Use of Machine Learning for Prediction of Patient Risk of Postoperative Complications After Liver, Pancreatic, and Colorectal Surgery.
Colorectal
Complications
Liver
Machine learning
Pancreas
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 2020
08 2020
Historique:
received:
20
05
2019
accepted:
21
07
2019
pubmed:
7
8
2019
medline:
15
4
2021
entrez:
7
8
2019
Statut:
ppublish
Résumé
Surgical resection is the only potentially curative treatment for patients with colorectal, liver, and pancreatic cancers. Although these procedures are performed with low mortality, rates of complications remain relatively high following hepatopancreatic and colorectal surgery. The American College of Surgeons (ACS) National Surgical Quality Improvement Program was utilized to identify patients undergoing liver, pancreatic and colorectal surgery from 2014 to 2016. Decision tree models were utilized to predict the occurrence of any complication, as well as specific complications. To assess the variability of the performance of the classification trees, bootstrapping was performed on 50% of the sample. Algorithms were derived from a total of 15,657 patients who met inclusion criteria. The algorithm had a good predictive ability for the occurrence of any complication, with a C-statistic of 0.74, outperforming the ASA (C-statistic 0.58) and ACS-Surgical Risk Calculator (C-statistic 0.71). The algorithm was able to predict with high accuracy thirteen out of the seventeen complications analyzed. The best performance was in the prediction of stroke (C-statistic 0.98), followed by wound dehiscence, cardiac arrest, and progressive renal failure (all C-statistic 0.96). The algorithm had a good predictive ability for superficial SSI (C-statistic 0.76), organ space SSI (C-statistic 0.76), sepsis (C-statistic 0.79), and bleeding requiring transfusion (C-statistic 0.79). Machine learning was used to develop an algorithm that accurately predicted patient risk of developing complications following liver, pancreatic, or colorectal surgery. The algorithm had very good predictive ability to predict specific complications and demonstrated superiority over other established methods.
Sections du résumé
BACKGROUND
Surgical resection is the only potentially curative treatment for patients with colorectal, liver, and pancreatic cancers. Although these procedures are performed with low mortality, rates of complications remain relatively high following hepatopancreatic and colorectal surgery.
METHODS
The American College of Surgeons (ACS) National Surgical Quality Improvement Program was utilized to identify patients undergoing liver, pancreatic and colorectal surgery from 2014 to 2016. Decision tree models were utilized to predict the occurrence of any complication, as well as specific complications. To assess the variability of the performance of the classification trees, bootstrapping was performed on 50% of the sample.
RESULTS
Algorithms were derived from a total of 15,657 patients who met inclusion criteria. The algorithm had a good predictive ability for the occurrence of any complication, with a C-statistic of 0.74, outperforming the ASA (C-statistic 0.58) and ACS-Surgical Risk Calculator (C-statistic 0.71). The algorithm was able to predict with high accuracy thirteen out of the seventeen complications analyzed. The best performance was in the prediction of stroke (C-statistic 0.98), followed by wound dehiscence, cardiac arrest, and progressive renal failure (all C-statistic 0.96). The algorithm had a good predictive ability for superficial SSI (C-statistic 0.76), organ space SSI (C-statistic 0.76), sepsis (C-statistic 0.79), and bleeding requiring transfusion (C-statistic 0.79).
CONCLUSION
Machine learning was used to develop an algorithm that accurately predicted patient risk of developing complications following liver, pancreatic, or colorectal surgery. The algorithm had very good predictive ability to predict specific complications and demonstrated superiority over other established methods.
