Machine Learning-Based Analysis in the Management of Iatrogenic Bile Duct Injury During Cholecystectomy: a Nationwide Multicenter Study.
Artificial neural network
Cholecystectomy
Iatrogenic bile duct injury
Machine learning
Journal
Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract
ISSN: 1873-4626
Titre abrégé: J Gastrointest Surg
Pays: United States
ID NLM: 9706084
Informations de publication
Date de publication:
08 2022
08 2022
Historique:
received:
24
03
2022
accepted:
17
06
2022
pubmed:
6
7
2022
medline:
9
9
2022
entrez:
5
7
2022
Statut:
ppublish
Résumé
Iatrogenic bile duct injury (IBDI) is a challenging surgical complication. IBDI management can be guided by artificial intelligence models. Our study identified the factors associated with successful initial repair of IBDI and predicted the success of definitive repair based on patient risk levels. This is a retrospective multi-institution cohort of patients with IBDI after cholecystectomy conducted between 1990 and 2020. We implemented a decision tree analysis to determine the factors that contribute to successful initial repair and developed a risk-scoring model based on the Comprehensive Complication Index. We analyzed 748 patients across 22 hospitals. Our decision tree model was 82.8% accurate in predicting the success of the initial repair. Non-type E (p < 0.01), treatment in specialized centers (p < 0.01), and surgical repair (p < 0.001) were associated with better prognosis. The risk-scoring model was 82.3% (79.0-85.3%, 95% confidence interval [CI]) and 71.7% (63.8-78.7%, 95% CI) accurate in predicting success in the development and validation cohorts, respectively. Surgical repair, successful initial repair, and repair between 2 and 6 weeks were associated with better outcomes. Machine learning algorithms for IBDI are a novel tool may help to improve the decision-making process and guide management of these patients.
Sections du résumé
BACKGROUND
Iatrogenic bile duct injury (IBDI) is a challenging surgical complication. IBDI management can be guided by artificial intelligence models. Our study identified the factors associated with successful initial repair of IBDI and predicted the success of definitive repair based on patient risk levels.
METHODS
This is a retrospective multi-institution cohort of patients with IBDI after cholecystectomy conducted between 1990 and 2020. We implemented a decision tree analysis to determine the factors that contribute to successful initial repair and developed a risk-scoring model based on the Comprehensive Complication Index.
RESULTS
We analyzed 748 patients across 22 hospitals. Our decision tree model was 82.8% accurate in predicting the success of the initial repair. Non-type E (p < 0.01), treatment in specialized centers (p < 0.01), and surgical repair (p < 0.001) were associated with better prognosis. The risk-scoring model was 82.3% (79.0-85.3%, 95% confidence interval [CI]) and 71.7% (63.8-78.7%, 95% CI) accurate in predicting success in the development and validation cohorts, respectively. Surgical repair, successful initial repair, and repair between 2 and 6 weeks were associated with better outcomes.
DISCUSSION
Machine learning algorithms for IBDI are a novel tool may help to improve the decision-making process and guide management of these patients.
Identifiants
pubmed: 35790677
doi: 10.1007/s11605-022-05398-7
pii: 10.1007/s11605-022-05398-7
pmc: PMC9439981
doi:
Types de publication
Journal Article
Multicenter Study
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1713-1723Informations de copyright
© 2022. The Author(s).
Références
Ann Surg. 2017 Jun;265(6):1045-1050
pubmed: 28486288
Surg Endosc. 2021 Apr;35(4):1651-1658
pubmed: 32306111
HPB (Oxford). 2011 Jan;13(1):1-14
pubmed: 21159098
J Hepatobiliary Pancreat Sci. 2019 Apr;26(4):123-127
pubmed: 30828991
Surg Endosc. 2021 Mar;35(3):1088-1092
pubmed: 32107631
Ann Surg. 2007 May;245(5):763-70
pubmed: 17457169
Ann Surg. 2013 Sep;258(3):490-9
pubmed: 24022441
J Gastrointest Surg. 2021 Nov;25(11):2796-2805
pubmed: 33532980
Br J Surg. 2014 Jan;101(2):63-8
pubmed: 24318962
Br J Surg. 2008 Apr;95(4):460-5
pubmed: 18161898
Ann Surg. 2020 Jul;272(1):3-23
pubmed: 32404658
Ann Surg. 2020 Mar;271(3):449-459
pubmed: 32106173
Surg Endosc. 2021 Jun;35(6):2879-2888
pubmed: 32572630
Surgery. 2018 May;163(5):1121-1127
pubmed: 29475612
Surg Clin North Am. 2019 Apr;99(2):283-299
pubmed: 30846035
Ann Surg. 2016 Sep;264(3):544-53
pubmed: 27433902
J Hepatobiliary Pancreat Sci. 2017 Nov;24(11):591-602
pubmed: 28884962
HPB (Oxford). 2019 Dec;21(12):1641-1647
pubmed: 31151812
Hepatobiliary Pancreat Dis Int. 2021 Feb;20(1):53-60
pubmed: 33268245
J Visc Surg. 2013 Jun;150(3 Suppl):S33-8
pubmed: 23791984
N Engl J Med. 2017 Jun 29;376(26):2507-2509
pubmed: 28657867
Surg Endosc. 2018 Mar;32(3):1184-1191
pubmed: 28840410
J Am Coll Surg. 2014 Nov;219(5):923-32.e10
pubmed: 25127511
Am J Gastroenterol. 2020 Aug;115(8):1191-1198
pubmed: 32483004
HPB (Oxford). 2017 Apr;19(4):352-358
pubmed: 28189346
EBioMedicine. 2019 Aug;46:27-29
pubmed: 31303500
Ann Surg. 1997 May;225(5):459-68; discussion 468-71
pubmed: 9193174
HPB (Oxford). 2016 Dec;18(12):1010-1016
pubmed: 27773464
Ann Surg. 2018 Jul;268(1):143-150
pubmed: 28426479
BMJ. 2020 Mar 25;368:m689
pubmed: 32213531