第一部分 重症急性心力衰竭中β受体阻滞剂应用与死亡风险的相关性分析

背景：β受体阻滞剂已被广泛证实可降低慢性心力衰竭患者死亡和再入院风险，并被指南推荐作为标准治疗。然而，在急性心力衰竭（acute heart failure, AHF），尤其是重症AHF患者中，其应用仍存在争议，尚缺乏基于大样本量研究的临床证据。

目的：在重症AHF患者中，探讨β受体阻滞剂院前、院内及出院时的不同应用状态与院内及院外死亡风险的相关性。

方法：本研究对重症监护医学信息库IV 2.2数据库的数据进行分析。研究对象为国际疾病分类编码诊断为AHF且入住重症监护室的成人患者。在针对院内死亡终点进行相关性分析时，依据β受体阻滞剂院前及院内的应用情况，把患者分为四组：未治疗（院前和院内均未应用β受体阻滞剂）、新启用（院前未应用但院内初次应用β受体阻滞剂）、停药（院前应用但院内停用β受体阻滞剂）、持续治疗（院前应用并在院内继续应用β受体阻滞剂）。采用经逆概率加权（inverse probability of treatment weighting，IPTW）方法构建的逻辑回归模型评估不同β受体阻滞剂在院前与院内的治疗策略与院内死亡风险的相关性。针对院外死亡的分析，选取幸存出院的人群，依据院前及出院β受体阻滞剂的应用情况，亦将患者分为四组：未治疗（院前和出院处方中均未应用β受体阻滞剂）、新启用（院前未应用但出院处方中应用β受体阻滞剂）、停药（院前应用但出院处方中停用β受体阻滞剂）、持续治疗（院前和出院处方中均应用β受体阻滞剂）。采用IPTW方法构建的Cox等比例风险回归模型，对院前与院外β受体阻滞剂不同治疗策略与出院后 1 年死亡风险的相关性进行评估。此外，在重大慢病国家注册登记研究心衰前瞻性队列研究（China Patient‐centered Evaluative Assessment of Cardiac Events Prospective Heart Failure，PEACE 5p-HF）中选取心功能IV级或院内静脉注射正性肌力药物/血管加压药物的人群，对出院后 1 年死亡风险的相关性进行验证。

结果：对于院内死亡的分析，纳入的5655例患者的中位年龄为75.5（65.4，84.3）岁，46.3%为女性患者，71.0%为急性失代偿性心力衰竭（acute decompensated heart failure，ADHF），68.5%为白人，中位住院时长为9.7（6.0，15.0）天，中位ICU时长为3.2（1.7，6.1）天。未治疗组有663人、新启用组有1581人、停药组有188人、持续治疗组有3223人。61.3%和85.0%的患者分别在院前和院内接受了β受体阻滞剂。在IPTW方法构建的逻辑回归模型中，与未治疗组相比，新启用组[比值比（odds ratio，OR）：0.45；95% CI：0.34–0.61；P＜0.001]和持续治疗组（OR：0.53；95% CI：0.41–0.69；P＜0.001）均与较低的院内死亡风险相关。相反，停药组与院内死亡风险增加有关（OR：2.59；95% CI：1.63–4.10；P＜0.001）。对于β受体阻滞剂应用和出院1年死亡风险的相关性分析，一共4798患者幸存出院并具备出院后一年的生存状态信息。其中未治疗组735人、新启用组1179人、停药组271人、持续治疗组2613人。在IPTW方法构建的Cox回归模型中，与未治疗组相比，新启用[危险比（hazard ratio，HR）：0.67；95% CI：0.54–0.85；P＜0.001]和持续治疗组（HR：0.73；95% CI：0.61–0.86；P＜0.001）与较低的出院1年的死亡风险相关；而停药组的出院1年的死亡风险未见显著差异（HR:1.12；95% CI：0.83–1.51；P＝0.456）。在PEACE 5p-HF队列中，与未治疗组相比，新启用组（HR：0.71；95% CI：0.52–0.97；P＝0.031）和持续治疗组（HR：0.68；95% CI：0.51–0.92；P＝0.011）均与较低的1年死亡风险相关；而停药组的 1 年死亡风险则未见差异（HR：0.74；95% CI：0.53–1.03；P＝0.070）。

