Scholarlink

Purpose <p>Incisional hernia (IH) is the most common long-term complication following open pancreatic surgery. This study aimed to evaluate the transportability of the Penn Hernia Risk Calculator (PHRC) Hepato-Pancreato-Biliary (HPB) model in a geo-temporally diverse population.</p> Methods <p>Adult patients undergoing non-palliative pancreatic surgery via a transverse laparotomy from April 2012 to December 2023 at a single tertiary academic center were included. Patients with prior IH diagnosis and/or repair were excluded. Demographics, medical history, lab, and operative data were collected. Composite IH risk scores were calculated using the PHRC. Model performance was assessed using the Area Under the Receiver Operating Characteristic Curve (AUROC) and positive likelihood ratios (LR+) across risk thresholds.</p> Results <p>A total of 403 patients were included with 4.2% developing IH. The study population differed significantly from the PHRC cohort in 19 of 20 predictive variables. The overall AUROC was 0.548, and LR + values were consistently below 1.8 at all risk thresholds. Calibration plots indicated systematic overprediction of IH risk. Subgroup analysis showed improved performance (AUROC = 0.875, LR + = 0.467) among patients with BMI &gt; 30&#xa0;kg/m². Malignancy was a protective factor, while undergoing left hemipancreatectomy or central pancreatectomy and BMI 20–30&#xa0;kg/m² were associated with IH development.</p> Conclusion <p>In this small single-center cohort, the PHRC HPB model demonstrated poor accuracy in predicting early IH following pancreatic surgery through transverse laparotomy. Performance improved in obese patients, suggesting case-mix and outcome incidence differences may largely explain the lack of model transportability.</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 Penn hernia risk calculator: an evaluation of model transportability for incisional hernia prediction in open pancreatic surgery

  • 期刊论文

  • 发表日期：

    2026年2月17日

查看全文
  • Senne Hellinck,
  • Margo Boedt,
  • Margaret M. Hornick,
  • Filip Gryspeerdt,
  • An-Verena Lerut,
  • Mathias Allaeys,
  • Niki Rashidian,
  • Hasan Eker,
  • Luis Abreu de Carvalho,
  • Ineke van Gremberghe,
  • Robyn B. Broach,
  • John P. Fischer,
  • Frederik Berrevoet

摘要

Purpose

Incisional hernia (IH) is the most common long-term complication following open pancreatic surgery. This study aimed to evaluate the transportability of the Penn Hernia Risk Calculator (PHRC) Hepato-Pancreato-Biliary (HPB) model in a geo-temporally diverse population.

Methods

Adult patients undergoing non-palliative pancreatic surgery via a transverse laparotomy from April 2012 to December 2023 at a single tertiary academic center were included. Patients with prior IH diagnosis and/or repair were excluded. Demographics, medical history, lab, and operative data were collected. Composite IH risk scores were calculated using the PHRC. Model performance was assessed using the Area Under the Receiver Operating Characteristic Curve (AUROC) and positive likelihood ratios (LR+) across risk thresholds.

Results

A total of 403 patients were included with 4.2% developing IH. The study population differed significantly from the PHRC cohort in 19 of 20 predictive variables. The overall AUROC was 0.548, and LR + values were consistently below 1.8 at all risk thresholds. Calibration plots indicated systematic overprediction of IH risk. Subgroup analysis showed improved performance (AUROC = 0.875, LR + = 0.467) among patients with BMI > 30 kg/m². Malignancy was a protective factor, while undergoing left hemipancreatectomy or central pancreatectomy and BMI 20–30 kg/m² were associated with IH development.

Conclusion

In this small single-center cohort, the PHRC HPB model demonstrated poor accuracy in predicting early IH following pancreatic surgery through transverse laparotomy. Performance improved in obese patients, suggesting case-mix and outcome incidence differences may largely explain the lack of model transportability.