HP - Patogenesis y Diagnostico
HP - Patogenesis y Diagnostico
HP - Patogenesis y Diagnostico
a,b c,d,
Laura Turco, MD , Guadalupe Garcia-Tsao, MD *
KEYWORDS
Cirrhosis Gastroesophageal varices Bleeding Ascites
Hepatic venous pressure gradient Liver stiffness
KEY POINTS
Portal hypertension is defined as an increase in portal pressure at any level in the portal
venous system. The site of increased resistance or obstruction is the basis of its
classification.
The most common cause of portal hypertension is cirrhosis in which the site of increased
resistance is intrahepatic, leading to an increase in portal pressure that, in turn, leads to
splanchnic vasodilation, increase in portal blood inflow, and a hyperdynamic circulation
that further aggravates portal hypertension.
The presence of varices (on endoscopy) and/or other abdominal portosystemic collaterals
(on imaging) establish the diagnosis of portal hypertension.
The cause of portal hypertension can be established by a combination of noninvasive and
imaging tests and, when uncertainty remains regarding its cause or severity (in cirrhosis),
measurement of sinusoidal pressure via catheterization of the hepatic vein and determina-
tion of the hepatic venous pressure gradient will be essential.
INTRODUCTION
Grants or Financial Support: L. Turco is a recipient of a grant from the University of Modena
and Reggio Emilia (PI Prof Filippo Schepis and Prof Erica Villa). This review was partially sup-
ported by the Yale Liver Center NIH P30 DK34989.
Conflicts of Interest: The authors disclose no conflicts.
a
Division of Gastroenterology, Azienda Ospedaliero-Universitaria di Modena, University of
Modena and Reggio Emilia, Via del Pozzo 71, Modena 41125, Italy; b PhD Program in Clinical
and Experimental Medicine, University of Modena and Reggio Emilia, Via del Pozzo n 71,
Modena 41125, Italy; c Section of Digestive Diseases, Yale School of Medicine, Yale University
School of Medicine, PO Box 208056, 333 Cedar Street -1080 LMP, New Haven, CT 06520-8056,
USA; d Section of Digestive Diseases, VA Connecticut Healthcare System, 950 Campbell Avenue,
West Haven, CT 06516, USA
* Corresponding author. Yale University School of Medicine, PO Box 208056, 333 Cedar Street
-1080 LMP, New Haven, CT 06520-8056.
E-mail address: [email protected]
vein thrombosis and at the hepatic sinusoids in the case of cirrhosis) and the inferior
vena cava.
The most common cause of PH is cirrhosis. Normal PPG values range between 1 to
5 mm Hg and values greater than 5 mm Hg indicate the presence of PH. In cirrhosis,
values greater than 5 but less than 10 mm Hg are considered mild or nonclinically sig-
nificant, whereas values greater than or equal to 10 mm Hg are considered clinically
significant because they are associated with the development of gastroesophageal
varices (GEVs) and clinical events that define decompensation (ascites, variceal hem-
orrhage, encephalopathy).1 As proof-of-concept, reductions in portal pressure
induced by pharmacologic therapy in patients with both compensated and decom-
pensated cirrhosis,2 or by placement of the transjugular intrahepatic portosystemic
shunt (TIPS) in patients with decompensated cirrhosis,3,4 have resulted in a reduction
in the development of decompensating events and/or in an improvement in survival.
PATHOGENESIS
As in any vascular system, pressure in the portal venous system follows dynamic
Ohm’s law in which pressure gradient (DP) is the result of the amount of blood flow
circulating through system (Q) and the resistance opposing this flow (R)
(DP 5 Q R). In all causes of PH, the primum movens is an increase in resistance
to portal blood flow followed by an increase in portal venous inflow.5 The site of
increased resistance is the basis of the classification of PH (see later discussion). In
the case of portal vein thrombosis (the second most common cause of PH), an
obstruction in the main portal vein (thrombus) is the initiating event and pathogenesis
is therefore clearer. Because cirrhosis is the main cause of PH and its pathogenesis is
more complex (Fig. 1), this article refers to the pathogenesis of PH in cirrhosis.
