The prognostic role of liver stiffness in patients with chronic liver disease: a systematic review and dose–response meta-analysis

• Published in: Hepatology international 2019-09, Vol.13 (5), p.560-572
• Main Authors: Shen, Yue, Wu, Sheng-Di, Wu, Ling, Wang, Si-Qi, Chen, Yao, Liu, Li-Li, Li, Jing, Yang, Chang-Qing, Wang, Ji-Yao, Jiang, Wei
• Format: Article
• Language: English
• Subjects: Colorectal Surgery; Confidence intervals; Elasticity Imaging Techniques; End Stage Liver Disease - diagnosis; End Stage Liver Disease - diagnostic imaging; End Stage Liver Disease - pathology; Hepatology; Humans; Liver; Liver - diagnostic imaging; Liver - pathology; Liver diseases; Medicine; Medicine & Public Health; Meta-analysis; Mortality; Nonlinear analysis; Nonlinear response; Original Article; Prognosis; Risk; Risk assessment; Splines; Statistical analysis; Stiffness; Surgery; Systematic review
• ISSN: 1936-0533; 1936-0541
• DOI: 10.1007/s12072-019-09952-5

Background and aims Liver stiffness measurement (LSM) by transient elastography (TE) has been assessed for the evaluation of clinically relevant outcomes in patients with chronic liver diseases (CLDs) while with variable results. This systematic review and meta–analysis aims to investigate the relationship between baseline LSM by TE and the development of clinically relevant outcomes. Methods The systematic review identified eligible cohorts reporting the association between baseline LSM by TE and risk of hepatic carcinoma (HCC), hepatic decompensation (HD), all–cause and/or liver–related mortality and liver–related events (LREs) in CLD patients. Summary relative risks (RRs) with 95% confidence intervals (CIs) were estimated using a random–effect model. The dose–response association was evaluated by generalized least squares trend (Glst) estimation and restricted cubic splines. Commands of GLST, MKSPLINE, MVMETA were applied for statistical analysis. Results 62 cohort studies were finally included, reporting on 43,817 participants. For one kPa (kilopascal) increment in baseline liver stiffness (LS), the pooled RR (95% CI) was 1.08 (1.05–1.11) for HCC, 1.08 (1.06–1.11) for all–cause mortality, 1.11 (1.05–1.17) for liver-related mortality, 1.08 (1.06–1.10) for HD and 1.07 (1.04–1.09) for LREs. Furthermore, the nonlinear dose–response analysis indicated that the significant increase in the risk of corresponding clinically relevant outcomes turned to a stable increase or a slight decrease with increasing baseline LS changing primarily in the magnitude of effect rather than the direction. Conclusions The dose–response meta-analysis presents a combination between the levels of baseline LS and RRs for each clinically relevant outcome. TE, which is noninvasive, might be a novel strategy for risk stratification and identification of patients at high risk of developing these outcomes.