Noninvasive Algorithm May Accurately Predict the Risk of Portal Hypertension Related to Primary Biliary Cholangitis

An algorithm combining liver stiffness and spleen stiffness values can improve risk stratification by predicting liver decompensation in patients with primary biliary cholangitis (PBC), according to a multicenter study presented at The Liver Meeting 2025 in Washington, D.C., hosted by the American Association for the Study of Liver Disease. 

PBC is a progressive cholestatic disease that may lead to clinically significant portal hypertension, which may occur even in the early stages of disease, before cirrhosis is evident. Patients with compensated advanced chronic liver disease, defined according to the Baveno VI criteria based on a liver stiffness measurement cut-off of ≥10 kPa, are at a higher risk of developing clinically significant portal hypertension and liver decompensation. Spleen stiffness measurements obtained by vibration-controlled transient elastography (VCTE) have been recently used as a noninvasive tool to predict the risk of portal hypertension in patients with a different etiology. 

A multicenter study was designed to assess the risk for clinically significant portal hypertension in patients with PBC using noninvasive algorithms based on liver and spleen stiffness measurements. The study included 100 patients with PBC and compensated advanced chronic liver disease from three hepatology centers in Italy, one clinic in Istanbul, and one clinic in Vienna, who underwent both liver stiffness and spleen stiffness measurements. The researchers used the Baveno VII spleen stiffness measurement cut-off of ≤40 kPa to rule out clinically significant portal hypertension, and they assessed the performance of the newly proposed Non-Invasive CSPH Estimated Risk (NICER) model in predicting the risk of clinical decompensation. 

“We previously found that liver stiffness equal to or higher than 10 kPa and spleen stiffness equal to or higher than 40 kPa equally stratified the risk of clinical decompensation in patients with PBC, and that the combination of those [two measurements] was better for predicting clinical decompensation,” presenting author Cristina Rigamonti, MD, associate professor of gastroenterology at Universita del Piemonte Orientale, in Novara, Italy, said in an interview. “[In this study], we decided to analyze only patients with compensated advanced chronic liver disease, so, only patients with liver stiffness equal to or higher than 10 kPa, to stratify the risk of clinically significant portal hypertension.” 

Clinically significant portal hypertension was ruled out in 47% of the included patients, according to the Baveno VII consensus criteria, whereas 46% of patients were considered at risk for portal hypertension and 7% of the patients remained in the grey zone. Patients with liver stiffness measurements under 15 kPa have a low risk of decompensation according to the Baveno VII criteria. Nevertheless, the researchers found that nearly a third of patients with liver stiffness values between 10 and 14.9 kPa also exhibited spleen stiffness values greater than 40 kPa, which are indicative of portal hypertension. 

“Spleen stiffness measurements demonstrated robust capability in identifying and stratifying the risk of clinically significant portal hypertension in PBC patients with compensated advanced chronic liver disease, confirming its applicability for the specific etiology of PBC,” Rigamonti and colleagues concluded. “Remarkably, almost one-quarter of patients with liver stiffness measurement values below 15 kPa had developed esophageal varices, which is uncommon in non-cholestatic etiologies, but this subgroup of patients was accurately detected by spleen stiffness measurements.” 
The NICER model also demonstrated accurate prognostic value for the risk of portal hypertension in patients with PBC. Twelve patients experienced liver decompensation during a median follow-up of 17.4 months after VCTE examination. The NICER model accurately predicted that the 2-year probability of liver decompensation was significantly higher in patients in the clinically significant portal hypertension risk category ≥75% than in those with a risk <75%. The findings suggest that noninvasive risk models based on spleen stiffness measurements may have clinical utility in estimating the risk for hepatic decompensation related to portal hypertension in patients with PBC.