Preventive strategies in hepatology: Role of multiparametric assessment of portal hypertension in chronic liver disease

Abstract

The burden of mortality caused by liver disease continues to increase, underscoring the need for preventive hepatology, an emerging subspecialty focused on reducing disease risk and preventing decompensation. Because most chronic liver diseases are incurable, prevention depends on early identification of compensated advanced liver disease and clinically significant portal hypertension to enable timely risk stratification and intervention. This review summarizes the recent evidence on noninvasive imaging for assessing portal hemodynamics, including Doppler ultrasound, contrast-enhanced ultrasound, elastography (liver and spleen stiffness), and endoscopic ultrasound. Multiparametric ultrasonography offers a bedside modality to screen for compensated advanced liver disease/clinically significant portal hypertension and to monitor portal hypertension risk over time. This modality will guide proactive care by identifying patients who need intensified surveillance, prophylaxis, endoscopy, hepatology/transplant referral. It can also assess the patient’s response to etiologic, lifestyle, and portal-pressure–targeted therapies to prevent decompensation and improve outcomes.

Key Words: Preventive hepatology; Advanced chronic liver disease; Clinically significant portal hypertension; Multiparametric ultrasound; Liver stiffness measurement; Spleen stiffness measurement

Core Tip: The goal of preventive hepatology is to reduce the burden of liver disease through early detection, risk stratification, and timely intervention. Assessing portal hypertension in patients with advanced chronic liver disease is essential for predicting decompensation and guiding clinical management. Noninvasive imaging techniques, particularly multiparametric ultrasound (MPUS), are valuable approaches for evaluating portal hemodynamics. Combining Doppler ultrasound, contrast-enhanced ultrasound, elastography, and endoscopic ultrasound, MPUS enables accurate, accessible, and radiation-free assessment of liver disease severity. The integration of MPUS into routine care will support effective screening and monitoring to ultimately improve outcomes.

INTRODUCTION

Chronic liver disease (CLD) is a major health concern that causes approximately 2 million deaths every year, contributing to 4% of total global deaths. Half of these deaths are due to hepatocellular carcinoma (HCC), and the other half are due to complications arising from cirrhosis. Two-thirds of total deaths from liver disease are in males[1]. Preventive hepatology (PH) is an approach to reduce mortality caused by liver diseases through early intervention and management. Noninvasive diagnostic techniques for portal hypertension (PHT) are necessary to achieve the objectives of PH through interventions to prevent complications, alleviate symptoms, provide optimal care, and educate patients[2].

The proper diagnosis of liver disease is critical for the management and early treatment of the disease. Liver stiffness measurement (LSM) and spleen stiffness measurement (SSM) are highly sensitive and specific noninvasive methods utilized to diagnose clinically significant PHT (CSPH), reducing the need for invasive diagnostic procedures like hepatic venous pressure gradient (HVPG) measurement[3,4]. Multiparametric ultrasound (MPUS) is the combination of various imaging techniques in one exam that provides a comprehensive and detailed evaluation of the liver status. MPUS is a major breakthrough in the diagnosis of liver disease by combining B-mode imaging, LSM, fat quantification, dispersion imaging, Doppler ultrasound (US), and contrast-enhanced US (CEUS)[5]. MPUS is useful in both screening and diagnosis of CLD[6].

Portal hemodynamics play a major role in the pathophysiology of cirrhosis because they are strongly correlated with the severity of the disease. In patients with advanced CLD (ACLD), (without “such as ACLD”), PHT results from increased intrahepatic vascular resistance caused by structural and functional alterations of the liver tissue. These alterations include the accumulation of fibrotic tissue within the sinusoids, the formation of regenerative nodules, and shift to a primarily vasoconstricted phenotype of the sinusoidal endothelial cells[7]. Gastroesophageal varices, ectopic varices, and hepatic encephalopathy are the major clinical manifestations of ACLD, and the formation of collateral vessels indicate portal circulation disorder[8].

Portal pressure can be indirectly measured by HVPG to establish the severity of PHT: Portal pressure > 10 mmHg indicates CSPH[9]; portal pressure > 12 mmHg indicates increased risk of bleeding; portal pressure > 16 mmHg indicates increased mortality; and portal pressure > 20 mmHg indicates failure to control bleeding[10]. However, HVPG measurement is invasive, and noninvasive methods are typically preferred in clinical practice.

This review summarizes the various strategies to accomplish the objectives of PH, emphasizing the multidisciplinary management of CLDs. We discuss the existing evidence on the utility of MPUS in the early detection and monitoring of PHT, and translating the risk profiles generated by MPUS into specific preventive actions, such as the initiation of nonselective beta blockers (NSBBs), the level of surveillance, and the ability to monitor the response to therapy.

PH

PH is based on the following principle: “Except for hepatitis C infection, all other forms of primary liver disease are not ‘curable’, and prevention becomes the backbone of care”. The prevalence of steatotic liver disease is increasing worldwide. Therefore, the field of PH strives to alleviate the burden of CLD by intervening at all stages of the natural history of the disease process itself, including primary prevention (prevention of the disease itself), secondary prevention (early detection and treatment of the disease), and tertiary prevention (limiting the complications of the disease and maintaining quality of life)[11] (Figure 1). Primary prevention involves vaccination against hepatitis A virus and hepatitis B virus (HBV), weight loss, and reduction of alcohol intake. Secondary prevention involves screening for viral hepatitis infection, routine blood and imaging studies, antiviral treatment of infections, and close follow-up of high-risk patients (e.g., HCC). Tertiary prevention involves preventing decompensation and the onset of complications by closely monitoring the patient and intervening appropriately.

