Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques

BackgroundComplications and mortality rates following gastrectomy for gastric cancer have improved over recent years; however, complications such as anastomotic leakage (AL) continue to significantly impact both immediate and long-term prognoses. This study aimed to develop a machine learning model...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuan Liu, Songyun Zhao, Xingchen Shang, Wei Shen, Wenyi Du, Ning Zhou
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-11-01
Series:Frontiers in Oncology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fonc.2024.1471137/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850225467776303104
author Yuan Liu
Songyun Zhao
Xingchen Shang
Wei Shen
Wenyi Du
Ning Zhou
author_facet Yuan Liu
Songyun Zhao
Xingchen Shang
Wei Shen
Wenyi Du
Ning Zhou
author_sort Yuan Liu
collection DOAJ
description BackgroundComplications and mortality rates following gastrectomy for gastric cancer have improved over recent years; however, complications such as anastomotic leakage (AL) continue to significantly impact both immediate and long-term prognoses. This study aimed to develop a machine learning model to identify preoperative and intraoperative high-risk factors and predict mortality in patients with AL after radical gastrectomy.MethodsFor this investigation, 906 patients diagnosed with gastric cancer were enrolled and evaluated, with a comprehensive set of 36 feature variables collected. We employed three distinct machine learning algorithms—extreme gradient boosting (XGBoost), random forest (RF), and k-nearest neighbor (KNN)—to develop our models. To ensure model robustness, we applied k-fold cross-validation for internal validation of the four models and subsequently validated them using independent datasets.ResultsIn contrast to the other machine learning models employed in this study, the XGBoost algorithm exhibited superior predictive performance in identifying mortality risk factors for patients with AL across one, three, and five-year intervals. The analysis identified several common risk factors affecting mortality rates at these intervals, including advanced age, hypoproteinemia, a history of anemia and hypertension, prolonged operative time, increased intraoperative bleeding, low intraoperative percutaneous arterial oxygen saturation (SPO2) levels, T3 and T4 tumors, tumor lymph node invasion, and tumor peripheral nerve invasion (PNI).ConclusionAmong the three machine learning models examined in this study, the XGBoost algorithm exhibited superior predictive capabilities concerning the prognosis of patients with AL following gastrectomy. Additionally, the use of machine learning models offers valuable assistance to clinicians in identifying crucial prognostic factors, thereby enhancing personalized patient monitoring and management.
format Article
id doaj-art-ee2efbc6cd464348a48aef26020cc94c
institution OA Journals
issn 2234-943X
language English
publishDate 2024-11-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Oncology
spelling doaj-art-ee2efbc6cd464348a48aef26020cc94c2025-08-20T02:05:20ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2024-11-011410.3389/fonc.2024.14711371471137Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniquesYuan Liu0Songyun Zhao1Xingchen Shang2Wei Shen3Wenyi Du4Ning Zhou5Department of General Surgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaDepartment of Neurosurgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaDepartment of General Surgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaDepartment of General Surgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaDepartment of General Surgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaDepartment of General Surgery, Wuxi People’s Hospital Affiliated to Nanjing Medical University, Wuxi, ChinaBackgroundComplications and mortality rates following gastrectomy for gastric cancer have improved over recent years; however, complications such as anastomotic leakage (AL) continue to significantly impact both immediate and long-term prognoses. This study aimed to develop a machine learning model to identify preoperative and intraoperative high-risk factors and predict mortality in patients with AL after radical gastrectomy.MethodsFor this investigation, 906 patients diagnosed with gastric cancer were enrolled and evaluated, with a comprehensive set of 36 feature variables collected. We employed three distinct machine learning algorithms—extreme gradient boosting (XGBoost), random forest (RF), and k-nearest neighbor (KNN)—to develop our models. To ensure model robustness, we applied k-fold cross-validation for internal validation of the four models and subsequently validated them using independent datasets.ResultsIn contrast to the other machine learning models employed in this study, the XGBoost algorithm exhibited superior predictive performance in identifying mortality risk factors for patients with AL across one, three, and five-year intervals. The analysis identified several common risk factors affecting mortality rates at these intervals, including advanced age, hypoproteinemia, a history of anemia and hypertension, prolonged operative time, increased intraoperative bleeding, low intraoperative percutaneous arterial oxygen saturation (SPO2) levels, T3 and T4 tumors, tumor lymph node invasion, and tumor peripheral nerve invasion (PNI).ConclusionAmong the three machine learning models examined in this study, the XGBoost algorithm exhibited superior predictive capabilities concerning the prognosis of patients with AL following gastrectomy. Additionally, the use of machine learning models offers valuable assistance to clinicians in identifying crucial prognostic factors, thereby enhancing personalized patient monitoring and management.https://www.frontiersin.org/articles/10.3389/fonc.2024.1471137/fullgastric tumorgastrectomyanastomotic leakageprognosisrisk factormachine learning
spellingShingle Yuan Liu
Songyun Zhao
Xingchen Shang
Wei Shen
Wenyi Du
Ning Zhou
Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
Frontiers in Oncology
gastric tumor
gastrectomy
anastomotic leakage
prognosis
risk factor
machine learning
title Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
title_full Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
title_fullStr Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
title_full_unstemmed Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
title_short Identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post-radical gastrectomy: an 8-year multicenter study utilizing machine learning techniques
title_sort identification of intraoperative hypoxemia and hypoproteinemia as prognostic indicators in anastomotic leakage post radical gastrectomy an 8 year multicenter study utilizing machine learning techniques
topic gastric tumor
gastrectomy
anastomotic leakage
prognosis
risk factor
machine learning
url https://www.frontiersin.org/articles/10.3389/fonc.2024.1471137/full
work_keys_str_mv AT yuanliu identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques
AT songyunzhao identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques
AT xingchenshang identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques
AT weishen identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques
AT wenyidu identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques
AT ningzhou identificationofintraoperativehypoxemiaandhypoproteinemiaasprognosticindicatorsinanastomoticleakagepostradicalgastrectomyan8yearmulticenterstudyutilizingmachinelearningtechniques