2024 AASLD | 肝硬化治疗的最新热点与挑战

2024年11月15日-19日，美国肝病研究学会年会(AASLD 2024)于美国圣迭戈举行。此次会议云集肝病研究领域的最新研究成果与前沿探索，为与会者搭建出一个深化肝病认知、拓宽专业视野的高端学术交流平台。

为便于广大读者及时捕捉学术精髓，肝胆相照平台特此甄选AASLD热点研究，精心翻译并呈现大会最新动态与核心看点。本篇聚焦“肝硬化”这一专题，深入剖析其最新研究进展与临床应用前景。

➢ 背景：门静脉血栓形成(PVT)在肝硬化患者中常见。许多风险因素已被报告，如CTP评分较高、门脉血流减少、侧支循环及β-受体阻滞剂的使用。然而，这些风险因素来源于小规模研究，不能精确预测PVT的发生。

• Background: Portal vein thrombosis (PVT) frequently occurs in cirrhosis. Many risk factors for PVT have been reported, such as higher CTP class, decreased portal blood flow, collateral circulation, and beta-blocker use. However, these risk factors are derived from small studies and do not precisely predict the occurrence of PVT.

➢ 方法：本研究旨在评估大规模前瞻性队列中肝硬化患者非肿瘤性PVT的发生率及其风险因素，并建立一个预测肝硬化患者PVT发生的模型。我们分析了电子病例数据，并对没有PVT的肝硬化患者的临床和生化特征进行了全面评估，这些患者均进行了定期影像学检查。基线时和每六个月进行了多普勒超声检查。PVT通过CT扫描确认。必要时，测定血清甲胎蛋白(AFP)和PIVKA II水平。

• Methods: The current study aimed to evaluate the incidence and risk factors for non-tumoral PVT development in a large prospective cohort of patients with cirrhosis and to create a model for predicting PVT in cirrhosis. We analyzed electronic data and thoroughly evaluated the clinical and biochemical profiles of cirrhosis patients without PVT who underwent serial imaging. Doppler ultrasound was performed at baseline and every six months or whenever clinically indicated. PVT was confirmed by computed tomography (CT) scan. Serum alpha-fetoprotein and PIVKA II levels were measured whenever necessary.

➢ 结果：我们筛查了2,776例肝硬化患者。在随访过程中，274例(9.8%)发展为非肿瘤性PVT，PVT发生率分别为CTP-A(16%)、CTP-B(15.4%)和CTP-C(14.3%)。其中78%为男性，中位年龄为48岁(范围39-57岁)。肝硬化的主要原因分别为酒精性(42.7%)、NASH(44.1%)、隐源性(6.25%)和乙型肝炎(5.7%)。单因素分析显示，以下因素与PVT的发生相关：CTP评分>9.3(OR 2.74)、有食道静脉曲张出血史(OR 15.6)、年龄较大(>47.6岁，OR 4.6)、及门体侧支循环(OR 5.9)。PVT的发生与较低的生存率相关。在PVT组中，15.9%的患者死亡，而在非PVT组中，仅有1.7%的患者死亡。

• Results: We screened 2,776 cirrhosis patients. Of these, 274 (9.8%) developed non-tumoral PVT over serial follow-up, with an incidence of PVT in CTP-A (16%), CTP-B (15.4%), and CTP-C (14.3%). Seventy-eight percent were male, and the median age was 48 years (range 39-57). The main causes of cirrhosis were ethanol-related (42.7%), NASH (44.1%), cryptogenic (6.25%), and hepatitis B (5.7%). Univariate analysis showed that the following were risk factors for PVT: a CTP score >9.3 (OR 2.74), a history of variceal bleeding (OR 15.6), advanced age (>47.6 years, OR 4.6), and portosystemic collaterals (OR 5.9). Development of PVT was associated with lower survival. In the PVT group, 15.9% of patients died, whereas in the non-PVT group, only 1.7% died.

➢ 结论：在肝硬化患者中，预测PVT发生的因素主要与门脉高压的严重程度相关。我们的结果显示年龄、食道静脉曲张出血史和门体侧支循环是肝硬化患者发生PVT的风险因素，但不支持CTP分级较高作为预测因素。PVT的发生是肝硬化患者死亡的风险因素。

• Conclusion:  In patients with cirrhosis, the factors predictive of PVT development are primarily those related to the severity of portal hypertension. Our results support age, variceal bleeding, and portosystemic collaterals as risk factors for PVT development, but not higher CTP class. The occurrence of PVT is a risk factor for death in cirrhosis patients.

