Scholarlink

Introduction <p>Several Fistula Risk Scores (FRS) have been developed to predict postoperative pancreatic fistula (POPF) after pancreatoduodenectomy (PD). This study aimed to externally assess and compare the performance of the original (o-FRS), alternative (a-FRS), updated alternative (ua-FRS), and auditing (Aud-FRS) scores in a Southeast Asian cohort undergoing open and minimally invasive PD (MIS-PD).</p> Methods <p>A post hoc analysis of a prospectively maintained database was performed including 724 consecutive patients who underwent PD (621 open, 103 MIS-PD) between 2005 and 2022 at a single tertiary institution. POPF was defined according to the 2016 ISGPS criteria (Grade B/C). Receiver operating characteristic analyses were used to assess predictive performance of each FRS. Associations between clinicopathologic variables and POPF were evaluated using logistic regression.</p> Results <p>The overall area under the curve (AUC) for o-FRS, a-FRS, ua-FRS, and Aud-FRS were 0.675, 0.668, 0.672, and 0.662, respectively, indicating modest discrimination. Similar performance was observed across open and MIS-PD subgroups. Higher risk categories in ua-FRS and Aud-FRS were associated with increasing POPF incidence (<i>p</i> &lt; 0.01). On multivariable analysis, smaller pancreatic duct diameter (OR 0.742, 95% CI 0.563–0.978) and pancreatojejunostomy compared with pancreatogastrostomy (OR 2.170, 95% CI 1.490–3.160) were independent predictors of POPF.</p> Conclusion <p>All four FRS demonstrated consistent directional risk stratification but only modest discriminative ability in this Southeast Asian cohort. These findings suggest limited clinical utility without recalibration and highlight the need for population-specific refinement and objective pre-operative risk prediction models.</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

External validation of the original, alternative, updated alternative, and auditing Fistula Risk Score for the prediction of post-operative pancreatic fistula after minimally invasive and open pancreatoduodenectomy

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Sabrina H. X. Cheok,
  • Teik Wen Lim,
  • Darren W. Chua,
  • Delphina B. X. Yeo,
  • Titus Q. W. Tan,
  • Jin-Yao Teo,
  • Ye-Xin Koh,
  • Peng-Chung Cheow,
  • London Lucien P. J. Ooi,
  • Alexander Y. F. Chung,
  • Brian K. P. Goh

摘要

Introduction

Several Fistula Risk Scores (FRS) have been developed to predict postoperative pancreatic fistula (POPF) after pancreatoduodenectomy (PD). This study aimed to externally assess and compare the performance of the original (o-FRS), alternative (a-FRS), updated alternative (ua-FRS), and auditing (Aud-FRS) scores in a Southeast Asian cohort undergoing open and minimally invasive PD (MIS-PD).

Methods

A post hoc analysis of a prospectively maintained database was performed including 724 consecutive patients who underwent PD (621 open, 103 MIS-PD) between 2005 and 2022 at a single tertiary institution. POPF was defined according to the 2016 ISGPS criteria (Grade B/C). Receiver operating characteristic analyses were used to assess predictive performance of each FRS. Associations between clinicopathologic variables and POPF were evaluated using logistic regression.

Results

The overall area under the curve (AUC) for o-FRS, a-FRS, ua-FRS, and Aud-FRS were 0.675, 0.668, 0.672, and 0.662, respectively, indicating modest discrimination. Similar performance was observed across open and MIS-PD subgroups. Higher risk categories in ua-FRS and Aud-FRS were associated with increasing POPF incidence (p < 0.01). On multivariable analysis, smaller pancreatic duct diameter (OR 0.742, 95% CI 0.563–0.978) and pancreatojejunostomy compared with pancreatogastrostomy (OR 2.170, 95% CI 1.490–3.160) were independent predictors of POPF.

Conclusion

All four FRS demonstrated consistent directional risk stratification but only modest discriminative ability in this Southeast Asian cohort. These findings suggest limited clinical utility without recalibration and highlight the need for population-specific refinement and objective pre-operative risk prediction models.