文献速递呼吸道感染

低剂量糖皮质激素治疗重度肺部感染:随机对照试验的荟萃分析

Low-dose corticosteroids in severe pulmonary infection: a meta-analysis of randomised controlled trials
RSS 订阅 · PubMed 高证据 · 肺炎 · 2026-07-09

该系统评价纳入12项随机对照试验共4622例重度肺部感染成人患者,比较低剂量糖皮质激素(每日≤400 mg氢化可的松等效剂量)与安慰剂或标准治疗。结果显示短期死亡率显著降低(OR 0.83,95%CI 0.70–0.98,p=0.029),社区获得性肺炎亚组获益更明显(OR 0.72,95%CI 0.54–0.98);疗程>7天者死亡率获益显著,≤7天者无显著差异。

要点

  • 低剂量糖皮质激素可降低重度肺部感染患者短期死亡率,NNT为37.84。
  • 社区获得性肺炎患者短期死亡率获益更显著,NNT为28.10。
  • 疗程>7天者死亡率获益显著,≤7天者无显著差异。

核心结论

  • 低剂量糖皮质激素可显著降低重度社区获得性肺炎90天死亡率,但对其他重度肺部感染及更短随访间隔的获益尚不确定。

研究设计与样本

  • 系统评价与荟萃分析,纳入12项随机对照试验,共4622例重度肺部感染成人患者,1203个事件。
  • 干预为低剂量糖皮质激素(每日≤400 mg氢化可的松等效剂量),对照为安慰剂或标准治疗。
  • 主要结局为短期死亡率(≤90天),次要结局包括28天死亡率、30天死亡率、至少一次严重不良事件患者数、住院时长和ICU停留时长。

主要发现

  • 短期死亡率合并OR为0.83(95%CI 0.70–0.98;p=0.029;I²=2.0%;NNT=37.84;GRADE证据质量低)。
  • 社区获得性肺炎患者短期死亡率OR为0.72(95%CI 0.54–0.98;p=0.034;NNT=28.10),且ICU停留时长平均减少0.78天(95%CI -1.46至-0.10;p=0.025)。
  • 糖皮质激素未显著影响住院时长或严重不良事件。
  • 疗程>7天者短期死亡率获益显著(OR 0.69;95%CI 0.53–0.89;p=0.039),≤7天者无显著差异(OR 0.96;95%CI 0.813–1.14;p=0.67),亚组差异p=0.03。

关键数值与推荐

  • 低剂量糖皮质激素定义:每日≤400 mg氢化可的松等效剂量。
  • 短期死亡率合并OR 0.83(95%CI 0.70–0.98;p=0.029;I²=2.0%;NNT=37.84;GRADE低证据质量)。
  • 社区获得性肺炎亚组短期死亡率OR 0.72(95%CI 0.54–0.98;p=0.034;NNT=28.10)。
  • 社区获得性肺炎患者ICU停留时长平均差异-0.78天(95%CI -1.46至-0.10;p=0.025)。
  • 疗程亚组:>7天OR 0.69(95%CI 0.53–0.89;p=0.039),≤7天OR 0.96(95%CI 0.813–1.14;p=0.67),亚组差异p=0.03。

临床怎么用

  • 对重度社区获得性肺炎成人患者,可考虑使用低剂量糖皮质激素以降低短期死亡率。
  • 若使用糖皮质激素,疗程>7天可能与死亡率获益相关,≤7天未见显著获益。
  • 其他重度肺部感染(如COVID-19、肺孢子菌肺炎、脓毒症、脓毒性休克、ARDS)的获益证据尚不充分,需个体化决策。

局限性

  • GRADE证据质量评为低确定性。
  • 对其他重度肺部感染及更短随访间隔的获益结论尚不确定。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BMJ Open Respir Res. 2026 Jul 9;13(1):e003934. doi: 10.1136/bmjresp-2025-003934. ABSTRACT PURPOSE: Pulmonary infections are one of the leading causes of intensive care unit (ICU) admission and contribute to high mortality rates. This paper aimed to determine the effect of low-dose corticosteroids on outcomes in patients with severe pulmonary infections, including coronavirus disease of 2019, community-acquired pneumonia (CAP), pneumocystis pneumonia, sepsis, septic shock, and acute respiratory distress syndrome. METHODS: We systematically reviewed randomised controlled trials (RCTs) comparing low-dose corticosteroids (≤400 mg hydrocortisone-equivalent daily) to placebo or standard care in adults with severe pulmonary infections. The primary outcome was short-term mortality (≤90 days). Secondary outcomes included 28-day mortality, 30-day mortality, number of patients with at least one serious adverse event, length of hospital stay and ICU stay. RESULTS: We analysed twelve RCTs including 4622 patients and 1203 events. The pooled OR showed a short-term mortality benefit: 0.83 (95% CI 0.70 to 0.98; p=0.029; I²=2.0%; number needed to treat: 37.84; Grading of Recommendations Assessment, Development and Evaluation (GRADE): low certainty). There was a reduction in length of ICU stay in patients with CAP, with a pooled mean difference of -0.78 days (95% CI -1.46 to -0.10, p=0.025). The use of corticosteroids did not significantly affect the length of hospital stay or severe adverse events. A significant difference remained in longer courses of steroids, whereas shorter courses had no significant difference (≤7 days vs >7 days; OR: 0.69; 95% CI 0.53 to 0.89; p=0.039 vs OR: 0.96; 95% CI 0.813 to 1.14; p=0.67; subgroup differences: p=0.03). Subgroup analyses in patients with CAP showed a significant benefit in short-term mortality (OR: 0.72, 95% CI 0.54 to 0.98; p=0.034; number needed to treat: 28.10). CONCLUSION: We found that low-dose corticosteroids significantly reduce 90-day mortality in severe CAP, although evidence for their benefit in other severe pulmonary infections and at shorter follow-up intervals remains inconclusive. PROSPERO REGISTRATION NUMBER: CRD42024627881. PMID: 42425738 | PMC: PMC13358306 | DOI: 10.1136/bmjresp-2025-003934

来源:PubMed 高证据 · 肺炎,查看原文。版权归原作者所有。

引用

  1. PubMed 高证据 · 肺炎:Low-dose corticosteroids in severe pulmonary infection: a meta-analysis of randomised controlled trials

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。