Vincenza Calvaruso

Clinically significant portal hypertension (CSPH) in patients with compensated advanced chronic liver disease indicates an increased risk of decompensation and death. While invasive methods like hepatic venous–portal gradient measurement is considered the gold standard, non-invasive tests (NITs) have emerged as valuable tools for diagnosing and monitoring CSPH. This review comprehensively explores non-invasive diagnostic modalities for portal hypertension, focusing on NITs in the setting of hepatitis B and hepatitis C virus-related cirrhosis. Biochemical-based NITs can be represented by single serum biomarkers (e.g., platelet count) or by composite scores that combine different serum biomarkers with each other or with demographic characteristics (e.g., FIB-4). On the other hand, liver stiffness measurement and spleen stiffness measurement can be assessed using a variety of elastography techniques, and they can be used alone, in combination with, or as a second step after biochemical-based NITs. The incorporation of liver and spleen stiffness measurements, alone or combined with platelet count, into established and validated criteria, such as Baveno VI or Baveno VII criteria, provides useful tools for the prediction of CSPH and for ruling out high-risk varices, potentially avoiding invasive tests like upper endoscopy. Moreover, they have also been shown to be able to predict liver-related events (e.g., the occurrence of hepatic decompensation). When transient elastography is not available or not feasible, biochemical-based NITs (e.g., RESIST criteria, that are based on the combination of platelet count and albumin levels) are valid alternatives for predicting high-risk varices both in patients with untreated viral aetiology and after sustained virological response. Ongoing research should explore novel biomarkers and novel elastography techniques, but current evidence supports the utility of routine blood tests, LSM, and SSM as effective surrogates in diagnosing and staging portal hypertension and predicting patient outcomes.