Advertisement

Multidimensional machine learning models predicting outcomes after trauma

Published:September 15, 2022DOI:https://doi.org/10.1016/j.surg.2022.08.007

      Abstract

      Background

      An emerging body of literature supports the role of individualized prognostic tools to guide the management of patients after trauma. The aim of this study was to develop advanced modeling tools from multidimensional data sources, including immunological analytes and clinical and administrative data, to predict outcomes in trauma patients.

      Methods

      This was a prospective study of trauma patients at Level 1 centers from 2015 to 2019. Clinical, flow cytometry, and serum cytokine data were collected within 48 hours of admission. Sparse logistic regression models were developed, jointly selecting predictors and estimating the risk of ventilator-associated pneumonia, acute kidney injury, complicated disposition (death, rehabilitation, or nursing facility), and return to the operating room. Model parameters (regularization controlling model sparsity) and performance estimation were obtained via nested leave-one-out cross-validation.

      Results

      A total of 179 patients were included. The incidences of ventilator-associated pneumonia, acute kidney injury, complicated disposition, and return to the operating room were 17.7%, 28.8%, 22.5%, and 12.3%, respectively. Regarding extensive resource use, 30.7% of patients had prolonged intensive care unit stay, 73.2% had prolonged length of stay, and 23.5% had need for prolonged ventilatory support. The models were developed and cross-validated for ventilator-associated pneumonia, acute kidney injury, complicated dispositions, and return to the operating room, yielding predictive areas under the curve from 0.70 to 0.91. Each model derived its optimal predictive value by combining clinical, administrative, and immunological analyte data.

      Conclusion

      Clinical, immunological, and administrative data can be combined to predict post-traumatic outcomes and resource use. Multidimensional machine learning modeling can identify trauma patients with complicated clinical trajectories and high resource needs.
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Surgery
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Belard A.
        • Buchman T.
        • Dente C.J.
        • Potter B.K.
        • Kirk A.
        • Elster E.
        The Uniformed Services University's Surgical Critical Care Initiative (SC2i): bringing precision medicine to the critically ill.
        Mil Med. 2018; 183: 487-495
        • Dente C.J.
        • Bradley M.
        • Schobel S.
        • et al.
        Towards precision medicine: Accurate predictive modeling of infectious complications in combat casualties.
        J Trauma Acute Care Surg. 2017; 83: 609-616
        • Bradley M.
        • Dente C.
        • Khatri V.
        • et al.
        Advanced modeling to predict pneumonia in combat trauma patients.
        World J Surg. 2020; 44: 2255-2262
        • Auslander N.
        • Gussow A.B.
        • Koonin E.V.
        Incorporating machine learning into established bioinformatics frameworks.
        Int J Mol Sci. 2021; 22: 2903
        • Leisman D.E.
        • Harhay M.O.
        • Lederer D.J.
        • et al.
        Development and reporting of prediction models: guidance for authors from editors of respiratory, sleep, and critical care journals.
        Crit Care Med. 2020; 48: 623-633
      1. Amico F, Efird JT, Briggs GD, Lott NJ, King KL, Hirani R, et al. Association between Blood Donor Demographics and Post-Injury Multiple Organ Failure after Polytrauma. Ann Surg. 2021. https://doi.org/10.1097/SLA.0000000000004754. Accessed September 9, 2022.