Fig. 1. Pathogenesis of PH. Na, Sodium; NHs, neurohumoral system; RAAs, renin-
angiotensin-aldosterone system.
Portal Hypertension 575
Architectural disruption
Progression of chronic liver disease of any cause contributes to protean changes in
the hepatic sinusoids,5 mostly due to phenotypic and functional alterations in hepatic
stellate cells (HSCs) and sinusoidal endothelial cells (SECs). HSCs respond to liver
injury6 with an activation that consists of a switch in activity from vitamin A production
or storage to matrix deposition in the hepatic sinusoids. Additionally. SECs, which nor-
mally contain fenestrae, also respond to injury by losing fenestrae, leading to capilla-
rization of the sinusoids by deposition of basement membrane. Capillarization may act
as a very early player both in increasing intrahepatic resistance and in promoting
fibrosis formation.6 Changes also occur at the level of the plasma membrane of hepa-
tocytes adjacent to the sinusoid that can lose their microvilli as a response to liver
injury.
The role of angiogenesis consists of the growth and proliferation of endothelial cells
that contribute to increased intrahepatic resistance and fibrosis.7 Angiogenic endo-
thelial cells may stimulate HSC activation through the vascular endothelial growth fac-
tor (VEGF), a potent angiogenic molecule.7
These changes and cross-talk between HSCs and the SECs, with resultant matrix
deposition and narrowing of the sinusoids, account for the fixed increase in intrahe-
patic resistance.
Intrahepatic microthrombi
The traditional view of cirrhosis as a prohemorrhagic condition has recently changed,
recognizing that cirrhosis is both a prohemorrhagic and prothrombotic entity in a very
precarious equilibrium.10 Several data indicate a close relationship between pro-
thrombotic status and worsening of hepatic fibrosis and PH, both in experimental
and clinical models.11 In patients with cirrhosis, the thrombotic occlusion of intrahe-
patic veins and sinusoids causes intimal fibrosis and vein obstruction, leading to
parenchymal extinction defined as the loss of contiguous hepatocytes and their
replacement by fibrous tissue.11 Moreover, the development of thrombi in the hepatic
sinusoids can activate HSCs by protease-activated receptors, responsible for extra-
cellular matrix deposition.11 These theories are supported by studies showing that
anticoagulation leads to a reduction in hepatic fibrosis and PH.12,13
Splanchnic Vasodilation
The increased intrahepatic vascular resistance leads to an increased pressure in the
portal vein system that induces shear stress in the splanchnic vessels and the release
of vasodilators such as NO.14 Consequent splanchnic arterial vasodilation15 is the
core factor in the progression and worsening of PH, leading to the development of
576 Turco & Garcia-Tsao
PH is caused by an obstruction of portal blood flow that can occur in different portions
of the portal venous system. The pressure proximal to the obstructed portion will be
hypertensive, whereas, distal to the obstruction, pressure will be normal. Based on
the site of obstruction, PH is classified in prehepatic (if the obstruction occurs in the
splenic vein, portal vein before the liver, or mesenteric vein; ie, before reaching the
liver), intrahepatic (if the obstruction is at the liver), or posthepatic (if the obstruction
involves the hepatic venous outflow; ie, after the liver). Intrahepatic PH is subclassified
according to the site of the obstruction within the hepatic sinusoid as presinusoidal
(portal-based illnesses; eg, schistosomiasis), sinusoidal (cirrhosis), or postsinusoidal
(illnesses affecting the central vein).
As shown in Table 1, different types of PH are caused by different etiological factors
and have different clinical presentations. In adults, cirrhosis accounts for more
than 90% of cases of PH, whereas the second most common cause is portal vein
thrombosis.
Because liver function and sinusoidal pressure are normal in patients with prehe-
patic or presinusoidal types of PH, ascites is not a main clinical feature of these entities
and they are, in general, associated with a better prognosis than sinusoidal and post-
sinusoidal or posthepatic causes of PH.