Figure 1 Preventive hepatology. Preventive hepatology is divided into primary, secondary, and tertiary levels of prevention. While primary prevention focuses on reducing liver disease risk, secondary prevention emphasizes early detection and management to halt disease progression. Tertiary prevention aims to minimize complications and enhance the quality of life in advanced liver disease. Key preventive strategies (highlighted in orange) can be implemented in primary care, by hepatologists, or through collaborative efforts. ALD: Alcohol-associated liver disease; cACLD: Compensated advanced chronic liver disease; CSPH: Clinically significant portal hypertension; HCC: Hepatocellular carcinoma; MASLD: Metabolic-dysfunction associated steatotic liver disease; NSBBs: Nonselective beta blockers; T2DM: Type 2 diabetes mellitus; TIPS: Transjugular intrahepatic portosystemic shunt.

Nonpharmacological and pharmacological approaches play a significant role in the prevention and treatment of CLDs, especially steatotic liver disease and HCC. Nonpharmacological measures include lifestyle changes that involve a well-balanced diet low in salt and weight control to prevent the progression of the disease. Abstinence from alcohol can prevent the progression of the disease[12]. Endoscopic surveillance to prevent the progression of esophageal varices (EVs) is essential in the prevention and treatment of PHT in patients with cirrhosis. Noninvasive tests to monitor the progression and risk of PHT are also essential. The use of specific drugs, including resmetirom, vitamin E, pioglitazone, and liraglutide, can prevent the progression of PHT by reducing the levels of liver enzymes and promoting weight loss. The combination of individualized nonpharmacological and pharmacological measures to prevent PHT is the most effective treatment.

Secondary/tertiary PH requires multidisciplinary care. Hepatologists and gastroenterologists are the primary care providers for the diagnosis and follow-up of patients with liver diseases and facilitate the early diagnosis and monitor the progression of liver disease[13]. Primary care physicians also play an important role in the care of patients with PH by promoting the prevention and early diagnosis of the disease[14]. Infectious disease experts assist in secondary prevention by diagnosing and treating early-stage viral hepatitis to prevent further progression or spread. Suppression of HBV infection and curing hepatitis C virus infection with antiviral drugs reduce the risk of advanced liver disease or HCC[15]. Endocrinologists guide the measures to control metabolic risk factors[16,17]. Nutritionists/dietitians help to manage metabolic dysfunction-associated steatotic liver disease by providing specific nutritional and lifestyle advice[18,19].

Pharmacists play a critical role in medication safety in liver disease because cirrhosis alters drug metabolism and drug interactions. They also assist in the management of medications to prevent overdosing or underdosing based on liver function[20]. Oncologists oversee the cancer care of patients with ACLD by guiding the extent of surgery, providing locoregional, systemic, and palliative treatments based on the residual liver function, and monitoring at-risk populations for HCC through imaging and biological markers. Close collaboration with hepatologists, interventional radiologists, and transplant teams is necessary[21]. Intensive care specialists provide early detection and management of acute liver failure and severe complications of cirrhosis with associated multiple system dysfunction to avoid further decompensation of the disease[22,23].

Clinicians in transplant centers recognize patients with advanced liver disease at high risk for liver failure or malignancies, evaluate suitability for liver transplantation, and assist hepatologists in managing patients to minimize transplant-triggering complications[24,25]. Endoscopists diagnose and treat PHT complications, including endoscopic variceal ligation (EVL) to treat EVs. Guidelines also recommend screening endoscopy for cirrhosis to stratify patients for bleeding risk and guide EVL or NSBBs administration[26]. The use of noninvasive diagnostic tools for early detection of PHT (prior to the onset of complications such as variceal bleeding or hepatic decompensation) represents a critical component of PH that focuses on proactive monitoring and management of CLD rather than on reactive, complication-driven care[27-29].

PH relies on experts in interventional procedures for the management of ACLD. Minimally invasive procedures, like transjugular intrahepatic portosystemic shunt (TIPS) for PHT, embolization for variceal bleeding, ablation for HCC, and biliary stenting for cholestasis, improve outcomes and prevent complications[30]. Lastly, radiologists perform and interpret noninvasive techniques, including MPUS and elastography, to manage various liver diseases[31] (Figure 2). The use of MPUS in managing liver diseases is a breakthrough in imaging techniques. Although radiologists play a major role in performing and interpreting various imaging techniques, gastroenterologists play a major role in performing these techniques.

Figure 2 Multidisciplinary approach in preventive hepatology. ACLD: Advanced chronic liver disease; LD: Liver disease; PHT: Portal hypertension; TIPS: Transjugular intrahepatic portosystemic shunt.

A collaborative strategy between disciplines and using various techniques improves the accuracy of liver disease assessment and provides comprehensive patient care. The incorporation of MPUS into clinical practice reflects the significance of interdisciplinary collaboration in the management of liver diseases[5].

INVASIVE DIAGNOSTIC TECHNIQUES FOR PHT ASSESSMENT

HVPG measurement is a widely accepted and validated method for the assessment of PHT. HVPG can be measured using a transjugular technique under fluoroscopic guidance. The procedure involves catheterization of the right internal jugular vein through which a balloon-tipped catheter is passed into the right hepatic vein. The free hepatic venous pressure (FHVP) is measured with the catheter tip placed freely in the hepatic vein, and the wedged hepatic venous pressure (WHVP) is measured after inflating the balloon to occlude the hepatic vein. The sinusoidal pressure is thus measured by equilibration with the portal venous system. HVPG is calculated as the difference between WHVP and FHVP (HVPG = WHVP − FHVP), and several measurements are typically averaged to improve accuracy. Clinically, PHT is defined as HVPG > 5 mmHg, and CSPH is defined as HVPG ≥ 10 mmHg (HVPG-defined CSPH). The procedure is generally safe, well tolerated, and takes 30-60 minutes. HVPG measurement is regarded as the gold standard for PHT assessment and provides critical prognostic information related to the risk of clinical decompensation, variceal bleeding, and death. HVPG is also used to assess treatment response and guide therapeutic decision-making, particularly in ACLD[32].

