Big data and machine learning in critical care: Opportunities for collaborative research.

The introduction of clinical information systems (CIS) in Intensive Care Units (ICUs) offers the possibility of storing a huge amount of machine-ready clinical data that can be used to improve patient outcomes and the allocation of resources, as well as suggest topics for randomized clinical trials. Clinicians, however, usually lack the necessary training for the analysis of large databases. In addition, there are issues referred to patient privacy and consent, and data quality. Multidisciplinary collaboration among clinicians, data engineers, machine-learning experts, statisticians, epidemiologists and other information scientists may overcome these problems. A multidisciplinary event (Critical Care Datathon) was held in Madrid (Spain) from 1 to 3 December 2017. Under the auspices of the Spanish Critical Care Society (SEMICYUC), the event was organized by the Massachusetts Institute of Technology (MIT) Critical Data Group (Cambridge, MA, USA), the Innovation Unit and Critical Care Department of San Carlos Clinic Hospital, and the Life Supporting Technologies group of Madrid Polytechnic University. After presentations referred to big data in the critical care environment, clinicians, data scientists and other health data science enthusiasts and lawyers worked in collaboration using an anonymized database (MIMIC III). Eight groups were formed to answer different clinical research questions elaborated prior to the meeting. The event produced analyses for the questions posed and outlined several future clinical research opportunities. Foundations were laid to enable future use of ICU databases in Spain, and a timeline was established for future meetings, as an example of how big data analysis tools have tremendous potential in our field.

Early prognostic value of an Algorithm based on spectral Variables of Ventricular fibrillAtion from the EKG of patients with suddEn cardiac death: A multicentre observational study (AWAKE).

OBJECTIVE: Ventricular fibrillation (VF)-related sudden cardiac death (SCD) is a leading cause of mortality and morbidity. Current biological and imaging parameters show significant limitations on predicting cerebral performance at hospital admission. The AWAKE study (NCT03248557) is a multicentre observational study to validate a model based on spectral ECG analysis to early predict cerebral performance and survival in resuscitated comatose survivors. METHODS: Data from VF ECG tracings of patients resuscitated from SCD will be collected using an electronic Case Report Form. Patients can be either comatose (Glasgow Coma Scale - GCS - ≤8) survivors undergoing temperature control after return of spontaneous circulation (RoSC), or those who regain consciousness (GCS=15) after RoSC; all admitted to Intensive Cardiac Care Units in 4 major university hospitals. VF tracings prior to the first direct current shock will be digitized and analyzed to derive spectral data and feed a predictive model to estimate favorable neurological performance (FNP). The results of the model will be compared to the actual prognosis. RESULTS: The primary clinical outcome is FNP during hospitalization. Patients will be categorized into 4 subsets of neurological prognosis according to the risk score obtained from the predictive model. The secondary clinical outcomes are survival to hospital discharge, and FNP and survival after 6 months of follow-up. The model-derived categorisation will be also compared with clinical variables to assess model sensitivity, specificity, and accuracy. CONCLUSIONS: A model based on spectral analysis of VF tracings is a promising tool to obtain early prognostic data after SCD.

Med-Score 24: A multivariable prediction model for poststernotomy mediastinitis 24 hours after admission to the intensive care unit.

OBJECTIVES: Mediastinitis is a serious complication of heart surgery. In this study, we developed a bedside risk score for poststernotomy mediastinitis. METHODS: Data were prospectively collected from 4625 patients admitted to our intensive care unit after heart surgery (January 2005-June 2011). Mediastinitis was defined according to Centers for Disease Control and Prevention criteria. A logistic model was constructed in a randomly selected subgroup of 2618 patients and validated in a second cohort of 1352, as well as in a prospective cohort of 2615 (June 2011-December 2015). Model discriminatory power was assessed according to the area under the receiver operating characteristic curve (AUROC). The ß coefficients of the model were used to define 3 levels of mediastinitis risk as a score designated Med-Score 24. Its performance to predict mediastinitis was compared with that of the logistic EuroSCORE and Society of Thoracic Surgeons score. RESULTS: Ninety-four (2.36%) patients developed mediastinitis. The risk factors identified as predictive of mediastinitis (AUROC 0.80) were 4 preoperative variables (age >70 years, chronic obstructive lung disease, obesity, and antiplatelet therapy) and 3 perioperative variables (prolonged ischemia, emergency reoperation, and prolonged intubation). AUROCs for the Society of Thoracic Surgeons score and logistic EuroSCORE were 0.63 and 0.55, respectively, both differing significantly from the area calculated for Med-Score 24 (P < .001). CONCLUSIONS: The score developed showed excellent predictive power 24 hours after admission to the intensive care unit for mediastinitis risk. This simple tool helps stratify patients according to this risk, thus identifying high-risk patients for preventive measures. In our patient cohort, Med-Score 24 performed better than other scores used for this purpose.

