2025 AASLD | 肝衰竭领域最新热点

肝衰竭领域

FRAILTY IS ASSOCIATED WITH THE DEVELOPMENT OF ACUTE-ON-CHRONIC LIVER FAILURE IN PATIENTS WITH END-STAGE LIVER DISEASE

作者：Avesh Thuluvath¹，Fawzy Barry²，Brendan Blackburn³，Reeti Gulati⁴，Angel Wilson⁵，Grace Johnson⁶，Avery Longdon⁶，Caroline Yu⁶，Anjana Rajan⁷，Yuan Zhang⁸，Robert Rahimi⁵，Andres Duarte-Rojo⁶，Matthew Kappus⁴，Jennifer Lai²，Elizabeth Verna¹

¹Division of Digestive and Liver Disease, Columbia University Irving Medical Center, New York, NY；²Division of Gastroenterology and Hepatology, University of California, San Francisco, CA；³Division of Digestive and Liver Disease, Columbia University Irving Medical Center, New York, NY；⁴Division of Gastroenterology and Hepatology, Duke University Hospital, Durham, NC；⁵Annette C. and Harold C. Simmons Transplant Institute, Baylor University Medical Center, Dallas, TX；⁶Division of Gastroenterology and Hepatology, Northwestern University Feinberg School of Medicine, Chicago, IL；⁷Division of Gastroenterology and Hepatology, Northwestern University Feinberg School of Medicine, Chicago, IL；⁸Mailman School of Public Health, Columbia University, New York, NY

在终末期肝病（ESLD）患者中，急性加重型慢性肝衰竭（ACLF）与显著的短期死亡风险密切相关。然而，ACLF 与衰弱（frailty）之间的关联尚未明确。本研究旨在探讨门诊环境下评估的衰弱程度与 ACLF 发生风险的关联。

Background：Acute-on-chronic liver failure (ACLF) is associated with significant short-term mortality in patients with end-stage liver disease (ESLD). The relationship between ACLF and frailty is unclear. The purpose of this study was to investigate the association between frailty measured in an ambulatory setting and the risk of ACLF..

本研究对前瞻性 FrAILT 研究中纳入的成年患者进行回顾性病历分析。2019 年 1 月 1 日至 2022 年 12 月 31 日期间，这些患者在五家肝移植（LT）中心门诊被列入肝移植等待名单。采用肝脏衰弱指数（LFI，涵盖握力、坐立测试及平衡测试）评估患者的衰弱程度，ACLF 的定义参照欧洲肝脏研究学会慢性肝衰竭联盟（EASL-CLIF）标准。主要研究终点为 LFI 评估后 1 年内发生≥1 级 ACLF。采用事件时间分析评估 LFI 与 ACLF 的关联，同时将死亡或肝移植作为竞争风险。

Methods：We performed a retrospective chart review of adult patients enrolled in the prospective FrAILT study, who were waitlisted for liver transplantation (LT) in an ambulatory setting at five LT centers between 1/1/2019-12/31/2022. Frailty was measured with the Liver Frailty Index (LFI: grip, chair stands, balance testing). ACLF was defined using the European Association for the Study of the Liver Chronic Liver Failure Consortium (EASL-CLIF) criteria. The primary outcome was the development of ≥ grade 1 ACLF within 1 year of LFI assessment. A time-to-event analysis was performed to evaluate the association between LFI and ACLF, with death or LT as competing risks.