Although HVPG provides important clinical information, it is an invasive procedure that requires specialized expertise, fluoroscopy, and catheterization facilities, limiting its general applicability in routine screening and repeated monitoring in many centers. It may also be subject to technical limitations, operator variability, and laboratory variability, emphasizing the need for procedure standardization. Notably, HVPG only indirectly reflects sinusoidal portal pressure. The value of WHVP in reflecting actual portal pressure may be limited in presinusoidal PHT as it may not be an accurate reflection of the actual pressure in this condition.

NONINVASIVE DIAGNOSTIC TECHNIQUES FOR PHT ASSESSMENT

Noninvasive techniques should be considered as complementary tests to diagnose PHT. Each test has slightly different benefits for screening and risk assessment, can be used to confirm an uncertain case, or provides follow-up over time, ultimately demonstrating its utility. US is a commonly used imaging modality in the management of CLD because of its simplicity and noninvasive nature. Doppler US enables real-time evaluation of blood flow under normal physiological conditions, and CEUS with microbubble contrast agents provides a detailed evaluation of peripheral blood circulation. Elastography was initially developed for fibrosis evaluation but can be used for the assessment of PHT by measuring LSM and SSM. These developments were made possible by the advancements in digital technology and the availability of information.

Abdominal US in B-mode

B-mode US is usually the first test to identify CLD. It is easy to access, quick to perform, and repeatable. On its own it does not measure portal pressure, but it gives a big picture of the changes seen with PHT (e.g., spleen enlargement, ascites, a wider portal vein, and signs of collateral circulation). A portal vein diameter above 13 mm, a splenic vein over 10-12 mm, or a superior mesenteric vein over 10-12 mm indicates likely PHT when seen with splenomegaly and/or ascites. In more advanced disease US can reveal gastroesophageal varices, reopening of the paraumbilical vein (often > 5 mm), or splenorenal shunts.

Gastroesophageal varices are among the most frequent portosystemic collaterals revealed on US. They show up as enlarged, tortuous (serpiginous) venous structures around the gastroesophageal junction. When PHT is more advanced, other collateral routes are visible [e.g., paraumbilical collaterals caused by reopening (recanalization) of the umbilical vein, often when the vein measures > 5 mm)]. Another classic pathway is the splenorenal shunt where blood is redirected from the splenic vein toward the left renal vein. Doppler US typically confirms this abnormal flow pattern. US is useful for revealing direct signs (collaterals/varices) and indirect findings, such as splenomegaly, ascites, and thrombocytopenia, for PHT screening.

The main limitation of standard B-mode is that it cannot grade the severity of PHT. It primarily indicates that PHT is present. It is best used for baseline assessment and screening, to document complications (ascites, collaterals), and to guide the next step (Doppler, elastography, endoscopy, or referral).

Color Doppler and pulsatile Doppler US

Doppler US is useful for understanding portal flow and the nature of the vessels. As liver disease advances vessels may dilate, portal velocity decreases, and the hepatic vein tracing becomes less phasic. We may also notice a higher hepatic artery resistive index although this is not specific because both active inflammation and fibrosis can shift it[33]. Doppler complements the structural assessment by providing hemodynamic evidence of collateralization (e.g., a prominent left gastric vein with hepatofugal flow or flow patterns compatible with collateral drainage via the azygos pathway). When used with B-mode US, Doppler can help clinicians interpret severity and anticipate potential complications. Doppler US is limited by its dependance on the operator (e.g., variable reproducibility across centers). Doppler US indicates abnormal findings but is less reliable for defining CSPH severity. Doppler US can be used as a follow-up tool to understand hemodynamics, improving risk stratification and identifying complication-related features (collaterals/shunts, altered portal flow) when B-mode findings are already suspicious for PHT.

CEUS

CEUS is minimally invasive and reveals hemodynamic information without the burden of liver biopsy or venous catheterization. In patients with ACLD CEUS can assess intrahepatic transit times with acceptable reproducibility and diagnostic performance. The intrahepatic transit time is usually expressed as the interval between hepatic artery arrival and hepatic vein arrival (hepatic vein arrival time-based metric)[34]. This parameter can discriminate the presence of EVs and high-risk varices with good accuracy [area under the receiver operating characteristic curve (AUROC): 0.883 and 0.915] using proposed cutoffs of 8.2 seconds and 7.0 seconds, respectively[35].

CEUS helps identify patients who are high risk for varices and will benefit from earlier preventive steps. CEUS can be helpful in revealing other complications. In portal vein thrombosis, CEUS improves characterization by distinguishing bland from tumor thrombus and can predict anticoagulation outcomes[36]. Quantitative approaches have been used for portal hypertensive gastropathy assessment with improved sensitivity in protocols using sonazoid[37].