Identifiants
pubmed: 31385172
doi: 10.1007/s11605-019-04338-2
pii: 10.1007/s11605-019-04338-2
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1843-1851Commentaires et corrections
Type : CommentIn
Références
Merath K, Chen Q, Bagante F, Sun S, Akgul O, Idrees JJ, et al. Variation in the cost-of-rescue among medicare patients with complications following hepatopancreatic surgery. HPB (Oxford) [Internet]. 2018 Sep; Available from: https://www.ncbi.nlm.nih.gov/pubmed/30266495
Ghaferi AA, Birkmeyer JD, Dimick JB. Complications, failure to rescue, and mortality with major inpatient surgery in medicare patients. Ann Surg. 2009;250(6):1029–34.
pubmed: 19953723
doi: 10.1097/SLA.0b013e3181bef697
Kohlnhofer BM, Tevis SE, Weber SM, Kennedy GD. Multiple complications and short length of stay are associated with postoperative readmissions. Am J Surg. 2014;207(4):449–56.
pubmed: 24524860
pmcid: 4822702
doi: 10.1016/j.amjsurg.2013.10.022
Mavros MN, de Jong M, Dogeas E, Hyder O, Pawlik TM. Impact of complications on long-term survival after resection of colorectal liver metastases. Br J Surg. 2013;100(5):711–8.
pubmed: 23364914
doi: 10.1002/bjs.9060
Healy MA, Mullard AJ, Campbell DA, Dimick JB. Hospital and payer costs associated with surgical complications. JAMA Surg. 2016;151(9):823–30.
pubmed: 27168356
doi: 10.1001/jamasurg.2016.0773
Idrees JJ, Johnston FM, Canner JK, Dillhoff M, Schmidt C, Haut ER, et al. Cost of major complications after liver resection in the United States: are high-volume centers cost-effective? Ann Surg [Internet]. 2017; Available from: https://www.ncbi.nlm.nih.gov/pubmed/29232212
Tevis SE, Kennedy GD. Postoperative complications and implications on patient-centered outcomes. J Surg Res. 2013;181(1):106–13.
pubmed: 23465392
pmcid: 3637983
doi: 10.1016/j.jss.2013.01.032
Silber JH, Rosenbaum PR, Trudeau ME, Chen W, Zhang X, Kelz RR, et al. Changes in prognosis after the first postoperative complication. Med Care. 2005;43(2):122–31.
pubmed: 15655425
doi: 10.1097/00005650-200502000-00005
Spolverato G, Yakoob MY, Kim Y, Alexandrescu S, Marques HP, Lamelas J, et al. Impact of complications on long-term survival after resection of intrahepatic cholangiocarcinoma. Cancer. 2015;121(16):2730–9.
pubmed: 25903409
doi: 10.1002/cncr.29419
Dorcaratto D, Mazzinari G, Fernandez M, Muñoz E, Garcés-Albir M, Ortega J, et al. Impact of postoperative complications on survival and recurrence after resection of colorectal liver metastases: systematic review and meta-analysis. Ann Surg. 2019;
Merath K, Chen Q, Bagante F, Akgul O, Idrees JJ, Dillhoff M, et al. Synergistic effects of perioperative complications on 30-day mortality following hepatopancreatic surgery. Journal of Gastrointestinal Surgery. 2018;22(10):1715–23.
pubmed: 29916105
doi: 10.1007/s11605-018-3829-3
Ejaz A, Kim Y, Spolverato G, Taylor R, Hundt J, Pawlik TM. Understanding drivers of hospital charge variation for episodes of care among patients undergoing hepatopancreatobiliary surgery. HPB (Oxford). 2015;17(11):955–63.
pubmed: 26256003
pmcid: 4605332
doi: 10.1111/hpb.12452
Abdelsattar ZM, Birkmeyer JD, Wong SL. Variation in Medicare payments for colorectal cancer surgery. J Oncol Pract. 2015;11(5):391–5.
pubmed: 26130817
pmcid: 4575403
doi: 10.1200/JOP.2015.004036
Pradarelli JC, Healy MA, Osborne NH, Ghaferi AA, Dimick JB, Nathan H. Variation in Medicare expenditures for treating perioperative complications: the cost of rescue. JAMA Surg. 2016;151(12):e163340.