In the remaining types of PH (sinusoidal, postsinusoidal, posthepatic), patients may
develop GEVs and variceal bleeding; however, ascites development is a sine qua non
and may be accompanied by liver insufficiency.
Table 1
Portal hypertension classification
Clinical Hemodynamic
Type Most Common Causes Presentation Characteristics
Prehepatic Portal vein occlusion Splenomegaly Normal HVPG
(thrombosis or GEVs Normal WHVP
neoplasm) Collaterals Normal FHVP
Splenic vein occlusion Variceal bleeding
(thrombosis or
neoplasm)
Portal vein stenosis
Intrahepatic
Presinusoidal Schistosomiasis Splenomegaly Normal HVPG
Primary biliary cholangitis GEVs Normal or slightly
(early stages) Collaterals elevated WHVP
Primary sclerosing Variceal bleeding Normal FHVP
cholangitis
Focal nodular hyperplasia
Idiopathic PH
Sarcoidosis
Sinusoidal Cirrhosis (viral, alcoholic, Splenomegaly Elevated HVPG
NASH-related) GEVs Elevated WHVP
Alcoholic hepatitis Collaterals Normal FHVP
Primary biliary cholangitis Variceal bleeding
(advanced stages) Ascites
Hepatic
encephalopathy
Postsinusoidal Venoocclusive disease Splenomegaly Normal HVPG
Collaterals Elevated WHVP
Ascites Elevated FHVP
Posthepatic
Vascular Hepatic vein thrombosis Ascites Not possible to
obstruction (Budd-Chiari syndrome) Intrahepatic catheterize hepatic
collaterals veins
Liver Chronic right heart Ascites Normal HVPG
congestion failure Elevated WHVP
Chronic constrictive Elevated FHVP
pericarditis
Restrictive
cardiomyopathy
Tricuspid insufficiency
Abbreviations: FHVP, free hepatic venous pressure; HVPG, hepatic venous pressure gradient; NASH,
nonalcoholic steatohepatitis; WHVP, wedged hepatic venous pressure.
investigations to diagnose cirrhosis and, if negative, imaging studies to rule out portal
vein thrombosis). Varices due to PH are larger in the lower esophagus and are of pro-
gressively smaller size as the endoscope is withdrawn. Varices that are larger in the
upper esophagus and smaller at the level of the lower esophageal sphincter (downhill
varices) are not due to PH but may be due to superior vena cava obstruction, which is
most commonly caused by tumors or thrombosis from vascular catheterization.
In patients with cirrhosis, finding GEVs establishes the presence of CSPH because it
has been shown that patients with GEVs have an hepatic venous pressure gradient
(HVPG) of at least 10 to 12 mm Hg.21
Patients with compensated cirrhosis who require prophylaxis to prevent bleeding
are those with high-risk varices (HRVs), defined as medium or large GEVs or any
size varices with red wale marks. In patients with decompensated cirrhosis, GEVs
of any size require prophylaxis.21
Compensated patients without GEVs at screening EGD should have it repeated
every 2 to 3 years according to the presence or not of ongoing liver injury (abstinence
in alcoholics, viral elimination in viral cirrhosis, absence of cofactors such as obesity).
In compensated patients with small varices at screening endoscopy, EGD should be
repeated every 1 to 2 years according to the presence or not of ongoing liver injury.
Decompensated patients without GEVs should have an EGD repeated every year.
In the last decade, a growing number of studies have suggested that endoscopic
ultrasound (EUS) may be useful in the evaluation of PH. Paraesophageal varices
and the left gastric vein can be better visualized by EUS, which not only correlates
with variceal size22 but seems to predict variceal recurrence after ligation or sclero-
therapy.23 Moreover, it has been recently shown in animal models24 and in humans25
that PPG can be measured directly by placing a needle through the gastric or
duodenal wall into the portal vein and then into the right hepatic vein. This method
seems to be safe and correlates well with pressure values obtained by the transjugular
approach. Although it has the potential of screening for varices and measuring portal
pressure (useful in those without varices) during the same procedure, it requires
further evaluation.