Because CEUS protocols and cutoffs vary across centers and depend on equipment and technique, CEUS is not an appropriate test for universal screening criteria. It is most useful as a problem-solving tool in the gray zone when B-mode/Doppler US and elastography are insufficient or when further hemodynamic information is required without invasive procedures. When used in the MPUS context, CEUS is a second-line tool for those in the elastography gray zone or when SSM is not available and aids in the decision to proceed to endoscopy or increase NSBB-based prophylaxis.

Elastography

Liver elastography is widely used to stage compensated ACLD (cACLD). It is important to note that stiffness can be caused by fibrosis as well as obesity, active necroinflammation, and cholestasis. Elastography results are typically interpreted within the context of the technique used [vibration-controlled transient elastography (VCTE), US-based shear wave elastography (SWE), or magnetic resonance elastography]. The FibroScan 630 Expert uses VCTE to quantify liver (50 Hz) and spleen (100 Hz) stiffness and supports clinical staging and monitoring in CLD. SWE is used with US to estimate stiffness by measuring how fast shear waves pass through the liver. Most systems create those waves with acoustic radiation force impulse (ARFI) as a single-point measurement or as a two-dimensional stiffness map over a chosen area. ARFI-based SWE can provide LSM in patients with ascites with good accuracy to predict CSPH and identify EVs[38,39]. Magnetic resonance elastography is a very accurate and reproducible, noninvasive option for assessing liver stiffness, steatosis, and iron, but it can be challenging to apply or interpret in patients with marked iron overload, large-volume ascites, severe obesity, or metallic stents[40].

LSM: It is often difficult to separate advanced fibrosis from early compensated cirrhosis in patients who are asymptomatic. The Baveno VI (2015) guidelines recommend the term cACLD rather than using a strict cirrhosis cutoff[27]. Baveno VII then proposed the VCTE rule of five (5 kPa, 10 kPa, 15 kPa, 20 kPa, and 25 kPa) interpreted with the platelet (PLT) count to quickly stratify risk: < 5 kPa is usually normal; < 10 kPa indicates cACLD unlikely; 10-15 kPa suggests possible cACLD; and > 15 kPa indicates cACLD. VCTE is commonly used as a quick, noninvasive way to gauge CSPH risk. LSM > 25 kPa generally indicates CSPH, while values of 15-25 kPa should be considered with other PHT markers (e.g., thrombocytopenia). When LSM is < 15 kPa and PLTs are ≥ 150 × 10⁹/L, CSPH is unlikely, and screening endoscopy can often be postponed if surveillance continues. Many centers annually perform VCTE and PLTs measurement to assess progression and update risk[41]. Severe obesity can lead to unreliable or unsuccessful readings, and right-lobe sampling may miss uneven disease.

The rule-of-4 is a practical interpretation method for liver stiffness values determined with ARFI-based elastography [point shear-wave elastography (pSWE) or two-dimensional SWE]. This approach assesses the degree of liver disease and the risk of PHT[42]. Liver stiffness values < 5 kPa are suggestive of normal liver parenchyma, while values between 5 kPa and 9 kPa exclude cACLD. Values between 9 kPa and 13 kPa suggest early cACLD, while values between 13 kPa and 17 kPa suggest a higher probability of cACLD and increased risk of liver-related complications in the absence of CSPH. Values > 17 kPa suggest CSPH and are associated with an increased risk of complications, such as variceal bleeding and ascites. The prognostic significance of LSM over time is also highlighted by the Baveno VII guidelines in the prediction of liver complications and mortality in patients with cACLD. The consensus agreed that a significant change in LSM values was a decrease of at least 20%, particularly if this results in values below 20 kPa, or any decrease to below 10 kPa[43] (Figure 3).

Figure 3 Liver stiffness measurement and dynamic change of liver stiffness. A: Use of the liver stiffness measurement (LSM) according to the rule five for transient elastography (TE) and rule of four for acoustic radiation force impulse (ARFI) to determine compensated advanced chronic liver disease (cACLD) and clinically significant portal hypertension (CSPH). Patients with an LSM < 10 kPa excludes cACLD in the absence of other clinical/imaging signs. LSM values between 10 kPa and 15 kPa for TE and 9 kPa and 13 kPa for ARFI are suggestive of cACLD. LSM measured by vibration-controlled TE (VCTE) > 15 kPa and > 13 kPa by ARFI are considered a high likelihood of cACLD for all etiologies. LSM ≤ 15 kPa for VCTE and ≤ 13 kPa for ARFI plus platelets ≥ 150 × 10⁹/L exclude CSPH for most etiologies. Patients with intermediate values of LSM between 15 kPa and 25 kPa for VCTE and 13 kPa and 21 kPa for ARFI are in the gray zone of CSPH. The best cutoff to determine the presence of CSPH is an LSM ≥ 25 kPa (specificity and positive predictive value > 90%) in alcoholic liver disease, chronic hepatitis B, chronic hepatitis C, and patients without obesity with hepatic steatosis; B: A dynamic change in liver stiffness is considered significant if there is a reduction of ≥ 20%, particularly when accompanied by an LSM < 20 kPa or any decrease that brings the LSM to < 10 kPa. TE: Transient elastography; cACLD: Compensated advanced chronic liver disease; CSPH: Clinically significant portal hypertension; LSM: Liver stiffness measurement; ALD: Alcohol-associated liver disease; ARFI: Acoustic radiation force impulse; HCV: Hepatic C virus; SVR: Sustained virological response.

