Prediction of thirty-day morbidity and mortality after duodenal switch using an artificial neural network.
Artificial neural network
Bariatric surgery
Duodenal switch
Outcomes
Journal
Surgical endoscopy
ISSN: 1432-2218
Titre abrégé: Surg Endosc
Pays: Germany
ID NLM: 8806653
Informations de publication
Date de publication:
02 2023
02 2023
Historique:
received:
10
03
2022
accepted:
03
06
2022
pubmed:
29
6
2022
medline:
25
2
2023
entrez:
28
6
2022
Statut:
ppublish
Résumé
Understanding factors that increase risk of both mortality and specific measures of morbidity after duodenal switch (DS) is important in deciding to offer this weight loss operation. Artificial neural networks (ANN) are computational deep learning approaches that model complex interactions among input factors to optimally predict an outcome. Here, a comprehensive national database is examined for patient factors associated with poor outcomes, while comparing the performance of multivariate logistic regression and ANN models in predicting these outcomes. 2907 DS patients from the 2019 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database were assessed for patient factors associated with the previously validated composite endpoint of 30-day postoperative reintervention, reoperation, readmission, or mortality using bivariate analysis. Variables associated (P ≤ 0.05) with the endpoint were imputed in a multivariate logistic regression model and a three-node ANN with 20% holdback for validation. Goodness-of-fit was assessed using area under receiver operating curves (AUROC). There were 229 DS patients with the composite endpoint (7.9%), and 12 mortalities (0.4%). Associated patient factors on bivariate analysis included advanced age, non-white race, cardiac history, hypertension requiring 3 + medications (HTN), previous foregut/obesity surgery, obstructive sleep apnea (OSA), and higher creatinine (P ≤ 0.05). Upon multivariate analysis, independently associated factors were non-white race (odds ratio 1.40; P = 0.075), HTN (1.55; P = 0.038), previous foregut/bariatric surgery (1.43; P = 0.041), and OSA (1.46; P = 0.018). The nominal logistic regression multivariate analysis (n = 2330; R Readily obtainable patient factors were identified that confer increased risk of the 30-day composite endpoint after DS. Moreover, use of an ANN to model these factors may optimize prediction of this outcome. This information provides useful guidance to bariatricians and surgical candidates alike.
Sections du résumé
BACKGROUND
Understanding factors that increase risk of both mortality and specific measures of morbidity after duodenal switch (DS) is important in deciding to offer this weight loss operation. Artificial neural networks (ANN) are computational deep learning approaches that model complex interactions among input factors to optimally predict an outcome. Here, a comprehensive national database is examined for patient factors associated with poor outcomes, while comparing the performance of multivariate logistic regression and ANN models in predicting these outcomes.
METHODS
2907 DS patients from the 2019 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database were assessed for patient factors associated with the previously validated composite endpoint of 30-day postoperative reintervention, reoperation, readmission, or mortality using bivariate analysis. Variables associated (P ≤ 0.05) with the endpoint were imputed in a multivariate logistic regression model and a three-node ANN with 20% holdback for validation. Goodness-of-fit was assessed using area under receiver operating curves (AUROC).
RESULTS
There were 229 DS patients with the composite endpoint (7.9%), and 12 mortalities (0.4%). Associated patient factors on bivariate analysis included advanced age, non-white race, cardiac history, hypertension requiring 3 + medications (HTN), previous foregut/obesity surgery, obstructive sleep apnea (OSA), and higher creatinine (P ≤ 0.05). Upon multivariate analysis, independently associated factors were non-white race (odds ratio 1.40; P = 0.075), HTN (1.55; P = 0.038), previous foregut/bariatric surgery (1.43; P = 0.041), and OSA (1.46; P = 0.018). The nominal logistic regression multivariate analysis (n = 2330; R
CONCLUSION
Readily obtainable patient factors were identified that confer increased risk of the 30-day composite endpoint after DS. Moreover, use of an ANN to model these factors may optimize prediction of this outcome. This information provides useful guidance to bariatricians and surgical candidates alike.
Identifiants
pubmed: 35764835
doi: 10.1007/s00464-022-09378-5
pii: 10.1007/s00464-022-09378-5
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1440-1448Informations de copyright
© 2022. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.
