¿Podrían ayudarnos los algoritmos de machine learning en la predicción de hemorragia masiva a nivel prehospitalario? / Could machine learning algorithms help us in predicting massive hemorrhage at the prehospital level?

Objetivo: Comparación de la capacidad predictiva de diferentes algoritmos de machine learning (AML) respecto a escalas tradicionales de predicción de hemorragia masiva en pacientes con enfermedad traumática grave (ETG). Diseño: Sobre una base de datos de una cohorte retrospectiva con variables clínicas prehospitalarias y de resultado de hemorragia masiva se realizó un tratamiento de la base de datos para poder aplicar los AML, obteniéndose un conjunto total de 473 pacientes (80% entrenamiento, 20% validación). Para la modelización se realizó imputación proporcional y validación cruzada. El poder predictivo se evaluó con la métrica ROC y la importancia de las variables mediante los valores Shapley. Ámbito: Atención extrahospitalaria del paciente con ETG. Pacientes: Pacientes con ETG atendidos en el medio extrahospitalario por un servicio médico extrahospitalario desde enero de 2010 hasta diciembre de 2015 y trasladados a un centro de trauma en Madrid. Intervenciones: Ninguna. Variables de interés principales: Obtención y comparación de la métrica ROC de 4 AML: random forest, support vector machine, gradient boosting machine y neural network con los resultados obtenidos con escalas tradicionales de predicción. Resultados: Los diferentes AML alcanzaron valores ROC superiores al 0,85, teniendo medianas cercanas a 0,98. No encontramos diferencias significativas entre los AML. Cada AML ofrece un conjunto de variables diferentes, pero con predominancia de las variables hemodinámicas, de reanimación y de deterioro neurológico. Conclusiones: Los AML podrían superar a las escalas tradicionales de predicción en la predicción de hemorragia masiva. (AU)

Objective: Comparison of the predictive ability of various machine learning algorithms (MLA) versus traditional prediction scales for massive hemorrhage in patients with severe traumatic injury (ETG). Design: On a database of a retrospective cohort with prehospital clinical variables and massive hemorrhage outcome, a treatment of the database was performed to be able to apply the different MLA, obtaining a total set of 473 patients (80% training and 20% validation). For modeling, proportional imputation and cross validation were performed. The predictive power was evaluated with the ROC metric and the importance of the variables using the Shapley values. Setting: Out-of-hospital care of patients with ETG. Participants: Patients with ETG treated out-of-hospital by a prehospital medical service from January 2010 to December 2015 and transferred to a trauma center in Madrid. Interventions: None. Main variables of interest: Obtaining and comparing the ROC curve metric of 4 MLAs: random forest, support vector machine, gradient boosting machine and neural network with the results obtained with traditional prediction scales. Results: The different MLA reached ROC values higher than 0.85, having medians close to 0.98. We found no significant differences between MLAs. Each MLA offers a different set of more important variables with a predominance of hemodynamic, resuscitation variables and neurological impairment. Conclusions: MLA may be helpful in patients with massive hemorrhage by outperforming traditional prediction scales. (AU)

Documento multidisciplinar de consenso sobre el manejo de la hemorragia masiva. Primera actualización 2023 (documento HEMOMAS-II) / Multidisciplinary consensus document on the management of massive haemorrhage. First update 2023 (document HEMOMAS-II)

El presente documento supone una puesta al día del documento multidisciplinar HEMOMAS, publicado en el año 2016 con el aval de las Sociedades Científicas Españolas de Anestesiología y Reanimación (SEDAR), Medicina Intensiva, Crítica y Unidades Coronarias (SEMICYUC) y de Trombosis y Hemostasia (SETH). El objetivo de este documento fue revisar y actualizar las recomendaciones existentes sobre el manejo de la hemorragia masiva (HM). Se siguió una metodología basada en elementos del método ADAPTE (búsqueda y adaptación de guías publicadas en el ámbito específico de la HM desde 2014, más búsqueda bibliográfica en PubMed y EMBASE desde enero-2014 hasta junio-2021). Tras la revisión de nueve guías y 207 artículos seleccionados, se actualizaron las 47 recomendaciones existentes en el artículo original, manteniendo, suprimiendo o modificando cada una de ellas y sus grados de recomendación y evidencia. Consensuadamente, los autores aprobaron la redacción final del artículo y las 41 recomendaciones resultantes.(AU)