SSM: Because LSM can be influenced by confounding factors, the addition of SSM can enhance noninvasive risk prediction for CSPH. In the Baveno VII guidelines for viral hepatitis-related cACLD with VCTE, an SSM < 21 kPa excludes CSPH, while an SSM > 50 kPa indicates CSPH[27,44-47] (Figure 4). SSM is affected by splenic congestion due to PHT and structural changes such as vascular remodeling and fibrosis, indicating that SSM reflects the effects of ongoing hemodynamic alterations and structural changes. Hyperdynamic circulation in cirrhosis can be detected by SSM and results from increased expression of vascular endothelial growth factor that causes splenic angiogenesis and fibrosis[48].

Figure 4 Spleen stiffness measurement and dynamic change of spleen stiffness. A: Spleen stiffness measurement (SSM). The transient elastography (TE) SSM value < 21 kPa can rule out clinically significant portal hypertension (CSPH), while a value of more than 50 kPa confirms CSPH in patients with viral hepatitis-related compensated advanced chronic liver disease. An SSM of ≤ 40 kPa, a liver stiffness measurement of < 20 kPa, and a platelet count of ≥ 150 × 10⁹/L allow for the omission of upper endoscopy screening; B: Dynamic change of spleen stiffness. A change in SSM ≥ 10% after the initiation of nonselective beta blockers (NSBBs) demonstrates excellent accuracy in identifying NSBBs responders. A SSM of ≥ 74 kPa by TE demonstrates outstanding performance in predicting a poor response to NSBBs. SSM is essential for noninvasively monitoring transjugular intrahepatic portosystemic shunt patency and identifying dysfunction. SSM is a useful tool for early prognosis and tracking liver function after liver transplantation. ARFI: Acoustic radiation force impulse; EGD: Esophagogastroduodenoscopy; TE: Transient elastography; LSM: Liver stiffness measurement; CSPH: Clinically significant portal hypertension; NSBB: Nonselective beta blockers; SSM: Spleen stiffness measurement; TIPS: Transjugular intrahepatic portosystemic shunt; SS: Spleen stiffness.

SSM is also highly sensitive to the hemodynamic effects of NSBBs. A reduction of 10% or more in SSM following the initiation of NSBBs identifies a response to treatment (AUROC: 0.973)[49,50]. If the SSM is 74 kPa or greater by VCTE, then the patient has not responded to NSBBs (100% sensitivity and negative predictive value for acute non-response and 87% sensitivity and 71% specificity for poor chronic response)[51]. These conditions identify patients who are unlikely to benefit from NSBB treatment. Wang et al[51] showed that increases in SSM over time can be a marker of deteriorating PHT and that SSM is a more dynamic measure than LSM. SSM was often superior for predicting decompensation and identifying EVs. LSM + SSM can enhance noninvasive risk stratification per the Baveno VII recommendations.

SSM evaluated by pSWE is an independent and strong predictor of survival in patients with cirrhosis undergoing TIPS. Zhu et al[52] showed that patients with higher preoperative SSM values had a significant association with mortality and liver failure after TIPS. With a cutoff value of 3.60 m/second, SSM has high specificity and predictive accuracy, outperforming LSM. The addition of SSM during the pre-TIPS assessment optimizes patient selection and risk stratification. Moreover, pSWE with conventional US is a useful noninvasive method for the diagnosis of TIPS dysfunction. SSM was significantly associated with shunt dysfunction and had high diagnostic accuracy, supporting its role in the follow-up of patients who had undergone TIPS[53].

Elastography examination of liver and spleen stiffness is a noninvasive tool to assess the post-transplant dynamics in living donor liver transplantation. In cases with a significant reduction in spleen stiffness, portal hemodynamics have improved. Liver graft stiffness is expected to remain constant in the absence of complications and increase in cases with adverse outcomes, indicating the advantages of this tool in the early detection of complications and treatment outcomes[54].

SSM is not universally available and is technique-dependent. The limitations of SSM include restricted acoustic windows, reliable readings in patients with obesity, and limited availability. Moreover, the values obtained from different elastography machines are not always comparable. SSM is most useful for screening/triage in combination with LSM and PLTs (Baveno VII) to refine the risk estimation. SSM can be useful when patients are in the gray zone in which the values are intermediate, when between two risk categories, or when LSM is not clearly above or below a threshold for a specific diagnosis. SSM, especially after the initiation of NSBBs, can be particularly useful for monitoring the hemodynamic response.

Endohepatology

Endohepatology involves the integration of endoscopic US (EUS) in the assessment and management of CLD. A multiparametric and real-time method of evaluation in hepatology is formed through the integration of imaging and minimally invasive interventions. Endohepatology involves various parameters that have increased the scope of practice in hepatology[55,56].

EUS precisely assesses the vascular dimensions of the portal venous system, which is an indicator of the static component of the hemodynamic response. EUS is very effective in identifying the presence of collateral circulation, such as varices, and intravascular complications, such as portal vein thrombosis. EUS is able to visualize peri-esophageal and para-esophageal collateral veins, which play a critical role in the occurrence of varices, their bleeding, and recurrence. EUS is particularly useful in identifying residual blood flow, assessing thrombosis after therapy, and determining an EVL or sclerotherapy course[57].

Para-EVs > 4 mm detected after endoscopic band ligation predict variceal recurrence within 1 year (sensitivity 70.6%, specificity 84.6%)[58]. A gastric cardia perforating vein diameter > 3 mm is likewise associated with EV recurrence within 3 months[59]. Recurrence within 1 year after eradication has also been linked to rapid hepatofugal flow in the left gastric vein and a dominant flow pattern in its anterior branch[60]. After TIPS placement EUS can assess shunt effectiveness by revealing reductions in collateral vessel caliber and flow, including changes in the thoracic duct and azygos vein. Standardization of EUS protocols and enhanced training of endosonographers are needed to improve diagnostic consistency and to facilitate reliable integration of artificial intelligence (AI) into EUS-based assessment. Overall, EUS is a valuable tool for safe, targeted, and individualized management of EVs[57].