Références
Roth AE, Thornley CJ, Blackstone RP (2020) Outcomes in Bariatric and metabolic surgery: an updated 5-year review. Curr Obes Re 9:380–389
doi: 10.1007/s13679-020-00389-8
Hedberg J, Sundstrom J, Sundbom M (2014) Duodenal switch versus Roux-en-Y gastric bypass for morbid obesity: systematic review and meta-analysis of weight results, diabetes resolution and early complications in single-centre comparisons. Obes Rev 15:555–563
doi: 10.1111/obr.12169
pubmed: 24666623
Topart P, Becouarn G, Ritz P (2013) Weight loss is more sustained after biliopancreatic diversion with duodenal switch than Roux-en-Y gastric bypass in superobese patients. Surg Obes Relat Dis 9:526–530
doi: 10.1016/j.soard.2012.02.006
pubmed: 22498360
Wise ES, Hocking KM, Kavic SM (2016) Prediction of excess weight loss after laparoscopic Roux-en-Y gastric bypass: data from an artificial neural network. Surg Endosc 30:480–488
doi: 10.1007/s00464-015-4225-7
pubmed: 26017908
Wise ES, Amateau SK, Ikramuddin S, Leslie DB (2020) Prediction of thirty-day morbidity and mortality after laparoscopic sleeve gastrectomy: data from an artificial neural network. Surg Endosc 34:3590–3596
doi: 10.1007/s00464-019-07130-0
pubmed: 31571034
Cao Y, Naslund I, Naslund E, Ottosson J, Montgomery S, Stenberg E (2021) Using a convolutional neural network to predict remission of diabetes after gastric bypass surgery: machine learning study from the scandinavian obesity surgery register. JMIR Med Inform 9:e25612
doi: 10.2196/25612
pubmed: 34420921
pmcid: 8414302
Telem DA, Dimick JB (2018) Practical guide to surgical data sets: metabolic and bariatric surgery accreditation and quality program (MBSAQIP). JAMA Surg 153:766–767
doi: 10.1001/jamasurg.2018.0495
pubmed: 29617538
Minhem MA, Safadi BY, Habib RH, Raad EPB, Alami RS (2018) Increased adverse outcomes after laparoscopic sleeve gastrectomy in older super-obese patients: analysis of American College of Surgeons National Surgical Quality Improvement Program Database. Surg Obes Relat Dis 14:1463–1470
doi: 10.1016/j.soard.2018.06.023
pubmed: 30449508
Wise ES, Beckman TR, Amateau SK, Ikramuddin S, Leslie DB (2021) The role of preoperative anemia in predicting short-term morbidity and mortality after Roux-En-Y gastric bypass. Am Surg 87:1926–1933
doi: 10.1177/0003134820982852
pubmed: 33502216
Hanley JA, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29–36
doi: 10.1148/radiology.143.1.7063747
pubmed: 7063747
Clapp B, Mehta K, Corbett J, El Badaoui J, Vahibe A, Ghanem OM (2022) Duodenal switch versus Roux-en-Y gastric bypass: a perioperative risk comparative analysis of the MBSAQIP Database (2015–2019). Surg Obes Relat Dis 18:253–259
doi: 10.1016/j.soard.2021.10.014
pubmed: 34789419
Landin MD, Gordee A, Lerebours RC, Kuchibhatla M, Eckhouse SR, Seymour KA (2022) Trends in risk factors for readmission after bariatric surgery 2015–2018. Surg Obes Relat Dis 15:159
Khrucharoen U, Juo YY, Wongpongsalee T, Chen Y, Dutson EP (2021) Risk factors for readmission for early small bowel obstruction following laparoscopic Roux-en-Y gastric bypass: an MBSAQIP analysis. Surg Obes Relat Dis 17:1041–1048
doi: 10.1016/j.soard.2021.02.031
pubmed: 33965351
Verhoeff K, Mocanu V, Dang J, Switzer NJ, Birch DW, Karmali S (2022) Characterization and risk factors for early biliary complications following elective bariatric surgery: an Mbsaqip analysis. Obes Surg 15:20
Yolsuriyanwong K, Ingviya T, Kongkamol C, Marcotte E, Chand B (2019) Effects of intraoperative leak testing on postoperative leak-related outcomes after primary bariatric surgery: an analysis of the MBSAQIP database. Surg Obes Relat Dis 15:1530–1540
