Development and validation of three machine-learning models for predicting multiple organ failure in moderately severe and severe acute pancreatitis

Qiu Qiu, Yong-Jian Nian, Yan Guo, Liang Tang, Nan Lu, Liang-Zhi Wen, Bin Wang, Dong-Feng Chen, Kai-Jun Liu, Qiu Qiu, Yong-Jian Nian, Yan Guo, Liang Tang, Nan Lu, Liang-Zhi Wen, Bin Wang, Dong-Feng Chen, Kai-Jun Liu

Abstract

Background: Multiple organ failure (MOF) is a serious complication of moderately severe (MASP) and severe acute pancreatitis (SAP). This study aimed to develop and assess three machine-learning models to predict MOF.

Methods: Patients with MSAP and SAP who were admitted from July 2014 to June 2017 were included. Firstly, parameters with significant differences between patients with MOF and without MOF were screened out by univariate analysis. Then, support vector machine (SVM), logistic regression analysis (LRA) and artificial neural networks (ANN) models were constructed based on these factors, and five-fold cross-validation was used to train each model.

Results: A total of 263 patients were enrolled. Univariate analysis screened out sixteen parameters referring to blood volume, inflammatory, coagulation and renal function to construct machine-learning models. The predictive efficiency of the optimal combinations of features by SVM, LRA, and ANN was almost equal (AUC = 0.840, 0.832, and 0.834, respectively), as well as the Acute Physiology and Chronic Health Evaluation II score (AUC = 0.814, P > 0.05). The common important predictive factors were HCT, K-time, IL-6 and creatinine in three models.

Conclusions: Three machine-learning models can be efficient prognostic tools for predicting MOF in MSAP and SAP. ANN is recommended, which only needs four common parameters.

Keywords: Machine learning; Multiple organ failure; Pancreatitis.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1
Fig. 1
The ROC curves of different models. a The ROC curves of different combinations of features from SVM for predicting MOF in MSAP and SAP. AUC of the optimal combination = 0.840 (95% CI: 0.783–0.896); AUC of single feature (BUN) = 0.702 (95% CI: 0.625–0.778); AUC of all features = 0.816 (95% CI: 0.755–0.876). b The ROC curves of different combinations of features from LRA for predicting MOF in MSAP and SAP. AUC of the optimal combination = 0.832 (95% CI: 0.773–0.890); AUC of single feature (IL-6) = 0.709 (95% CI: 0.642–0.775); AUC of all features = 0.783 (95% CI: 0.714–0.853). c The ROC curves of different combinations of features from ANN for predicting MOF in MSAP and SAP. AUC of the optimal combination = 0.834 (95% CI: 0.777–0.890); AUC of single feature (IL-6) = 0.705 (95% CI: 0.639–0.772); AUC of all features = 0.789 (95% CI: 0.723–0.856). d The ROC curves of three models and the APACHE II score for predicting MOF in MSAP and SAP. AUC of SVM = 0.840 (95% CI: 0.783–0.896); AUC of LRA = 0.832 (95% CI: 0.773–0.890); AUC of ANN = 0.834 (95% CI: 0.777–0.890); AUC of APACHE II score = 0.814 (95% CI: 0.759–0.869)