研究共纳入 460 例患者，其中男性占 62%，平均年龄为 56.3±6.9 岁。最常见的病因是酒精相关性肝病（32%）和代谢功能障碍相关脂肪性肝病（26%），平均 MELD 评分为 16.3±6.4，平均 LFI 为 3.90±0.74。根据既定 LFI 截断值，19% 的患者为衰弱型，64% 为衰弱前期，17% 为稳健型。1 年随访期间，87 例（19%）患者发生 ACLF，271 例接受肝移植，13 例死亡。多变量分析（校正年龄、性别及 MELD 评分）显示，LFI 与 ACLF 发生风险显著相关（每增加 1.0 个 LFI 单位，HR=1.40，95% CI：1.06~1.86，P=0.02）。按衰弱状态分层的单变量分析显示，与衰弱组相比，稳健组患者的 ACLF 发生风险显著降低（HR=0.45，95% CI：0.22~0.95，P=0.04）；但在多变量分析中，这一负相关趋势未达到统计学显著性（HR=0.44，95% CI：0.18~1.03，P=0.06）。

Results：460 patients were included; 62% were male and the mean age was 56.3 (±6.9) years. Alcohol-associated liver disease (32%) and metabolic-dysfunction associated steatotic liver disease (26%) were the most common etiologies of ESLD. Mean MELD was 16.3 (±6.4). Mean LFI was 3.90 (±0.74), with 19% classified as frail, 64% pre-frail and 17% robust (using previously established LFI cut-offs). 87 (19%) patients developed ACLF during the 1-year follow-up period, 271 underwent LT and 13 died. On multivariable analysis adjusted for age, sex and MELD, LFI was associated with the development of ACLF (HR 1.40, 95% CI 1.06-1.86, p=0.02, per 1.0 LFI). When categorized by frailty status, robust patients were less likely to develop ACLF compared to frail patients on univariable analysis (HR 0.45, 95% CI 0.22-0.95, p=0.04) (Figure). However, on multivariable analysis, robust status was no longer inversely associated with the development of ACLF (HR 0.44, 95% CI 0.18-1.03, p=0.06).

本项多中心研究表明，基于 LFI 评估的衰弱状态与 EASL-CLIF 标准定义的 ACLF 发生风险显著相关。研究结果提示，衰弱对 ESLD 患者死亡风险的影响，可能归因于体能减退患者更高的 ACLF 发生风险。

Conclusion： In this multi-center study, frailty measured by LFI was associated with the development of ACLF by EASL-CLIF criteria. Our findings suggest that frailty’s known effect on mortality in ESLD could be in part due to an increased risk of ACLF in deconditioned patients.

ESTABLISHING A RANK-ORDERED HEIRARCHY OF ACLF RELATED MORTALITY RISK ASSOCIATED WITH ORGAN SUPPORT IN REAL WORLD CLINICAL PRACTICE

作者：Kavin Parmar¹，Ana Oliveira¹，Vinay Jahagirdar¹，Ekaterina Smirnova¹，Katharina Staufer²，Juan Pablo Arab¹，Richard Moreau³，Jasmohan Bajaj¹，Arun Sanyal¹

¹Virginia Commonwealth University, ²GENFIT, ³European Foundation for the Study of Chronic Liver Failure

器官衰竭（OF）的数量及器官支持需求均与急性加重型慢性肝衰竭（ACLF）患者的死亡率相关，但在真实临床实践中，不同器官衰竭组合模式及器官支持需求对患者死亡率的具体影响尚未明确。

Background：The number of organ failures (OF) and need for organ support are linked to mortality in ACLF. The impact of varying combinations of OFs and need for organ support on mortality in real-world clinical practice is unknown.

器官衰竭的数量、类型以及器官支持需求，与 ACLF 患者的 30 天死亡率存在差异化关联。

Hypothesis: The number and type of OFs and need for organ support are differentially related to 30-day mortality in ACLF.

本研究旨在明确两点核心内容：（1）在真实临床实践中，单一器官衰竭及多器官衰竭联合发生时，对 ACLF 患者死亡率的影响；（2）当患者存在 2-4 种器官衰竭联合发生的情况时，器官支持需求与死亡率之间的关系。

Aims: To define (1) the impact of OFs, alone and in combination, on mortality in ACLF in real-world clinical practice and (2) within 2-4 OF combinations, the relationship of need for organ support to mortality.

