre-sinusoidal disease within the liver and may extend neoplastic infiltration of the hepatic veins, caval inferior occlusion
(thrombosis, neoplasms), cardiac diseases: chronic right ventricular
failure, chronic constrictive pericarditis, tricuspid insufficiency
PHT = portal hypertension

109
REVIEWS IMAJ VOL 15 February 2013

Sinusoidal PHT: Elevated wedge pressure and normal Systemic circulation: Most patients with cirrhosis and por-
free hepatic venous pressure (high HVPG) tal hypertension have a hyperdynamic circulation, defined
Post-sinusoidal PHT: Elevated wedge and free hepatic by a high cardiac output and index, low blood pressures,
vein pressure (normal HVPG). high blood volume but with low effective blood volume,
and a low systemic vascular resistance [5]. Due to the low
The next section will focus on portal hypertension in effective blood volume, volume receptors are stimulated
patients with liver cirrhosis, the resultant complications, (despite high blood volume), which leads to neurohor-
especially esophageal varices, and the treatments that can be monal activation, salt and water retention in the renal
offered. system and, in many cases, ascites [6], and later on, renal
failure, probably due to vasoconstricted glomerular afferent
arterioles. Hepatic ascites can be infected spontaneously
From pathophysiology to the clinical syndrome by bacteria colonizing the intestine, causing spontaneous
Cirrhotic PHT is a vascular disease that affects many body bacterial peritonitis.
organs. The development of portal hypertension in patients Renal abnormalities can develop in the cirrhotic
with cirrhosis is a poor prognostic sign. The first event in the patient, especially if PHT already exists. There are many
cascade of portal hypertension development is increased vas- causes of renal failure in cirrhosis, but the most ominous
cular resistance to portal blood flow. The reason is not merely and lethal is the hepatorenal syndrome, which is a func-
mechanical; there is also a contributing dynamic component tional renal disease (at least in the beginning) due to an
that is created by contracting portal myofibroblasts, activated imbalance between vasodilating and vasoconstricting fac-
hepatic stellate cells and vascular tors in the systemic circulation. Due
smooth cells in portal venules caus- Portal hypertension is defined as to this imbalance, vasoconstrictive
ing elevated vascular resistance. As a an elevated portal pressure and is factors become dominant and lead to
result, the pressure gradient between the main cause of morbidity and afferent arteriole vasoconstriction in
the portal vein and the hepatic vein mortality in liver cirrhosis the renal glomeruli and deteriorating
is elevated to more than 5 mmHg. renal function that can lead to death
When the pressure gradient between the portal vein and the within days to weeks [7].
hepatic veins (HVPG) exceeds 10 mmHg, clinical pathology Cardiopulmonary system: Cirrhotic portal hypertensive
emerges. Such an elevation is defined as clinical significant patients often exhibit high cardiac output and index, as
portal hypertension. CSPHT already exists in about 60% of
3

mentioned before. In addition, a long QT interval may also

histologically proven, well-compensated cirrhotic patients [3]. appear as part of other electrophysiological disturbances
As a consequence of elevated liver resistance, systemic and tendency to cardiac arrhythmias. In later stages of
levels of vasoactive mediators are increased, including decompensated cirrhosis, especially if accompanied by hep-
vasodilators such as nitric oxide on one hand, and vasocon- atorenal syndrome, cardiac output may in fact decrease for
strictors such as thromboxane A2, endothelin, epinephrine, as yet unknown reasons [8]. These disturbances are called
angiotensin and renin on the other. The interaction between cirrhotic cardiomyopathy. In the pulmonary system, vaso-
these opposing vasoactive substances and their combined dilatation of pulmonary arterioles and shunting of blood
effect is responsible for some of the clinical phenomena to the left system, accompanied by ventilation/perfusion
caused by PHT: mismatch, is not uncommon. This phenomenon is defined
Splanchic vascular bed: Cirrhosis can cause splanchic as the hepatopulmonary syndrome. The opposite phenom-
vasodilatation, decreased responsiveness to vasoconstric- enon is the portopulmonary syndrome, characterized by
tors, and formation of new blood vessels (angiogenesis), pulmonary artery hypertension due to high pulmonary
which contributes to the increase in splanchic blood flow vascular resistance for which there is no explanation to
and portosystemic collaterals [4] including esophageal date. A possible cause is pulmonary endothelial dysfunction
and gastric varices. These varices tend to bleed when leading to vascular remodelling of the pulmonary system.
large enough or when portal pressure is high enough. Blood abnormalities: Leukopenia, thrombocytopenia
Bleeding esophageal or gastric varices are the final and and anemia are frequent findings in PHT patients, in part
most important clinical event resulting from portal because of splenomegaly and hypersplenism and, in the
hypertension, and will be discussed in detail further on. case of hemoglobin and platelets, because of low produc-
Portosystemic collaterals and shunting are also involved tion. It is well known that the thrombopoietin level is
in liver encephalopathy. reduced or inappropriately low in PHT patients, which
contributes to the thrombocytopenia that already exists
CSPHT = clinical significant portal hypertension due to the enlarged spleen and splenic sequestration [9].