This document is an update of the multidisciplinary document HEMOMAS, published in 2016 with the endorsement of the Spanish Scientific Societies of Anaesthesiology (SEDAR), Intensive Care (SEMICYUC) and Thrombosis and Haemostasis (SETH). The aim of this document was to review and update existing recommendations on the management of massive haemorrhage. The methodology of the update was based on several elements of the ADAPTE method by searching and adapting guidelines published in the specific field of massive bleeding since 2014, plus a literature search performed in PubMed and EMBASE from January 2014 to June 2021. Based on the review of 9 guidelines and 207 selected articles, the 47 recommendations in the original article were reviewed, maintaining, deleting, or modifying each of them and the accompanying grades of recommendation and evidence. Following a consensus process, the final wording of the article and the resulting 41 recommendations were approved by all authors.(AU)

Multidisciplinary consensus document on the management of massive haemorrhage. First update 2023 (document HEMOMAS-II).

This document is an update of the multidisciplinary document HEMOMAS, published in 2016 with the endorsement of the Spanish Scientific Societies of Anaesthesiology (SEDAR), Intensive Care (SEMICYUC) and Thrombosis and Haemostasis (SETH). The aim of this document was to review and update existing recommendations on the management of massive haemorrhage. The methodology of the update was based on several elements of the ADAPTE method by searching and adapting guidelines published in the specific field of massive bleeding since 2014, plus a literature search performed in PubMed and EMBASE from January 2014 to June 2021. Based on the review of 9 guidelines and 207 selected articles, the 47 recommendations in the original article were reviewed, maintaining, deleting, or modifying each of them and the accompanying grades of recommendation and evidence. Following a consensus process, the final wording of the article and the resulting 41 recommendations were approved by all authors.

Multidisciplinary consensus document on the management of massive haemorrhage. First update 2023 (document HEMOMAS-II).

This document is an update of the multidisciplinary document HEMOMAS, published in 2016 with the endorsement of the Spanish Scientific Societies of Anaesthesiology (SEDAR), Intensive Care (SEMICYUC) and Thrombosis and Haemostasis (SETH). The aim of this document was to review and update existing recommendations on the management of massive haemorrhage. The methodology of the update was based on several elements of the ADAPTE method by searching and adapting guidelines published in the specific field of massive bleeding since 2014, plus a literature search performed in PubMed and EMBASE from January 2014 to June 2021. Based on the review of 9 guidelines and 207 selected articles, the 47 recommendations in the original article were reviewed, maintaining, deleting, or modifying each of them and the accompanying grades of recommendation and evidence. Following a consensus process, the final wording of the article and the resulting 41 recommendations were approved by all authors.

Documento multidisciplinar de consenso sobre el manejo de la hemorragia masiva. Primera actualización 2023 (documento HEMOMAS-II) / Multidisciplinary consensus document on the management of massive haemorrhage. First update 2023 (document HEMOMAS-II)

El presente documento supone una puesta al día del documento multidisciplinar HEMOMAS, publicado en el año 2016 con el aval de las Sociedades Científicas Españolas de Anestesiología y Reanimación (SEDAR), Medicina Intensiva, Crítica y Unidades Coronarias (SEMICYUC) y de Trombosis y Hemostasia (SETH). El objetivo de este documento fue revisar y actualizar las recomendaciones existentes sobre el manejo de la hemorragia masiva (HM). Se siguió una metodología basada en elementos del método ADAPTE (búsqueda y adaptación de guías publicadas en el ámbito específico de la HM desde 2014, más búsqueda bibliográfica en PubMed y EMBASE desde enero-2014 hasta junio-2021). Tras la revisión de nueve guías y 207 artículos seleccionados, se actualizaron las 47 recomendaciones existentes en el artículo original, manteniendo, suprimiendo o modificando cada una de ellas y sus grados de recomendación y evidencia. Consensuadamente, los autores aprobaron la redacción final del artículo y las 41 recomendaciones resultantes (AU)