Early prognostic value of an Algorithm based on spectral Variables of Ventricular fibrillAtion from the EKG of patients with suddEn cardiac death: A multicentre observational study (AWAKE) / Valor pronóstico precoz de un algoritmo basado en variables espectrales de la fibrilación ventricular del ECG de pacientes con muerte súbita cardiaca: estudio observacional multicéntrico (AWAKE)

Abstract Objective: Ventricular fibrillation (VF)-related sudden cardiac death (SCD) is a leading cause of mortality and morbidity. Current biological and imaging parameters show significant limitations on predicting cerebral performance at hospital admission. The AWAKE study (NCT03248557) is a multicentre observational study to validate a model based on spectral ECG analysis to early predict cerebral performance and survival in resuscitated comatose survivors. Methods: Data from VF ECG tracings of patients resuscitated from SCD will be collected using an electronic Case Report Form. Patients can be either comatose (Glasgow Coma Scale GCS --- ≤8) survivors undergoing temperature control after return of spontaneous circulation (RoSC), or those who regain consciousness (GCS = 15) after RoSC; all admitted to Intensive Cardiac Care Units in 4 major university hospitals. VF tracings prior to the first direct current shock will be digitized and analyzed to derive spectral data and feed a predictive model to estimate favorable neurological performance (FNP). The results of the model will be compared to the actual prognosis. Results: The primary clinical outcome is FNP during hospitalization. Patients will be categorized into 4 subsets of neurological prognosis according to the risk score obtained from the predictive model. The secondary clinical outcomes are survival to hospital discharge, and FNP and survival after 6 months of follow-up. The model-derived categorisation will be also compared with clinical variables to assess model sensitivity, specificity, and accuracy. Conclusions: A model based on spectral analysis of VF tracings is a promising tool to obtain early prognostic data after SCD.

Resumen Objetivo: La muerte súbita (MS) por fibrilación ventricular (FV) es una importante causa de morbilidad y mortalidad. Los métodos biológicos y de imagen actuales muestran limitaciones para predecir el pronóstico cerebral al ingreso hospitalario. AWAKE es un estudio observacional, multicéntrico, con el objetivo de validar un modelo basado en el análisis espectral del elec- trocardiograma (ECG), que predice precozmente el pronóstico cerebral y la supervivencia en pacientes resucitados y en estado de coma. Métodos: Se recogerán datos de los ECG con FV de pacientes reanimados de MS. Los pacientes pueden ser tanto supervivientes en estado de coma (Glasgow Coma Scale GCS ≤ 8) sometidos a control de temperatura tras la recuperación de circulación espontánea (RCE), como aquellos que recuperan la consciencia (GCS = 15) tras RCE; todos ellos ingresados en unidades de terapia intensiva cardiológica de 4 hospitales de referencia. Los registros de FV previos al primer choque se digitalizarán y analizarán para obtener datos espectrales que se incluirán en un modelo predictivo que estime el pronóstico neurológico favorable (PNF). El resultado del modelo se comparará con el pronóstico real. Resultados: El objetivo principal es el PNF durante la hospitalización. Los pacientes se categorizarán en 4 subgrupos de pronóstico neurológico según la estimación de riesgo obtenida en el modelo predictivo. Los objetivos secundarios son supervivencia al alta hospitalaria, y PNF y supervivencia a los 6 meses. El resultado de este modelo también se comparará con el pronóstico según variables clínicas. Conclusiones: Un modelo basado en el análisis espectral de registros de FV es una herramienta prometedora para obtener datos pronósticos precoces tras MS por FV.

Resultados de un programa mixto de implantación de prótesis aórtica por vía transfemoral con los dispositivos de Edwards y CoreValve / Findings of a mixed transfemoral aortic valve implantation program using edwards and CoreValve devices