结论：对于重症AHF患者，与未治疗相比，β受体阻滞剂的新启用或持续治疗与较低的院内及出院1年死亡风险相关，而院内停用β受体阻滞剂则与较高的院内死亡风险相关。未来仍需进一步的前瞻性研究加以验证。

第二部分 重症急性心力衰竭中β受体阻滞剂的启用时机和剂量与院内死亡风险的相关性分析

背景：β受体阻滞剂在AHF患者中的最佳启用时机及启用剂量尚无明确指南推荐，临床实践中的应用仍存在较大异质性。

目的：本研究旨在重症AHF人群中，探索（1）β受体阻滞剂不同启用时机与院内死亡风险的相关性；（2）β受体阻滞剂不同启用剂量与院内死亡风险的相关性。

方法：本研究基于重症监护医学信息库IV 2.2数据库，纳入国际疾病分类编码诊断为AHF且进入重症监护室并在住院期间应用β受体阻滞剂的患者。对于β受体阻滞剂启用时机和院内死亡风险的相关性分析，依据院前β受体阻滞剂的应用情况以及以入院 48 小时为界限的院内启用时间早晚，将患者分为四组：组1（院前应用β受体阻滞剂且在入院48小时内启用）、组2（院前应用β受体阻滞剂且在入院48小时后启用）、组3（院前未应用β受体阻滞剂且在入院48小时内启用）、组4（院前未应用β受体阻滞剂且在入院48小时后启用）。采用IPTW方法构建的逻辑回归模型评估β受体阻滞剂启用时机与院内死亡风险的相关性。在针对β受体阻滞剂不同启用剂量与院内死亡风险的相关性分析方面，依照既往文献的等效剂量转换比例，选取起始口服用药为美托洛尔、比索洛尔和卡维地洛的患者，对其进行卡维地洛等效剂量的转换，进而分为高启用剂量（卡维地洛等效剂量≥10mg）以及低启用剂量组（卡维地洛等效剂量＜10 mg）。采用IPTW方法构建的逻辑回归模型评估β受体阻滞剂启用剂量与院内死亡风险的相关性。

结果：在院内β受体阻滞剂启用时机与院内死亡风险相关性分析中，共纳入4804例患者，中位年龄为75.4（65.6，84.1）岁，45.8%为女性患者，70.9%为ADHF，68.6%为白人，中位入院时长为9.9（6.2，15.5）天，中位ICU时长为3.2（1.7，6.2）天。组1～组4分别为2644、579、926和655人。共有519例发生院内死亡。从入院到启用β受体阻滞剂的中位时间为12.0（4.2，49.8）小时。在IPTW方法构建的模型中，与组1相比，组2（OR：0.44；95% CI：0.30–0.64；P＜0.001）和组3（OR：0.65；95% CI：0.48–0.86；P＝0.003）均与较低的院内死亡风险相关；而组4的院内死亡风险未见显著差异（OR：0.82；95% CI：0.60–1.11；P＝0.202）。在启用剂量与院内死亡风险相关性分析中，纳入的3606例患者，中位启用剂量为10.0（5.0，15.0）mg。1700人为低启动剂量，1907人为高启用剂量。高低启用剂量并未观察到院内死亡风险的差异（OR：0.92；95% CI：0.70–1.21；P＝0.534）。

结论：在重症AHF人群中，相较于院前应用β受体阻滞剂且在入院48小时内启用的患者，院前应用但在入院48小时后启用以及院前未使用但在入院48小时内新启用的患者，均与较低的院内死亡风险相关。β受体阻滞剂的不同起始剂量与院内死亡风险未见明显相关性。本研究为重症AHF患者中β受体阻滞剂的个体化用药策略提供了依据。

Part 1 Association Between Beta-Blocker Use and    Mortality Risk in Acute Heart Failure with Critical Ill

Background: Beta-blockers have been widely demonstrated to reduce mortality and rehospitalization rates in patients with chronic heart failure and are recommended as standard therapy by major guidelines. However, in acute heart failure (AHF), particularly in critically ill AHF patients, the use of beta-blockers remains controversial, and large-scale data-driven studies are lacking to clarify their clinical value in this population.

