Development and Validation of a Novel Risk Prediction Model Using Recursive Feature Elimination Algorithm for Acute-on-Chronic Liver Failure in Chronic Hepatitis B Patients With Severe Acute Exacerbation

Abstract

Background: Patients with chronic hepatitis B (CHB) with severe acute exacerbation (SAE) are at a progression stage of acute-on-chronic liver failure (ACLF) but uniform models for predicting ACLF occurrence are lacking. We aimed to present a risk prediction model to early identify the patients at a high risk of ACLF and predict the survival of the patient.Methods: We selected the best variable combination using a novel recursive feature elimination algorithm to develop and validate a classification regression model and also an online application on a cloud server from the training cohort with a total of 342 patients with CHB with SAE and two external cohorts with a sample size of 96 and 65 patients, respectively.Findings: An excellent prediction model called the PATA model including four predictors, prothrombin time (PT), age, total bilirubin (Tbil), and alanine aminotransferase (ALT) could achieve an area under the receiver operating characteristic curve (AUC) of 0.959 (95% CI 0.941–0.977) in the development set, and AUC of 0.932 (95% CI 0.876–0.987) and 0.905 (95% CI 0.826–0.984) in the two external validation cohorts, respectively. The calibration curve for risk prediction probability of ACLF showed optimal agreement between prediction by PATA model and actual observation. After predictive stratification into different risk groups, the C-index of predictive 90-days mortality was 0.720 (0.675–0.765) for the PATA model, 0.549 (0.506–0.592) for the end-stage liver disease score model, and 0.648 (0.581–0.715) for Child–Turcotte–Pugh scoring system.Interpretation: The highlypredictive risk model and easy-to-use online application can accurately predict the risk of ACLF with a poor prognosis. They may facilitate risk communication and guidetherapeutic options.

Low lymphocyte-to-monocyte ratio, calcitriol level, and CD206 level predict the development of acute-on-chronic liver failure in patients with cirrhosis with acute decompensation

“…54 Yu et al developed another four-variable model, including age, prothrombin time, serum total bilirubin level, and alanine aminotransferase level, abbreviated as the PATA model, for risk identification of ACLF. 55 However, these models have limitations. First, they were derived from retrospective rather than prospective studies.…”

“…ACLF = acute-on-chronic liver failure. 55 However, these models have limitations. First, they were derived from retrospective rather than prospective studies.…”

“…The AARC score, comprising bilirubin, creatinine, international normalized ratio (INR), lactate levels, and HE grade, has recently been shown to have the highest sensitivity and specificity. 53 Nevertheless, none of the scoring systems or predictive variables have been universally accepted for the prediction of ACLF 55 occurrence. Li et al proposed a combination score incorporating age, INR, serum total bilirubin level, direct-to-total bilirubin ratio, serum sodium level, and hepatitis B virus (HBV) DNA level.…”

Acute decompensation of cirrhosis versus acute‐on‐chronic liver failure: What are the clinical implications?

“…39 Real-life cohorts have confirmed that the pre-ACLF phenotype is characterized by heightened systemic inflammation. 31,40 Notably, the [43][44][45] More recently, Dual-energy CT quantification of extracellular liver volume has been associated with such progression independently from liver disease severity parameters. 46 However, most of the predictive models are derived from single-center retrospective studies, and…”

“…Studies investigating factors associated with APASL‐ACLF (mainly in the setting of severe acute exacerbation of chronic hepatitis B infection) occurrence have also been conducted. They have identified that age, liver disease severity (bilirubin, prothrombin time, MELD score) or HBV viral load were associated with the risk of progression toward ACLF 43–45 . More recently, Dual‐energy CT quantification of extracellular liver volume has been associated with such progression independently from liver disease severity parameters 46 .…”