References

    1. Banks PA, Bollen TL, Dervenis C, Gooszen HG, Johnson CD, Sarr MG, et al. Classification of acute pancreatitis--2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013;62(1):102–111. doi: 10.1136/gutjnl-2012-302779.
    1. Mole DJ, Olabi B, Robinson V, Garden OJ, Parks RW. Incidence of individual organ dysfunction in fatal acute pancreatitis: analysis of 1024 death records. HPB (Oxford) 2009;11(2):166–170. doi: 10.1111/j.1477-2574.2009.00038.x.
    1. Doctor N, Agarwal P, Gandhi V. Management of severe acute pancreatitis. Indian J Surg. 2012;74(1):40–46. doi: 10.1007/s12262-011-0384-5.
    1. Mc Kay CJ, Buter A. Natural history of organ failure in acute pancreatitis. Pancreatology. 2003;3(2):111–114. doi: 10.1159/000070078.
    1. Lisman T, Porte RJ. Activation and regulation of hemostasis in acute liver failure and acute pancreatitis. Semin Thromb Hemost. 2010;36(4):437–443. doi: 10.1055/s-0030-1254052.
    1. Dumnicka Paulina, Maduzia Dawid, Ceranowicz Piotr, Olszanecki Rafał, Drożdż Ryszard, Kuśnierz-Cabala Beata. The Interplay between Inflammation, Coagulation and Endothelial Injury in the Early Phase of Acute Pancreatitis: Clinical Implications. International Journal of Molecular Sciences. 2017;18(2):354. doi: 10.3390/ijms18020354.
    1. Wu BU, Bakker OJ, Papachristou GI, Besselink MG, Repas K, van Santvoort HC, et al. Blood urea nitrogen in the early assessment of acute pancreatitis: an international validation study. Arch Intern Med. 2011;171(7):669–676. doi: 10.1001/archinternmed.2011.126.
    1. Muddana V, Whitcomb DC, Khalid A, Slivka A, Papachristou GI. Elevated serum creatinine as a marker of pancreatic necrosis in acute pancreatitis. Am J Gastroenterol. 2009;104(1):164–170. doi: 10.1038/ajg.2008.66.
    1. Librenza-Garcia D, Kotzian BJ, Yang J, Mwangi B, Cao B, Pereira Lima LN, et al. The impact of machine learning techniques in the study of bipolar disorder: a systematic review. Neurosci Biobehav Rev. 2017;80:538–554. doi: 10.1016/j.neubiorev.2017.07.004.
    1. Tenner S, Baillie J, DeWitt J, Vege SS. American College of G. American College of Gastroenterology guideline: management of acute pancreatitis. Am J Gastroenterol. 2013;108(9):1400–15; 16. doi: 10.1038/ajg.2013.218.
    1. Yadav D, Lowenfels AB. The epidemiology of pancreatitis and pancreatic cancer. Gastroenterology. 2013;144(6):1252–1261. doi: 10.1053/j.gastro.2013.01.068.
    1. Wan J, He W, Zhu Y, Zhu Y, Zeng H, Liu P, et al. Stratified analysis and clinical significance of elevated serum triglyceride levels in early acute pancreatitis: a retrospective study. Lipids Health Dis. 2017;16(1):124. doi: 10.1186/s12944-017-0517-3.
    1. Batty GD, Barzi F, Huxley R, Chang CY, Jee SH, Jamrozik K, et al. Obesity and liver cancer mortality in Asia: the Asia Pacific cohort studies collaboration. Cancer Epidemiol. 2009;33(6):469–472. doi: 10.1016/j.canep.2009.09.010.
    1. Petrov MS, Shanbhag S, Chakraborty M, Phillips AR, Windsor JA. Organ failure and infection of pancreatic necrosis as determinants of mortality in patients with acute pancreatitis. Gastroenterology. 2010;139(3):813–820. doi: 10.1053/j.gastro.2010.06.010.
    1. Afghani E, Pandol SJ, Shimosegawa T, Sutton R, Wu BU, Vege SS, et al. Acute pancreatitis-Progress and challenges: a report on an international symposium. Pancreas. 2015;44(8):1195–1210. doi: 10.1097/MPA.0000000000000500.
    1. Mole DJ, Webster SP, Uings I, Zheng X, Binnie M, Wilson K, et al. Kynurenine-3-monooxygenase inhibition prevents multiple organ failure in rodent models of acute pancreatitis. Nat Med. 2016;22(2):202–209. doi: 10.1038/nm.4020.
    1. Savage N. Machine learning: calculating disease. Nature. 2017;550(7676):S115–S1S7. doi: 10.1038/550S115a.
    1. Koutroumpakis E, Wu BU, Bakker OJ, Dudekula A, Singh VK, Besselink MG, et al. Admission hematocrit and rise in blood urea nitrogen at 24 h outperform other laboratory markers in predicting persistent organ failure and pancreatic necrosis in acute pancreatitis: a post hoc analysis of three large prospective databases. Am J Gastroenterol. 2015;110(12):1707–1716. doi: 10.1038/ajg.2015.370.