110
IMAJ VOL 15 February 2013 REVIEWS

due to splanchic arteriolar vasodilatation, contributes to the

Diagnosis and Evaluation of Cirrhotic PHT elevated portal pressure. The reason for the splanchic arteriolar
HVPG measurement vasodilatation is probably increased release of some endog-
Since the practical definition of sinusoidal portal hyperten- enous vasodilators, among which nitric oxide is best known
sion is based on an elevated pressure gradient between the [13]. Vasodilatation of the splanchic system is accompanied, as
sinusoidal pressure and the hepatic vein pressure, measuring mentioned earlier, by hyperdynamic circulation, high cardiac
this gradient is the gold standard method of diagnosis. Apart output, low effective blood volume, hypervolemia, and salt
from being objective, HVPG serves as an independent tool and water reabsorbtion in the renal system, leading to ascites.
to assess prognosis in cirrhotic patients with varying clinical Therefore, it is not surprising that treatment of PHT can be
conditions [10]. Elevated HVPG above 6 mmHg indicates applied by mechanical TIPS (transjugular intrahepatic porto-
sinusoidal portal hypertension. Further elevation above 10 systemic shunt) or by pharmacological means (such as non-
mm Hg indicates CSPHT, as mentioned earlier. selective beta blockers, nitrates and diuretics). To complete
the therapeutic profile, an arm of endoscopic therapy, such
Endoscopy as variceal band ligation, even though it does not reduce the
There is a consensus that all cirrhotic patients should undergo portal pressure, provides another tool to control and prevent
an upper endoscopy as a screening tool for cirrhosis close to rebleeding from esophageal varices in cirrhotic patients.
the time of diagnosis. This allows the addition of prophylactic
therapy if esophageal or gastric varices are found. In a patient Prevention of varices
without varices on the initial assessment, endoscopy should At the time of diagnosis about 60% of cirrhotic patients have
be repeated approximately every 2 years since 510% of these esophageal varices of different grades. In patients without
patients will develop esophageal varices every year [3]. varices, the rate of developing esophageal varices is about
5% annually. This is the rationale for the recommendation to
Non-invasive clinical laboratory and imaging techniques screen for variceal varices every 23 years, and perhaps every
Splenomegaly is the most frequently reported clinical sign with year in high risk patients.
a correlation to portal hypertension. It can be found on physical Experimental models demonstrated the ability of beta block-
examination, in upper abdominal ers to delay and attenuate the
ultrasonography or computed The main treatments for esophageal development of collaterals [14].
tomography. CT and ultrasound varices are still beta blockers (with or In clinical terms, however, beta
can also recognize abdominal without nitrates) and esophageal variceal blockers did not succeed in reduc-
venous collaterals, esophageal band ligation, although carvedilol ing or delaying the appearance of
varices, liver nodularity, changes emerges as a treatment for esophageal esophageal varices. [1]. Thus, the
in size of the liver, enlarged por- use of beta blockers as a preven-
varices in Child A cirrhotic patients
tal vein diameter, and changes in tive treatment in the cirrhotic
portal vein flow velocity, which have been found to correlate patient without esophageal varices cannot be recommended.
with the existence of CSPHT [11]. Other measurements such as One approach that proved successful in preventing the
platelet count and a low platelets-to-spleen ratio probably reflect progression of cirrhosis is to treat the basic etiology (weight
portal hypertension. loss in non-alcoholic steatohepatitis, steroids in autoimmune