Objective: To investigate the association between different pre-hospital, in-hospital, and discharge statuses of beta-blockers use and the risk of in-hospital and one-year post-discharge mortality in critically ill patients with AHF.

Methods: This study analyzed data from the Medical Information Mart for Intensive Care IV version 2.2 (MIMIC-IV 2.2) database. The study population consisted of adult patients diagnosed with AHF based on International Classification of Diseases codes and admitted to the intensive care unit (ICU). For the analysis of in-hospital mortality, patients were categorized into four groups based on their pre-admission and in-hospital beta-blocker use: Not Treated (no beta-blockers pre-admission and in-hospital), Newly Started (no pre-admission beta-blockers but initiated in-hospital), Withdrawn (pre-admission beta-blockers use but discontinued in-hospital), and Continued (beta-blockers used both pre-admission and in-hospital). Inverse probability of treatment weighting (IPTW) weighted logistic regression was applied to assess the association between beta-blocker use and in-hospital mortality. For the analysis of post-discharge mortality, patients were categorized into four groups based on their pre-admission and discharge beta-blocker use: Not Treated (no beta-blockers pre-admission and at discharge), Newly Started (no pre-admission beta-blockers but initiated at discharge), Withdrawn (pre-admission beta-blockers use but discontinued at discharge), and Continued (beta-blockers used both pre-admission and at discharge). IPTW weighted Cox proportional hazards regression was used to evaluate the association between beta-blocker use and one-year post-discharge mortality. In addition, we validated the association with 1-year post-discharge mortality in a high-risk population selected from the China Patient-centered Evaluative Assessment of Cardiac Events Prospective Heart Failure study (PEACE 5p-HF), which included patients with New York Heart Association class IV heart failure or those who received intravenous inotropes or vasopressors.

Results: For the in-hospital mortality analysis, a total of 5655 patients were included, with a median age of 75.5 (65.4, 84.3) years, 46.3% being female, 71.0% admitted with acute decompensated heart failure (ADHF), and 68.5% identified as White. The median hospital length of stay was 9.7 (6.0, 15.0) days, and ICU stay was 3.2 (1.7, 6.1) days. The Not Treated group included 663 patients, Newly Started 1581 patients, Withdrawn 188 patients, and Continued 3223 patients. Beta-blockers were used in 61.3% pre-admission and in 85.0% in-hospital. In the IPTW weighted logistic regression model, compared to the Not Treated group, both the Newly Started (odds ratio [OR]：0.45；95% CI：0.34–0.61；P < 0.001) and Continued (OR：0.53；95% CI：0.41–0.69；P < 0.001) groups were associated with a lower risk of in-hospital mortality. Conversely, the Withdrawn group was associated with an increased risk of in-hospital mortality (OR: 2.59; 95% CI: 1.63–4.10; P < 0.001). For the analysis of the association between beta-blocker use and one-year mortality, a total of 4798 patients who survived to discharge. The Not Treated group included 735 patients, Newly Started 1179 patients, Withdrawn 271 patients, and Continued 2613 patients. In the IPTW weighted Cox regression model, compared to the Not Treated group, both the Newly Started (hazard ratio [HR]: 0.67; 95% CI: 0.54–0.85; P < 0.001) and Continued (HR: 0.73; 95% CI: 0.61–0.86; P < 0.001) groups were associated with a lower risk of one-year post-discharge mortality. However, the Withdrawn group did not show a significant difference in one-year post-discharge mortality risk (HR: 1.12; 95% CI: 0.83–1.51; P = 0.456). In the PEACE 5P-HF cohort, compared with the Not Treated group, both the Newly Started (HR: 0.71; 95% CI: 0.52–0.97; P = 0.031) and Continued (HR: 0.68; 95% CI: 0.51–0.92; P = 0.011) were associated with a lower 1-year mortality risk, whereas no significant difference was observed in the Withdrawn group (HR: 0.74; 95% CI: 0.53–1.03; P = 0.070).

