J-S David1,2, E-J Voiglio2,3, E Cesareo4,5, O Vassal1,2, E Decullier6,7, P-Y Gueugniaud4,5, S Peyrefitte8, K Tazarourte4,5. 1. Department of Anesthesiology and Critical Care Medicine, Hospices Civils de Lyon (HCL), Lyon-Sud Hospital, Pierre Bénite, France. 2. Lyon Est School of Medicine, University Lyon 1, Lyon, France. 3. Department of Surgery, Hospices Civils de Lyon, Lyon-Sud Hospital, Pierre Bénite, France. 4. SAMU de Lyon and Department of Emergency Medicine, Hospices Civils de Lyon, Edouard Herriot Hospital, Lyon, France. 5. Lyon Sud School of Medicine, University Lyon 1, Oullins, France. 6. Pôle Information Médicale Evaluation Recherche, Hospices Civils de Lyon, Lyon, France. 7. EA Santé Individu Société, Université de Lyon, Lyon, France. 8. Antenne Médicale Spécialisée, Base des Fusiliers Marins et des Commandos, Lanester, France.
Abstract
BACKGROUND: This study aimed to evaluate the accuracy of prehospital parameters, including vital signs and resuscitation (fluids, vasopressor), to predict trauma-induced coagulopathy (TIC, fibrinogen <1·5 g/l or PTratio > 1·5 or platelet count <100 × 109 /l), and a massive transfusion (MT, ≥10 RBC units within the first 24 h). METHODS: From a trauma registry (2011-2015), in which patients are prospectively included, we retrospectively retrieved the heart rate (HR), systolic blood pressure (SBP), volume of prehospital fluids and administration of noradrenaline. We calculated the shock index (SI: HR/SBP), the MGAP prehospital triage score and the Injury Severity Score (ISS). We also identified patients who had positive criteria from the Resuscitation Outcome Consortium (ROC, SBP < 70 mmHg or SBP 70-90 and HR > 107 pulse/min). For these parameters, we drew a ROC curve and defined a cut-off value to predict TIC or MT. The strength of association between prehospital parameters and TIC as well as MT was assessed using logistic regression, and cut-off values were determined using ROC curves. RESULTS: Among the 485 patients included in the study, TIC was observed in 112 patients (23%) and MT in 22 patients (5%). For the prediction of TIC, ISS had good accuracy (AUC: 0·844, 95% confidence interval, CI: 0·799-0·879), as did the volume of fluids (>1000 ml) given during prehospital care (AUC: 0·801, 95% CI: 0·752-0·842). For the prediction of MT, ISS had excellent accuracy (AUC: 0·932, 95% CI: 0·866-0·966), whereas good accuracy was found for SI (> 0·9; AUC: 0·859, 95% CI: 0·705-0·936), vasopressor administration (AUC: 0·828, 95% CI: 0·736-0·890) and fluids (>1000 ml; AUC: 0·811, 95% CI: 0·737-0·867). Vasopressor administration, ISS and SI were independent predictors of TIC and MT, whereas fluid volume and ROC criteria were independent predictor of TIC but not MT. No independent relationship was found between MGAP and TIC or MT. CONCLUSIONS: Prehospital parameters including the SI and resuscitation may help to better identify the severity of bleeding in trauma patients and the need for blood product administration at admission.
BACKGROUND: This study aimed to evaluate the accuracy of prehospital parameters, including vital signs and resuscitation (fluids, vasopressor), to predict trauma-induced coagulopathy (TIC, fibrinogen <1·5 g/l or PTratio > 1·5 or platelet count <100 × 109 /l), and a massive transfusion (MT, ≥10 RBC units within the first 24 h). METHODS: From a trauma registry (2011-2015), in which patients are prospectively included, we retrospectively retrieved the heart rate (HR), systolic blood pressure (SBP), volume of prehospital fluids and administration of noradrenaline. We calculated the shock index (SI: HR/SBP), the MGAP prehospital triage score and the Injury Severity Score (ISS). We also identified patients who had positive criteria from the Resuscitation Outcome Consortium (ROC, SBP < 70 mmHg or SBP 70-90 and HR > 107 pulse/min). For these parameters, we drew a ROC curve and defined a cut-off value to predict TIC or MT. The strength of association between prehospital parameters and TIC as well as MT was assessed using logistic regression, and cut-off values were determined using ROC curves. RESULTS: Among the 485 patients included in the study, TIC was observed in 112 patients (23%) and MT in 22 patients (5%). For the prediction of TIC, ISS had good accuracy (AUC: 0·844, 95% confidence interval, CI: 0·799-0·879), as did the volume of fluids (>1000 ml) given during prehospital care (AUC: 0·801, 95% CI: 0·752-0·842). For the prediction of MT, ISS had excellent accuracy (AUC: 0·932, 95% CI: 0·866-0·966), whereas good accuracy was found for SI (> 0·9; AUC: 0·859, 95% CI: 0·705-0·936), vasopressor administration (AUC: 0·828, 95% CI: 0·736-0·890) and fluids (>1000 ml; AUC: 0·811, 95% CI: 0·737-0·867). Vasopressor administration, ISS and SI were independent predictors of TIC and MT, whereas fluid volume and ROC criteria were independent predictor of TIC but not MT. No independent relationship was found between MGAP and TIC or MT. CONCLUSIONS: Prehospital parameters including the SI and resuscitation may help to better identify the severity of bleeding in traumapatients and the need for blood product administration at admission.
Authors: David N Naumann; James M Hancox; James Raitt; Iain M Smith; Nicholas Crombie; Heidi Doughty; Gavin D Perkins; Mark J Midwinter Journal: BMJ Open Date: 2018-01-23 Impact factor: 2.692
Authors: Mathieu Hylands; Augustin Toma; Nicolas Beaudoin; Anne Julie Frenette; Frédérick D'Aragon; Émilie Belley-Côté; Emmanuel Charbonney; Morten Hylander Møller; Jon Henrik Laake; Per Olav Vandvik; Reed Alexander Siemieniuk; Bram Rochwerg; François Lauzier; Robert S Green; Ian Ball; Damon Scales; Srinivas Murthy; Joey S W Kwong; Gordon Guyatt; Sandro Rizoli; Pierre Asfar; François Lamontagne Journal: BMJ Open Date: 2017-11-17 Impact factor: 2.692