Transient elastography (FibroScan , France) is a rela-
tively new non-invasive technique developed to assess liver
hepatitis, copper chelators in Wilson disease, abstinence from
alcohol in alcoholic liver disease, and antiviral therapy in
fibrosis. It appears that FibroScan is a useful tool to reach, patients with hepatitis C virus infection). This proved suc-
or rule out, the diagnosis of CSPHT. Still, it is inaccurate in cessful in alcoholic cirrhosis where abstinence from alcohol
about 50% of patients with cirrhosis. Another disadvantage decreased the complications of cirrhosis, and with hepatitis C
of FibroScan is that there is a poor correlation between the virus cirrhosis where achieving a sustained virologic response
absolute Fibroscan result and the HVPG in patients with correlated with reducing the portal pressure [15].
CSPHT (HVPG > 10 mmHg) [12].
Primary prevention of variceal bleeding
From pathophysiology to treatment In patients with existing esophageal varices, the annual rate
As noted earlier, the first event in the cascade of portal hyper- of variceal bleeding is about 4%. Acute variceal bleeding is a
tension development is increased vascular resistance to portal medical emergency and a life-threatening event, with a mortal-
blood flow. This elevated vascular resistance can be treated by ity of about 25% within 6 weeks after the bleeding episode.
mechanical or pharmacological means. In addition, a second Most of the deaths are related to liver failure and secondary
component of increased blood flow through the portal system, complications such as hepatorenal syndrome, spontaneous

111
REVIEWS IMAJ VOL 15 February 2013

bacterial peritonitis and sepsis. Preventing or reducing the rate tic step in primary prevention of variceal bleeding, relying
of variceal bleeding was therefore a target for many trials using largely on the fact that about 25% patients cannot tolerate
pharmacological, mechanical and endoscopic treatments. beta blockers and that many do not take them without
Beta blockers act by decreasing HVPG and were shown informing their physician; thus, the number of patients
to attenuate the risk of variceal bleeding from 24% to 15% actually not treated with the drug may well be higher than
during a follow-up of 2 years [16]. The most studied beta we assume. Moreover, of those who do take beta blockers,
blockers were propranolol and timolol and both were found more than 30% are early non-responders in terms of reducing
to be effective. Mortality was reduced as well, but statistical HVPG [18], which reduces further the percentage of patients
significance was not achieved. who benefit from beta blockers.
In the past it was recommended that only patients with Combined treatment with EVL and beta blockers for pri-
medium to large varices be treated with beta blockers. Due mary prevention cannot be justified. A study that compared
to the fact that grading variceal size is subjective, and that EVL plus beta blockers to EVL alone showed a small advantage
small varices with red signs or in Child-Pugh class C cirrhotic over the combined treatment but at the cost of more side effects
patients have a similar bleeding rate as medium and large in the combined treatment arm [21]. Further studies are needed
varices, it is now recommended that beta blockers be started to clarify if there is a real advantage to this mode of combined
as a preventive therapy in these groups as well [17]. treatment for primary prevention of variceal bleeding.
It was recently shown in a hemodynamic trial that patients In the same context, there is an emerging debate on the
in whom HVPG decreased by 12% after receiving intravenous preferred primary prevention therapy for patients who are can-
propranolol (responders) are at low risk of suffering a first didates for liver transplantation. This population of patients is
episode of variceal bleeding or variceal rebleeding, compared unique, since their liver disease is usually more advanced on
with those not given propranolol (non-responders) [18]. There the one hand, meaning that the first variceal bleeding is a real
is still debate about which treatment strategy should be applied life-threatening event, probably with higher mortality. In addi-
for the non-responders to i.v. propranolol, who, as shown in tion, they might not need long-term medical treatment with
another trial, do not respond as well to EVL in terms of reducing beta blockers, if hopefully transplanted. Also, the fact that EVL
the bleeding rate. More trials are doesn not actually reduce portal
needed to answer this question. Transjugular intrahepatic portosystemic hypertension, (so, esophageal
The addition of nitrates shunt (TIPS) seems a rational treatment varices can theoreticaly develop
to beta blockers as a primary in a selective group of cirrhotic patients again) does not have a long-term
prevention was assessed in two with bleeding esophageal varices importance if the patient is sched-
studies but did not show any uled for transplantation soon.
advantage over beta blockers alone in reducing the rate of On the other hand, EVL-related esophageal bleeding from a
the first variceal bleeding in cirrhotic patients [19]. Therefore, variceal ulcer might worsen these patients medical condition
adding nitrates cannot be recommended for this indication. dramatically, even making them unacceptable candidates for
Endoscopic variceal band ligation, a procedure that liver transplantation. Some trials assessed the effectiveness of
includes banding varices with rubber rings endoscopically, EVL in this group, but with conflicting results [22,23]. The
was also the subject of numerous studies on preventing the question of EVL or beta blockers for primary prevention of
first variceal bleeding episode. EVL was compared with beta
4