➢ 背景：肝硬化患者经常出现门脉高压，这会导致腹水和容量超负荷。肝硬化患者通常采用呋塞米(furosemide)和螺内酯(spironolactone)联合利尿治疗。根据美国肝脏病研究学会(AASLD)指南，联合使用这两种药物被认为是实现排钠和维持正常钾水平的首选方案。对于心力衰竭或肾功能不全的患者，SGLT2抑制剂(SGLT2-I)已显示出改善治疗效果和降低死亡率的潜力。这促使人们探索将SGLT2-I添加到肝硬化患者标准利尿方案中的疗效如何，这是一个相对未开发的领域。

• Background: Patients with cirrhosis often experience portal hypertension, which can lead to ascites and volume overload. A common diuretic regimen for these patients includes furosemide and spironolactone. According to the American Association for the Study of Liver Diseases (AASLD) guidelines, initiating both drugs together is the preferred approach to achieve natriuresis and maintain normokalemia. The use of SGLT2 inhibitors (SGLT2-I) in patients with heart failure or kidney dysfunction has shown promising results in improving outcomes and reducing mortality. This has led to interest in exploring the potential benefits of adding SGLT2-I to the standard diuretic regimen for patients with cirrhosis, a relatively unexplored area.

➢ 方法：我们从TriNetX去标识的研究数据库中查询了全球合作网络的数据。将接受常规利尿治疗(呋塞米和螺内酯)的失代偿期肝硬化患者分为两组：第一组为使用SGLT2-I(达格列净或恩格列净)的患者，第二组为未使用任何SGLT2-I的患者。排除了酒精性肝硬化患者。使用1:1倾向评分匹配(PSM)方法，根据患者的特征和合并症匹配两组。我们比较了1、3、6个月时的新发失代偿(需要进行腹腔穿刺的腹水、上消化道出血、黄疸、肝性脑病或肝肾综合征)、死亡率、腹腔穿刺需求和住院情况。

• Methods: We queried the Global Collaborative Network from the TriNetX de-identified research database. Patients with decompensated cirrhosis receiving standard diuretic therapy (furosemide and spironolactone) were divided into two cohorts: the first cohort included patients on SGLT2-I (dapagliflozin or empagliflozin), while the second cohort included patients not on any SGLT2-I. Patients with alcoholic cirrhosis were excluded. Propensity score matching (PSM) was performed on a 1:1 basis to match the cohorts based on patients’ characteristics and comorbidities. We compared the rate of new decompensation (ascites requiring paracentesis, upper gastrointestinal bleed, jaundice, hepatic encephalopathy, or hepatorenal syndrome), mortality rate, need for paracentesis, and hospital admissions at 1, 3, and 6 months.

• Results: A total of 62,491 patients with decompensated cirrhosis on furosemide and spironolactone from April 2004 to April 2024 were identified. Among these, 6.4% (n = 4,023) received SGLT2-I (dapagliflozin or empagliflozin) and 93.6% (n = 58,459) did not. The SGLT2-I cohort showed a lower rate of new decompensation at 1 month (28% vs 39.2%, p < 0.001), 3 months (38% vs 49.6%, p < 0.001), and 6 months (43.6% vs 55.2%, p < 0.001). Additionally, patients on SGLT2-I required significantly fewer paracentesis procedures at 1 month (9.9% vs 14.2%, p < 0.001), 3 months (13.1% vs 18.4%, p < 0.001), and 6 months (14.9% vs 20.5%, p < 0.001), with risk reductions ranging from 4.3% to 5.5%. The SGLT2-I group also had a significantly lower hospital admission rate at 1 month (36.4% vs 43.8%, p < 0.001), 3 months (41.2% vs 48.9%, p < 0.001), and 6 months (45% vs 52.4%, p < 0.001), with a relative risk reduction of 7.3% to 7.7%. Moreover, patients on SGLT2-I experienced lower mortality rates at 1 month (5.5% vs 9.9%, p < 0.001), 3 months (11.1% vs 19.2%, p < 0.001), and 6 months (15.3% vs 26%, p < 0.001).