Introducción y objetivos. La implantación de prótesis valvular aórtica por vía transfemoral con los dispositivos Edwards-SAPIEN (ES) y Medtronic-CoreValve (MCV) es una alternativa al reemplazo quirúrgico en pacientes con estenosis aórtica severa y alto riesgo quirúrgico. Nuestro objetivo es comparar los resultados obtenidos con uno y otro dispositivo. Métodos. Estudio prospectivo de las prótesis implantadas por vía transfemoral en nuestro centro. Resultados. De los 76 pacientes (edad, 83±6 años; el 63% mujeres; EuroSCORE logístico, 18±9), Se seleccionó a 50 para ES y a 26 para MCV. No se observaron diferencias entre grupos en edad, sexo, clase funcional, área valvular, enfermedades asociadas o EuroSCORE. Se consiguió implantar la prótesis en el 84% del grupo ES y el 100% del grupo MCV (p=0,04); hubo 3 casos de taponamiento, 2 disecciones aórticas y 1 malposición en la serie ES. Las complicaciones vasculares fueron similares (el 26 frente al 23%), pero la necesidad de marcapasos fue mayor con la MCV (el 10 frente al 39%; p=0,003). Las mortalidades totales a 30 días fueron del 12 y el 20% (sin diferencia significativa) y a 1 año, del 24 y el 20% en los grupos ES y MCV respectivamente. Tras un seguimiento de 367±266 días para ES y 172±159 para MCV, 3 pacientes fallecieron; los demás mantienen la mejoría clínica y no se observan cambios ecocardiográficos. Conclusiones. La mortalidad hospitalaria, la tasa de complicaciones y la evolución a medio plazo han sido similares con ambos dispositivos; las únicas diferencias encontradas han sido una tasa de implantación con éxito mayor con la MCV, aunque a expensas de una mayor frecuencia de bloqueo auriculoventricular (AU)

Introduction and objectives: Transfemoral implantation of an Edwards SAPIEN (ES) or Medtronic CoreValve (MCV) aortic valve prosthesis is an alternative to surgical replacement for patients with severe aortic stenosis and a high surgical risk. The study’s aim was to compare results obtained with these two devices. Methods: Prospective observational study of transfemoral prosthesis implantation performed at our center. Results: Of the 76 patients (age 83 +/- 6 years, 63% female, logistic EuroSCORE 18 +/- 9) included, 50 were assigned the ES and 26 the MCV device. There was no difference between the groups in age, sex, functional class, valve area, associated conditions, or EuroSCORE. Implantation was successful in 84% of the ES group and 100% of the MCV group (P = .04). There were three cases of tamponade, two aortic dissections and one valve malposition in the ES group. The two groups had similar vascular access complication rates (26% vs. 23%; P = NS), but pacemaker need was greater with the MCV (10% vs. 39%; P = .003). Mortality rates at 30 days were 12% and 20% (P = NS) in the ES and MCV groups, respectively, and at 1 year, 24% and 20% (P = NS), respectively. After a follow-up of 367 +/- 266 days in the ES group and 172 +/- 159 days in the MCV group, three patients died. Clinical improvement was maintained in other patients and no echocardiographic changes were observed. Conclusions: In-hospital mortality, the complication rate and medium-term outcomes were similar with the two devices. The only difference observed was a higher implantation success rate with the MCV, although at the expense of a greater frequency of atrioventricular block (AU)

[Repercussion on the hemodynamic measurements across a Swan-Ganz catheter with the postural changes]. / Repercusión en las mediciones hemodinámicas a través de un Swan-Ganz con los cambios posturales.

AIM: Usually hemodynamic measures are done with the patient in dorsal decubitus and the bedside at 0 degrees. Our aim has been to evaluate the influence that postural changes has in the hemodynamic measures which were carried out with a pulmonary artery catheter, so as called Swan-Ganz. MATERIAL AND METHOD: It's a prospective study. The same patient is control group and study group. There were done tree consecutive measures in each patient. Firstly in dorsal decubitus, then right lateral decubitus and finally in left lateral decubitus. Before doing the measures after change of posture a thirty minutes period was left in order to stabilise the hemodynamical flow. The items of study were, a part of demographic ones, cardiac index, pulmonary artery systolic pressure, pulmonary artery diastolic pressure, pulmonary artery mean pressure, pulmonary artery occlusion pressure, right atrial pressure, systolic arterial pressure, diastolic arterial pressure, mean arterial pressure and heart rate. 28 patients were included in the study. RESULTS: The age average was 62.5 years (27.05-67.05); a 78.6% were male. Who had a NEMS average of 42.4 (39.9-44.9). No difference was found between hemodynamic measures in the different postures. CONCLUSIONS: Postural changes in stable patients have no influence in pressures and other hemodynamic variables measures.

Repercusión de las mediciones hemodinámicas a través de un Swan-Ganz con los cambios posturales / Repercussion on the hemodynamic measurements across a Swan-Ganz catheter with the postural changes

