Objectives Randomised controlled trials (RCTs) have demonstrated conflicting results regarding the effects of corticosteroids on the treatment of severe community-acquired pneumonia (CAP). We aimed to investigate the efficacy and safety of different corticosteroids on patients who were hospitalised for severe CAP. Methods We performed a systematic search through PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science, and Scopus from inception to May 2023. The primary outcome was all-cause mortality. Data analysis was performed using a random-effects model. Results A total of 10 RCTs comprising 1962 patients were included. Corticosteroids were associated with a lower rate of all-cause mortality (risk ratio (RR), 0.70 (95% CI 0.54 to 0.90); I2=0.00%). When stratified into different corticosteroid types, hydrocortisone was associated with an approximately 50% lower mortality risk (RR, 0.48 (95% CI 0.32 to 0.72); I2=0.00%). However, dexamethasone, methylprednisolone or prednisolone were not associated with an improvement in mortality. Furthermore, hydrocortisone was associated with a reduction in the rate of mechanical ventilation, acute respiratory distress syndrome, shock and duration of intensive care unit stay. These trends were not observed for dexamethasone, methylprednisolone or prednisolone. Corticosteroids were not associated with an increased risk of adverse events including gastrointestinal bleeding, secondary infection or hyperglycaemia. Conclusions The use of hydrocortisone, but not other types of corticosteroids, was associated with a reduction in mortality and improvement in pneumonia outcomes among patients hospitalised with severe CAP. PROSPERO registration number CRD42023431360. Data are available upon reasonable request. The authors confirm that the data supporting the findings of this study are available within the article or its supplementary materials. The template data collection forms and data used for all analyses including the analytic code can be obtained upon reasonable request from the corresponding author.

中文翻译：

---

不同皮质类固醇对严重社区获得性肺炎的影响：系统评价和荟萃分析

目的 随机对照试验 (RCT) 已证明关于皮质类固醇治疗严重社区获得性肺炎 (CAP) 效果的相互矛盾的结果。我们的目的是研究不同皮质类固醇对因重症 CAP 住院患者的疗效和安全性。方法 从成立到 2023 年 5 月，我们通过 PubMed、Embase、Cochrane 对照试验中央注册库、Web of Science 和 Scopus 进行了系统检索。主要结果是全因死亡率。使用随机效应模型进行数据分析。结果 共纳入 10 项随机对照试验，共纳入 1962 名患者。皮质类固醇与较低的全因死亡率相关（风险比 (RR)，0.70（95% CI 0.54 至 0.90）；I2=0.00%）。当分层为不同的皮质类固醇类型时，氢化可的松与死亡风险降低约 50% 相关（RR，0.48（95% CI 0.32 至 0.72）；I2=0.00%）。然而，地塞米松、甲泼尼龙或泼尼松龙与死亡率的改善无关。此外，氢化可的松与机械通气、急性呼吸窘迫综合征、休克和重症监护病房住院时间的减少有关。地塞米松、甲泼尼龙或泼尼松龙没有观察到这些趋势。皮质类固醇与胃肠道出血、继发感染或高血糖等不良事件风险增加无关。结论 使用氢化可的松（而非其他类型的皮质类固醇）与重症 CAP 住院患者死亡率降低和肺炎结局改善相关。PROSPERO 注册号 CRD42023431360。数据可根据合理要求提供。作者确认，支持本研究结果的数据可在文章或其补充材料中找到。模板数据收集表格和用于所有分析的数据（包括分析代码）可以根据相应作者的合理要求获得。