文献速递肝病

Meta分析:ELF检测识别MASLD患者显著纤维化、进展期纤维化及肝硬化的价值

Meta-Analysis: Enhanced Liver Fibrosis (ELF) Test for Identifying Significant Fibrosis, Advanced Fibrosis, and Cirrhosis in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease.
PubMed 文献 · Aliment Pharmacol Ther · 2026-08-24

该Meta分析纳入34项MASLD研究(37个队列),评估ELF检测识别≥F2、≥F3及F4纤维化的最佳截断值。Youden截断值分别为9.37、9.64和9.99,对应AUROC为0.818、0.818和0.774。ELF<9.00或可用于排除≥F2纤维化,ELF≥9.80和≥11.30分别为≥F3和F4的平衡截断值,ELF≥10.50或可作为抗纤维化治疗决策上限,而7.70无实际意义。

要点

  • ELF<9.00可用于排除MASLD患者≥F2纤维化,敏感性0.98但特异性仅0.14。
  • ELF≥9.80和≥11.30分别为识别≥F3和F4纤维化的平衡截断值。
  • ELF≥10.50或可作为MASLD相关抗纤维化治疗决策的上限参考。

核心结论

  • ELF<9.00可能有助于排除MASLD患者≥F2纤维化。
  • ELF≥9.80和≥11.30分别是识别≥F3和F4纤维化的平衡截断值。
  • ELF≥10.50或可作为MASLD相关抗纤维化治疗决策的上限。
  • ELF 7.70截断值对任何目的均无实际意义。

研究设计与样本

  • 研究为基于四个数据库的系统检索与Meta分析,纳入2004至2026年间34项MASLD研究(37个队列)。
  • 研究采用QUADAS-2工具评估质量,并运用分层汇总受试者工作特征分析和Steinhauser分层多截断随机效应模型整合多个截断值。

主要发现

  • Youden截断值及AUROC(95%CI):≥F2为9.37(0.818[0.781-0.850]);≥F3为9.64(0.818[0.782-0.850]);F4为9.99(0.774[0.711-0.827])。
  • 在既定截断值下,≥F2(9.00)的敏感性和特异性分别为0.86和0.61。
  • ≥F3(9.80)的敏感性和特异性分别为0.70和0.79。
  • F4(11.30)的敏感性和特异性分别为0.21和0.96;F4(10.50)的敏感性和特异性分别为0.50和0.86。
  • F2排除截断值(7.70)的敏感性和特异性分别为0.98和0.14。

关键数值与推荐

  • ELF<9.00用于排除≥F2纤维化时,敏感性0.98,特异性0.14。
  • ELF≥9.80识别≥F3纤维化时,敏感性0.70,特异性0.79。
  • ELF≥11.30识别F4肝硬化时,敏感性0.21,特异性0.96。
  • ELF≥10.50识别F4肝硬化时,敏感性0.50,特异性0.86。

临床怎么用

  • 临床可考虑用ELF<9.00初步排除MASLD患者≥F2纤维化,但需注意其低特异性。
  • ELF≥9.80和≥11.30可分别辅助识别≥F3和F4纤维化,用于风险分层。
  • ELF≥10.50或可作为启动MASLD相关抗纤维化治疗的决策参考上限。
  • 不应将ELF 7.70作为任何临床目的的截断值。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Non-invasive tests are increasingly employed to evaluate liver fibrosis associated with metabolic dysfunction-associated steatotic liver disease (MASLD). Accurate cutoffs for significant fibrosis (≥F2), advanced fibrosis (≥F3), and cirrhosis (F4) are critical for establishing diagnostic guidelines and qualification for MASLD-related anti-fibrotic therapy. AIMS: Determine optimal Enhanced Liver Fibrosis (ELF) Test MASLD cutoffs for ≥F2, ≥F3, and F4 identification and decision-making, evaluate the appropriateness of guideline-recommended cutoffs, and assess ELF performance. METHODS: Using four databases, we identified 34 MASLD-focused studies conducted between 2004 and 2026 evaluating ELF cutoffs (37 cohorts): quality was assessed using Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2). Multiple cutoffs were integrated using hierarchical summary receiver operating characteristic analysis. Sensitivity (Sens) and specificity (Spec) at Youden cutoffs and guideline-recommended cutoffs were estimated using Steinhauser's hierarchical multi-cutoff random-effects model and R diagmeta package v0.5-1. RESULTS: Youden cutoffs (AUROCs [95% CIs]) were: ≥F2: 9.37 (0.818 [0.781-0.850]); ≥F3: 9.64 (0.818 [0.782-0.850]); F4: 9.99 (0.774 [0.711-0.827]). Sens and Spec at established cutoffs were: ≥F2 (9.00) 0.86 Sens, 0.61 Spec; ≥F3 (9.80) 0.70 Sens, 0.79 Spec; F4 (11.30) 0.21 Sens, 0.96; and F4 (10.50) 0.50 Sens, 0.86 Spec. For F2 rule-out (7.70), Sens = 0.98 and Spec = 0.14. CONCLUSION: ELF < 9.00 may be useful to rule out ≥ F2 fibrosis in MASLD. ELF ≥ 9.80 and ≥ 11.30 represent balanced cutoffs for ≥F3 and F4 fibrosis, respectively. ELF ≥ 10.50 could represent an upper limit for MASLD-related anti-fibrotic therapeutic decision-making. ELF at 7.70 is not a meaningful cutoff for any purpose. TRIAL REGISTRATION: PROSPERO registration: CRD420251145859.

来源:Aliment Pharmacol Ther,查看原文。版权归原作者所有。

引用

  1. Gee MFW, Chalfin DB, De Vol EB, et al. Meta-Analysis: Enhanced Liver Fibrosis (ELF) Test for Identifying Significant Fibrosis, Advanced Fibrosis, and Cirrhosis in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease. Aliment Pharmacol Ther. 2026. PMID: 42638152. doi:10.1111/apt.70846

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。