variceal bleeding in this group of patients will likely remain an

blockers with regard to reducing the rate of first variceal open question, but it is suggested by the author that the time
bleeding events and was found as effective as beta blockers if spent on the waiting list for liver transplantation in each center
only large randomized well-designed studies with more than or country is an important consideration. In centers where the
100 patients in each study were included [20]. Interestingly, average waiting time for transplantation is long, beta blockers
patients treated with beta blockers suffered more side effects should be preferred, as in most cirrhotic patients. In centers
than those treated with EVL, but while adverse events of beta where the bioavailability of donor organs is high, both beta
blockers were almost always reversible and usually mild, blockers and EVL are considered reasonable options in view
those of EVL were much more serious, including recurrent of the short waiting time for transplantation.
bleeding from banding-related variceal ulcers, even leading Recently, carvedilol, a non-selective beta blocker with an
to death. This is the reason why EVL is not uniformly recom- alpha blocker effect as well, was studied and proved to be
mended by the author as a first-line treatment, unless beta more effective in reducing portal pressure than proprano-
blockers are not tolerated or contraindicated. Nonetheless, lol, in the set up of primary prophylaxis.. A study compar-
some physicians prefer band ligation as the first therapeu- ing carvedilol to EVL showed a lower rate of first variceal
bleeding with carvedilol and no difference in mortality [24].
EVL = endoscopic variceal band There is no study comparing carvedilol and beta blockers

