Non-selective beta-blockers (NSBBs) are highly effective in preventing variceal bleeding. They are therefore widely used in patients with liver cirrhosis and portal hypertension [1]. Since the landmark study by Lebrec et al. many studies have investigated NSBB use in cirrhosis, including the controversial paper by Serste et al. which suggested increased mortality in decompensated cirrhosis due to NSBB use [2]. Despite its methodological limitations, this study sparked extensive research on NSBBs in decompensated cirrhosis. Krag et al. thereafter introduced the ‘window hypothesis’, proposing that NSBBs might harm patients with advanced liver cirrhosis by reducing cardiac output in those already at risk of low organ perfusion [3]. However, NSBBs not only reduce portal pressure, they may also possess anti-inflammatory effects [4]. Therefore, NSBB use is advocated in all patients with clinically significant portal hypertension, not just in those with oesophageal varices [5]. Nonetheless, the safety of NSBBs in patients with decompensated cirrhosis, particularly those with acute kidney injury (AKI)/hepatorenal syndrome (HRS), and refractory ascites, remains controversial.

The study by Wang et al. explores NSBB safety in a hospitalised Asian population with already decompensated cirrhosis and finds that NSBBs did not affect further decompensation nor overall survival [6]. However, a significant interaction between admission MELD-score and NSBB use justified a stratified analysis by MELD-score (low [≤ 9] vs. high MELD-scores [> 9]). NSBB use was associated with a lower risk of further decompensation in the low MELD group but a higher risk of further decompensation in the high MELD group. This stratified analysis did not show an association with overall survival. Advantages of this study are the population size (N = 332) with a sufficient number of events and median follow-up of almost 4 years. It highlights a very relevant and timely topic. However, the limitations of this study must also be considered. As this was an all-Asian population using low-dose propranolol, the generalisability of this study is limited. Also, many studies have shown an advantage of carvedilol over traditional NSBBs, which is now also advocated by the BAVENO VII consensus [5, 7, 8]. Furthermore, the retrospective study design may have introduced bias. Finally, clearer justification for the MELD-9 cut-off, perhaps using visual splines, would have been helpful.

Building upon this current study there seems to be a non-linear association between NSBB benefit and liver disease severity. Efforts have been made to define this specific turning point in which NSBB may be detrimental rather than beneficial. In the current BAVENO VII consensus this point is defined by a persistent systolic blood pressure of < 90 mmHg or the presence of AKI/HRS [5]. It is difficult to draw definitive conclusions on this issue based on previous studies due to heterogeneity in NSBB type and dosage, aetiology-skewed or selected study populations, in definition of portal hypertension and/or in design (often retrospective and non-randomised). Also, further studies on the safety of stopping and restarting NSBBs are needed, while there is limited evidence on this topic [9]. In conclusion, the study by Wang et al. highlights an important and timely issue, raising new and relevant questions for future research.

非选择性 β 受体阻滞剂 （NSBBs） 在预防静脉曲张出血方面非常有效。因此，它们被广泛用于肝硬化和门静脉高压症患者 [1]。自 Lebrec 等人进行具有里程碑意义的研究以来，许多研究调查了 NSBB 在肝硬化中的应用，包括 Serste 等人有争议的论文，该论文表明使用 NSBB 会增加失代偿期肝硬化的死亡率 [2]。尽管存在方法学上的局限性，但这项研究引发了对失代偿性肝硬化中 NSBB 的广泛研究。Krag 等人随后提出了“窗口假说”，提出 NSBB 可能通过减少已经存在低器官灌注风险患者的心输出量来伤害晚期肝硬化患者 [3]。然而，NSBBs 不仅可以降低门静脉压力，还可以具有抗炎作用 [4]。因此，提倡所有有临床意义的门静脉高压患者都使用NSBB，而不仅仅是食管静脉曲张患者[5]。尽管如此，NSBBs 在失代偿期肝硬化患者中的安全性仍然存在争议，尤其是急性肾损伤 （AKI）/肝肾综合征 （HRS） 和难治性腹水患者。

Wang 等人的研究探讨了 NSBB 在已经失代偿性肝硬化的亚洲住院人群中的安全性，发现 NSBBs 不会影响进一步的失代偿或总生存期 [6]。然而，入院 MELD 评分和 NSBB 使用之间的显着交互证明了按 MELD 评分 （低 [≤ 9] 与高 MELD 评分 [> 9] 进行分层分析是合理的。NSBB 使用与低 MELD 组进一步失代偿的风险较低相关，但高 MELD 组进一步失代偿的风险较高。这种分层分析未显示与总生存期相关。这项研究的优点是人群规模 （N = 332） 和足够的事件数量和近 4 年的中位随访。它突出了一个非常相关和及时的主题。但是，还必须考虑这项研究的局限性。由于这是使用低剂量普萘洛尔的全亚洲人群，因此本研究的普遍性有限。此外，许多研究表明卡维地洛优于传统的 NSBBs，现在 BAVENO VII 共识也倡导这一点 [5， 7， 8]。此外，回顾性研究设计可能引入了偏倚。最后，更清晰地证明 MELD-9 临界值的理由，也许使用可视样条，会有所帮助。

基于目前的这项研究，NSBB 益处与肝病严重程度之间似乎存在非线性关联。已经努力定义 NSBB 可能是有害而不是有益的这个特定转折点。在目前的 BAVENO VII 共识中，这一点定义为持续收缩压 <、90 mmHg 或存在 AKI/HRS [5]。由于 NSBB 类型和剂量的异质性、病因学偏倚或选定的研究人群、门静脉高压症的定义和/或设计（通常是回顾性和非随机性的）存在异质性，因此很难根据以前的研究就此问题得出明确的结论。此外，需要进一步研究停止和重新启动 NSBB 的安全性，但关于该主题的证据有限 [9]。总之，Wang 等人的研究强调了一个重要而及时的问题，为未来的研究提出了新的相关问题。