Hepatic venous pressure gradient The HVPG is the most commonly used indirect
method to assess portal pressure and is considered the gold standard. HVPG is
calculated by subtracting the free hepatic venous pressure (FHVP), a measure of sys-
temic pressure, from the wedged hepatic venous pressure (WHVP), a measure of he-
patic sinusoidal pressure. It was first described in 1951,29 when a small catheter was
advanced into the hepatic vein until it could not be advanced further; that is, the
catheter was wedged into the most distal portion of the hepatic vein branch. The
catheter thus wedged measures the pressure in the hepatic sinusoid, which in
cirrhosis is a measure of main portal vein pressure. In fact, it has been shown that
WHVP correlates very closely (r 5 0.95) with the pressure measured by direct cath-
eterization of the main portal vein in alcoholic and viral cirrhosis.30 The advantage is
that measures of the pressure when the catheter is free in the hepatic vein allows a
systemic pressure that acts as an internal zero and corrects for extravenous factors
such as ascites.
In 1979, Groszmann and colleagues31 described a method that involved the use of a
balloon catheter instead of a straight catheter that would occlude rather than wedge
the hepatic vein and, by inflating the balloon, repeated measures of free and occluded
pressure could be obtained without the need to move the catheter.
Normal HVPG values are between 3 and 5 mm Hg. Values above 5 mm Hg indicate
PH. In a patient with chronic liver disease, the presence of PH should be diagnostic of
cirrhosis. However, there is a percentage of patients (w10%)32 with histologic
cirrhosis with a normal HVPG. The reason for this is unclear and may be related to
venovenous shunts or to a poor HVPG technique. As previously mentioned, an
HVPG equal or greater than 10 mm Hg in patients with compensated cirrhosis without
GEVs predicts development of GEVs, clinical decompensation, and hepatocellular
carcinoma.1 In compensated patients with CSPH (with or without varices), an HVPG
greater than 16 mm Hg predicts clinical decompensation,16 whereas in patients
with acute variceal hemorrhage, an HVPG greater than 20 mm Hg33,34 predicts vari-
ceal rebleeding, treatment failure, and higher mortality.
As mentioned previously, in patients with prehepatic or presinusoidal PH, in which
pressure in the sinusoids is normal, the HVPG will be normal or only slightly elevated
(see Table 1). In patients with posthepatic PH, mainly in those with heart failure, HVPG
will also be normal because the hepatic sinusoidal pressure is high as a reflection of
high systemic pressure (see Table 1).
Other than for diagnostic purposes, HVPG is not widely used and is not recommen-
ded in the clinical evaluation of patients with cirrhosis.21 However, its use is essential in
proof-of-concept studies investigating drugs with a potential portal pressure lowering
effect.14
Noninvasive Tests
Determining the presence of GEVs is essential in patients with cirrhosis and deter-
mining the severity of PH is useful in their management and in discussions regarding
prognosis. However, EGD is not free of risk and performance of HVPG is generally
limited to specialized centers (and specialized providers) because it is a nuanced pro-
cedure that can lead to erroneous values in inexperienced hands. Therefore, many
studies have looked for NITs to determine the presence of varices or PH.
and has been very useful in staging liver fibrosis in patients with chronic liver
disease.35
It has also been used to assess the presence of varices or degree of PH in patients
with cirrhosis.