112
IMAJ VOL 15 February 2013 REVIEWS

in terms of clinical endpoints, but it seems that treatment demonstrate superiority over conventional therapy with pro-
with carvedilol in patients with variceal bleeding is a rational pranolol and EVL in terms of survival and was associated with
choice, especially if begun at a low dosage in patients with higher encephalopathy rates when compared to conventional
Child-Pugh class A cirrhosis. treatment. Recent evidence shows that in a select group of cir-
Simvastatin emerges as a future drug for portal hyperten- rhotic patients with first variceal bleeding TIPS emerges as an
sion. Simvastatin was shown to reduce HVPG in patients acceptable treatment. When measuring the HVPG of cirrhotic
with portal hypertension in both primary and secondary patients with first esophageal variceal bleeding, a value > 20
prophylaxis. Interestingly, the portal hypertension reduction mmHg is significantly correlated with a high risk of short-
effect was achieved due to reduced hepatic resistance and not term rebleeding [29]. It was recently shown that when placed
to reduced hepatic blood flow [25]. A study evaluating the wihin the first 3 days after the bleeding episode TIPS reduced
effect of simvastatin on cirrhotic portal hypertension in terms rebleeding rates, improved 2 year survival, and was associated
of clinical endpoints is already ongoing. with a reduced risk of hepatic encephalopathy in a select group
of patients with cirrhosis Child-Pugh class B or C with acute
Secondary prevention of variceal rebleeding esophageal bleeding for 3 days after the bleeding episode [30].
Patients surviving the first variceal bleeding have a 60% chance This new evidence may shift the pendulum towards acceptance
of rebleeding in the next year. Since variceal bleeding is a life- of TIPS as a therapeutic procedure for the first esophageal
threatening event, with mortality of about 25% within the first bleeding episode in a select group of cirrhotic patients.
6 weeks, it is clear why a second prophylaxis is mandatory. Beta Finally, it is important to remember that the only defini-
blockers were assessed in many randomized control trials and tive therapy to prevent variceal rebleeding in a cirrhotic
were shown to reduce the rate of rebleeding and mortality [16]. patient is liver transplantation, so every patient with a history
It was also shown that the combined effect of beta blockers and of esophageal variceal bleeding should be evaluated, and if
nitrates is stronger in reducing portal pressure. For this rea- found suitable, should be offered this treatment option.
son, the combination of these drugs was assessed in two trials
on reducing the rebleeding rate, with conflicting results. No Treatment of acute esophageal variceal bleeding
doubt the data on this combination therapy are not well proven. Acute variceal bleeding is a medical emergency and a life-
Nonetheless, since this combination was shown to be more threatening event with a mortality of about 25%. The treat-
effective than sclerotherapy in reducing the rebleeding rate [26], ment of acute variceal bleeding should be a combination of
and since the adverse effects are not major, it is recommended treatments given by hepatologists, gastroenterologists, anes-
that this combination treatment after the first variceal bleeding thesiologists and infectiologists. A good cooperation between
be considered, if there are no contraindications. these experts might lead to a better outcome for the patient.
Sclerotherapy, the injection of a sclerosing agent into the The patient with variceal bleeding should be treated in the
varices, was shown to reduce the rebleeding rate but, due to intensive care unit. The first step is to restore the patients hemo-
unacceptable side effects such as esophageal stenosis, dys- dynamic status and ensure that major organs are not affected by
phagia and more, and due to the fact that endoscopic EVL the continuous bleeding. The patient should be given full oxy-
proved more potent in reducing the rebleeding rate than genation and, if unstable hemodynamically, should receive fluid
sclerotherapy [27], sclerotherapy is no longer recommended resuscitation. Anemia should be corrected to a level of about
as first-line endoscopic therapy in these patients. 25% hematocrit. A correction to a higher level is not recom-
The combined endoscopic and pharmacological treatment mended due to the risk of rebleeding in these patients because
with EVL and beta blockers has been shown to reduce the of volume overload. A nasogastric tube should be inserted into
rebleeding rate more than EVL alone [28]. For this reason, the stomach for drainage, as well as a urinary catheter to the
it is recommended that in a patient who survived the first bladder to measure urine output. The patient should be fully
variceal bleeding event, this combined treatment should be monitored and if possible should have a central vein catheter
initiated as a first-line therapy, with the addition of nitrates to assess his/her hemodynamic status accurately. Patients with
as well. The varices should be banded in a few separate gas- reduced conscious level are at risk of aspiration and should be
troscopy sessions until disappearance of all varices, or at least intubated and ventilated. Coagulation disturbances, which are
the not significant varices. The rate of banding sessions is a fairly common in these non-compensated cirrhotic patients,
question of debate, but it is accepted that an interval of 2 should be corrected with fresh frozen plasma.
weeks to 1 month is reasonable. There is an increased rate of infections in patients suffering
TIPS is a very effective way to reduce variceal rebleeding
5

from acute variceal bleeding, with pneumonia, spontaneous

after the first variceal bleeding. Until recently TIPS did not bacterial peritonitis and urinary tract infections being the most
common. Antibiotic therapy should be started on a preventive
TIPS = transjugular intrahepatic portosystemic shunt basis as it was shown to reduce the rate of infections, reduce