        • Perkins Z.B.
        • Yet B.
        • Sharrock A.
        • et al.
        Predicting the outcome of limb revascularization in patients with lower-extremity arterial trauma: development and external validation of a supervised machine-learning algorithm to support surgical decisions.
        Ann Surg. 2020; 272: 564-572
        • Munoz B.
        • Schobel S.A.
        • Lisboa F.A.
        • Khatri V.
        • Grey S.F.
        • Dente C.J.
        • et al.
        Clinical risk factors and inflammatory biomarkers of post-traumatic acute kidney injury in combat patients.
        Surgery. 2020; 168: 662-670
        • Gelbard R.B.
        • Hensman H.
        • Schobel S.
        • et al.
        An integrative model using flow cytometry identifies nosocomial infection after trauma.
        J Trauma Acute Care Surg. 2021; 91: 47-53
        • Gelbard R.B.
        • Hensman H.
        • Schobel S.
        • et al.
        Random forest modeling can predict infectious complications following trauma laparotomy.
        J Trauma Acute Care Surg. 2019; 87: 1125-1132
        • Udelsman R.
        • Ramp J.
        • Gallucci W.T.
        • et al.
        Adaptation during surgical stress. A reevaluation of the role of glucocorticoids.
        J Clin Invest. 1986; 77: 1377-1381
        • Skelton J.K.
        • Purcell R.
        Preclinical models for studying immune responses to traumatic injury.
        Immunology. 2021; 162: 377-388
        • Hazeldine J.
        • Hampson P.
        • Lord J.M.
        The diagnostic and prognostic value of systems biology research in major traumatic and thermal injury: a review.
        Burns Trauma. 2016; 4: 33
        • Pallister I.
        Current concepts of the inflammatory response after major trauma: an update.
        Injury. 2005; 36 (author reply 9–30): 227-229
        • Hazeldine J.
        • Naumann D.N.
        • Toman E.
        • et al.
        Prehospital immune responses and development of multiple organ dysfunction syndrome following traumatic injury: a prospective cohort study.
        PLoS Med. 2017; 14e1002338
        • Cabrera C.P.
        • Manson J.
        • Shepherd J.M.
        • et al.
        Signatures of inflammation and impending multiple organ dysfunction in the hyperacute phase of trauma: a prospective cohort study.
        PLoS Med. 2017; 14e1002352
        • Hofman M.
        • Andruszkow H.
        • Kobbe P.
        • Poeze M.
        • Hildebrand F.
        Incidence of post-traumatic pneumonia in poly-traumatized patients: identifying the role of traumatic brain injury and chest trauma. Eur.
        J Trauma Emerg Surg. 2020; 46: 11-19
        • Meagher A.D.
        • Lind M.
        • Senekjian L.
        • et al.
        Ventilator-associated events, not ventilator-associated pneumonia, is associated with higher mortality in trauma patients.
        J Trauma Acute Care Surg. 2019; 87: 307-314
        • Khwaja A.
        KDIGO clinical practice guidelines for acute kidney injury.
        Nephron Clin Pract. 2012; 120: c179-184
        • Hastie T.
        • Tibshirani R.
        • Friedman J.H.
        The Elements of Statistical Learning: Data Mining, Inference, and Prediction.
        Springer, New York (NY)2009
        • Fawcett T.
        An introduction to ROC analysis.
        Pattern Recogn Lett. 2006; 27: 861-874
        • Youden W.J.
        Index for rating diagnostic tests.
        Cancer. 1950; 3: 32-35
        • Collins G.S.
        • Reitsma J.B.
        • Altman D.G.
        • Moons K.G.
        Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): the TRIPOD statement.
        Ann Intern Med. 2015; 162: 55-63
        • Hwang S.Y.
        • Lee J.H.
        • Lee Y.H.
        • Hong C.K.
        • Sung A.J.
        • Choi Y.C.
        Comparison of the Sequential Organ Failure Assessment, Acute Physiology and Chronic Health Evaluation II scoring system, and Trauma and Injury Severity Score method for predicting the outcomes of intensive care unit trauma patients.
        Am J Emerg Med. 2012; 30: 749-753
        • Fueglistaler P.
        • Amsler F.
        • Schuepp M.
        • et al.
        Prognostic value of Sequential Organ Failure Assessment and Simplified Acute Physiology II Score compared with trauma scores in the outcome of multiple-trauma patients.
        Am J Surg. 2010; 200: 204-214
        • Gursel G.
        • Demirtas S.
        Value of APACHE II, SOFA and CPIS scores in predicting prognosis in patients with ventilator-associated pneumonia.
        Respiration. 2006; 73: 503-508
        • Kuwabara K.
        • Matsuda S.
        • Imanaka Y.
        • et al.
        Injury severity score, resource use, and outcome for trauma patients within a Japanese administrative database.
        J Trauma. 2010; 68: 463-470
        • Christensen M.C.
        • Nielsen T.G.
        • Ridley S.
        • Lecky F.E.
        • Morris S.
        Outcomes and costs of penetrating trauma injury in England and Wales.
        Injury. 2008; 39: 1013-1025
        • Stoecklein V.M.
        • Osuka A.
        • Lederer J.A.
        Trauma equals danger--damage control by the immune system.
        J Leukoc Biol. 2012; 92: 539-551
        • Mackenzie E.J.
        • Rivara F.P.
        • Jurkovich G.J.
        • et al.
        The National Study on Costs and Outcomes of Trauma.
        J Trauma. 2007; 63 (discussion S81–6): S54-S67
        • Ranson T.
        • Vosshenrich C.A.
        • Corcuff E.
        • Richard O.
        • Muller W.