pubmed: 27706473
doi: 10.1001/jamasurg.2016.3340
Copeland GP, Jones D, Walters M. POSSUM: a scoring system for surgical audit. Br J Surg. 1991;78(3):355–60.
pubmed: 2021856
doi: 10.1002/bjs.1800780327
Gawande AA, Kwaan MR, Regenbogen SE, Lipsitz SA, Zinner MJ. An Apgar score for surgery. J Am Coll Surg. 2007;204(2):201–8.
pubmed: 17254923
doi: 10.1016/j.jamcollsurg.2006.11.011
Bilimoria KY, Liu Y, Paruch JL, Zhou L, Kmiecik TE, Ko CY, et al. Development and evaluation of the universal ACS NSQIP surgical risk calculator: a decision aid and informed consent tool for patients and surgeons. J Am Coll Surg. 2013;217(5):833–842.e1-3.
pubmed: 24055383
pmcid: 3805776
doi: 10.1016/j.jamcollsurg.2013.07.385
Beal EW, Saunders ND, Kearney JF, Lyon E, Wei L, Squires MH, et al. Accuracy of the ACS NSQIP online risk calculator depends on how you look at it: results from the United States Gastric Cancer Collaborative. Am Surg. 2018;84(3):358–64.
pubmed: 29559049
Gleeson EM, Shaikh MF, Shewokis PA, Clarke JR, Meyers WC, Pitt HA, et al. WHipple-ABACUS, a simple, validated risk score for 30-day mortality after pancreaticoduodenectomy developed using the ACS-NSQIP database. Surgery. 11;160(5):1279–87.
Beal EW, Lyon E, Kearney J, Wei L, Ethun CG, Black SM, et al. Evaluating the American College of Surgeons National Surgical Quality Improvement project risk calculator: results from the U.S. Extrahepatic Biliary Malignancy Consortium. HPB (Oxford). 2017;19(12):1104–11.
pubmed: 28890310
pmcid: 5915623
doi: 10.1016/j.hpb.2017.08.009
Bertsimas D, Dunn J, Velmahos GC, Kaafarani HMA. Surgical risk is not linear: derivation and validation of a novel, user-friendly, and machine-learning-based Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) Calculator. Annals of Surgery. 2018;268(4):574–83.
pubmed: 30124479
doi: 10.1097/SLA.0000000000002956
Carlos RC, Kahn CE, Halabi S. Data science: big data, machine learning, and artificial intelligence. J Am Coll Radiol. 2018;15(3 Pt B):497–8.
pubmed: 29502583
doi: 10.1016/j.jacr.2018.01.029
Linden A, Yarnold PR. Modeling time-to-event (survival) data using classification tree analysis. J Eval Clin Pract. 2017;23(6):1299–308.
pubmed: 28670833
doi: 10.1111/jep.12779
Hu Y-J, Ku T-H, Jan R-H, Wang K, Tseng Y-C, Yang S-F. Decision tree-based learning to predict patient controlled analgesia consumption and readjustment. BMC Med Inform Decis Mak. 2012;12:131.
pubmed: 23148492
pmcid: 3507711
doi: 10.1186/1472-6947-12-131
Khuri SF, Daley J, Henderson W, Hur K, Demakis J, Aust JB, et al. The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA Surgical Quality Improvement Program. Ann Surg. 1998;228(4):491–507.
pubmed: 9790339
Hall BL, Hamilton BH, Richards K, Bilimoria KY, Cohen ME, Ko CY. Does surgical quality improve in the American College of Surgeons National Surgical Quality Improvement Program: an evaluation of all participating hospitals. Ann Surg. 2009;250(3):363–76.
pubmed: 19644350
Fink AS, Campbell DA, Mentzer RM, Henderson WG, Daley J, Bannister J, et al. The National Surgical Quality Improvement Program in non-veterans administration hospitals: initial demonstration of feasibility. Ann Surg. 2002;236(3):344–53; discussion 353–354.