113
REVIEWS IMAJ VOL 15 February 2013

the rebleeding rate and improve survival in these patients [31]. possible by EVL after initial resuscitation. Several trials have
Usually quinolones are the antibiotics of choice, but in patients demonstrated that starting a vasoactive drug before endo-
with spontaneous bacterial peritonitis in the past or patients scopic therapy improves bleeding control and reduces the
with Child-Pugh class C, ceftriaxone should be introduced due rebleeding rate [36,37].
to its superiority [32]. Balloon tamponade is not an ideal therapeutic option and
About 50% of all esophageal variceal bleeding episodes should be used only in the event of massive bleeding. The
stop spontaneously [26], but the rebleeding rate is high with balloon achieves hemostasis in about 6090% of cases but
about 50% of patients experiencing a second episode, usu- cannot be in place for more than 24 hours. After the balloon
ally within 2 weeks after the first episode. A second episode is removed, bleeding starts in at least 50% of patients. In addi-
of bleeding puts the patient at a high mortality risk and is tion, balloon tamponade might cause severe side effects such
thus the reason for starting pharmacologic therapy as soon as as esophageal rupture. This is why this option is used only as
possible. Pharmacologic therapy in acute esophageal variceal a bridge until a more definitive therapy such as TIPS [38].
bleeding is based on the pathophysiology of portal hyperten- TIPS and surgical shunts are another option for a salvage
sion. The target is to reduce the portal pressure by reducing therapy in the resistant bleeder. TIPS is preferred over
the flow to the portal system. surgical therapy due to the advanced stage of cirrhosis in
Terlipressin, an analogue of vasopressin, has an effect of these patients. Still, it should be kept in mind that the rate
relatively selective splanchic vasoconstriction and is the drug of of encephalopathy after TIPS placement is high and, when
choice for this purpose, since it was shown to improve outcome inserting TIPS in a patient with Child-Hugh C cirrhosis,
in these patients in terms of bleeding control and survival [33]. liver failure almost always occurs. Thus, a liver transplanta-
Therapy with terlipressin should be started as soon as possible, tion should be planned for these patients, if possible, after
even if variceal bleeding is only suspected and not proven. The placing the TIPS.
starting dosage is 2 mg every 6 hours for the first 48 hours and Despite all of the above mentioned treatments for portal
can be reduced later to half the dose for the next 3 days. It is hypertensive complications in the cirrhotic patient, it is impor-
important to continue therapy for 5 days even after the bleeding tant to remember that the only definitive treatment for portal
has stopped, due to the high rate of rebleeding in these patients. hypertension due to liver cirrhosis is liver transplantation.
Side effects of terlipressin are usually minor. Rarely, peripheral
or myocardial ischemia and abdominal pain may occur. Corresponding author:
Dr. E. Ashkenazi
Another therapeutic option is to start somatostatin at a
Liver Unit, Carmel Medical Center, Haifa 34362, Israel
starting dose of 250 g followed by a 250 g/hour infusion, Phone: (972-4) 825-0052
which is maintained until 24 hours bleed free is achieved. email: [email protected], [email protected]
Treatment can be prolonged to 5 days to prevent rebleeding.
It is worth noting that despite the effect of somatostatin on References
bleeding control, no impact on survival was shown [16]. For 1. Valla DC, Condat B. Portal vein thrombosis in adults: pathophysiology,
pathogenesis and management. J Hepatol 2000; 32: 865-71.
this reason somatostatin is recommended as a second-line
2. Plessier A, Sibert A, Consigny Y, et al. Aiming at minimal invasiveness as a
therapy for this condition. therapeutic strategy for Budd Chiari syndrome. Hepatology 2006; 44: 1308-16.
Octreotide, a somatostatin analogue, is another option 3. Groszman RJ, Garcia-Tsao G, Bosch J, et al. Beta blockers to prevent
for second-line therapy for variceal bleeding. The dose is an gastroesophageal varices in patients with cirrhosis. N Engl J Med 2005; 353:
2254-61.
initial bolus of 50 g, followed by 25 g/hour for up to 5
4. Fernandez M, Vizzuti F, Garcia Pagan JC, Rodes J, Bosch J. Anti VEGF
days. The efficacy of octreotide to stop variceal bleeding is receptor-2 monoclonal antibody prevents portal-systemic collateral vessel
controversial but was superior to sclerotherapy in preventing formation in portal hypertensive mice. Gastroenterology 2004; 126: 886-94.
early rebleeding [34]. 5. Bosch J, Abraldes JG, Groszman RJ. Current management of portal
hypertension. J Hepatol 2003; 38: S54-68.
EVL and sclerotherapy were both validated as effective
6. Schrier RW, Arroyo V, Bernardi M, Eppstein M, Henriksen JH, Rodes J.
treatments for the control of esophageal variceal bleeding. EVL Peripheral arterial vasodilatation hypothesis: a proposal for the initiation of
was more effective than sclerotherapy [35] and is associated renal sodium and water retention in cirrhosis. Hepatology 1988; 8: 1151-7.
with fewer side effects. Therefore, EVL should be the first- 7. Gines P, Guevara M, Arroyo V, Rodes J. Hepatorenal syndrome. Lancet 2003;
362: 1819-27.
line endoscopic therapy. Sclerotherapy can be a therapeutic
8. Ruiz-del-Arbol L, Monescillo A, Arocena C, et al. Circulatory function and
option in cases where EVL is not available. Endoscopy should hepatorenal syndrome in cirrhosis. Hepatology 2005; 42: 439-47.
be performed early after admission to hospital, assuring that 9. Rois R, Sangro B, Herrero I, Quiroga J, Prieto J. The role of thrombopoietin in
the patient was resuscitated and is stable hemodynamically. thrombocytopenia of patients with liver cirrhosis. Am J Gastroenterol 2005;
100: 1311-16.
The recommendation for the initial treatment is to com-
10. Bosch J, Abraldez JG, Berzigotti A, Garcia-Pagan JC. The clinical use of
bine medical and endoscopic therapy, starting terlipressin HVPG measurements in chronic liver disease. Nat Rev Gastroenterol Hepatol
immediately on admission to hospital, followed as soon as 2009; 6: 573-82.