This document is an update of the multidisciplinary document HEMOMAS, published in 2016 with the endorsement of the Spanish Scientific Societies of Anaesthesiology (SEDAR), Intensive Care (SEMICYUC) and Thrombosis and Haemostasis (SETH). The aim of this document was to review and update existing recommendations on the management of massive haemorrhage. The methodology of the update was based on several elements of the ADAPTE method by searching and adapting guidelines published in the specific field of massive bleeding since 2014, plus a literature search performed in PubMed and EMBASE from January 2014 to June 2021. Based on the review of 9 guidelines and 207 selected articles, the 47 recommendations in the original article were reviewed, maintaining, deleting, or modifying each of them and the accompanying grades of recommendation and evidence. Following a consensus process, the final wording of the article and the resulting 41 recommendations were approved by all authors (AU)

Could machine learning algorithms help us predict massive bleeding at prehospital level?

OBJECTIVE: Comparison of the predictive ability of various machine learning algorithms (MLA) versus traditional prediction scales (TPS) for massive hemorrhage (MH) in patients with severe traumatic injury (STI). DESIGN: On a database of a retrospective cohort with prehospital clinical variables and MH outcome, a treatment of the database was performed to be able to apply the different AML, obtaining a total set of 473 patients (80% training, 20% validation). For modeling, proportional imputation and cross validation were performed. The predictive power was evaluated with the ROC metric and the importance of the variables using the Shapley values. SETTING: Out-of-hospital care of patients with STI. PARTICIPANTS: Patients with STI treated out-of-hospital by a out-of-hospital medical service from January 2010 to December 2015 and transferred to a trauma center in Madrid. INTERVENTIONS: None. MAIN VARIABLES OF INTEREST: Obtaining and comparing the "Receiver Operating Characteristic curve" (ROC curve) metric of four MLAs: "random forest" (RF), "vector support machine" (SVM), "gradient boosting machine" (GBM) and "neural network" (NN) with the results obtained with TPS. RESULTS: The different AML reached ROC values higher than 0.85, having medians close to 0.98. We found no significant differences between AMLs. Each AML offers a different set of more important variables with a predominance of hemodynamic, resuscitation variables and neurological impairment. CONCLUSIONS: MLA may be helpful in patients with HM by outperforming TPS.

Prehospital and hospital shock indices as predictors of massive blood transfusion during the initial treatment of polytrauma patients. / Índices de shock prehospitalario y hospitalario como predictores de transfusión masiva en la atención inicial del paciente politraumático.

OBJECTIVES: To explore a possible association between the shock index and a need for massive blood transfusion, duration of hospital stay in the critical care unit, and mortality. MATERIAL AND METHODS: Observational study of data for all patients over the age of 18 years with multiple high-energy injuries included in the TraumCat Registry who were treated in Hospital Universitario de Bellvitge between 2012 and 2016. We calculated shock index values before hospital emergency department arrival, on arrival at the hospital, and on admission to the critical care unit for resuscitation. The amount of blood transfused in the first 24 hours was also obtained from the registry. RESULTS: Of 184 polytrauma patients, 75 (41%) received blood transfusions. Median (interquartile range) shock indices were as follows: prehospital, 0.77 (0.61-1.01); on hospital arrival, 0.78 (0.64-1); and on critical care admission, 0.92 (0.76-1.13). Forty-six patients (25%) died. A prehospital shock index of 0.9 was significant, differentiating the amount of blood transfused. The specificity and sensitivity of the cutoff were 73% and 66%, respectively, at the prehospital recording and 74% and 80% on hospital arrival. The areas under the receiver operating characteristic curve and 95% CIs were as follows for prehospital and on-arrival shock indices: 68% (61%- 75%) and 72% (65%-79%). Mortality and hospital stay were not significantly associated with shock indices. CONCLUSION: The shock index is a useful, easy-to-obtain predictor to identify polytrauma patients who need early blood transfusion for optimal treatment. Hospital stay and mortality might be better predicted by other indicators.