        • Di Santo J.P.
        IL-15 is an essential mediator of peripheral NK-cell homeostasis.
        Blood. 2003; 101: 4887-4893
        • Souza-Fonseca-Guimaraes F.
        • Parlato M.
        • Fitting C.
        • Cavaillon J.M.
        • Adib-Conquy M.
        NK cell tolerance to TLR agonists mediated by regulatory T cells after polymicrobial sepsis.
        J Immunol. 2012; 188: 5850-5858
        • Jiang W.
        • Li X.
        • Wen M.
        • et al.
        Increased percentage of PD-L1(+) natural killer cells predicts poor prognosis in sepsis patients: a prospective observational cohort study.
        Crit Care. 2020; 24: 617
        • Gogos C.
        • Kotsaki A.
        • Pelekanou A.
        • et al.
        Early alterations of the innate and adaptive immune statuses in sepsis according to the type of underlying infection.
        Crit Care. 2010; 14: R96
        • Li Y.T.
        • Wang Y.C.
        • Lee H.L.
        • Tsao S.C.
        • Lu M.C.
        • Yang S.F.
        Monocyte chemoattractant protein-1, a possible biomarker of multiorgan failure and mortality in ventilator-associated pneumonia.
        Int J Mol Sci. 2019; 20: 2218
        • Dries D.J.
        • Jurkovich G.J.
        • Maier R.V.
        • et al.
        Effect of interferon gamma on infection-related death in patients with severe injuries. A randomized, double-blind, placebo-controlled trial.
        Arch Surg. 1994; 129 (discussion 42): 1031-1041
        • Shankar-Hari M.
        • Fear D.
        • Lavender P.
        • et al.
        Activation-associated accelerated apoptosis of memory B cells in critically ill patients with sepsis.
        Crit Care Med. 2017; 45: 875-882
        • Monserrat J.
        • de Pablo R.
        • Diaz-Martin D.
        • et al.
        Early alterations of B cells in patients with septic shock.
        Crit Care. 2013; 17: R105
        • Yang S.
        • Ding W.
        • Feng D.
        • et al.
        Loss of B cell regulatory function is associated with delayed healing in patients with tibia fracture.
        APMIS. 2015; 123: 975-985
        • Muire P.J.
        • Mangum L.H.
        • Wenke J.C.
        Time course of immune response and immunomodulation during normal and delayed healing of musculoskeletal wounds.
        Front Immunol. 2020; 11: 1056
        • O'Sullivan S.T.
        • Lederer J.A.
        • Horgan A.F.
        • Chin D.H.
        • Mannick J.A.
        • Rodrick M.L.
        Major injury leads to predominance of the T helper-2 lymphocyte phenotype and diminished interleukin-12 production associated with decreased resistance to infection.
        Ann Surg. 1995; 222 (discussion 90–2): 482-490
        • Lederer J.A.
        • Rodrick M.L.
        • Mannick J.A.
        The effects of injury on the adaptive immune response.
        Shock. 1999; 11: 153-159
        • Salazar-Mather T.P.
        • Lewis C.A.
        • Biron C.A.
        Type I interferons regulate inflammatory cell trafficking and macrophage inflammatory protein 1alpha delivery to the liver.
        J Clin Invest. 2002; 110: 321-330
        • Cagliani J.
        • Yang W.L.
        • McGinn J.T.
        • Wang Z.
        • Wang P.
        Anti-interferon-alpha receptor 1 antibodies attenuate inflammation and organ injury following hemorrhagic shock.
        J Trauma Acute Care Surg. 2019; 86: 881-890
        • Gentile L.F.
        • Cuenca A.G.
        • Efron P.A.
        • et al.
        Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care.
        J Trauma Acute Care Surg. 2012; 72: 1491-1501
        • Vanzant E.L.
        • Lopez C.M.
        • Ozrazgat-Baslanti T.
        • et al.
        Persistent inflammation, immunosuppression, and catabolism syndrome after severe blunt trauma.
        J Trauma Acute Care Surg. 2014; 76 (discussion 9–30): 21-29
        • Orr M.T.
        • Lanier L.L.
        Natural killer cell education and tolerance.
        Cell. 2010; 142: 847-856
        • Feng T.
        • Liao X.
        • Yang X.
        • et al.
        A shift toward inhibitory receptors and impaired effector functions on NK cells contribute to immunosuppression during sepsis.
        J Leukoc Biol. 2020; 107: 57-67
        • Callahan T.J.
        • Tripodi I.J.
        • Pielke-Lombardo H.
        • Hunter L.E.
        Knowledge-based biomedical data science.
        Annu Rev Biomed Data Sci. 2020; 3: 23-41
        • Haendel M.A.
        • Chute C.G.
        • Robinson P.N.
        Classification, ontology, and precision medicine.
        N Engl J Med. 2018; 379: 1452-1462
        • Hulsen T.
        • Jamuar S.S.
        • Moody A.R.
        • et al.
        From big data to precision medicine.
        Front Med (Lausanne). 2019; 6: 34
        • Buchman T.G.
        • Billiar T.R.
        • Elster E.
        • et al.
        Precision medicine for critical illness and injury.
        Crit Care Med. 2016; 44: 1635-1638
        • Belard A.
        • Buchman T.
        • Forsberg J.
        • et al.
        Precision diagnosis: a view of the clinical decision support systems (CDSS) landscape through the lens of critical care.
        J Clin Monit Comput. 2017; 31: 261-271
        • Nederpelt C.J.
        • Mokhtari A.K.
        • Alser O.
        • et al.
        Development of a field artificial intelligence triage tool: confidence in the prediction of shock, transfusion, and definitive surgical therapy in patients with truncal gunshot wounds.
        J Trauma Acute Care Surg. 2021; 90: 1054-1060