114
IMAJ VOL 15 February 2013 REVIEWS

11. Zoli M, Iervese T, Merkel C, et al. Prognostic significance of portal hemody- bleed. Hepatology 2009; 50: 825-33.
namics in patients with compensated cirrhosis. J Hepatol 1993; 17: 56-61. 25. Abraldes J.G, Abillos A, Banares R, et al. Simvastatin lowers portal pressure
12. Vizutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement predicts in patients with cirrhosis and portal hypertension: a randomized control trial
severe portal hypertension in patients with HCV related cirrhosis. Hepatology Gastroenterology 2009; 136: 1651-8.
2007; 45: 1290-7. 26. Villanueva C, Balanzo J, Novella MT, et al. Nadolol plus isosorbide
13. Pizcueta MP, Pique JM, Bosch J, Whittle BJ, Moncada S. Effects of inhibiting mononitrate compared with sclerotherapy for the prevention of variceal
nitric oxide biosynthesis on the systemic and splanchic circulation of rats bleeding. N Engl J Med 1996; 334: (25): 1624-9.
with portal hypertension. Br J Pharmacol 1992; 105: 184-90.
27. Garcia-Pagan JC, Bosch J. Endoscopic band ligation in the treatment of
14. Sarin SK, Groszman RJ, Mosca PG, et al. Propranolol ameliorates the portal hypertension. Nat Clin Pract Gastroenterol Hepatol 2005; 2: 526-35.
development of porto-systemic shunting in a chronic murine schistosomiasis
28. Lo GH, Lai KH, Cheng JS, et al. Endoscopic variceal ligation plus nadolol
model of portal hypertension. J Clin Invest 1991; 87 (3): 1032-6.
and sucralfate compared with ligation alone for the prevention of variceal
15. Rincon C, Ripoli C, Lo lacono O, et al. Antiviral therapy decreases hepatic rebleeding: a prospective randomized trial. Hepatology 2000; 32: 461-5.
venous pressure gradient in patients with chronic hepatitis C and advanced
29. Moithino E, Escorsell A, Bandi JC, et al. Prognostic value of early measurements
fibrosis. Am J Gastroenterol 2006; 101 (10): 2269-74.
of portal pressure in acute variceal bleeding. Gastroenterology 1999; 117: 626-31.
16. Damico G, Pagliaro L, Bosch J. Pharmacological treatment of portal hyper-
tension: an evidence based approach. Semin Liver Dis 1999; 19 (4): 475-505. 30. Garcia-Pagan JC, Caca K, Bureau C, et al. Early use of TIPS in patients with
cirrhosis and variceal bleeding. N Engl J Med 2010; 362: 2370-9.
17. De Francis R. Revising consensus in portal hypertension: Report of the
Baveno V consensus workshop on methodology of diagnosis and therapy in 31. Bernard B, Grange JD, Khac EN, Amiot X, Opolon P, Poynard T. Antibiotic
portal hypertension. J Hepatol 2010; 53: 762-8. prophylaxis for the prevention of bacterial infections in cirrhotic patients with
gastrointestinal bleeding: a meta-analysis. Hepatology 1999; 29: 1655-61.
18. La Mura V, Abraldes JG, Raffa S, et al. Prognostic value of acute hemodynamic
response to i.v. propranolol in patients with cirrhosis and portal hypertension. 32. Farnandez J, Ruiz-Del-Arbol L, Gomez C, et al. Norfloxacin vs ceftriaxine
J Hepatol 2009; 51 (2): 279-87. in the prophylaxis of infections in patients with advanced cirrhosis and
hemorrhage. Gastroenterology 2006; 131: 1049-56.
19. Garcia-Pagan JC, Morillas R, Banares R, et al. Propranolol plus placebo
versus propranolol plus isosorbide-5-mononitrate for prevention of first 33. Ioannou GN, Doust J, Rockey DC. Systemic review: terlipressin in acute
variceal bleed: a double blind RCT. Hepatology 2003; 37 (6): 1260-6. esophageal variceal hemorrhage. Aliment Pharmacol Ther 2003; 17: 53-64.
20. Gludd LL, Klingenberg S, Nikolova D, Gludd C. Banding ligation versus beta- 34. Corley DA, Cello JP, Adkisson W, Ko WF, Kerlikowske K. Octreotide for acute
blockers as primary prophylaxis in esophageal varices: systemic review of esophageal bleeding: a meta-analysis. Gastroenterology 2001; 120: 946-54.
randomised trials. Am J Gastroenterol 2007; 102 (12): 2842-8; quiz 2841, 2849. 35. Lo GH, Lai KH, Cheng JS, et al. Emergency banding ligation versus
21. Sarin SK, Wadhawan M, Agarwal SR, Tyagi P, Sharma BC. Endoscopic variceal sclerotherapy for the control of active bleeding from esophageal varices.
ligation plus propranolol versus endoscopic variceal ligation alone in primary Hepatology 1997; 25: 1101-4.
prophylaxis of variceal bleeding. Am J Gastroenterol 2005; 100 (4):797-804. 36. Levacher S, Letoumelin P, Pateron D, Blaise M, Lapandry C, Pourriat JL.
22. Lim EJ, Gow PJ, Angus PW. Endoscopic variceal ligation for primary prophylaxis Early administration of terlipressin plus glycerol trinitrate to control active
of esophageal variceal hemorrhage in pre-liver transplant patients. Liver Transpl upper gastrointestinal bleeding in cirrhotic patients. Lancet 1995; 346: 865-8.
2009; 15 (11): 1508-13. 37. Villanueva C, Ortiz J, Sabat M, et al. Somatostatin alone or combined with
23. Norberto L, Polese L, Cillo U, et al. A randomized study comparing ligation emergency sclerotherapy in the treatment of acute esophageal variceal
with propranolol for primary prophylaxis of variceal bleeding in candidates bleeding: a prospective randomized trial. Hepatology 1999; 30: 384-9.
for liver transplantation. Liver Transpl 2007; 13 (9): 1272-8. 38. Goykhman Y, Ben Haim M, Rosen G, et al. Transjugular intrahepatic
24. Tripathy D, Ferguson JW, Kochar N, et al. Randomised controlled trial of portosystemic shunt: current indications, patient selection and results. IMAJ
carvedilol versus variceal band ligation for the prevention of the first variceal Isr Med Assoc J 2010; 12 (11): 687-91.

