微创与胸骨切开术在再次二尖瓣手术中的比较：基于meta回归和事件时间分析的短期及长期结局。
完整摘要 · 中文译文
背景：再次二尖瓣手术风险较高，而微创二尖瓣手术（MI-MVS）与胸骨切开二尖瓣手术相比的相对获益，尤其是远期生存方面的获益，仍不明确。 方法：检索电子数据库，寻找在成人再次二尖瓣手术中比较MI-MVS与胸骨切开二尖瓣手术的研究。对Kaplan-Meier曲线进行数字化处理，重建个体患者数据用于时间-事件分析，并使用Cox脆弱模型生成研究层面的风险比（HR）。采用随机效应模型汇总围手术期结局。亚组分析评估了倾向性匹配队列、内镜/视频辅助小切口开胸以及直视小切口开胸。Meta回归探讨了研究层面协变量对死亡率的影响。 结果：共分析15项研究，包括3457例患者。与胸骨切开二尖瓣手术相比，MI-MVS与较低的围手术期死亡率相关（比值比[OR]，0.33[95% CI，0.23-0.48]；P<0.001），并降低再次手术发生率（OR，0.65[95% CI，0.44-0.98]；P=0.04）和急性肾损伤发生率（OR，0.58[95% CI，0.39-0.87]；P=0.008）；卒中和心律失常发生率相似。在亚组分析中，死亡率效应的方向和大小均一致有利于MI-MVS。时间-事件分析显示MI-MVS可改善生存率（HR，0.74[95% CI，0.62-0.89]；P=0.001），但这一效应在倾向性匹配分析中有所减弱。Meta回归显示，性别分布、平均年龄和基线左心室射血分数并未显著调节死亡率效应。 结论：在再次 MVS、MI-MVS 与改善的围手术期结局相关，且不损害长期生存；然而，倾向匹配后生存差异的减弱提示，这种长期优势可能反映了基线风险而非手术操作本身的效果。
完整摘要 · 英文原文
BACKGROUND: Redo mitral valve surgery carries substantial risk, and the relative benefits of minimally invasive (MI-MVS) versus sternotomy MVS, particularly regarding long-term survival, remain uncertain. METHODS: Electronic databases were searched for studies comparing MI-MVS with sternotomy MVS in adult redo MVS. Kaplan-Meier curves were digitized to reconstruct individual patient data for time-to-event analysis, with study-level hazard ratios (HR) generated using a Cox frailty model. Perioperative outcomes were pooled using random-effects models. Subgroup analyses evaluated propensity-matched cohorts, endoscopic/video-assisted minithoracotomy, and direct-vision minithoracotomy. Meta-regression explored the impact of study-level covariates on mortality. RESULTS: Fifteen studies including 3457 patients were analyzed. MI-MVS was associated with lower perioperative mortality compared with sternotomy MVS (odds ratio [OR], 0.33 [95% CI, 0.23-0.48]; P<0.001), reduced incidence of reoperation (OR, 0.65 [95% CI, 0.44-0.98]; P=0.04), and acute kidney injury (OR, 0.58 [95% CI, 0.39-0.87]; P=0.008); rates of stroke and arrhythmia were similar. The direction and magnitude of the mortality effect consistently favored MI-MVS in the subgroup analyses. Time-to-event analysis demonstrated improved survival with MI-MVS (HR, 0.74 [95% CI, 0.62-0.89]; P=0.001), but this was attenuated in the propensity-matched analysis. Meta-regression showed that sex distribution, mean age, and baseline left ventricular ejection fraction did not significantly moderate mortality effects. CONCLUSIONS: In redo MVS, MI-MVS is associated with improved perioperative outcomes without compromising long-term survival; however, attenuation of the survival difference after propensity matching suggests that this long-term advantage may reflect baseline risk rather than procedural effect.