单孔与多孔机器人辅助胸外科手术用于解剖性肺切除：围手术期恢复与手术质量的系统评价和荟萃分析。
完整摘要 · 中文译文
单孔机器人辅助胸外科手术（uRATS）旨在将机器人平台的操作灵巧性与减少的入路创伤相结合，但其相较于多孔机器人辅助胸外科手术（mRATS）的比较价值仍不确定。我们系统评估了解剖性肺切除术后的围手术期恢复和手术质量。检索了PubMed/MEDLINE、Web of Science 和 Cochrane Library，检索时间截止至2026年。纳入了单孔/单切口与多孔机器人辅助解剖性肺切除术的比较研究。采用随机效应荟萃分析汇总风险比（RRs）和均数差（MDs）。使用ROBINS-I V2评估偏倚风险，使用GRADE评估证据确定性。共纳入9项非随机研究，涉及813例患者，其中7项研究被评为存在严重偏倚风险。单孔组术后并发症发生率为16.0%，多孔组为30.2%，但合并差异无统计学意义（RR 0.49，95% CI 0.22-1.06；p = 0.069；I²=66%）。在主要并发症、中转率、死亡率、手术时间、失血量、胸管留置时间、住院时间、淋巴结清扫数目或R0切除方面均未发现显著差异。单孔手术与更短的操控台时间（MD - 23.5 分钟，95% CI - 35.6 至 - 11.4）和更低的术后第1天疼痛评分（MD - 0.6 分，95% CI - 0.9 至 - 0.3）相关。0.5分的疼痛差异低于临床重要性常用的阈值，且在排除数字化数值后，该差异失去统计学显著性（p = 0.092）。现有的观察性证据未显示出明确的差异。 单孔/单端口与多端口机器人辅助胸外科手术在围手术期安全性或手术质量方面没有差异。探索性亚组分析提示，住院时间和手术时间的差异可能与对照端口数量有关，这可能反映了手术时代和经验带来的混杂因素。观察到控制台时间和早期术后疼痛可能减少，但这些发现基于极低确定性的证据，并且对单项研究敏感。
完整摘要 · 英文原文
Uniportal robotic-assisted thoracic surgery (uRATS) seeks to combine the dexterity of robotic platforms with reduced access trauma, yet its comparative value over multiportal RATS (mRATS) remains uncertain. We systematically evaluated perioperative recovery and surgical quality after anatomical pulmonary resection. PubMed/MEDLINE, Web of Science, and the Cochrane Library were searched through 2026. Comparative studies of uniportal/single-port versus multiportal robotic anatomical lung resection were included. Random-effects meta-analyses pooled risk ratios (RRs) and mean differences (MDs). Risk of bias was assessed using ROBINS-I V2 and certainty of evidence using GRADE. Nine non-randomised studies involving 813 patients were included, with seven studies rated at critical risk of bias. Postoperative complications occurred in 16.0% of uniportal and 30.2% of multiportal patients, with no statistically significant pooled difference (RR 0.49, 95% CI 0.22-1.06; p = 0.069; I²=66%). No significant differences were identified in major complications, conversion, mortality, operative time, blood loss, chest-tube duration, length of stay, or lymph-node yield, or R0 resection. Uniportal surgery was associated with shorter console time (MD - 23.5 min, 95% CI - 35.6 to - 11.4) and lower postoperative day-1 pain (MD - 0.6 points, 95% CI - 0.9 to - 0.3). The 0.5- point pain difference was below commonly used thresholds for clinical importance and lost statistical significance when digitised values were excluded (p = 0.092). Available observational evidence does not demonstrate a clear difference in perioperative safety or surgical quality between uniportal/single-port and multiportal robotic-assisted thoracic surgery. Exploratory subgroup analyses suggested differences in length of stay and operative time by comparator port count, potentially reflecting confounding by operative era and experience. A possible reduction in console time and early postoperative pain was observed, but these findings were based on very-low-certainty evidence and were sensitive to individual studies.