Prehospital continuous vital signs predict need for resuscitative endovascular balloon occlusion of the aorta and resuscitative thoracotomy prehospital continuous vital signs predict resuscitative endovascular balloon occlusion of the aorta.
Adult
Aorta
/ surgery
Balloon Occlusion
/ statistics & numerical data
Female
Humans
Injury Severity Score
Male
Middle Aged
Resuscitation
/ methods
Retrospective Studies
Risk Assessment
/ statistics & numerical data
Shock, Hemorrhagic
/ diagnosis
Thoracic Injuries
/ complications
Thoracotomy
/ statistics & numerical data
Triage
/ statistics & numerical data
Vital Signs
Journal
The journal of trauma and acute care surgery
ISSN: 2163-0763
Titre abrégé: J Trauma Acute Care Surg
Pays: United States
ID NLM: 101570622
Informations de publication
Date de publication:
01 11 2021
01 11 2021
Historique:
pubmed:
3
4
2021
medline:
11
11
2021
entrez:
2
4
2021
Statut:
ppublish
Résumé
Rapid triage and intervention to control hemorrhage are key to survival following traumatic injury. Patients presenting in hemorrhagic shock may undergo resuscitative thoracotomy (RT) or resuscitative endovascular balloon occlusion of the aorta (REBOA) as adjuncts to rapidly control bleeding. We hypothesized that machine learning along with automated calculation of continuously measured vital signs in the prehospital setting would accurately predict need for REBOA/RT and inform rapid lifesaving decisions. Prehospital and admission data from 1,396 patients transported from the scene of injury to a Level I trauma center via helicopter were analyzed. Utilizing machine learning and prehospital autonomous vital signs, a Bleeding Risk Index (BRI) based on features from pulse oximetry and electrocardiography waveforms and blood pressure (BP) trends was calculated. Demographics, Injury Severity Score and BRI were compared using Mann-Whitney-Wilcox test. Area under the receiver operating characteristic curve (AUC) was calculated and AUC of different scores compared using DeLong's method. Of the 1,396 patients, median age was 45 years and 68% were men. Patients who underwent REBOA/RT were more likely to have a penetrating injury (24% vs. 7%, p < 0.001), higher Injury Severity Score (25 vs. 10, p < 0.001) and higher mortality (44% vs. 7%, p < 0.001). Prehospital they had lower BP (96 [70-130] vs. 134 [117-152], p < 0.001) and higher heart rate (106 [82-118] vs. 90 [76-106], p < 0.001). Bleeding risk index calculated using the entire prehospital period was 10× higher in patients undergoing REBOA/RT (0.5 [0.42-0.63] vs. 0.05 [0.02-0.21], p < 0.001) with an AUC of 0.93 (95% confidence interval [95% CI], 0.90-0.97). This was similarly predictive when calculated from shorter periods of transport: BRI initial 10 minutes prehospital AUC of 0.89 (95% CI, 0.83-0.94) and initial 5 minutes AUC of 0.90 (95% CI, 0.85-0.94). Automated prehospital calculations based on vital sign features and trends accurately predict the need for the emergent REBOA/RT. This information can provide essential time for team preparedness and guide trauma triage and disaster management. Therapeutic/care management, Level IV.
Sections du résumé
BACKGROUND
Rapid triage and intervention to control hemorrhage are key to survival following traumatic injury. Patients presenting in hemorrhagic shock may undergo resuscitative thoracotomy (RT) or resuscitative endovascular balloon occlusion of the aorta (REBOA) as adjuncts to rapidly control bleeding. We hypothesized that machine learning along with automated calculation of continuously measured vital signs in the prehospital setting would accurately predict need for REBOA/RT and inform rapid lifesaving decisions.
METHODS
Prehospital and admission data from 1,396 patients transported from the scene of injury to a Level I trauma center via helicopter were analyzed. Utilizing machine learning and prehospital autonomous vital signs, a Bleeding Risk Index (BRI) based on features from pulse oximetry and electrocardiography waveforms and blood pressure (BP) trends was calculated. Demographics, Injury Severity Score and BRI were compared using Mann-Whitney-Wilcox test. Area under the receiver operating characteristic curve (AUC) was calculated and AUC of different scores compared using DeLong's method.
RESULTS
Of the 1,396 patients, median age was 45 years and 68% were men. Patients who underwent REBOA/RT were more likely to have a penetrating injury (24% vs. 7%, p < 0.001), higher Injury Severity Score (25 vs. 10, p < 0.001) and higher mortality (44% vs. 7%, p < 0.001). Prehospital they had lower BP (96 [70-130] vs. 134 [117-152], p < 0.001) and higher heart rate (106 [82-118] vs. 90 [76-106], p < 0.001). Bleeding risk index calculated using the entire prehospital period was 10× higher in patients undergoing REBOA/RT (0.5 [0.42-0.63] vs. 0.05 [0.02-0.21], p < 0.001) with an AUC of 0.93 (95% confidence interval [95% CI], 0.90-0.97). This was similarly predictive when calculated from shorter periods of transport: BRI initial 10 minutes prehospital AUC of 0.89 (95% CI, 0.83-0.94) and initial 5 minutes AUC of 0.90 (95% CI, 0.85-0.94).
CONCLUSION
Automated prehospital calculations based on vital sign features and trends accurately predict the need for the emergent REBOA/RT. This information can provide essential time for team preparedness and guide trauma triage and disaster management.
LEVEL OF EVIDENCE
Therapeutic/care management, Level IV.