Correlation with hepatic venous pressure gradient Overall, a good correlation has
been reported between HVPG and liver stiffness measurement (LSM) in patients
with advanced liver fibrosis or cirrhosis (r 5 0.55–0.86).35
The best correlation between HVPG and LSM occurs with HVPG values between 5
and 10 to 12 mm Hg (r 5 0.91), which is the range described for mild PH, the stage
driven by increase in intrahepatic resistance.36 The correlation persists but decreases
markedly with levels above 12 mm Hg (r 5 0.17) when increased hepatic blood flow is
an additional factor contributing to PH (CSPH).36
Ability to identify the presence of clinically significant portal
hypertension Importantly, LSM can discriminate between patients with and without
CSPH with an area under the receiver operating characteristic (AUROC) curve ranging
from 0.82 to 0.94.35 Cut off values of LSM greater than 13.6 kPa36 or LSM greater than
21 Kpa35 have been found to have a 90% sensitivity and a 90% specificity in the diag-
nosis of CSPH. Considering all studies, which were mostly performed in subjects with
hepatitis C virus (HCV)-related cirrhosis, the most recent Baveno VI Consensus Con-
ference recommended that an LSM greater than or equal to 21 kPa be used rule in
CSPH in these subjects.37
Ability to identify the presence high-risk varices Many studies have aimed at identi-
fying patients with HRVs. A multicenter large cohort study provided the most validated
criteria38 that led to the Baveno VI consensus recommendation to consider patients
with an LSM less than 20 kPa and a platelet count greater than 150.000/mm3 as being
very unlikely to have HRV.37 With these criteria, 21% EGDs could be avoided and only
less than 5% patients with HRV would be missed. Other criteria have been proposed
in studies that have included subjects with all causes of cirrhosis and in studies
including subjects with specific causes of cirrhosis (Table 2). Notably, these studies
included mostly subjects with HCV-related cirrhosis and were carried out before the
availability of effective antiviral therapy. Because viral elimination has been shown
to decrease LSM and increase platelet count,46 concerns have been raised regarding
the applicability of these criteria in the setting of patients with HCV-related cirrhosis
who have attained sustained virological response (SVR). However, a recent study per-
formed in 94 subjects with HCV-related cirrhosis who achieved SVR and in 98 treated
subjects with hepatitis B virus (HBV)-related cirrhosis, demonstrated that Baveno VI
criteria can also be applied to these subjects, showing that 25% of EGDs could
have been avoided and only 1% subjects with HRV would have been missed.45
Spleen stiffness measurements
With the development of PH, there is a progressive increase in spleen size due not only
to backflow into the spleen but also due to hyperplasia, increased angiogenesis, and
fibrogenesis of the spleen that increases its stiffness.35 Spleen stiffness measurement
(SSM) can be detected by TE and has been used to assess the presence of varices or
degree of PH in patients with cirrhosis. Importantly, SSM seem to be useful in prehe-
patic and presinusoidal causes of PH.35
Correlation with hepatic venous pressure gradient In a cohort of subjects with HCV-
related cirrhosis, spleen stiffness was found to correlate significantly with HVPG
(multivariate coefficient [r2] 5 0.78).47
Table 2
Noninvasive criteria to rule out high-risk varices in compensated cirrhosis
Portal Hypertension
XL probe)
Moctezuma-Velazquez 227 PBC and PSC 13% LSM <20 kPa and PLT 30%–40% 0%
et al,44 2019 >150.000/mm3
Thabut et al,45 2019 200 HCV and HBV 8% LSM <20 kPa and PLT 25% 1%
after SVR >150.000/mm3
Abbreviations: EGDs, esophagogastroduodenoscopies; HBV, hepatitis B virus; M, medium; MELD, model for end-stage liver disease; PBC, primary biliary cholan-
gitis; PLT, platelet (count); PSC, primary sclerosing cholangitis; SSM, spleen stiffness measurement; SVR, sustained viral response; XL, extra large; US, United States.
581
582 Turco & Garcia-Tsao
Ability to identify the presence high-risk varices SSM less than our equal to 46 kPa
rules out HRV with a high sensitivity (98%).42 As shown in Table 2, a stepwise
approach of Baveno VI and SSM criteria increases the number of subjects (up to
44%) in whom EGD can be safely avoided while missing only less than 5% of the sub-
jects with HRV.
One of the main limitations of SSM by TE is the high rate of unsuccessful examina-
tions (15%–20%), which can be avoided by using ultrasound guidance, which is the
case with point shear wave elastography when using acoustic radiation force impulse
(ARFI). SSM by ARFI has been shown to correlate with various markers of PH, such as
GEVs,35 HVPG,35 and HVPG changes pre-TIPS and post-TIPS.35 It has also been
shown to predict decompensation (cutoff 3.25 m/s) and death (cutoff 3.43 m/s).35
However, the number of subjects studied is not large and it has not been used to iden-
tify HRVs.
