Construction and evaluation of gastrointestinal bleeding nomogram after laparoscopic pancreaticoduodenectomy for patients with periampullary space occupying lesion
10.3760/cma.j.cn113884-20250108-00007
- VernacularTitle:壶腹周围占位性病变患者腹腔镜胰十二指肠切除术术后消化道出血列线图构建及评估
- Author:
Shuai WANG
1
;
Dongrui LI
1
;
Jianhua LIU
1
;
Chengxu DU
1
;
Qiusheng LI
1
;
Jianzhang QIN
1
;
Haitao LYU
1
Author Information
1. 河北医科大学第二医院肝胆外科，石家庄 050000
- Publication Type:Journal Article
- Keywords:
Pancreaticoduodenectomy;
Laparoscopes;
Ampulla of Vater;
Postoperative hemorrhage;
Nomogram
- From:
Chinese Journal of Hepatobiliary Surgery
2025;31(3):182-187
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a nomogram model for predicting the risk factors of gastrointestinal bleeding following laparoscopic pancreaticoduodenectomy (LPD) based on relevant risk factors and evaluate its predictive value.Methods:A retrospective analysis was conducted on 466 patients with periampullary space occupying lesion who underwent LPD at the Department of Hepatobiliary Surgery, the Second Hospital of Hebei Medical University, from January 2021 to December 2024. Among them, there were 284 males and 182 females, aged (59.9±10.7) years. Patients were randomly divided into a training cohort ( n=326) and a validation cohort ( n=140) using a random number table (7: 3 ratio). Based on whether patients suffered gastrointestinal bleeding, the training cohort was further stratified into a gastrointestinal bleeding group ( n=23) and control group ( n=303). Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for gastrointestinal bleeding. A nomogram was constructed based on multivariate results, and its predictive performance was evaluated by receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Results:Compared to the control group, the gastrointestinal bleeding group exhibited significantly higher age, higher rates of postoperative pancreatic fistula (POPF) and intra-abdominal infection, along with lower body mass index, and lower levels of fibrinogen and albumin (all P<0.05). Multivariate analysis identified age ( OR=1.065, 95% CI: 1.002-1.132), fibrinogen ( OR=0.486, 95% CI: 0.243-0.969), albumin ( OR=0.840, 95% CI: 0.741-0.953), POPF ( OR=4.299, 95% CI: 1.348-13.716), and postoperative intra-abdominal infection ( OR=6.352, 95% CI: 1.476-27.341) as independent predictors of gastrointestinal bleeding (all P<0.05). The nomogram demonstrated robust discrimination, with an AUC of 0.861 (95% CI: 0.784-0.939), sensitivity of 82.6%, and specificity of 82.2% in the training cohort. In the validation cohort, the AUC was 0.824 (95% CI: 0.675-0.973), with sensitivity and specificity of 80.0% and 83.8%, respectively. Calibration curves indicated excellent agreement between predicted and observed outcomes. DCA revealed superior net clinical benefit of the nomogram over " treat-all" or " treat-none" strategies within threshold probabilities of 0-0.9 (training) and 0-0.75 (validation). Conclusion:The nomogram based on age, fibrinogen, albumin, POPF, and intra-abdominal infection provides accurate prediction of gastrointestinal bleeding after LPD and demonstrates high clinical utility for risk stratification and decision-making in periampullary space occupying lesion patients.