Objetivo. Habitualmente, las mediciones hemodinámicas se realizan con el paciente en decúbito supino y la cabecera de la cama a 0°.Nuestro objetivo ha sido evaluar la influencia de los cambios posturales en las mediciones hemodinámicas efectuadas a través de un catéter de arteria pulmonar, también llamado de Swan-Ganz. Material y método. Estudio prospectivo, el mismo paciente es el grupo control y el grupo estudio. Se realizaron 3 mediciones consecutivas en cada paciente, comenzando en decúbito supino, posteriormente en decúbito lateral derecho y en decúbito lateral izquierdo. Antes de realizar las mediciones tras el cambio de postura, se dejó un período de 30 min de estabilización hemodinámica. Las variables a estudio, a parte de las demográficas, fueron: índice cardíaco, presión arteria pulmonar sistólica, presión arteria pulmonar diastólica, presión arteria pulmonar media, presión capilar pulmonar, presión de la aurícula derecha, presión arterial sistólica, presión arterial diastólica, presión arterial media y frecuencia cardíaca. Se incluyeron 28 pacientes. Resultados. La media de edad de los pacientes fue de 62,05 años (57,05-67,05), el 78,6 por ciento fueron varones y tuvieron un NEMS medio de 42,4 (39,9-44,9). No encontramos diferencias en las mediciones hemodinámicas en las diferentes posturas. Conclusiones. Los cambios posturales en pacientes estables, no influyen en las mediciones de las presiones y valores hemodinámicos(AU)

[Training in cardiopulmonary reanimation with early defibrillation to nurses from the 7th area of Madrid]. / Formación en reanimación cardiopulmonar básica con desfibrilación precoz a enfermeros/as del área 7 de Madrid.

UNLABELLED: Ventricular fibrillation and ventricular tachicardia without pulse are the most frequent causes of suddenly death, therefore, it was considered that training hospitals and healthcentre nurses in identifying arritms and using properly the defibrilators was necessary. It was made by practical-theoric courses. The main aim of this study has been to know does the nurses value these type of courses and its repercussion on their level of knowledge. This is a descriptive study. The course consisted of 2 parts, one theoric and other practical. The valuation was made with a practical exam (A, B o C, being C a fail) and the other was written. To obtain the certificate of the course, you need to get at least 70% in the written part, and an A or a B in the practical exam. An exam was set at the beginning of the course and other at the end of it, in order to see the improvement. To value the course, we passed one anonimous poll, for this, it was used one scale from 1-5. we have gone 8 courses with 226 pupils, and the 74.3% of them, have passed. The level of knowledges has considerably increased p < 0.0001 the level of satisfaction was 4.7 0.5, of the teachers was of 4.6 0.5 and theirs enthusiasm on the course was 4.7 0.6. For them, theirs level of theorical and practical knowledges were 2.7 0.7 and 2.5 0.7 respectively. CONCLUSIONS: 1) Nurses have had a great interest in this course. 2) After the course, the level of knowledge was increased. 3) The califications of the pupils on the teachers and the course has been high. 4) The training of nursery in RCP and early defibrillation is absolutely essential as it is shawn in their low level of theorical and practical knowledge.

Formación en reanimacióncardiopulmonar básica condesfibrilación precoz aenfermeros/as del área 7de Madrid / Training in cardiopulmonary reanimation with early defibrillation to nurses from the 7th area of Madrid

La fibrilación ventricular y la taquicardia ventricular sin pulso son la causa más frecuente de muerte súbita, por ello se consideró necesario formar a la enfermería de hospitalización y ambulatorios, en la identificación de estas arritmias y en la utilización del desfibrilador, mediante cursos teórico-prácticos. El objetivo ha sido conocer cómo valora la enfermería este tipo de cursos y su repercusión en el nivel de conocimientos. Estudio descriptivo. El curso constó de 2 partes, una teórica y otra práctica. La evaluación se realizó mediante examen práctico (A, B o C, siendo C suspenso) y escrito. Para obtener la certificación del curso, debían obtener una puntuación superior o igual al 70 por ciento en el examen escrito y "A o B" en el práctico. Para valorar la probable mejoría en el nivel de conocimientos, se realizó un examen al inicio del curso, que fue comparado con el posterior. La evaluación del curso y profesorado se realizó mediante encuesta anónima, se utilizó una escala de valoración del 1 al 5. Se han impartido 8 cursos con un total de 226 alumnos, han obtenido certificado de capacitación el 74,3 por ciento. El aumento en el nivel de conocimientos ha sido estadísticamente significativo p > 0,0001. El nivel de satisfacción obtuvo una media de 4,7 ñ 0,5, el profesorado 4,6 ñ 0,5 y su interés por el curso fue 4,7 ñ 0,6. Según ellos, su nivel de conocimientos teóricos y prácticos fue 2,7 ñ 0,7 y 2,5 ñ 0,7 respectivamente. Como conclusión podemos decir: 1) El interés de la enfermería en este curso es alto; 2) Existe un aumento de conocimientos tras la realización del curso; 3) La valoración que el alumno hace tanto del curso como del profesorado es alto; 4) La formación de la enfermería en RCP y desfibrilación precoz es imprescindible a la vista de los escasos conocimientos teóricos y prácticos que tienen (AU)