OBJETIVO: Establecer la posible relación entre el Índice de Shock (IS) con los requerimientos de transfusión masiva, estancia hospitalaria y en unidad de críticos, y mortalidad. METODO: Estudio observacional de los pacientes mayores de 18 años con traumatismos de alta energía del registro TraumCat atendidos en el Hospital Universitario de Bellvitge entre 2012 y 2016. Se recogió el IS prehospitalario (PH), a la llegada al hospital (H) y en la unidad de reanimación (IS-C), y la cantidad de transfusión las primeras 24 horas. RESULTADOS: Se recogieron 184 pacientes y 75 (41%) recibieron transfusión sanguínea. Las medianas de los IS para todos los pacientes del estudio fueron: IS-PH 0,77 (Q1-Q3; 0,61-1,01), IS-H 0,78 (Q1-Q3; 0,64-1), IS-C 0,92 (Q1-Q3; 0,76-1,13). Fallecieron 46 pacientes (25%). El IS-PH y el IS-H fueron los que diferenciaron de manera significativa la cantidad de transfusión. El valor 0,9 mostró una especificidad/sensibilidad del 73%/66% para el IS-PH y del 74%/80% para el IS-H. El área bajo la curva ROC para el IS-PH y el IS-H fue del 68% (IC 95% 61-75) y del 72% (IC 95% 65-79) respectivamente. No hubo relación significativa de los IS con la mortalidad y la estancia hospitalaria. CONCLUSIONES: El IS es una herramienta útil y accesible para identificar pacientes politraumatizados con requerimientos transfusionales de manera temprana y optimizar el tratamiento. Para evaluar estancias hospitalarias o mortalidad, podrían ser más útiles otros índices.

Prediction of massive bleeding in a prehospital setting: validation of six scoring systems. / Predicción de hemorragia masiva a nivel extrahospitalario: validación de seis escalas.

OBJECTIVE: To validate the diagnostic ability of six different scores to predict massive bleeding in a prehospital setting. DESIGN: Retrospective cohort. SETTING: Prehospital attention of patients with severe trauma. SUBJECTS: Subjects with more than 15 years, a history of severe trauma (defined by code 15 criteria), that were initially assisted in a prehospital setting by the emergency services between January 2010 and December 2015 and were then transferred to a level one trauma center in Madrid. VARIABLES: To validate: 1. Trauma Associated Severe Haemorrhage Score. 2. Assessment of Blood Consumption Score. 3. Emergency Transfusión Score. 4. Índice de Shock. 5. Prince of Wales Hospital/Rainer Score. 6. Larson Score. RESULTS: 548 subjects were studied, 76,8% (420) were male, median age was 38 (interquartile range [IQR]: 27-50). Injury Severity Score was 18 (IQR: 9-29). Blunt trauma represented 82,5% (452) of the cases. Overall, frequency of MB was 9,2% (48), median intensive care unit admission days was 2,1 (IQR: 0,8 - 6,2) and hospital mortality rate was 11,2% (59). Emergency Transfusión Score had the highest precisions (AUC 0,85), followed by Trauma Associated Severe Haemorrhage score and Prince of Wales Hospital/Rainer Score (AUC 0,82); Assessment of Blood Consumption Score was the less precise (AUC 0,68). CONCLUSION: In the prehospital setting the application of any the six scoring systems predicts the presence of massive hemorrhage and allows the activation of massive transfusion protocols while the patient is transferred to a hospital.