Capsule

Lymph node T cell responses predict the efficacy of live attenuated SIV vaccines
Live attenuated simian immunodeficiency virus (SIV) vaccines is associated with persistent LAV replication in the lymph
(LAVs) remain the most efficacious of all vaccines in non- node, which occurs almost exclusively in follicular helper T
human primate models of human immunodeficiency virus and cells. Thus, effective LAVs maintain lymphoid tissue-based,
AIDS, yet the basis of their robust protection remains poorly effector-differentiated, SIV-specific T cells that intercept and
understood. Fukazawa et al. show that the degree of LAV- suppress early wild-type SIV amplification and, if present in
mediated protection against intravenous wild-type SIVmac239 sufficient frequencies, can completely control and perhaps
challenge strongly correlates with the magnitude and function clear infection, an observation that provides a rationale for
of SIV-specific, effector-differentiated T cells in the lymph node the development of safe, persistent vectors that can elicit and
but not with the responses of such T cells in the blood or with maintain such responses.
other cellular, humoral and innate immune parameters. The Nature Med 2012; 18: 1637
authors found that maintenance of protective T cell responses Eitan Israeli

There may be times when we are powerless to prevent injustice, but there must never
be a time when we fail to protest
Elie Wiesel (born 1928), Romanian-born Jewish-American writer, professor, political
activist, Nobel Laureate, and Holocaust survivor

115