    1. Li N, Wang BM, Cai S, Liu PL. The role of serum high mobility group box 1 and Interleukin-6 levels in acute pancreatitis: a meta-analysis. J Cell Biochem. 2018;119(1):616–624. doi: 10.1002/jcb.26222.
    1. Merza M, Hartman H, Rahman M, Hwaiz R, Zhang E, Renström E, et al. Neutrophil extracellular traps induce trypsin activation, inflammation, and tissue damage in mice with severe acute pancreatitis. Gastroenterology. 2015;149(7):1920–1931. doi: 10.1053/j.gastro.2015.08.026.
    1. Hong SS, Choi JH, Lee SY, Park YH, Park KY, Lee JY, et al. A novel small-molecule inhibitor targeting the IL-6 receptor beta subunit, glycoprotein 130. J Immunol. 2015;195(1):237–245. doi: 10.4049/jimmunol.1402908.
    1. Zhang H, Neuhofer P, Song L, Rabe B, Lesina M, Kurkowski MU, et al. IL-6 trans-signaling promotes pancreatitis-associated lung injury and lethality. J Clin Invest. 2013;123(3):1019–1031. doi: 10.1172/JCI64931.
    1. Radenkovic D, Bajec D, Ivancevic N, Milic N, Bumbasirevic V, Jeremic V, et al. D-dimer in acute pancreatitis: a new approach for an early assessment of organ failure. Pancreas. 2009;38(6):655–660. doi: 10.1097/MPA.0b013e3181a66860.
    1. Liu H, Li J, Yu J, Yuan T. Research into the predictive effect of TEG in the changes of coagulation functions of the patients with traumatic brain hemorrhage. Open Med (Wars) 2015;10(1):399–404.
    1. Liao CY, Huang SC, Lin CH, Wang CC, Liu MY, Ben RJ, et al. Successful resolution of symmetrical peripheral gangrene after severe acute pancreatitis: a case report. J Med Case Rep. 2015;9:213. doi: 10.1186/s13256-015-0688-3.
    1. Ou ZB, Miao CM, Ye MX, Xing DP, He K, Li PZ, et al. Investigation for role of tissue factor and blood coagulation system in severe acute pancreatitis and associated liver injury. Biomed Pharmacother. 2017;85:380–388. doi: 10.1016/j.biopha.2016.11.039.
    1. Siemiatkowski A, Wereszczynska-Siemiatkowska U, Mroczko B, Galar M, Maziewski T. Circulating endothelial mediators in human pancreatitis-associated lung injury. Eur J Gastroenterol Hepatol. 2015;27(6):728–734. doi: 10.1097/MEG.0000000000000338.
    1. Danckwardt S, Hentze MW, Kulozik AE. Pathologies at the nexus of blood coagulation and inflammation: thrombin in hemostasis, cancer, and beyond. J Mol Med (Berl) 2013;91(11):1257–1271. doi: 10.1007/s00109-013-1074-5.
    1. Wang X, Xu Y, Qiao Y, Pang X, Hong L, Fu J, et al. An evidence-based proposal for predicting organ failure in severe acute pancreatitis. Pancreas. 2013;42(8):1255–1261. doi: 10.1097/MPA.0b013e3182a5d6a7.
    1. Deng L-H, Xue P, Xia Q, Yang X-N, Wan M-H. Effect of admission hypertriglyceridemia on the episodes of severe acute pancreatitis. World J Gastroenterol. 2008;14(28):4558. doi: 10.3748/wjg.14.4558.
    1. Wang Y, Sternfeld L, Yang F, Rodriguez JA, Ross C, Hayden MR, et al. Enhanced susceptibility to pancreatitis in severe hypertriglyceridaemic lipoprotein lipase-deficient mice and agonist-like function of pancreatic lipase in pancreatic cells. Gut. 2009;58(3):422–430. doi: 10.1136/gut.2007.146258.
    1. Bexelius TS, Ljung R, Garcia Rodriguez LA. Type 2 diabetes, high blood pressure and acute pancreatitis. Hepatobiliary Pancreat Dis Int. 2016;15(4):443–445. doi: 10.1016/S1499-3872(16)60117-0.
    1. Mole DJ, Gungabissoon U, Johnston P, Cochrane L, Hopkins L, Wyper GM, et al. Identifying risk factors for progression to critical care admission and death among individuals with acute pancreatitis: a record linkage analysis of Scottish healthcare databases. BMJ Open. 2016;6(6):e011474. doi: 10.1136/bmjopen-2016-011474.
    1. Fei Y, Hu J, Li WQ, Wang W, Zong GQ. Artificial neural networks predict the incidence of portosplenomesenteric venous thrombosis in patients with acute pancreatitis. J Thromb Haemost. 2017;15(3):439–445. doi: 10.1111/jth.13588.

Source: PubMed

Sponsoren und Mitarbeiter

Krankheiten

Drogeninterventionen

3
Abonnieren