➢ 结论：将SGLT2-I加于常规利尿治疗(螺内酯和呋塞米)似乎可以改善失代偿期肝硬化患者的死亡率。此外，它还降低了6个月内的住院率，并减少了腹腔穿刺的需求。在此期间，SGLT2-I组的新发失代偿事件也较少。

• Conclusion: The addition of SGLT2-I to the standard diuretic therapy (spironolactone and furosemide) appears to offer mortality benefits for patients with decompensated cirrhosis. It also reduced hospital admissions over the 6-month period and decreased the need for paracentesis. Additionally, fewer decompensation events were observed in the SGLT2-I group during this timeframe.

➢ 背景：终末期肝病模型(MELD)3.0 近期被提出用于预测终末期肝病患者的非肝移植生存率。本研究旨在评估 MELD 3.0 在死亡率预测中的表现，并将其与 Child-Pugh 评分、白蛋白-胆红素(ALBI)等级、MELD 和 MELD 钠(MELDNa)评分进行比较，并评估其与肝脏相关并发症的关联。

• Background: The Model for End-Stage Liver Disease (MELD) 3.0 was recently proposed as a predictive model for liver transplantation-free survival in patients with end-stage liver disease. This study aimed to evaluate the performance of MELD 3.0 in predicting mortality, comparing it with the Child-Pugh class, albumin-bilirubin (ALBI) grade, MELD, and MELD sodium (MELDNa) scores, as well as assessing its association with liver-related complications.

➢ 方法：我们开展了一项多中心回顾性队列研究，纳入了2013至2019年间的肝硬化患者。通过比较MELD 3.0与其他评分模型的受试者工作特征曲线下面积(AUROC)，评估其对3个月死亡率的预测能力。此外，还根据 MELD 3.0 的评分评估了肝硬化相关并发症的风险。

• Methods: We conducted a multicenter, retrospective cohort study of patients with cirrhosis from 2013 to 2019. We compared the area under the receiver operating characteristic curve (AUROC) for MELD 3.0 and other scoring models in predicting 3-month mortality. Additionally, we evaluated the risk of cirrhosis-related complications according to MELD 3.0 scores.

➢ 结果：共纳入3,314名患者，平均年龄为55.9±11.3岁，其中70.2%为男性。入院时Child-Pugh、ALBI、MELD、MELDNa和MELD 3.0的平均评分分别为7.4±1.6、-1.5±0.7、14.7±6.3、15.7±7.1和16.8±7.1。在前3个月内，共有220名患者(6.6%)死亡。MELD 3.0显示出最佳的预测表现，AUROC达到0.851，优于Child-Pugh评分(0.754)、ALBI等级(0.704)、MELD(0.846)和MELDNa(0.849)。此外，MELD 3.0与肝硬化相关并发症的风险增加相关，尤其是肝肾综合征的风险增加最为显著，调整后风险比(HR)为1.149[95%置信区间(CI)：1.111-1.188]，其次为肝性脑病(调整后 HR：1.077，95% CI：1.063-1.090)和腹水(调整后 HR：1.040，95% CI：1.028-1.053)。

• Results: A total of 3,314 patients were included, with an average age of 55.9 ± 11.3 years; 70.2% of them were male. The average scores at admission for Child-Pugh, ALBI, MELD, MELDNa, and MELD 3.0 were 7.4 ± 1.6, -1.5 ± 0.7, 14.7 ± 6.3, 15.7 ± 7.1, and 16.8 ± 7.1, respectively. During the first three months, 220 patients (6.6%) died. MELD 3.0 demonstrated the best predictive performance, with an AUROC of 0.851, surpassing the Child-Pugh class (0.754), ALBI grade (0.704), MELD (0.846), and MELDNa (0.849). MELD 3.0 was also associated with a 1.033-fold increased risk of cirrhosis-related complications. The risk of developing hepatorenal syndrome showed the greatest increase, with an adjusted hazard ratio (HR) of 1.149 [95% confidence interval(CI): 1.111-1.188], followed by hepatic encephalopathy (adjusted HR: 1.077, 95% CI: 1.063-1.090) and ascites (adjusted HR: 1.040, 95% CI: 1.028-1.053).

➢ 结论：MELD 3.0可以很好地预测肝硬化患者死亡率。此外，MELD 3.0评分还与肝硬化相关并发症的风险增加有关，尤其是肝肾综合征和腹水。

• Conclusion: MELD 3.0 demonstrated strong prognostic performance in predicting mortality in patients with cirrhosis. Additionally, it was associated with an increased risk of cirrhosis-related complications, especially those affecting kidney function, such as hepatorenal syndrome and ascites.