Contrast-enhanced EUS (CE-EUS) integrates CEUS with EUS to assess portal-systemic hemodynamics and to improve detection of clinically meaningful changes. Compared with standard color Doppler-EUS, CE-EUS increases the detection of perforating veins[61]. It may also identify portal hypertensive gastropathy with greater sensitivity than conventional esophagogastroduodenoscopy, the current diagnostic gold standard[62]. CE-EUS can evaluate intrahepatic hemodynamic alterations related to PHT. Hepatic vein arrival time is the interval between microbubble injection and the appearance in the hepatic vein. It is inversely associated with fibrosis severity, reflecting fibrosis-driven circulatory changes. A hepatic vein arrival time < 14 second was associated with a positive likelihood ratio of 6.95 for diagnosing CSPH[63].

EUS-guided SWE (EUS-SWE) has a diagnostic accuracy comparable with VCTE for identifying cACLD and CSPH. It is particularly useful when VCTE is unreliable or fails because of body habitus or other technical limitations. EUS-SWE can be used for staging fibrosis. The AUROC values were 0.87 for VCTE and 0.80 for EUS-SWE in the left lobe (EUS-SWE 0.78 in the right lobe), and for cirrhosis the AUROC was 0.90 for VCTE and 0.96 for EUS-SWE in the left lobe (EUS-SWE 0.90 in the right lobe)[64]. EUS-SWE also yields reliable LSM in patients with unsuccessful VCTE. Higher EUS-SWE stiffness values have been observed in patients with varices and hepatic encephalopathy, suggesting potential utility for decompensation risk stratification. Moreover, EUS-SWE can track changes in liver stiffness following portal pressure-modifying and metabolic interventions, including NSBBs and glucagon-like peptide-1 agonists, for ACLD[65].

The EUS-guided portal pressure gradient measurement is an emerging technique to directly assess portal pressure. Under EUS guidance the portal and hepatic veins are punctured with a 25-gauge aspiration needle, and pressures are recorded via noncompressible tubing connected to a compact manometer to calculate the gradient. Pooled evidence from eight cohort studies revealed high technical success (94.6%) and clinical success (85.4%) with a low overall adverse event rate (10.9%), most of which were mild (93.7%). EUS-guided portal pressure gradient appears to be a safe and effective method for evaluating PHT with the key advantage of accurate, direct portal pressure measurement and a feasible alternative to the transjugular approach when conventional assessment is limited or contraindicated[66].

Endohepatology provides both diagnostic and therapeutic value in the management of PHT. A key therapeutic application is EUS-guided obliteration of gastric varices using coils and cyanoacrylate, enabling precise targeting of the variceal complex. Compared with conventional endoscopic therapy, EUS guidance improves visualization and lowers rebleeding rates. Although TIPS remains the gold standard radiologic method for portal decompression in patients with severe complications (e.g., refractory ascites or gastroesophageal variceal bleeding), EUS-guided alternatives, such as EUS-guided intrahepatic portosystemic shunt creation, are under investigation. Preclinical porcine studies have demonstrated that it is technically feasible to form a functional intrahepatic shunt via a transgastric EUS approach. While still experimental, these developments highlight the expanding interventional potential of EUS and demonstrate a strategy integrating diagnostic assessment and therapy within a single session to optimize patient care and resource utilization[67-69] (Figure 5).

Figure 5 Endohepatology assessment of portal hypertension. Endoscopic ultrasound (EUS) is equivalent to conventional upper endoscopy for detecting esophageal varices but is superior for identifying gastric varices and portal hypertensive gastropathy. Evaluating para-esophageal collateral veins (ECVs), peri-ECVs, and left gastric vein predicts the risk of variceal rebleeding and recurrence. EUS has potential utility in monitoring the effectiveness of transjugular intrahepatic portosystemic shunt. When combined with contrast-enhanced ultrasound, it is a valuable tool for assessing portal hypertension. EUS-shear wave elastography is comparable with transient elastography for evaluating compensated advanced chronic liver disease and clinically significant portal hypertension. EUS-portal pressure gradient directly measures portal vein pressure. EUS: Enhanced endoscopic ultrasound; ACLD: Advanced chronic liver disease; CE-EUS: Contrast-enhanced endoscopic ultrasound; CSPH: Clinically significant portal hypertension; GLP-1: Glucagon-like peptide-1; HE: Hepatic encephalopathy; HVAT: Hepatic venous arrival time; LGV: Left gastric vein; LR: Likelihood ratio; LSM: Liver stiffness measurement; NSBBs: Nonselective beta blockers; PHG: Portal hypertensive gastropathy; PHT: Portal hypertension; PPG: Portal pressure gradient; PVT: Portal vein thrombosis; SWE: Shear wave elastography; TIPS: Transjugular intrahepaticportosystemic shunt; VCTE: Vibration-controlled transient elastography; ECVs: Esophageal collateral veins; IPSS: Intrahepatic portosystemic shunt.