本研究基于 TriNetX 全球协作网络数据，对住院患者开展回顾性分析。ACLF 的定义依据如下：患者存在肝硬化失代偿（表现为腹水、肝性脑病或食管胃底静脉曲张出血）的诊断编码，且同时合并一种或多种器官衰竭（OF），具体包括循环系统衰竭（如低血压、弥散性血管内凝血（DIC）等）、脑衰竭（显性肝性脑病）、肾衰竭（急性肾损伤（AKI））及肺衰竭（呼吸衰竭）。其中，升压药使用、肾脏替代治疗（RRT）及呼吸机支持的应用，代表患者对器官支持的需求。研究选取仅存在肝硬化失代偿但未合并 ACLF 的患者作为对照组，通过匹配年龄、种族 / 民族、性别、体质指数（BMI）、吸烟状况、2 型糖尿病、心力衰竭、慢性肾病、高血压及慢性阻塞性肺病等因素，计算合并一种或多种器官衰竭的 ACLF 患者与对照组的 30 天死亡率比值比（OR）；同时在不同器官衰竭组合亚组中，分析死亡率与所需器官支持类型的关联。

Methods：A retrospective analysis of hospitalized patients was performed using the TriNetX global collaborative network. ACLF was defined by diagnosis codes for cirrhosis with decompensation (ascites, encephalopathy or variceal hemorrhage) with one or more OFs, i.e. circulatory (hypotension, DIC etc.), brain (overt hepatic encephalopathy), renal (acute kidney injury [AKI]) and lung (respiratory failure). The use of pressors, renal replacement therapy (RRT), and ventilator support represented need for organ support. Odds ratio (OR) for 30-day mortality for those with one or combinations of OF versus controls admitted with cirrhosis decompensation without ACLF matched for age, race/ethnicity, sex, BMI, smoking status, type 2 Diabetes Mellitus, heart failure, chronic kidney disease, hypertension and chronic obstructive pulmonary disease were computed. Within combinations of OFs, the relationship between mortality and need for types of organ support was measured.

研究共纳入 54,592 例合并不同器官衰竭（OF）组合的 ACLF 患者，并完成与对照组的匹配。单一器官衰竭患者的 30 天死亡率 OR 值（95% CI）排序为：脑衰竭 1.42（1.3-1.6）、循环衰竭 1.87（1.7-2）、肾衰竭 3.2（3-3.3）、肺衰竭 4（3.7-4.3）；两种器官衰竭联合患者的 30 天死亡率 OR 值（95% CI）排序为：循环 - 脑衰竭 2.9（2.5-3.4）、肾 - 脑衰竭 5.2（4.6-5.8）、肾 - 循环衰竭 5.5（5.1-5.8）、肺 - 循环衰竭 6.2（5.5-6.9）、肺 - 脑衰竭 8.9（7.1-11.3）、肺 - 肾衰竭 11.9（11.2-12.7）。此外，三种器官衰竭（肺 - 肾 - 脑衰竭）患者的死亡率 OR 值为 18.1（15.4-21.1），四种器官衰竭（肺 - 肾 - 脑 - 循环衰竭）患者的死亡率 OR 值最高，达 19.7（18-21.5）。

Results：Up to 54,592 cases of ACLF with various combinations of OF were identified with matched controls. The rank-order of outcomes in single OF was: OR (95% CI): brain-1.42 (1.3-1.6), circulatory-1.87 (1.7-2), kidney-3.2 (3-3.3), lung-4 (3.7-4.3). The rank-order of outcomes for 2 OF were: circulatory-brain-2.9 (2.5-3.4), kidney-brain-5.2 (4.6-5.8), kidney-circulatory-5.5 (5.1-5.8), lung-circulatory-6.2 (5.5-6.9), lung-brain-8.9 (7.1-11.3), lung-kidney-11.9 (11.2-12.7). Those with 3-OF lung-kidney-brain-18.1 (15.4-21.1) and 4-OF lung-kidney-brain-circulatory-19.7 (18-21.5) had the highest mortality. The impact of type of organ support for each 2-4 OF combinations on mortality is shown in the Table.