Combination of tests
Combinations of NITs have been proposed to detect the presence and severity of PH
in patients with cirrhosis. A combination of LSM by TE, platelet count and spleen diam-
eter (calculated as LSM spleen size or platelet count [LSPS]) at cutoffs of 1.72 and of
3.21 correctly identifies patients with CSPH and GEVs, respectively, with good sensi-
tivity and specificity (both >80%).49 In the ANTICIPATE study,38 LSPS showed an
excellent ability in identifying HRV (AUROC 5 0.79) with a cutoff of 1.33 avoiding
26% of EGDs. Platelet count or spleen ratio was found having good power in predict-
ing the presence of HRV (cutoff of 1.64 with AUROC 5 0.74),38 with the advantage of
not requiring TE, technology that is not widely available. Other criteria that have been
proposed in the absence of TE is the combination of MELD and platelet count that was
able to avoid up to 54% endoscopies while missing only 1% of patients with HRV38
(see Table 2).
rule out (100% sensitivity) and rule in (73% specificity) the presence large varices.51
The number of subjects included in these studies is not large and the technique is
not well established or standardized in most centers.
ICG-r15 together with HVPG measurement and GEVs have been shown to be inde-
pendent predictors of decompensation in patients with compensated cirrhosis.52
However, in Child B and C patients, in which the ICG clearance could be influenced
by the progressive deterioration of liver function, ICG-r15 shows less accuracy in pre-
dicting CSPH than in Child A patients.53
MRI
Liver architecture, perfusion of the liver and spleen, and blood flow in the splenic artery
determined by MRI have been shown to correlate with portal pressure assessed by
HVPG.
In a study performed in subjects with advanced fibrosis and cirrhosis (60%), liver
longitudinal relaxation time and splenic artery velocity were found to correlate signif-
icantly with HVPG (r 5 0.90) and with CSPH (r 5 0.85).54 MRI could be integrated with
MR elastography of both the liver and the spleen. SSM by MRI can identify patients
with cirrhosis and an HVPG greater than 12 mm Hg (AUROC 5 0.81) and with HRV
(AUROC 5 0.93).55
These exploratory studies need to be confirmed in a larger number of subjects and
issues of cost and feasibility (TE being a point-of-care test) require evaluation.
SUMMARY
Patients with PH are identified by the presence of varices or variceal hemorrhage and/
or by the presence of portosystemic collaterals on imaging (Fig. 2). Because the main
cause of PH is cirrhosis, clinical tests, imaging studies, and LSMs should be first used
to rule out or rule in cirrhosis. If negative, imaging studies to rule out portal vein throm-
bosis should be performed and, if these are negative, HVPG measurements should be
PORTAL HYPERTENSION
(paent without ascites, VH, HE)
Yes No
performed with simultaneous liver biopsy. In cirrhosis, NITs can safely rule out the
presence of HRVs and avoid endoscopy; however, in general, EGD is still necessary
to screen for varices. The presence of varices or collaterals establishes the presence
of CSPH. In the absence of varices, only HVPG can accurately determine the presence
of CSPH.
REFERENCES
19. Taura K, De Minicis S, Seki E, et al. Hepatic stellate cells secrete angiopoietin 1
that induces angiogenesis in liver fibrosis. Gastroenterology 2008;135:1729–38.
20. Albillos A, de la Hera A, González M, et al. Increased lipopolysaccharide binding
protein in cirrhotic patients with marked immune and hemodynamic derange-
ment. Hepatology 2003;37:208–17.
21. Garcia-Tsao G, Abraldes JG, Berzigotti A, et al. Portal hypertensive bleeding in
cirrhosis: risk stratification, diagnosis, and management: 2016 practice guidance
by the American Association for the study of liver diseases. Hepatology 2017;
65(1):310–35.