Hastie T, Tibshirani R, Friedman JH. The elements of statistical learning: data mining, inference, and prediction. 2nd ed. New York, NY: Springer; 2009. 745 p. (Springer series in statistics).
doi: 10.1007/978-0-387-84858-7
Pencina MJ, D’Agostino RB. Evaluating discrimination of risk prediction models: the C statistic. JAMA. 2015;314(10):1063–4.
pubmed: 26348755
doi: 10.1001/jama.2015.11082
Merath K, Bagante F, Beal EW, Lopez-Aguiar AG, Poultsides G, Makris E, et al. Nomogram predicting the risk of recurrence after curative-intent resection of primary non-metastatic gastrointestinal neuroendocrine tumors: an analysis of the U.S. Neuroendocrine Tumor Study Group. J Surg Oncol [Internet]. 2018; Available from: https://www.ncbi.nlm.nih.gov/pubmed/29448303
Efron B, Tibshirani R. An introduction to the bootstrap. New York: Chapman & Hall; 1993. 436 p. (Monographs on statistics and applied probability).
doi: 10.1007/978-1-4899-4541-9
Obermeyer Z, Emanuel EJ. Predicting the future - big data, machine learning, and clinical medicine. N Engl J Med. 2016;375(13):1216–9.
pubmed: 27682033
pmcid: 5070532
doi: 10.1056/NEJMp1606181
Zhao B, Waterman RS, Urman RD, Gabriel RA. A machine learning approach to predicting case duration for robot-assisted surgery. J Med Syst. 2019;43(2):32.
pubmed: 30612192
doi: 10.1007/s10916-018-1151-y
Harvin JA, Green CE, Pedroza C, Tyson JE, Moore LJ, Wade CE, et al. Using machine learning to identify change in surgical decision making in current use of damage control laparotomy. J Am Coll Surg. 2019;228(3):255–64.
pubmed: 30639299
pmcid: 6391184
doi: 10.1016/j.jamcollsurg.2018.12.025
Canchi T, Kumar SD, Ng EYK, Narayanan S. A review of computational methods to predict the risk of rupture of abdominal aortic aneurysms. Biomed Res Int. 2015;2015:861627.
Yeong E-K, Hsiao T-C, Chiang HK, Lin C-W. Prediction of burn healing time using artificial neural networks and reflectance spectrometer. Burns. 2005;31(4):415–20.
pubmed: 15896502
doi: 10.1016/j.burns.2004.12.003
Chen JH, Asch SM. Machine learning and prediction in medicine - beyond the peak of inflated expectations. N Engl J Med. 2017;376(26):2507–9.
pubmed: 28657867
pmcid: 5953825
doi: 10.1056/NEJMp1702071
James G, Witten D, Hastie T, Tibshirani R, editors. An introduction to statistical learning: with applications in R. New York: Springer; 2013. 426 p. (Springer texts in statistics).
Parikh SA, Gomez R, Thirugnanasambandam M, Chauhan SS, De Oliveira V, Muluk SC, et al. Decision tree-based classification of abdominal aortic aneurysms using geometry quantification measures. Ann Biomed Eng. 2018;46(12):2135–47.
pubmed: 30132212
pmcid: 6249073
doi: 10.1007/s10439-018-02116-w
Wynter-Blyth V, Moorthy K. Prehabilitation: preparing patients for surgery. BMJ. 8;358:j3702.
Minnella EM, Liberman AS, Charlebois P, Stein B, Scheede-Bergdahl C, Awasthi R, et al. The impact of improved functional capacity before surgery on postoperative complications: a study in colorectal cancer. Acta Oncol. 2019;1–6.
Raval MV, Pawlik TM. Practical guide to surgical data sets: national surgical quality improvement program (NSQIP) and pediatric NSQIP. JAMA Surg [Internet]. 2018; Available from: https://www.ncbi.nlm.nih.gov/pubmed/29617521