Risk prediction of in-hospital major adverse events of postoperative Stanford type A aortic dissection based on machine learning
Author:
Affiliation:

1.Department of Physiology and Pathophysiology of Tianjin Medical University & Collaborative Innovation Center of Tianjin for Medical Epigenetics, Tianjin 300070, China;2.Beijing Anzhen Hospital Affiliated to Capital Medical University & Department of Vascular Biology, Beijing Institute of Heart, Lung and Blood Vessel Disease & Key Laboratory of Remodeling-related Cardiovascular Diseases, Ministry of Education & Collaborative Innovation Center for Cardiovascular Disorders, Beijing 100029, China;3.Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing 100191, China)

Clc Number:

R54;R654

  • Article
  • | |
  • Metrics
  • |
  • Reference [20]
  • | | | |
  • Comments
    Abstract:

    Aim To build a model based on general clinical variables and machine learning method to predict the risk of in-hospital major adverse events (MAE) in patients with Stanford a type aortic dissection (TAAD) after surgery. Methods A total of 1 641 patients with TAAD who underwent surgical treatment in Beijing Anzhen Hospital from January 2013 to December 2017 were included in this study. The individual characteristic variables, clinical signs and the first clinical serum markers on admission were collected. The outcome was defined as in-hospital MAE, including in-hospital death, new acute heart failure after mezzanine, respiratory failure, nervous system disorders, acute renal failure, infection, and unplanned secondary chest opening. The model was constructed after using machine learning to screen variables. Receiver operating characteristic curve (ROC) was used to analyze the ability of the model to predict in-hospital MAE. Net reclassification index (NRI) and integrated discrimination index (IDI) were used to compare the new model with the commonly used clinical models to evaluate the improvement effect of the new model in predicting the prognosis of postoperative TAAD. Finally, the nomogram was established to predict the risk of MAE in patients after TAAD operation. Results The risk prediction model of in-hospital MAE after TAAD operation was determined by using machine learning screening variables, which consisted of D-dimer, creatine kinase isoenzyme, urea, leukocyte count, age, abnormal electrocardiogram and operation time. The area under curve of ROC of in-hospital MAE predicted by the model was 0.776 (95%CI 0.718-0.734, P<0.001). Compared with the commonly used clinical models, the NRI of our model was 0.654 (95%CI 0.540-0.750, P<0.001), and the IDI was 0.136 (95%CI 0.117-0.155, P<0.001), which improved the predictive ability for in-hospital MAE after TAAD operation. The model was presented in the form of nomogram, and the score of nomogram model could evaluate the risk of in-hospital MAE after TAAD operation. ConclusionsBased on machine learning, a model is constructed by using clinical variables of patients. The model can comprehensively evaluate individual characteristic variables, inflammation level, organ damage status and operation status of patients, which has predictive value for postoperative in-hospital MAE of patients with TAAD.

    Reference
    [1] Erbel R, Aboyans V, Boileau C, et al.2014 ESC guidelines on the diagnosis and treatment of aortic diseases:document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult.The task force for the diagnosis and treatment of aortic diseases of the European Society of Cardiology (ESC).Eur Heart J, 4,5(41):2873-2926.
    [2] Goldfinger JZ, Halperin JL, Marin ML, et al.Thoracic aortic aneurysm and dissection.J Am Coll Cardiol, 4,4(16):1725-1739.
    [3] Pape LA, Awais M, Woznicki EM, et al.Presentation, diagnosis, and outcomes of acute aortic dissection:17-year trends from the international registry of acute aortic dissection.J Am Coll Cardiol, 5,6(4):350-358.
    [4] 周楚芝, 杨建安, 温隽珉, 等.急性A型主动脉夹层患者术后谵妄发生的危险因素分析.中国动脉硬化杂志, 6,4(12):1253-1257.
    [5] Evangelista A, Isselbacher EM, Bossone E, et al.Insights from the international registry of acute aortic dissection:a 20-year experience of collaborative clinical research.Circulation, 8,7(17):1846-1860.
    [6] Mehta RH, Suzuki T, Hagan PG, et al.Predicting death in patients with acute type a aortic dissection.Circulation, 2,5(2):200-206.
    [7] Al'Aref SJ, Anchouche K, Singh G, et al.Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging.Eur Heart J, 9,0(24):1975-1986.
    [8] Angraal S, Mortazavi BJ, Gupta A, et al.Machine learning prediction of mortality and hospitalization in heart failure with preserved ejection fraction.JACC Heart Fail, 0,8(1):12-21.
    [9] Al'Aref SJ, Singh G, Choi JW, et al.A boosted ensemble algorithm for determination of plaque stability in high-risk patients on coronary cta.JACC Cardiovasc Imaging, 0,3(10):2162-2173.
    [10] Al'Aref SJ, Maliakal G, Singh G, et al.Machine learning of clinical variables and coronary artery calcium scoring for the prediction of obstructive coronary artery disease on coronary computed tomography angiography:analysis from the confirm registry.Eur Heart J, 0,1(3):359-367.
    [11] Huo D, Kou B, Zhou Z, et al.A machine learning model to classify aortic dissection patients in the early diagnosis phase.Sci Rep, 9,9(1):2701.
    [12] Suzuki T, Distante A, Zizza A, et al.Diagnosis of acute aortic dissection by D-dimer:the international registry of acute aortic dissection substudy on biomarkers (IRAD-Bio) experience.Circulation, 9,9(20):2702-2707.
    [13] Weber T, Rammer M, Auer J, et al.Plasma concentrations of D-dimer predict mortality in acute type a aortic dissection.Heart, 6,2(6):836-837.
    [14] Czerny M, Schoenhoff F, Etz C, et al.The impact of pre-operative malperfusion on outcome in acute type a aortic dissection:results from the geraada registry.J Am Coll Cardiol, 5,5(24):2628-2635.
    [15] Fan X, Huang B, Lu H, et al.Impact of admission white blood cell count on short-and long-term mortality in patients with type a acute aortic dissection:an observational study.Medicine (Baltimore), 5,4(42):e1761.
    [16] Ehrlich M, Fang WC, Grabenwger M, et al.Perioperative risk factors for mortality in patients with acute type a aortic dissection.Circulation, 8,8(19 Suppl):II294-II298.
    [17] Nienaber CA, Fattori R, Mehta RH, et al.Gender-related differences in acute aortic dissection.Circulation, 4,9(24):3014-3021.
    [18] Mortazavi BJ, Downing NS, Bucholz EM, et al.Analysis of machine learning techniques for heart failure readmissions.Circ Cardiovasc Qual Outcomes, 6,9(6):629-640.
    [19] 熊日新, 林英忠.人工智能在心血管疾病风险评估中的应用研究进展.中国临床新医学, 0,3(5):537-540.
    [20] Connor CW.Artificial intelligence and machine learning in anesthesiology.Anesthesiology, 9,1(6):1346-1359.
    Related
    Cited by
    Comments
    Comments
    分享到微博
    Submit
Get Citation

PEI Wang, WANG Xue, JIANG Wenxi, AN Meiyu, XUE Bingjie, GAO Pei, WANG Yuan, DU Jie. Risk prediction of in-hospital major adverse events of postoperative Stanford type A aortic dissection based on machine learning[J]. Editorial Office of Chinese Journal of Arteriosclerosis,2021,29(4):332-338.

Copy
Share
Article Metrics
  • Abstract:1252
  • PDF: 835
  • HTML: 0
  • Cited by: 0
History
  • Received:February 22,2021
  • Revised:March 17,2021
  • Online: April 14,2021
Article QR Code