22. Hino S, Kakutani H, Ikeda K, et al. Hemodynamic assessment of the left gastric
vein in patients with esophageal varices with color Doppler EUS: factors affecting
development of esophageal varices. Gastrointest Endosc 2002;55(04):512–7.
23. Kuramochi A, Imazu H, Kakutani H, et al. Color Doppler endoscopic ultrasonog-
raphy in identifying groups at a high-risk of recurrence of esophageal varices af-
ter endoscopic treatment. J Gastroenterol 2007;42(03):219–24.
24. Huang JY, Samarasena JB, Tsujino T, et al. EUS-guided portal pressure gradient
measurement with a novel 25-gauge needle device versus standard transjugular
approach: a comparison animal study. Gastrointest Endosc 2016;84(02):358–62.
25. Huang JY, Samarasena JB, Tsujino T, et al. EUS-guided portal pressure gradient
measurement with a simple novel device: a human pilot study. Gastrointest En-
dosc 2017;85(05):996–1001.
26. Thompson WP, Caughey JL, Whipple AO, et al. Splenic vein pressure in conges-
tive splenomegaly. J Clin Invest 1937;16:571–2.
27. Gonzalez CO. Portography: a preliminary report of a new technique via the um-
bilical vein. Clin Proc Child Hosp Dist Columbia 1959;15:120–2.
28. Boyer TD, Triger DR, Horisawa M, et al. Direct transhepatic measurement of por-
tal vein pressure using a thin needle. Comparison with wedged hepatic vein pres-
sure. Gastroenterology 1977;72:584–9.
29. Myers JD, Taylor WJ. An estimation of portal venous pressure by occlusive cath-
eterization of a hepatic venule. J Clin Invest 1951;30:662–3.
30. Perelló A, Escorsell A, Bru C, et al. Wedged hepatic venous pressure adequately
reflects portal pressure in hepatitis C virus-related cirrhosis. Hepatology 1999;
30(6):1393–7.
31. Groszmann RJ, Glickman M, Blei AT, et al. Wedged and free hepatic venous
pressure measured with a balloon catheter. Gastroenterology 1979;76:253–8.
32. Calvaruso V, Burroughs AK, Standish R, et al. Computer-assisted image analysis
of liver collagen: relationship to Ishak scoring and hepatic venous pressure
gradient. Hepatology 2009;49(4):1236–44.
33. Monescillo A, Martı́nez-Lagares F, Ruiz-del-Arbol L, et al. Influence of portal hy-
pertension and its early decompression by TIPS placement on the outcome of
variceal bleeding. Hepatology 2004;40(4):793–801.
34. Abraldes JG, Villanueva C, Banares R, et al. Hepatic venous pressure gradient
and prognosis in patients with acute variceal bleeding treated with pharmaco-
logic and endoscopic therapy. J Hepatol 2008;48:229–36.
35. Ferraioli G, Wong VW, Castera L, et al. Liver ultrasound elastography: an update
to the World Federation for Ultrasound in Medicine and Biology guidelines and
recommendations. Ultrasound Med Biol 2018;44(12):2419–40.
36. Vizzutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement predicts se-
vere portal hypertension in patients with HCV-related cirrhosis. Hepatology 2007;
45:1290–7.
586 Turco & Garcia-Tsao
53. Pind ML, Bendtsen F, Kallemose T, et al. Indocyanine green retention test (ICG-
r15) as a noninvasive predictor of portal hypertension in patients with different
severity of cirrhosis. Eur J Gastroenterol Hepatol 2016;28(8):948–54.
54. Palaniyappan N, Cox E, Bradley C, et al. Non-invasive assessment of portal hy-
pertension using quantitative magnetic resonance imaging. J Hepatol 2016;
65(6):1131–9.
55. Ronot M, Lambert S, Elkrief L, et al. Assessment of portal hypertension and high-
risk oesophageal varices with liver and spleen three-dimensional multifrequency
MR elastography in liver cirrhosis. Eur Radiol 2014;24(6):1394–402.