BLOOD-BASED NONINVASIVE TESTS FOR PHT ASSESSMENT

Various blood-based noninvasive tests, such as PLT count, fibrosis scores, and patented biomarkers such as Enhanced Liver Fibrosis test score, have been considered for CSPH evaluation in patients with cACLD. The individual diagnostic capability is modest, but the combination of these blood-based tests has practical applications. PLT count is one of the most readily available surrogate tests because thrombocytopenia is indicative of splenic sequestration and bone marrow suppression due to PHT. Nevertheless, its diagnostic validity is poor when used alone (AUROC: 0.787)[29]. Composite scores such as aspartate aminotransferase-to-PLT ratio index (APRI), fibrosis-4 (FIB-4), and the Lok index have been considered for CSPH prediction. Their diagnostic validity has been dependent on the population of patients and underlying causes of liver disease. These tests are often combined in complex multivariable models, such as Baveno VII[27] and ANTICIPATE[70]. These models improve diagnostic validity by combining LSM, SSM, PLT, and body mass index with other variables.

Machine learning-derived composite indexes using PLT and liver function tests have been proposed using the 3P model (PLT, bilirubin, and international normalized ratio) and the 5P model (PLT, bilirubin, cholinesterase, gamma-glutamyl transferase, and activated partial thromboplastin time). However, the inconsistent results of these models during validation studies have limited their utility in the clinic. Fibrosis scores, such as FIB-4 and APRI, have a poor correlation with HVPG due to their association with PLT levels. These models have better accuracy when used to predict outcomes in heterogeneous populations with both early-stage and late-stage disease[71].

The circulating biomarker von Willebrand factor to PLT ratio has promising utility. It is particularly useful in refining the diagnosis of patients in the gray zone with the elastography-based criteria, such as the Baveno VII consensus. The von Willebrand factor-to-PLT ratio score is highly effective in diagnosing CSPH among patients with cACLD and has consistent findings in an external validation cohorts[72]. This biomarker could serve as a potential substitute for elastography.

However, the variability of these assays among different analytical platforms is a challenge and underlines the need for a standardized calibration procedure before the widespread use of these biomarkers in the clinical setting. These blood tests are not reliable enough to completely replace HVPG measurement for clinical decision-making, especially regarding the risk of decompensation or the response to treatment. Blood-based tests can serve as a low-cost entry point for early findings and risk stratification, allowing for timely referral to elastography-based assessment and preventive interventions for CSPH and decompensation.

AI-BASED METHODS FOR PHT ASSESSMENT

AI is revolutionizing noninvasive PHT assessments by enhancing the capabilities of imaging modalities like US, computed tomography (CT), magnetic resonance imaging (MRI), and endohepatology[73]. AI integration has led to the development of novel strategies for PHT risk stratification and monitoring. AI-assisted computerized diagnostic systems can extract features from images independently, minimizing human error. Deep learning algorithms and image analysis can detect PHT-related clinical indicators at an early stage, and the characteristics of associated lesions may be accurately determined. Moreover, the capacity of AI systems to handle complex imaging data facilitates timely and precise decision making to maximize the utility of noninvasive imaging modalities in PHT management.

Noninvasive diagnostic models for PHT have been developed as AI-based image processing techniques have advanced. Traditional radiomics techniques have also been used to identify a broad range of features from medical images with useful diagnostic information[74,75]. Machine learning algorithms, particularly supervised learning algorithms, can analyze large volumes of clinical data and predict CSPH in combination with routine lab values and elastography/radiomics features.

Marozas et al[76] reported 90% accuracy using a clinical model augmented with transient elastography, while Liu et al[77] achieved area under the curve values ≥ 0.90 by applying deep convolutional neural networks to CT/MRI data. Bosch et al[78] demonstrated that CSPH can be predicted from histology using a biopsy-based machine learning approach. Despite these promising results, the clinical utility and generalizability of AI-driven tools for PHT are insufficiently validated. Rigorous external validation is still needed before routine implementation. Continued progress will require close collaboration among hepatologists, radiologists, and pathologists to develop robust diagnostic and prognostic models[79].

Ethical and legal considerations are necessary for the clinical implementation of AI in PHT management. AI systems should be designed as decision-support tools with clinical accountability and decision-making responsibility residing with the physician and multidisciplinary team. Transparency is needed for the use and limitations of AI systems as well as the origin of the information. Mechanisms to ensure an unbiased approach and differential performance in subgroups of patients are needed. Clinical implementation should be restricted to AI systems that are compliant with regulatory requirements (e.g., medical device approval) and are backed by both external and local validation. There is also a need for quality assurance, audit trails, and adverse event reporting.

CRITICAL SYNTHESIS: STRENGTHS, LIMITS, AND COMBINING TESTS

The best approach for diagnosing PHT is combining noninvasive tests because each method captures part of the picture but none measure the portosystemic pressure gradient directly. A multiparametric approach supports a pragmatic risk-based pathway to stratify patients as low risk, intermediate risk, and high risk to guide the next clinical step (closer follow-up, endoscopy, medical prophylaxis, or referral).

In the clinic we typically use these tools in three steps. The first step is screening and initial triage. LSM and the PLT count (the Baveno VII approach) gives the most information, and SSM is helpful for further risk estimation. If the patient is in the gray zone, SSM, CEUS, or an integrated score can further guide clinicians. During follow-up trends should be prioritized (e.g., SSM after NSBBs initiation, LSM over time, and detecting decompensation, variceal bleeding, ascites, encephalopathy). Currently, there is no universal best method. Frontline tests are typically affordable and easy to implement, but their performance may vary with operator skill and local resources. More advanced techniques are usually more consistent and informative, yet they require higher costs, dedicated equipment, and experienced teams.

