Results: Overall mortality at 14 days was 22.8%. Models had a high prediction performance, with the best prediction for overall mortality achieved through Naive Bayes (area under the curve = 0.906). The most significant predictors were the GCS at admission and prehospital GCS, age, and pupil reaction. When predicting the length of stay at the intensive care unit, the Conditional Inference Tree model had the best performance (root mean square error = 1.011), with the most important variable across all models being the GCS at scene.Conclusions: Models for early mortality and hospital length of stay using Machine Learning can achieve high performance when based on registry data even in LMICs. These models have the potential to inform treatment decisions and counsel family members.Level of evidence: This observational study provides a level IV evidence on prognosis after TBI.

Background Lung ultrasound is increasingly used in critically ill patients as an alternative to bedside chest radiography, but the best training method remains uncertain. This study describes a training curriculum allowing trainees to acquire basic competence. Methods This multicenter, prospective, and educational study was conducted in 10 Intensive Care Units in Brazil, China, France and Uruguay. One hundred residents, respiratory therapists, and critical care physicians without expertise in transthoracic ultrasound (trainees) were trained by 18 experts. The main study objective was to determine the number of supervised exams required to get the basic competence, defined as the trainees’ ability to adequately classify lung regions with normal aeration, interstitial–alveolar syndrome, and lung consolidation. An initial 2-h video lecture provided the rationale for image formation and described the ultrasound patterns commonly observed in critically ill and emergency patients. Each trainee performed 25 bedside ultrasound examinations supervised by an expert. The progression in competence was assessed every five supervised examinations. In a new patient, 12 pulmonary regions were independently classified by the trainee and the expert. Results Progression in competence was derived from the analysis of 7,330 lung regions in 2,562 critically ill and emergency patients. After 25 supervised examinations, 80% of lung regions were adequately classified by trainees. The ultrasound examination mean duration was 8 to 10 min in experts and decreased from 19 to 12 min in trainees (after 5 vs. 25 supervised examinations). The median training duration was 52 (42, 82) days. Conclusions A training curriculum including 25 transthoracic ultrasound examinations supervised by an expert provides the basic skills for diagnosing normal lung aeration, interstitial–alveolar syndrome, and consolidation in emergency and critically ill patients. Editor’s Perspective What We Already Know about This Topic What This Article Tells Us That Is New

A previously healthy 27-year-old female had 3 days of fever (40°C), headache, and myalgia. She had not been previously vaccinated for YF. Initial workup revealed 2,150 leukocytes/mm , 83,000 platelets/mm , AST 8462U/L and ALT 5249U/L (Figure 1). She was icteric with a heart rate of 60 bpm. The following day, a generalized seizure led to intubation, and renal failure led to hemodialysis. Transcranial doppler ultrasound showed signs of intracranial hypertension, cranial CT scan showed diffuse hypoattenuation and loss of grey-white differentiation; abdominal doppler ultrasound, and echocardiogram were normal. This article is protected by copyright. All rights reserved.

Background Geoeconomic variations in epidemiology, the practice of ventilation, and outcome in invasively ventilated intensive care unit (ICU) patients without acute respiratory distress syndrome (ARDS) remain unexplored. In this analysis we aim to address these gaps using individual patient data of four large observational studies. MethodsIn this pooled analysis we harmonised individual patient data from the ERICC, LUNG SAFE, PRoVENT, and PRoVENT-iMiC prospective observational studies, which were conducted from June, 2011, to December, 2018, in 534 ICUs in 54 countries. We used the 2016 World Bank classification to define two geoeconomic regions: middleincome countries (MICs) and high-income countries (HICs). ARDS was defined according to the Berlin criteria. Descriptive statistics were used to compare patients in MICs versus HICs. The primary outcome was the use of low tidal volume ventilation (LTVV) for the first 3 days of mechanical ventilation. Secondary outcomes were key ventilation parameters (tidal volume size, positive end-expiratory pressure, fraction of inspired oxygen, peak pressure, plateau pressure, driving pressure, and respiratory rate), patient characteristics, the risk for and actual development of acute respiratory distress syndrome after the first day of ventilation, duration of ventilation, ICU length of stay, and ICU mortality. Findings Of the 7608 patients included in the original studies, this analysis included 3852 patients without ARDS, of whom 2345 were from MICs and 1507 were from HICs. Patients in MICs were younger, shorter and with a slightly lower body-mass index, more often had diabetes and active cancer, but less often chronic obstructive pulmonary disease and heart failure than patients from HICs. Sequential organ failure assessment scores were similar in MICs and HICs. Use of LTVV in MICs and HICs was comparable (42•4% vs 44•2%; absolute difference -1•69 [-9•58 to 6•11] p=0•67; data available in 3174 [82%] of 3852 patients). The median applied positive end expiratory pressure was lower in MICs than in HICs (5 [IQR 5-8] vs 6 [5-8] cm H 2 O; p=0•0011). ICU mortality was higher in MICs than in HICs (30•5% vs 19•9%; p=0•0004; adjusted effect 16•41% [95% CI 9•52-23•52]; p<0•0001) and was inversely associated with gross domestic product (adjusted odds ratio for a US$10 000 increase per capita 0•80 [95% CI 0•75-0•86]; p<0•0001).Interpretation Despite similar disease severity and ventilation management, ICU mortality in patients without ARDS is higher in MICs than in HICs, with a strong association with country-level economic status.

Efeitos da posição sentada na força de músculos respiratórios durante o desmame de pacientes sob ventilação mecânica prolongada no pós-operatório de cirurgia cardiovascular Concluímos que pacientes em pós-operatório de cirurgia cardiovascular em ventilação mecânica podem se beneficiar da posição sentada durante o desmame do suporte ventilatório, observado pela melhora da força muscular inspiratória (PImáx). Além disso, não foram observadas alterações hemodinâmicas com a troca de postura, sendo considerado um procedimento seguro. DEScRitORES: cirurgia torácica; ventilação mecânica; desmame; posicionamento do paciente. ABStRAct:The purpose was to study the effects of sitting position, in ventilatory and hemodynamic parameters, in patients under prolonged mechanical ventilation, in the postoperative of cardiovascular surgery. Participated 40 postoperative cardiovascular patients, intubated and mechanically ventilated, were randomized into control group (n=17) and intervention group (n=23). Hemodynamic parameters, arterial blood gases, respiratory muscle strength and ventilometry, were measured in two moments: first and second evaluation, with 30-minute interval between measurements. In the control group, both evaluations were performed at head of the bed elevated at 30°. In the sitting group, the first evaluation was developed at 30° and the second, (30 minutes after transfer to the armchair) in the sitting position (90°). The mean age of patients was 64.7±11.2 years. There were no differences regarding the hemodynamic parameters, blood gas analysis, forced vital capacity, minute ventilation and tidal volume. However, a significant increase in maximal inspiratory pressure (MIP) in the intervention group was observed (p<0.01). We conclude that stable patients in the postoperative period of cardiovascular surgery, mechanically ventilated, maybe benefits of the sitting position observed of increased inspiratory muscle strength. Furthermore, no significant hemodynamic changes, and it was well tolerated when the patients were transferred to the sitting position, so it was considered a safe procedure.