ACLF 患者的结局存在显著异质性，其死亡风险根据器官衰竭的组合类型及器官支持类型呈现不同层级。该研究结果可为 ACLF 相关临床研究的试验设计及分级终点制定提供参考依据。

Conclusion：ACLF is heterogeneous in its outcome-profile with a hierarchy of mortality risk based on types of OF and organ support. These can inform trial design and development of hierarchical endpoints.

HDL-C, CHOLESTEROL AND THE ALT/HDL-C RATIO ARE PREDICTORS OF SURVIVAL IN PATIENTS WITH ACUTE-ON-CHRONIC LIVER FAILURE

作者：Georg Kramer¹、Vlad Taru¹、Benedikt Simbrunner¹、Benedikt Hofer¹、Nina Dominik¹、Lorenz Balcar¹、Lukas Hartl¹、Mathias Jachs¹、Michael Schwarz¹、Christian Sebesta¹、Georg Semmler¹、Paul Thoene¹、Philipp Schwabl¹、Michael Trauner¹、Mattias Mandorfer¹、Thomas Reiberger¹

1Christian Sebesta1 Georg Semmler1 Paul Thoene1 Philipp Schwabl1 Michael Trauner1 Mattias Mandorfer1 Thomas Reiberger1, 1Medical University of Vienna

肝硬化进展常伴随高密度脂蛋白胆固醇（HDL-C）、低密度脂蛋白胆固醇（LDL-C）及总胆固醇水平降低，这一变化可反映肝功能障碍程度。鉴于 HDL-C 具有调节全身炎症的作用，低脂蛋白血症或可对急性加重型慢性肝衰竭（ACLF）的病情产生影响。本研究旨在描述 ACLF 患者血脂谱（包括 HDL-C、LDL-C、总胆固醇及甘油三酯）的变化轨迹，并评估其对 ACLF 患者的预后价值。

Background：Progression to cirrhosis is commonly linked to reduced high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and total cholesterol, reflecting hepatic dysfunction. As HDL-C modulates systemic inflammation, hypolipoproteinemia may impact on acute-on-chronic liver failure (ACLF). This study aimed to characterize lipid profile trajectories (HDL-C, LDL-C, total cholesterol, and triglycerides) and assess their prognostic utility in ACLF.

回顾性分析 2003 年 11 月至 2022 年 11 月期间于维也纳综合医院确诊的 ACLF 患者（符合 EASL-CLIF 标准）。数据收集时间节点包括 ACLF 诊断前（诊断前 30~365 天）、诊断时（D0）及诊断后第 7 天（D7）、第 28 天（D28）、第 90 天（D90）。主要研究终点为患者 D28 及 D90 死亡率。采用受试者工作特征曲线下面积（AUROC）评估指标的鉴别诊断效能，通过 DeLong 检验进行组间比较；采用 Cox 回归模型（校正 MELD 评分、年龄、性别及 C 反应蛋白（CRP）水平）筛选死亡率的独立预测因子。

Methods：We retrospectively identified patients with ACLF (EASL-CLIF criteria) at the Vienna General Hospital between 11/2003 and 11/2022. Data were collected at pre-ACLF (30-365 days before diagnosis), at diagnosis (D0), and days 7 (D7), 28 (D28), and 90 (D90) thereafter. Primary endpoints were mortality at D28 and D90. Discrimination was evaluated using area under the receiver operating characteristic curve (AUROC) compared via DeLong’s test. Cox regression models – adjusted for MELD, age, sex and C-reactive protein (CRP) – were used to identify independent predictors of mortality.