Identifiants
pubmed: 33797486
doi: 10.1097/TA.0000000000003171
pii: 01586154-202111000-00004
doi:
Types de publication
Journal Article
Research Support, U.S. Gov't, Non-P.H.S.
Langues
eng
Sous-ensembles de citation
IM
Pagination
798-802Informations de copyright
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
Références
Centers for Disease Control and Prevention. WISQARS leading causes of death reports. Available at https://webappa.cdc.gov/sasweb/ncipc/leadcause.html . Updated February 20, 2020. Accessed December 5, 2020.
Pfeifer R, Tarkin IS, Rocos B, Pape HC. Patterns of mortality and causes of death in polytrauma patients—has anything changed? Injury . 2009;40(9):907–911.
Callcut RA, Kornblith LZ, Conroy AS, et al. The why and how our trauma patients die: a prospective Multicenter Western Trauma Association study. J Trauma Acute Care Surg . 2019;86(5):864–870.
Rhee PM, Acosta J, Bridgeman A, Wang D, Jordan M, Rich N. Survival after emergency department thoracotomy: review of published data from the past 25 years. J Am Coll Surg . 2000;190(3):288–298.
Brenner M, Moore L, Dubose J, Scalea T. Resuscitative endovascular balloon occlusion of the aorta (REBOA) for use in temporizing intra-abdominal and pelvic hemorrhage: physiologic sequelae and considerations. Shock . 2020;54(5):615–622.
Brenner M, Inaba K, Aiolfi A, et al. Resuscitative endovascular balloon occlusion of the aorta and resuscitative thoracotomy in select patients with hemorrhagic shock: early results from the American Association for the Surgery of Trauma’s Aortic Occlusion in Resuscitation for Trauma and Acute Care Surgery Registry. J Am Coll Surg . 2018;226(5):730–740.
Moore LJ, Fox EE, Meyer DE, et al. Prospective observational evaluation of the ER-REBOA catheter at 6 U.S. trauma centers. Ann Surg . 2020; Publish Ah(Xx):1–7.
Yang S, Mackenzie CF, Rock P, Lin C, Floccare D, Scalea T, Stumpf F, Winans C, Galvagno S, Miller C, Stein D, Hu PF. Comparison of massive and emergency transfusion prediction scoring systems after trauma with a new Bleeding Risk Index score applied in-flight. J Trauma Acute Care Surg . 2021;90:268–273.
Hu P, Galvagno SM Jr., Sen A, Dutton R, Jordan S, Floccare D, Handley C, Shackelford S, Pasley J, MacKenzie C. Identification of dynamic prehospital changes with continuous vital signs acquisition. Air Med J . 2014;33(1):27–33.
Woodford MR, Mackenzie CF, DuBose J, Hu P, Kufera J, Hu EZ, Dutton RP, Scalea TM. Continuously recorded oxygen saturation and heart rate during prehospital transport outperform initial measurement in prediction of mortality after trauma. J Trauma Acute Care Surg . 2012;72(4):1006–1012.
Mackenzie CF, Wang Y, Hu PF, et al. Automated prediction of early blood transfusion and mortality in trauma patients. J Trauma Acute Care Surg . 2014;76(6):1379–1385.
Shackelford S, Yang S, Hu P, et al. Predicting blood transfusion using automated analysis of pulse oximetry signals and laboratory values. J Trauma Acute Care Surg . 2015;79(4):S175–S180.
Parimi N, Hu PF, MacKenzie CF, Yang S, Bartlett ST, Scalea TM, Stein DM. Automated continuous vital signs predict use of uncrossed matched blood and massive transfusion following trauma. J Trauma Acute Care Surg . 2016;80(6):897–906.
Liu NT, Holcomb JB, Wade CE, Darrah MI, Salinas J. Utility of vital signs, heart rate variability and complexity, and machine learning for identifying the need for lifesaving interventions in trauma patients. Shock . 2014;42(2):108–114.
Guyette FX, Gomez H, Suffoletto B, Quintero J, Mesquida J, Kim HK, Hostler D, Puyana JC, Pinsky MR. Prehospital dynamic tissue oxygen saturation response predicts in-hospital lifesaving interventions in trauma patients. J Trauma Acute Care Surg . 2012;72(4):930–935.
Reljin N, Zimmer G, Malyuta Y, Mendelson Y, Darling CE, Chon KH. Detection of blood loss in trauma patients using time-frequency analysis of photoplethysmographic signal. 3rd IEEE EMBS Int Conf Biomed Heal Informatics, BHI 2016 . 2016;(Iv):118–121.
MacKenzie CF, Gao C, Hu PF, et al. Comparison of decision-assist and clinical judgment of experts for prediction of lifesaving interventions. Shock . 2015;43(3):238–243.
Yang S, Hu P, Mackenzie CF. Trauma Outcome Prediction in the Era of Big Data: From Data Collection to Analytics. In: Signal Processing and Machine Learning for Biomedical Big Data . CRC Press; 2018.
Shackelford SA, Colton K, Stansbury LG, Galvagno SM, Anazodo AN, Dubose JJ, Hess JR, Mackenzie CF. Early identification of uncontrolled hemorrhage after trauma: current status and future direction. J Trauma Acute Care Surg . 2014;77(3 Suppl 2):222–227.
Mutschler M, Nienaber U, Münzberg M, Wölfl C, Schoechl H, Paffrath T, Bouillon B, Maegele M. The Shock Index revisited - a fast guide to transfusion requirement? A retrospective analysis on 21,853 patients derived from the TraumaRegister DGU®. Crit Care . 2013;17(4):R172.