Assessment of portal hypertension severity using machine learning models in patients with compensated cirrhosis.

BACKGROUND & AIMS: In individuals with compensated advanced chronic liver disease (cACLD), the severity of portal hypertension (PH) determines the risk of decompensation. Invasive measurement of the hepatic venous pressure gradient (HVPG) is the diagnostic gold standard for PH. We evaluated the utility of machine learning models (MLMs) based on standard laboratory parameters to predict the severity of PH in individuals with cACLD. METHODS: A detailed laboratory workup of individuals with cACLD recruited from the Vienna cohort (NCT03267615) was utilised to predict clinically significant portal hypertension (CSPH, i.e., HVPG ≥10 mmHg) and severe PH (i.e., HVPG ≥16 mmHg). The MLMs were then evaluated in individual external datasets and optimised in the merged cohort. RESULTS: Among 1,232 participants with cACLD, the prevalence of CSPH/severe PH was similar in the Vienna (n = 163, 67.4%/35.0%) and validation (n = 1,069, 70.3%/34.7%) cohorts. The MLMs were based on 3 (3P: platelet count, bilirubin, international normalised ratio) or 5 (5P: +cholinesterase, +gamma-glutamyl transferase, +activated partial thromboplastin time replacing international normalised ratio) laboratory parameters. The MLMs performed robustly in the Vienna cohort. 5P-MLM had the best AUCs for CSPH (0.813) and severe PH (0.887) and compared favourably to liver stiffness measurement (AUC: 0.808). Their performance in external validation datasets was heterogeneous (AUCs: 0.589-0.887). Training on the merged cohort optimised model performance for CSPH (AUCs for 3P and 5P: 0.775 and 0.789, respectively) and severe PH (0.737 and 0.828, respectively). CONCLUSIONS: Internally trained MLMs reliably predicted PH severity in the Vienna cACLD cohort but exhibited heterogeneous results on external validation. The proposed 3P/5P online tool can reliably identify individuals with CSPH or severe PH, who are thus at risk of hepatic decompensation. IMPACT AND IMPLICATIONS: We used machine learning models based on widely available laboratory parameters to develop a non-invasive model to predict the severity of portal hypertension in individuals with compensated cirrhosis, who currently require invasive measurement of hepatic venous pressure gradient. We validated our findings in a large multicentre cohort of individuals with advanced chronic liver disease (cACLD) of any cause. Finally, we provide a readily available online calculator, based on 3 (platelet count, bilirubin, international normalised ratio) or 5 (platelet count, bilirubin, activated partial thromboplastin time, gamma-glutamyltransferase, choline-esterase) widely available laboratory parameters, that clinicians can use to predict the likelihood of their patients with cACLD having clinically significant or severe portal hypertension.

Assessment of portal hypertension and high-risk oesophageal varices with liver and spleen three-dimensional multifrequency MR elastography in liver cirrhosis.

OBJECTIVE: To assess the value of the liver and spleen viscoelastic parameters at multifrequency MR elastography to determine the degree of portal hypertension and presence of high-risk oesophageal varices in patients with cirrhosis. METHODS: From January to September 2012, 36 consecutive patients with cirrhosis evaluated for transplantation were prospectively included. All patients underwent hepatic venous pressure gradient (HVPG) measurements and endoscopy to assess oesophageal varices. Multifrequency MR elastography was performed within the liver and spleen. The shear, storage and loss moduli were calculated and compared to the HVPG with Spearman coefficients and multiple regressions. Patients with and without severe portal hypertension and high-risk varices were compared with Mann-Whitney tests, logistic regression and ROC analysis. RESULTS: The liver storage and loss moduli and the spleen shear, storage and loss moduli correlated with the HVPG. At multiple regression, only the liver and the spleen loss modulus correlated with the HVPG (r = 0.44, p = 0.017, and r = 0.57, p = 0.002, respectively). The spleen loss modulus was the best parameter for identifying patients with severe portal hypertension (p = 0.019, AUROC = 0.81) or high-risk varices (p = 0.042, AUROC = 0.93). CONCLUSIONS: The spleen loss modulus appears to be the best parameter for identifying patients with severe portal hypertension or high-risk varices. KEY POINTS: 1. Noninvasive HVPG assessment can be performed with liver and spleen MR elastography 2. The spleen loss modulus enables the detection of high-risk oesophageal varices 3. The spleen loss modulus enables the detection of severe portal hypertension.