Coagulopatía Aguda Traumática y hemorragia masiva: una relación que debemos conocer / Acute traumatic coagulopathy and massive hemorrhage: a relationship that we should know

Los traumatismos y accidentes ocupan una proporción relevante de las causas de muerte para el grupo de edad que abarca desde la primera infancia hasta la cuarta década de la vida. En los últimos diez años la mortalidad mundial por traumatismos se incrementó en el 20 por ciento. En esos casos se estima que las hemorragias aportan entre el 30 y 70 por ciento de la mortalidad. Además, la hemorragia masiva es conocida por ser una de las principales causas de muerte intraoperatoria en pacientes sin traumatismos, sobre todo en aquellos sometidos a intervenciones que implican lesiones tisulares importantes. En alrededor de un tercio o más de esos pacientes se desarrolla una coagulopatía aguda asociada al trauma o coagulopatía aguda traumática, que obedece a mecanismos fisiopatogénicos complejos, superpuestos y dependientes en gran medida de la naturaleza del evento que originó la pérdida sanguínea. Esta coagulopatía se define básicamente como una reducción funcional de la fortaleza del coágulo con cambios mínimos en los tiempos de coagulación. Su aparición incrementa el riesgo de sangrado masivo, la utilización de mayores volúmenes de componentes sanguíneos e implica una mayor probabilidad de fallecer por hemorragia. Su identificación temprana mediante técnicas de evaluación hemostáticas como el TP/INR, y pruebas viscoelásticas (tromboelastometría rotacional y tromboelastografía), modifica los riesgos y aporta un blanco terapéutico factible que constituye la base de las nuevas estrategias transfusionales en el enfrentamiento a los sangrados masivos(AU)

Accidents and different kind of injuries account for a relevant proportion of death causes from childhood to the fourth decade of life. Death from injury has increased by 20 percent over the last decade. Hemorrhages are estimated to cause between 30 to 70 percent of those deaths. Massive bleeding is also known for being one of the main causes of death during surgery in cases with extensive tissue damage from not traumatic origin. In around one third or more of such patients an acute traumatic coagulopathy or trauma associated coagulopathy is developed. This phenomenon obeys to complex, overlapping mechanisms which would be in its major part dependent on the event that caused the blood lost. This kind of coagulopathy can be defined as a functional reduction in clot strength with only minimal changes in clotting times. Its presence increases the risk of a massive bleeding, the use of larger volumes of blood components and the likelihood of dying by the hemorrhage. The early identification of acute traumatic coagulopathy through hemostatic tests such as TP/INR and viscoelastic assays as thrombelastography or rotational thrombelastometry modifies those risks since it brings a feasible therapeutic target to aim at, becoming the cornerstone for the newer transfusional strategies while facing massive bleedings(AU)

Prediction of massive bleeding. Shock index and modified shock index. / Predicción de hemorragia masiva. Índice de shock e índice de shock modificado.

OBJECTIVE: To determine the predictive value of the Shock Index and Modified Shock Index in patients with massive bleeding due to severe trauma. DESIGN: Retrospective cohort. SETTING: Severe trauma patient's initial attention at the intensive care unit of a tertiary hospital. SUBJECTS: Patients older than 14 years that were admitted to the hospital with severe trauma (Injury Severity Score >15) form January 2014 to December 2015. VARIABLES: We studied the sensitivity (Se), specificity (Sp), positive and negative predictive value (PV+ and PV-), positive and negative likelihood ratio (LR+ and LR-), ROC curves (Receiver Operating Characteristics) and the area under the same (AUROC) for prediction of massive hemorrhage. RESULTS: 287 patients were included, 76.31% (219) were male, mean age was 43,36 (±17.71) years and ISS was 26 (interquartile range [IQR]: 21-34). The overall frequency of massive bleeding was 8.71% (25). For Shock Index: AUROC was 0.89 (95% confidence intervals [CI] 0.84 to 0.94), with an optimal cutoff at 1.11, Se was 91.3% (95% CI: 73.2 to 97.58) and Sp was 79.69% (95% CI: 74.34 to 84.16). For the Modified Shock Index: AUROC was 0.90 (95% CI: 0.86 to 0.95), with an optimal cutoff at 1.46, Se was 95.65% (95% CI: 79.01 to 99.23) and Sp was 75.78% (95% CI: 70.18 to 80.62). CONCLUSION: Shock Index and Modified Shock Index are good predictors of massive bleeding and could be easily incorporated to the initial workup of patients with severe trauma.

Implementation of a management protocol for massive bleeding reduces mortality in non-trauma patients: Results from a single centre audit.

