文献速递心外重症与机械循环支持肺外科

非择期胸科手术围术期VV-ECMO与内科适应证的倾向性匹配双中心比较

Perioperative Veno-Venous Extracorporeal Membrane Oxygenation in Non-Elective Thoracic Surgery: A Propensity-Matched Bicentric Comparison With Medical Indication.
PubMed 文献 · Interdiscip Cardiovasc Thorac Surg · 2025-01-01

该回顾性双中心队列纳入2011年1月至2025年1月两家大学ICU因急性呼吸衰竭接受VV-ECMO的成人患者,按围术期胸科手术或内科适应证分组,经1:4倾向性匹配比较90天全因死亡率及出血和血栓事件。结果显示围术期组90天死亡率与内科组无差异,匹配前后均无统计学意义,大出血和血栓事件亦无显著差异。

要点

  • 非择期胸科手术围术期VV-ECMO与内科VV-ECMO的90天死亡率无显著差异。
  • 匹配后围术期组与内科组大出血和血栓事件发生率均无显著差异。
  • 围术期组呼吸机相关性肺炎和血胸发生率高于内科组。

核心结论

  • 非择期胸科手术围术期VV-ECMO与内科VV-ECMO相比,90天死亡率、大出血和血栓并发症均无增加,匹配前后结果一致。

研究设计与样本

  • 回顾性双中心队列研究,纳入2011年1月至2025年1月两家大学ICU因急性呼吸衰竭接受VV-ECMO的成人患者372例,其中44例(12%)为围术期胸科手术适应证。
  • 采用多变量logistic回归构建倾向性评分,行1:4最近邻匹配(不放回),主要终点为90天死亡率,次要终点为大出血和血栓事件。

主要发现

  • 围术期组呼吸机相关性肺炎发生率78% vs 内科组57%(P = .01),血胸发生率36% vs 2%(P < .001)。
  • 90天死亡率匹配前54% vs 51%(P = .71),匹配后log-rank P = .95;多变量Cox分析显示手术ECMO与死亡率无关(HR = 1.003,95% CI [0.64-1.57])。
  • 大出血发生率48% vs 47%(P = 1.00),血栓事件发生率41% vs 32%(P = .31),均无显著差异。

临床怎么用

  • 在经仔细筛选的患者中,非择期胸科手术围术期使用VV-ECMO不增加90天死亡率、大出血或血栓风险,可支持其临床应用。

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

OBJECTIVES: Perioperative veno-venous extracorporeal membrane oxygenation (VV-ECMO) can be used during non-elective thoracic surgery to manage hypoxaemia and facilitate complex surgical procedures. However, outcome data in this setting remain limited, particularly regarding bleeding and thrombotic risks. We aimed to evaluate whether perioperative VV-ECMO during non-elective thoracic surgery is associated with increased 90-day all-cause mortality compared with VV-ECMO used for medical indications, using a propensity score-matched analysis. Secondary outcomes included major bleeding and thrombotic events. METHODS: This retrospective bi-centre cohort study included adults treated with VV-ECMO for acute respiratory failure between January 2011 and January 2025 in 2 university intensive care units. Patients were categorized according to ECMO indication: perioperative thoracic surgery or medical. A propensity score was built using multivariable logistic regression, followed by 1:4 nearest-neighbor matching without replacement. The primary endpoint was 90-day mortality; secondary endpoints included major bleeding and thrombotic events. RESULTS: Among 372 patients, 44 (12%) received perioperative VV-ECMO. Ventilator-associated pneumonia (78% vs 57%, P = .01) and haemothorax (36% vs 2%, P < .001) were more frequent in the perioperative group. Ninety-day mortality did not differ between groups, both before (54% vs 51%, P = .71) and after matching (log-rank P = .95). Multivariable Cox analysis confirmed no association between surgical ECMO and mortality (HR = 1.003, 95% CI [0.64-1.57]). No significant differences were found in major bleeding (48% vs 47%, P = 1.00) or thrombotic events (41% vs 32%, P = .31) between groups. CONCLUSIONS: Perioperative VV-ECMO during non-elective thoracic surgery was not associated with increased 90-day mortality, bleeding, or thrombotic complications compared to medical VV-ECMO, both before and after matching. These findings support its use in carefully selected patients. CLINICAL TRIAL REGISTRATION NUMBER: NCT07016685.

来源:Interdiscip Cardiovasc Thorac Surg,查看原文。版权归原作者所有。

引用

  1. Beyls C, Soulier-Zaninka Q, Georges O, et al. Perioperative Veno-Venous Extracorporeal Membrane Oxygenation in Non-Elective Thoracic Surgery: A Propensity-Matched Bicentric Comparison With Medical Indication. Interdiscip Cardiovasc Thorac Surg. 2025. PMID: 41134639. doi:10.1093/icvts/ivaf255

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。