研究共纳入 164 例 ACLF 患者，其中 G1 期 80 例（48.8%）、G2 期 53 例（32.3%）、G3 期 31 例（18.9%）。随访至 D28 时，60 例（36.6%）患者死亡；随访至 D90 时，85 例（51.8%）患者死亡。与 D0 相比，ACLF 诊断前患者的 HDL-C 及总胆固醇浓度显著更高（P < 0.001），而 D90 时上述指标恢复至正常水平（与 D0 相比，P < 0.001）。

LDL-C 呈现相似的动态变化趋势（ACLF 诊断前 vs D0：P = 0.022；D0 vs D90：P = 0.009），甘油三酯水平则基本维持稳定。D0 时，HDL-C 对 D28（AUROC = 0.724）及 D90（AUROC = 0.736）死亡率的预测效能，与 CLIF-C ACLF 评分（D28：AUROC = 0.787；D90：AUROC = 0.752）相当。

由于晚期 ACLF 患者（G3 期）的 ALT 水平显著高于早期患者（G1 期 vs G3 期：P < 0.001；G2 期 vs G3 期：P = 0.020），进一步分析 ALT/HDL-C 比值的预后价值，结果显示：该比值对 D28（AUROC = 0.817）及 D90（AUROC = 0.816）死亡率的判别效能，在数值上高于 CLIF-C ACLF 评分。多变量 Cox 回归分析显示，ALT/HDL-C 比值是 D28（校正风险比 aHR = 1.028，P < 0.001）及 D90（aHR = 1.030，P < 0.001）死亡率的独立预测因子。此外，按 ALT/HDL-C 比值（< 2 vs ≥ 2）、HDL-C（< 20 vs ≥ 20 mg/dL）及总胆固醇（< 80 vs ≥ 80 mg/dL）进行分层，均能有效区分患者死亡风险（所有对数秩检验 P < 0.01）。

Results：The study included 164 patients with ACLF (G1: 80 (48.8%), G2: 53 (32.3%), G3: 31 (18.9%)). By D28 and D90 respectively, 60 (36.6%) and 85 (51.8%) patients died. HDL-C and total cholesterol concentrations were significantly higher pre-ACLF compared to D0 (p < 0.001) and recovered by D90 (p <0.001 vs. D0). LDL-C showed similar dynamics (pre-ACLF vs. D0: p =0.022; D0 vs. D90: p =0.009), while triglyceride levels remained largely stable. At D0, HDL-C exhibited comparable AUROCs to the CLIF-C ACLF score for D28 (0.724 vs. 0.787) and D90 (0.736 vs. 0.752) mortality. With higher alanine aminotransferase (ALT) levels in advanced ACLF (G1 vs. G3: p< 0.001; G2 vs. G3: p = 0.020), prognostic impact of the ALT/HDL-C ratio was analyzed, yielding numerically higher AUROCs for D28 (0.817) and D90 (0.816) mortality discrimination than the CLIF-C ACLF score. ALT/HDL-C independently predicted both D28 (aHR: 1.028, p <0.001) and D90 (aHR: 1.030, p < 0.001) mortality in multivariable models. Stratification using the ALT/HDL-C ratio ( < 2 vs. ≥ 2), HDL-C ( < 20 vs. ≥20 mg/dL) and total cholesterol ( <80 vs. ≥ 80 mg/dL), proved effective (all log-rank p <0.01).

HDL-C 及总胆固醇水平降低在 ACLF 患者中较为常见，且可反映 ACLF 的病程进展轨迹。ALT/HDL-C 比值是一种可靠且易获取的 ACLF 预后生物标志物，基于常规实验室参数即可计算，适用于临床床边风险分层。

Conclusion： Reductions in HDL-C and total cholesterol are common in ACLF and reflect ACLF trajectory. The ALT/HDL-C ratio represents a robust and easily accessible prognostic ACLF biomarker score enabling bed-side risk stratification using routine laboratory parameters.