文献速递肝胆胰外科围术期管理与 ERAS

早期采用者350例机器人肝切除的临床结局与学习曲线评估

Evaluation of 350 robotic liver resections performed by an early adopter in laparoscopic liver surgery: clinical outcomes and critical appraisal of the learning curve.
PubMed 文献 · Surg Endosc · 2025-10-01

本研究评估一位具有丰富开腹和腹腔镜经验的高容量肝胆胰外科医生完成的350例机器人肝切除的围术期结局、学习曲线、器械使用及费用。结果显示中转开腹率0.6%,主要并发症6%,90天死亡率1.1%,学习曲线在轻微、解剖性大和技術性大切除中分别于28、15和54例达到平台。腹腔镜难度评分(IWATE、Southampton、IMM)能强预测手术时间和出血量,而病例量无关联。

要点

  • 机器人肝切除中转开腹率0.6%,主要并发症6%,90天死亡率1.1%。
  • 学习曲线平台:轻微切除28例,解剖性大切除15例,技术性大切除54例。
  • IWATE、Southampton和IMM评分能强预测手术时间和出血量(p<0.001)。

核心结论

  • 在有经验的术者手中,机器人肝切除安全、可行且成本效益好,学习曲线短,腹腔镜难度评分预测价值强。

研究设计与样本

  • 单中心回顾性研究,纳入2020年2月至2025年4月由单一术者完成的350例机器人肝切除术,分析围术期结局、学习曲线(累积和分析手术时间)、器械使用和费用。

主要发现

  • 350例中,轻微切除147例,解剖性大切除43例,技术性大切除160例。
  • 中位手术时间145分钟(IQR 122-188),中位估计出血量30 mL(IQR 20-90)。
  • 中转开腹率0.6%(n=2),主要并发症(Clavien-Dindo ≥ III)发生率6%,中位住院时间3天,90天死亡率1.1%。
  • 学习曲线平台:轻微切除28例,解剖性大切除15例,技术性大切除54例。
  • 病例量与手术时间(p=0.79)或出血量(p=0.93)无关;IWATE、Southampton和IMM评分强预测两者(p<0.001)。

临床怎么用

  • 机器人肝切除可用于轻微、解剖性大和技术性大切除,学习曲线较短。
  • 腹腔镜难度评分(IWATE、Southampton、IMM)可用于预测机器人肝切除的手术时间和出血量。
  • 多数机器人肝切除需准备3-4件器械,平均住院费用约5408.50欧元。

局限性

  • 单术者经验,结果可能受术者学习曲线和技能影响,外推性受限。
  • 回顾性研究设计,存在选择偏倚和信息偏倚可能。

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

INTRODUCTION: Robotic-assisted liver surgery is increasingly adopted, offering many advantages over conventional laparoscopy. This study evaluates perioperative outcomes, learning curves, instrument use, and cost price of the first 350 robotic liver resections (RLRs) performed by a high-volume HPB surgeon with extensive open and laparoscopic experience. Additionally, we assess whether established laparoscopic difficulty scoring systems can reliably predict surgical complexity and outcomes in robotic procedures. METHODS: All RLRs performed by a single surgeon from February 2020 to April 2025 were identified from a prospectively maintained database. Perioperative outcomes were analyzed, and learning curves for minor, anatomically major, and technically major resections were defined using cumulative sum analysis of operative time. Multivariate regression was used to assess the impact of case volume and preoperative difficulty scores on operative time and blood loss. And a descriptive analysis of instrument use, robotic techniques, and cost price is provided. RESULTS: Of the 350 RLRs, 147 were minor, 43 were anatomically major, and 160 were technically major resections. Median operative time was 145 min (IQR: 122-188), with median estimated blood loss of 30 mL (IQR: 20-90). Our conversion rate to open surgery was 0.6% (n = 2). Major complications (Clavien-Dindo ≥ III) occurred in 6%, with a median hospital stay of 3 days and a 90-day mortality rate of 1.1%. Learning plateaus were reached at 28 (minor) cases, 15 (anatomically major) cases, and 54 (technically major) cases. Case volume was not associated with operative time (p = 0.79) or blood loss (p = 0.93). In contrast, IWATE, Southampton, and IMM scores strongly predicted both (p < 0.001). A clear instrument preference was observed; most procedures required the use of 3-4 instruments, and the mean hospitalization cost was €5,408.50. CONCLUSION: Robotic liver resection is safe, feasible, and cost-effective in experienced hands, with a short learning curve and strong predictive value from laparoscopic difficulty scores.

来源:Surg Endosc,查看原文。版权归原作者所有。

引用

  1. Vandaele T, Chaoui I, De Meyere C, et al. Evaluation of 350 robotic liver resections performed by an early adopter in laparoscopic liver surgery: clinical outcomes and critical appraisal of the learning curve. Surg Endosc. 2025. PMID: 41034574. doi:10.1007/s00464-025-12274-3

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。