AUTHOR=Wu Daxian , Zhang Lingjian , Ma Shanshan , Zhao Yalei , Chen Ronggao , Zhang Fen , Liu Qiuhong , Xu Xiaowei , Xie Zhongyang
TITLE=Low Growth Hormone Levels Predict Poor Outcome of Hepatitis B Virus-Related Acute-on-Chronic Liver Failure
JOURNAL=Frontiers in Medicine
VOLUME=8
YEAR=2021
URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2021.655863
DOI=10.3389/fmed.2021.655863
ISSN=2296-858X
ABSTRACT=Background and Aims: Hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) remains a serious entity with high mortality. Growth hormone (GH) is related to the liver metabolism and regeneration. The present study aimed to explore the changes and prognostic efficacy of GH on the outcome of HBV-ACLF.
Methods: A prospective cohort of 124 patients and a cross-sectional cohort of 142 subjects were enrolled. GH and insulin-like growth factor-1(IGF-1) were detected by ELISA. Thirty-day survival was collected and the association between GH and the 30-day mortality of HBV-ACLF was analyzed.
Results: The mean age of the whole prospective cohort was 46.61 ± 12.71 years, and 19 (15.3%) patients were female. The median (IQR) of GH levels in non-survivors were 1106.55 (674.25, 1922.4) pg/ml, which were significantly lower than in survivors (p < 0.001). In the cross-sectional cohort, GH level was significantly higher in liver cirrhosis - acute decompensation (LC-AD) group than liver cirrhosis (LC) group (p < 0.001) while IGF-1 decreased significantly in LC, LC-AD, ACLF groups than health control (HC) and chronic Hepatitis B (CHB) groups (p < 0.001). The area under the receiver operating characteristic curve (AUROC) of GH for predicting 30-day mortality was 0.793. We built a new prognostic model, namely MELD-GH, which showed better predictive efficacy than Child-Pugh, MELD, CLIF-SOFA, and CLIF-C ACLF scores.
Conclusions: Low GH predicted the poor outcome of HBV-ACLF patients. GH and IGF-1 levels were differently distributed among HC, CHB, LC, LC-AD, and ACLF patients. MELD-GH had better predictive accuracy when compared to Child-Pugh, MELD, CLIF-SOFA, and CLIF-C ACLF scores.