OBJECTIVE: To audit the impact upon mortality of a massive bleeding management protocol (MBP) implemented in our center since 2007. DESIGN: A retrospective, single-center study was carried out. Patients transfused after MBP implementation (2007-2012, Group 2) were compared with a historical cohort (2005-2006, Group 1). BACKGROUND: Massive bleeding is associated to high mortality rates. Available MBPs are designed for trauma patients, whereas specific recommendations in the medical/surgical settings are scarce. PATIENTS: After excluding patients who died shortly (<6h) after MBP activation (n=20), a total of 304 were included in the data analysis (68% males, 87% surgical). INTERVENTIONS: Our MBP featured goal-directed transfusion with early use of adjuvant hemostatic medications. VARIABLES OF INTEREST: Primary endpoints were 24-h and 30-day mortality. Fresh frozen plasma-to-red blood cells (FFP:RBC) and platelet-to-RBC (PLT:RBC) transfusion ratios, time to first FFP unit and the proactive MBP triggering rate were secondary endpoints. RESULTS: After MBP implementation (Group 2; n=222), RBC use remained stable, whereas FFP and hemostatic agents increased, when compared with Group 1 (n=82). Increased FFP:RBC ratio (p=0.053) and earlier administration of FFP (p=0.001) were also observed, especially with proactive MBP triggering. Group 2 patients presented lower rates of 24-h (0.5% vs. 7.3%; p=0.002) and 30-day mortality (15.9% vs. 30.2%; p=0.018) - the greatest reduction corresponding to non-surgical patients. Logistic regression showed an independent protective effect of MBP implementation upon 30-day mortality (OR=0.3; 95% CI 0.15-0.61). CONCLUSIONS: These data suggest that the implementation of a goal-directed MBP for prompt and aggressive management of non-trauma, massive bleeding patients is associated to reduced 24-h and 30-day mortality rates.

Massive obstetric hemorrhage: Current approach to management. / Hemorragia masiva obstétrica: enfoque terapéutico actual.

Massive obstetric hemorrhage is a major cause of maternal mortality and morbidity worldwide. It is defined (among others) as the loss of>2,500ml of blood, and is associated to a need for admission to critical care and/or hysterectomy. The relative hemodilution and high cardiac output found in normal pregnancy allows substantial bleeding before a drop in hemoglobin and/or hematocrit can be identified. Some comorbidities associated with pregnancy can contribute to the occurrence of catastrophic bleeding with consumption coagulopathy, which makes the situation even worse. Optimization, preparation, rational use of resources and protocolization of actions are often useful to improve outcomes in patients with postpartum hemorrhage. Using massive obstetric hemorrhage protocols is useful for facilitating rapid transfusion if needed, and can also be cost-effective. If hypofibrinogenemia during the bleeding episode is identified, early fibrinogen administration can be very useful. Other coagulation factors in addition to fibrinogen may be necessary during postpartum hemorrhage replacement measures in order to effectively correct coagulopathy. A hysterectomy is recommended if the medical and surgical measures prove ineffective.

Black esophagus (acute esophageal necrosis) after spinal anesthesia.

Acute esophagic necrosis or black esophagus is an uncommon clinical entity that owes its name to the endoscopic view of the necrotic esophageal mucosa. It is always related with a critical medical condition and usually has an ischemic etiology. We report the first case of acute esophageal necrosis after a spinal anesthetic for partial hip joint arthroplasty. We discuss the underlying pathophysiological mechanisms.

[Iliac artery occlusion balloons for suspected placenta accreta during cesarean section]. / Inserción de balones intraarteriales ilíacos previa a la realización de cesárea ante la sospecha de acretismo placentario.

Massive obstetric hemorrhage still remains a major cause of maternal mortality and morbidity. The risk factors associated with this pathology must be identified in order to schedule the appropriate delivery with the necessary resources. A case is presented of an iliac artery occlusion with intravascular balloons for suspected placenta accreta during cesarean section. The perioperative treatment, as well as an analysis of the treatment options is described, along with their advantages and disadvantages, from the use of postpartum hemorrhage protocols, blood transfusion and procoagulant factors, and other maneuvers to control bleeding, until the hysterectomy.