Scholarlink

Background <p>Postoperative pulmonary complications (PPCs) can significantly reduce quality of life and may be life-threatening. While predictors of PPCs have been studied in various surgeries, specific data on pancreaticoduodenectomy is limited.</p> Methods <p>We conducted a retrospective analysis of 338 patients who underwent pancreaticoduodenectomy for biliary, pancreatic, and duodenal tumors at our hospital from 2017 to 2023. We compared demographic, clinical, and laboratory variables between patients with and without PPCs within 30 days post-surgery. Independent risk factors were identified using multivariate logistic regression analysis, and a nomogram was developed to predict PPC risk. Model accuracy was validated with the concordance index (C-index) and calibration curves.</p> Results <p>Of the 338 patients, 149 (44.1%) developed PPCs. Significant variables associated with PPCs included age, smoking, alcohol consumption, type and duration of surgery, timing of nasogastric tube removal, tumor size, tumor location, and preoperative albumin levels. Multivariate logistic regression analysis identified independent risk factors: type and duration of surgery, tumor location, maximum tumor diameter, and preoperative albumin levels. The nomogram incorporating these factors had a C-index of 0.749. The ROC curve showed an AUC of 0.75, indicating satisfactory predictive performance. Calibration curves, decision curve analysis (DCA), and clinical impact curves confirmed the model's validity and reliability.</p> Conclusion <p>Type and duration of surgery, tumor location, maximum tumor diameter, and preoperative albumin levels are independent risk factors for PPCs after pancreaticoduodenectomy. We developed a new nomogram to predict PPCs in these patients.</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

Risk factors and predictive model for postoperative pulmonary complications following pancreaticoduodenectomy: a retrospective cohort study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年10月16日

查看全文
  • Xiaohan Kong,
  • Quan Yuan,
  • Heng Wang,
  • Meng Liu,
  • Zheyu Niu,
  • Jun Lu,
  • XU Zhou,
  • Huaqiang Zhu

摘要

Background

Postoperative pulmonary complications (PPCs) can significantly reduce quality of life and may be life-threatening. While predictors of PPCs have been studied in various surgeries, specific data on pancreaticoduodenectomy is limited.

Methods

We conducted a retrospective analysis of 338 patients who underwent pancreaticoduodenectomy for biliary, pancreatic, and duodenal tumors at our hospital from 2017 to 2023. We compared demographic, clinical, and laboratory variables between patients with and without PPCs within 30 days post-surgery. Independent risk factors were identified using multivariate logistic regression analysis, and a nomogram was developed to predict PPC risk. Model accuracy was validated with the concordance index (C-index) and calibration curves.

Results

Of the 338 patients, 149 (44.1%) developed PPCs. Significant variables associated with PPCs included age, smoking, alcohol consumption, type and duration of surgery, timing of nasogastric tube removal, tumor size, tumor location, and preoperative albumin levels. Multivariate logistic regression analysis identified independent risk factors: type and duration of surgery, tumor location, maximum tumor diameter, and preoperative albumin levels. The nomogram incorporating these factors had a C-index of 0.749. The ROC curve showed an AUC of 0.75, indicating satisfactory predictive performance. Calibration curves, decision curve analysis (DCA), and clinical impact curves confirmed the model's validity and reliability.

Conclusion

Type and duration of surgery, tumor location, maximum tumor diameter, and preoperative albumin levels are independent risk factors for PPCs after pancreaticoduodenectomy. We developed a new nomogram to predict PPCs in these patients.