Lower Serum Albumin Level: A Prospective Risk Predictor of Hepatocellular Carcinoma in Patients With Chronic Hepatitis B Virus Infection

• DOI: 10.1111/jvh.14008
• Publication Date: 2024-09-16T12:27:05Z
• Authors: Yusheng Song; Haiyan Chen; Jinlong Li; Feifei Zhong; Yunbing Liu; Hui Liu; Shaogui Wan
• Keywords: Hepatitis virus; Serum Albumin

ABSTRACT Hepatocellular carcinoma (HCC) is one of the most common malignant tumours in China, at high annual incidence and mortality. Chronic hepatitis B virus infection (CHB) considered as a leading cause to bring about HCC China. Serum albumin (ALB) level has been adopted verify its risk with development combination variable other factors. However, predictive value single ALB on HBV‐related remained unclear. The aim this study was evaluate prediction ability serum concentration development. A prospectively enrolled clinical cohort compromising 2932 cases CHB patients least 1‐year exclusion window selected explore role incident risk. Baseline data including host characters laboratory test were collected initial period hospitalisation. hazard ratio associated assessed by Cox proportional hazards regression model using univariate multivariate analyses. We evaluated discrimination accuracy predicting receiver operating characteristic (ROC) curves. Dose‐dependent time‐dependent effects demonstrated, respectively, restricted cubic spline Fine Grey competing model. Referred higher level, those lower exhibited significantly increased after adjustment for variables (dichotomised analyses: = 3.12, 95% confidence interval 1.63–5.97, p 8.23 × 10 −4 , log‐rank 5.97 ; tertile 2.07, 1.63–2.64, 3.77 −9 < 2.00 −16 quartile 2.10, 1.56–2.84, 9.87 −7 ). There statistically increasing trend which found following decrease ( 0.0001). Similar findings present Kaplan–Meier analysis, cumulative incidences levels, obtained from all 0.0001. result dose‐dependent effect showed (HR) gradually decreasing non‐linear correlation being significant (Wald χ 2 20.59, 0.000). HR persistently prominent fluctuating around 2.73 whole follow‐up time adjusting variables. Sub‐cohort analysis ROC revealed that performed better than Child‐Pugh (C‐P) both (area under curve [AUC] 0.762 vs. 0.720) 2‐year (AUC 0.768 0.728). AUC added demonstrated full Lower an could provide extra useful utility features, might be promising non‐invasive indicator surveillance