IDDF2024-ABS-0027 The prognostic value of serum metabolites in predicting hepatic decompensation among patients with cirrhosis

<h3>Background</h3> This study aimed to evaluate the clinical significance of metabolomics in stratifying the risk of decompensation in cirrhosis and identifying differential serum metabolites implicated in this process. <h3>Methods</h3> This prospective study examined the alterations in serum metabolites among cirrhotic patients with and without ascites at baseline, as well as those with and without decompensation at the 1-year mark, through the employment of liquid chromatography/mass spectrometry. Univariate and multivariate analyses were utilized to unveil differential metabolites. The identified differential metabolites, along with clinical parameters, underwent Cox regression analyses to construct predictive models for 1-year decompensation. These models were subsequently validated using <i>C</i>-index, Kaplan-Meier analysis, receiver operating characteristic (ROC) curves, and calibration curves to assess their effectiveness. <h3>Results</h3> This study included a cohort of 58 cirrhotic patients and followed up for 1 year. In contrast to their respective control groups, 134 differential metabolites were identified at baseline, while 22 were at the 1-year mark. The chenodeoxycholic acid glycine conjugate (GCDCA) and bilirubin diglucuronide (BDG) were consistently detected at two-time points, exhibiting enrichment in the primary bile acid biosynthesis and porphyrin metabolism pathways, respectively (IDDF2024-ABS-0027 Figure 1). The prognostic prediction models, which utilized differential metabolites and clinical parameters, demonstrated high discriminatory capability and notable consistency (IDDF2024-ABS-0027 Figure 2). Moreover, the area under the ROC for these models exceeded that of other noninvasive parameters. <h3>Conclusions</h3> During the progression of cirrhosis, fluctuations in the levels of GCDCA and BDG might be detected, while relevant models could be employed to predict the probability of decompensation within one year.