Scholarlink

Background <p>Intraoperative bleeding severity and hemostasis quality have not been objectively evaluated.</p> Methods <p>Robot-assisted pancreatoduodenectomy (RPD) cases between April 2021 and June 2023 were selected. Intraoperative bleeding scenes were extracted from the resection videos. Additionally, bleeding scenes leading to open conversion due to difficulty in achieving hemostasis were extracted from cases between October 2020 and March 2025. In each bleeding scene, the pixel index (PI), defined as the number of red pixels 1&#xa0;s after bleeding onset, and the hemostatic method were used to grade the bleeding severity. Hemostatic time was also assessed to evaluate hemostasis quality.</p> Results <p>A total of 885 bleeding scenes and four scenes requiring open conversion were included. The areas under the curve (AUC) of the PI predicting non-cauterization and open conversion procedures were 0.846 (95% confidence interval [CI]: 0.802–0.890; cutoff: 12041) and 0.990 (95% CI: 0.977–1.000; cutoff: 62084), respectively. Bleeding was graded as grade 1 (PI &lt; 12000, cauterization), grade 2 (12000 ≤ PI &lt; 60000, other procedures), or grade 3 (PI ≥ 60000, open laparotomy). The receiver operating characteristic curve for the number of hemostatic procedures ≥ 15&#xa0;s and total blood loss ≥ 100&#xa0;ml had the highest AUC at 0.879. Hemostasis requiring &lt; 15&#xa0;s was considered effective, whereas hemostasis requiring ≥ 15&#xa0;s was considered difficult.</p> Conclusions <p>We proposed a grading system for intraoperative bleeding and hemostasis during RPD. This study provides a quantitative framework for future research, enabling objective recommendations for surgical hemostatic strategies. This grading classification system was developed based on a dataset from a single-institution. Therefore, it has not yet been validated using data from external institutions or sources. Accordingly, this grading classification should be regarded as a proposed system, and this study should be considered exploratory.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Proposal for grading of intraoperative bleeding and the assessment of hemostasis quality during robot-assisted pancreatoduodenectomy

  • 期刊论文

  • 发表日期：

    2025年11月7日

查看全文
  • Yui Sawa,
  • Yosuke Inoue,
  • Sho Kiritani,
  • Kosuke Kobayashi,
  • Atsushi Oba,
  • Yoshihiro Ono,
  • Hiromichi Ito,
  • Yu Takahashi

摘要

Background

Intraoperative bleeding severity and hemostasis quality have not been objectively evaluated.

Methods

Robot-assisted pancreatoduodenectomy (RPD) cases between April 2021 and June 2023 were selected. Intraoperative bleeding scenes were extracted from the resection videos. Additionally, bleeding scenes leading to open conversion due to difficulty in achieving hemostasis were extracted from cases between October 2020 and March 2025. In each bleeding scene, the pixel index (PI), defined as the number of red pixels 1 s after bleeding onset, and the hemostatic method were used to grade the bleeding severity. Hemostatic time was also assessed to evaluate hemostasis quality.

Results

A total of 885 bleeding scenes and four scenes requiring open conversion were included. The areas under the curve (AUC) of the PI predicting non-cauterization and open conversion procedures were 0.846 (95% confidence interval [CI]: 0.802–0.890; cutoff: 12041) and 0.990 (95% CI: 0.977–1.000; cutoff: 62084), respectively. Bleeding was graded as grade 1 (PI < 12000, cauterization), grade 2 (12000 ≤ PI < 60000, other procedures), or grade 3 (PI ≥ 60000, open laparotomy). The receiver operating characteristic curve for the number of hemostatic procedures ≥ 15 s and total blood loss ≥ 100 ml had the highest AUC at 0.879. Hemostasis requiring < 15 s was considered effective, whereas hemostasis requiring ≥ 15 s was considered difficult.

Conclusions

We proposed a grading system for intraoperative bleeding and hemostasis during RPD. This study provides a quantitative framework for future research, enabling objective recommendations for surgical hemostatic strategies. This grading classification system was developed based on a dataset from a single-institution. Therefore, it has not yet been validated using data from external institutions or sources. Accordingly, this grading classification should be regarded as a proposed system, and this study should be considered exploratory.

Graphical Abstract