MPUS can be incorporated into a practical decision algorithm for screening and preventive care of cACLD/CSPH. First, a baseline MPUS evaluation (B-mode and Doppler) should be conducted in conjunction with routine laboratory work, recognizing the PLT count as a proxy for PHT. Second, elastography should be performed to measure LSM and if possible SSM. Third, patients should be stratified into three risk groups that can be directly linked to preventive measures. In individuals with noninvasive confirmation of CSPH (e.g., highly elevated LSM values or elevated LSM values in combination with low PLT counts and SSM is used as a confirmatory test where available), secondary preventive measures are instituted (e.g., NSBBs, endoscopy, and follow-up). For individuals in whom CSPH has been noninvasively excluded, a conservative approach should be adopted. Annual reassessment of LSM and PLT count (and potentially SSM) is appropriate. For individuals in the gray zone, escalation using SSM, CEUS-derived transit time (if available), or an integrated multivariable approach (e.g., ANTICIPATE or NICER) should be undertaken. The process through which MPUS operates clearly defines and specifies the preventive outputs (i.e., who should be treated with NSBBs, who should be referred for endoscopic evaluation, who should be referred for routine HCC surveillance, and how frequently patients should be reassessed) (Figure 6).

Figure 6 Multiparametric ultrasound-based algorithm for noninvasive risk stratification of compensated advanced liver disease and clinically significant portal hypertension and linkage to preventive interventions. Proposed stepwise decision algorithm for ambulatory patients with compensated advanced liver disease. Step 1: Baseline multiparametric ultrasonography, including B-mode and Doppler ultrasound, along with routine blood tests with platelet count as a surrogate of portal hypertension. Step 2: Liver stiffness measurement and spleen stiffness measurement. Step 3: Risk stratification into low risk, gray zone/indeterminate risk, and high risk. Step 4: Associated preventive outputs, including initiation of nonselective beta blockers and/or endoscopy with intensified follow-up for patients with confirmed clinically significant portal hypertension. PLT: Platelet; CSPH: Clinically significant portal hypertension; SSM: Spleen stiffness measurement; LSM: Liver stiffness measurement; NSBBs: Nonselective beta blockers; cACLD: Compensated advanced chronic liver disease; CEUS: Contrast-enhanced ultrasound; MPUS: Multiparametric ultrasound; TE: Transient elastography.

MPUS can be utilized as a dynamic decision-making process in patients with cACLD undergoing treatment and monitoring by measuring changes in the liver and spleen stiffness. If the LSM and/or SSM decreases significantly, the patient’s risk is lowered and monitoring can be reduced accordingly (e.g., longer time between clinic visits and MPUS). Conversely, if the LSM and/or SSM increases, the patient’s risk has increased and care can be escalated accordingly. The status of PHT and the development of decompensating features such as varices, ascites, and encephalopathy must also be considered.

IMPLEMENTATION IN REAL-WORLD PHT CARE

Translating MPUS into clinic practice must be done in a stepwise fashion that fits local needs while maintaining the triage → confirm → act pathway. In clinical settings, especially environments without easy access to advanced imaging, endoscopy facilities, or HVPG, MPUS can act as a gatekeeper for prioritizing at-risk patients for preventive interventions.

At the primary level of first contact care (primary care or general gastroenterology), clinicians should be able to identify patients with suspected cACLD/CSPH and determine the next course of action. Tools to identify these patients are: (1) Basic US examination by B-mode for liver morphology, presence of ascites, splenomegaly, and portal systemic collaterals; (2) Doppler US for portal vessel direction/velocity and waveform of hepatic veins when technically possible; (3) PLT count; and (4) LSM by transient elastography/SWE when facilities are available. Patients in the high-risk group should be prioritized for further hepatology input, preventive pharmacotherapy when appropriate, and endoscopy when appropriate and available.

At the secondary/tertiary levels (confirmation and refinement). The power of MPUS is increased with the addition of tests providing more personalized recommendations. SSM should be added to the noninvasive assessment of CSPH, including cases in the gray zone. CEUS may provide additional information, and EUS (including elastography or specific assessment of variceal anatomy) may enable precise risk assessment. At these levels, MPUS can be used for diagnosis and operational management, including the selection of patients for intense surveillance or therapy.

It is expected that many facilities will not have one or more of these technologies available. In that event, decision making should be focused on LSM + PLT count + simple US/Doppler assessment with repeated measures over time. Indeterminate results should prompt one of three options: (1) Repeat elastography and PLT count after a certain period instead of waiting 1 year; (2) Refer patients to a facility with available SSM/endoscopy tools; or (3) Use a hybrid model that incorporates all available variables (LSM/PLT ± body mass index and routine lab parameters).

The success of MPUS use will be greatly influenced by the personnel involved in its execution and reporting. An acceptable approach is the shared care approach in which the hepatology or gastroenterology team is trained to perform the US/Doppler study and incorporates the elastography results to interpret the best course for prevention. Standardization of MPUS reporting, including LSM (method, type of probe, reliability), PLT count, B-mode imaging (presence of ascites, splenomegaly), Doppler imaging (direction of portal flow, presence of gross abnormalities), SSM, and CEUS/EUS, must be achieved.

CONCLUSION

PH has great significance due to the increasing disease burden and a scarcity of liver disease specialists. Early, noninvasive detection of CSPH in patients with ACLD is crucial for the prevention of decompensation and improving disease prognosis. Routine management of liver disease with a combination of US/Doppler US, liver and spleen elastography, and escalation in indeterminate cases with the aid of complementary tools and multivariable models is necessary. This approach ensures that risk prediction is directly linked with prevention strategies, including the initiation of NSBBs, and changes the intensity of follow-up as the disease is monitored. This approach must be standardized to improve the efficiency of referrals and prevention of complications of PHT in liver disease.