Conclusion: Among critically ill with AHF, Newly Started or Continued use of beta-blockers was associated with a lower risk of in-hospital and 1-year post-discharge mortality compared to Not Treated, whereas in-hospital withdrawal of beta-blockers was associated with an increased risk of in-hospital mortality. Further prospective studies are warranted to confirm these findings.

Part 2 The Association Between the Timing and Dose of Beta-Blocker Initiation and In-Hospital Mortality in Acute Heart Failure with Critical Ill

Background: The optimal timing and dosage of beta-blocker initiation in patients with AHF remain unclear, particularly in those with hemodynamic instability, where clinical practice varies widely.

Objective: This study aimed to explore the association between (1) different beta-blocker initiation timing and in-hospital mortality, and (2) different beta-blocker initiation doses and in-hospital mortality in AHF with critically ill.

Methods: This study utilized data from the Medical Information Mart for Intensive Care version 2.2 database. Adult patients diagnosed with AHF based on International Classification of Diseases codes who were admitted to the ICU and received beta-blocker therapy during hospitalization were included. For the analysis of beta-blocker initiation timing and in-hospital mortality, patients were categorized into four groups based on prior beta-blocker use and the timing of in-hospital initiation (using 48 hours post-admission as a cutoff): Group 1(pre-admission beta-blocker use with initiation within 48 hours of admission), Group 2 (pre-admission beta-blocker use with initiation after 48 hours of admission), Group 3 (no pre-admission beta-blocker use with initiation within 48 hours of admission), and Group 4 (no pre-admission beta-blocker use with initiation after 48 hours of admission). IPTW weighted logistic regression was used to evaluate the association between beta-blocker initiation timing and in-hospital mortality. For the analysis of beta-blocker initiation dosage and in-hospital mortality, patients who were orally initiated on metoprolol, bisoprolol, or carvedilol were included, and carvedilol-equivalent doses were calculated. Patients were then categorized into a high-dose initiation group (carvedilol equivalent dose ≥ 10 mg) and a low-dose initiation group (carvedilol equivalent dose < 10 mg). IPTW weighted logistic regression was used to assess the association between beta-blocker initiation dose and in-hospital mortality.

Results: For the analysis of the association between initiation timing and in-hospital mortality, 4804 eligible patients were included, with a median age of 75.4 (65.6, 84.1) years; 45.8% were female, 70.9% were ADHF, and 68.6% were White. The median length of hospital stay was 9.9 (6.2, 15.5) days, and ICU stay was 3.2 (1.7, 6.2) days. The distribution of patients was as follows: Group 1 (n=2644), Group 2 (n=579), Group 3 (n=926), and Group 4 (n=655). A total of 519 patients experienced in-hospital mortality. The median time from admission to beta-blocker initiation was 12.0 (4.2, 49.8) hours. After weighting, compared to Group 1, both Group 2 (OR: 0.44; 95% CI: 0.30–0.64; P < 0.001) and Group 3 (OR: 0.65; 95% CI: 0.48–0.86; P = 0.003) were associated with a lower in-hospital mortality risk. However, Group 4 showed no significant difference from Group 1 in in-hospital mortality (OR: 0.82; 95% CI: 0.60–1.11; P = 0.202). For the analysis of the association between initiation dosage and in-hospital mortality, 3606 patients were included, with a median initiation dose of 10.0 (5.0, 15.0) mg. There were 1700 patients in the low-dose group and 1907 in the high-dose group. No significant difference in in-hospital mortality was observed between high-dose and low-dose initiation groups (OR: 0.92; 95% CI: 0.70–1.21; P = 0.534).

Conclusion: In critically ill patients with AHF, compared with early initiation (< 48 hours) in patients who had prior beta-blocker use, both late initiation (≥ 48 hours) in prior users and early initiation in beta-blocker-naïve patients were significantly associated with a lower risk of in-hospital mortality. Moreover, higher initiation doses of beta-blockers were not associated with in-hospital mortality. These findings provide novel and clinically relevant evidence to optimize beta-blocker initiation strategies in critically ill patients with AHF.