Matthew Bradley1,2,3, Christopher Dente4,5,3, Vivek Khatri1,3,6, Seth Schobel1,3,6, Felipe Lisboa1,3,6, Audrey Shi3,7, Hannah Hensman3,7, Allan Kirk8,3, Timothy G Buchman4,7, Eric Elster9,10. 1. Department of Surgery, Uniformed Services University of the Health Sciences and Walter Reed National Military Medical Center, 4301 Jones Bridge Road, Bethesda, MD, 20814, USA. 2. Department of Regenerative Medicine, Naval Medical Research Center, Silver Spring, MD, USA. 3. Surgical Critical Care Initiative (SC2i), Bethesda, MD, USA. 4. Emory University, Atlanta, GA, USA. 5. Grady Memorial Hospital, Atlanta, GA, USA. 6. Henry M. Jackson Foundation for the Advancement of Military Sciences, Bethesda, MD, USA. 7. DecisionQ Corporation, Arlington, VA, USA. 8. Duke University, Durham, NC, USA. 9. Department of Surgery, Uniformed Services University of the Health Sciences and Walter Reed National Military Medical Center, 4301 Jones Bridge Road, Bethesda, MD, 20814, USA. eric.elster@usuhs.edu. 10. Surgical Critical Care Initiative (SC2i), Bethesda, MD, USA. eric.elster@usuhs.edu.
Abstract
BACKGROUND: Tools to assist clinicians in predicting pneumonia could lead to a significant decline in morbidity. Therefore, we sought to develop a model in combat trauma patients for identifying those at highest risk of pneumonia. METHODS: This was a retrospective study of 73 primarily blast-injured casualties with combat extremity wounds. Binary classification models for pneumonia prediction were developed with measurements of injury severity from the Abbreviated Injury Scale (AIS), transfusion blood products received before arrival at Walter Reed National Military Medical Center (WRNMMC), and serum protein levels. Predictive models were generated with leave-one-out-cross-validation using the variable selection method of backward elimination (BE) and the machine learning algorithms of random forests (RF) and logistic regression (LR). BE was attempted with two predictor sets: (1) all variables and (2) serum proteins alone. RESULTS: Incidence of pneumonia was 12% (n = 9). Different variable sets were produced by BE when considering all variables and just serum proteins alone. BE selected the variables ISS, AIS chest, and cryoprecipitate within the first 24 h following injury for the first predictor set 1 and FGF-basic, IL-2R, and IL-6 for predictor set 2. Using both variable sets, a RF was generated with AUCs of 0.95 and 0.87-both higher than LR algorithms. CONCLUSION: Advanced modeling allowed for the identification of clinical and biomarker data predictive of pneumonia in a cohort of predominantly blast-injured combat trauma patients. The generalizability of the models developed here will require an external validation dataset.
BACKGROUND: Tools to assist clinicians in predicting pneumonia could lead to a significant decline in morbidity. Therefore, we sought to develop a model in combat traumapatients for identifying those at highest risk of pneumonia. METHODS: This was a retrospective study of 73 primarily blast-injured casualties with combat extremity wounds. Binary classification models for pneumonia prediction were developed with measurements of injury severity from the Abbreviated Injury Scale (AIS), transfusion blood products received before arrival at Walter Reed National Military Medical Center (WRNMMC), and serum protein levels. Predictive models were generated with leave-one-out-cross-validation using the variable selection method of backward elimination (BE) and the machine learning algorithms of random forests (RF) and logistic regression (LR). BE was attempted with two predictor sets: (1) all variables and (2) serum proteins alone. RESULTS: Incidence of pneumonia was 12% (n = 9). Different variable sets were produced by BE when considering all variables and just serum proteins alone. BE selected the variables ISS, AIS chest, and cryoprecipitate within the first 24 h following injury for the first predictor set 1 and FGF-basic, IL-2R, and IL-6 for predictor set 2. Using both variable sets, a RF was generated with AUCs of 0.95 and 0.87-both higher than LR algorithms. CONCLUSION: Advanced modeling allowed for the identification of clinical and biomarker data predictive of pneumonia in a cohort of predominantly blast-injured combat traumapatients. The generalizability of the models developed here will require an external validation dataset.
Authors: Diego A Vicente; Seth A Schobel; Simone Anfossi; Hannah Hensman; Felipe Lisboa; Henry Robertson; Vivek Khatri; Matthew J Bradley; Masayoshi Shimizu; Timothy G Buchman; Thomas A Davis; Christopher J Dente; Allan D Kirk; George A Calin; Eric A Elster Journal: J Trauma Acute Care Surg Date: 2022-07-07 Impact factor: 3.697
Authors: Jennifer M Leonard; Christina X Zhang; Liang Lu; Mark H Hoofnagle; Anja Fuchs; Regina A Clemens; Sarbani Ghosh; Shin-Wen Hughes; Grant V Bochicchio; Richard Hotchkiss; Isaiah R Turnbull Journal: J Trauma Acute Care Surg Date: 2021-06-01 Impact factor: 3.697
Authors: Charles J Gerardo; Elizabeth Silvius; Seth Schobel; John C Eppensteiner; Lauren M McGowan; Eric A Elster; Allan D Kirk; Alexander T Limkakeng Journal: Front Immunol Date: 2021-03-15 Impact factor: 7.561
Authors: Aram Avila-Herrera; James B Thissen; Nisha Mulakken; Seth A Schobel; Michael D Morrison; Xiner Zhou; Scott F Grey; Felipe A Lisboa; Desiree Unselt; Shalini Mabery; Meenu M Upadhyay; Crystal J Jaing; Eric A Elster; Nicholas A Be Journal: Sci Rep Date: 2022-08-15 Impact factor: 4.996