doi: 10.1016/j.soard.2019.06.008
pubmed: 31474524
Castro A, Cassinello N, Alfonso R, Ortega J (2022) Preoperative risk factors for early hemorrhagic complications in bariatric surgery: a case-control study. Surg Endosc 36:430–434
doi: 10.1007/s00464-021-08302-7
pubmed: 33523271
Wang A, Wang H, Abdurakhmanov A, Vijayanagar V, Thompson KJ, McKillop IH, Barbat S, Bauman R, Gersin KS, Kuwada TS, Nimeri A (2022) Safety of Primary Versus Revisional Biliopancreatic Diversion with Duodenal Switch in Patients with Super Obesity Using the MBSAQIP database. Obes Surg 20:5–9
Haim A, Daniel S, Hershkovitz E, Goldbart AD, Tarasiuk A (2021) Obstructive sleep apnea and metabolic disorders in morbidly obese adolescents. Pediatr Pulmonol 56:3983–3990
doi: 10.1002/ppul.25652
pubmed: 34499813
Mocanu V, Dang JT, Delisle M, Switzer NJ, Birch DW, Karmali S (2022) 30-day Major Adverse Cardiac Events (MACE) after elective bariatric surgery: the development and validation of the Bari-MACE clinical predictive tool. Surg Obes Relat Dis 18:357–364
doi: 10.1016/j.soard.2021.12.008
pubmed: 35016838
Wise ES, Stonko DP, Glaser ZA, Garcia KL, Huang JJ, Kim JS, Kallos JA, Starnes JR, Fleming JW, Hocking KM, Brophy CM, Eagle SS (2017) Prediction of prolonged ventilation after coronary artery bypass grafting: data from an artificial neural network. Heart Surg Forum 20:E007-E014
doi: 10.1532/hsf.1566
pubmed: 28263144
Wise ES, Hocking KM, Brophy CM (2015) Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network. J Vasc Surg 62:8–15
doi: 10.1016/j.jvs.2015.02.038
pubmed: 25953014
pmcid: 4484301
Zhao Y, Hu B, Wang Y, Yin X, Jiang Y, Zhu X (2022) Identification of gastric cancer with convolutional neural networks: a systematic review. Multimed Tools Appl 15:1–20
Lui TKL, Tsui VWM, Leung WK (2020) Accuracy of artificial intelligence-assisted detection of upper GI lesions: a systematic review and meta-analysis. Gastrointest Endosc 92(821–830):e829
Morris RS, Tignanelli CJ, deRoon-Cassini T, Laud P, Sparapani R (2022) Improved prediction of older adult discharge after trauma using a novel machine learning paradigm. J Surg Res 270:39–48
doi: 10.1016/j.jss.2021.08.021
pubmed: 34628162
Nagai S, Nallabasannagari AR, Moonka D, Reddiboina M, Yeddula S, Kitajima T, Francis I, Abouljoud M (2022) Use of neural network models to predict liver transplant waitlist mortality. Liver Transpl 25:1–9
Min JK, Kwak MS, Cha JM (2019) Overview of Deep Learning in Gastrointestinal Endoscopy. Gut Liver 13:388–393
doi: 10.5009/gnl18384
pubmed: 30630221
pmcid: 6622562
Grieco A, Huffman KM, Cohen ME, Hall BL, Morton JM, Ko CY (2021) The metabolic and bariatric surgery accreditation and quality improvement program bariatric surgical risk/benefit calculator: 30-day risk. Surg Obes Relat Dis 25:8
van Olst N, van Rijswijk AS, Mikdad S, Schoonmade LJ, van de Laar AW, Acherman YIZ, Bruin SC, van der Peet DL, de Brauw LM (2021) Long-term emergency department visits and readmissions after laparoscopic Roux-en-Y gastric bypass: a systematic review. Obes Surg 5:1–9
Mocanu V, Dang J, Ladak F, Switzer N, Birch DW, Karmali S (2019) Predictors and outcomes of leak after Roux-en-Y gastric bypass: an analysis of the MBSAQIP data registry. Surg Obes Relat Dis 15:396–403
doi: 10.1016/j.soard.2019.01.012
pubmed: 30782473
Tripepi G, Jager KJ, Dekker FW, Zoccali C (2010) Selection bias and information bias in clinical research. Nephron Clin Pract 115:c94-99
doi: 10.1159/000312871
pubmed: 20407272