Prehospital triage of acute aortic syndrome using a machine learning algorithm.

BACKGROUND: Acute aortic syndrome (AAS) comprises a complex and potentially fatal group of conditions requiring emergency specialist management. The aim of this study was to build a prediction algorithm to assist prehospital triage of AAS. METHODS: Details of consecutive patients enrolled in a regional specialist aortic network were collected prospectively. Two prediction algorithms for AAS based on logistic regression and an ensemble machine learning method called SuperLearner (SL) were developed. Undertriage was defined as the proportion of patients with AAS not transported to the specialist aortic centre, and overtriage as the proportion of patients with alternative diagnoses but transported to the specialist aortic centre. RESULTS: Data for 976 hospital admissions between February 2010 and June 2017 were included; 609 (62·4 per cent) had AAS. Overtriage and undertriage rates were 52·3 and 16·1 per cent respectively. The population was divided into a training cohort (743 patients) and a validation cohort (233). The area under the receiver operating characteristic (ROC) curve values for the logistic regression score and the SL were 0·68 (95 per cent c.i. 0·64 to 0·72) and 0·87 (0·84 to 0·89) respectively (P < 0·001) in the training cohort, and 0·67 (0·60 to 0·74) and 0·73 (0·66 to 0·79) in the validation cohort (P = 0·038). The logistic regression score was associated with undertriage and overtriage rates of 33·7 (bootstrapped 95 per cent c.i. 29·3 to 38·3) and 7·2 (4·8 to 9·8) per cent respectively, whereas the SL yielded undertriage and overtriage rates of 1·0 (0·3 to 2·0) and 30·2 (25·8 to 34·8) per cent respectively. CONCLUSION: A machine learning prediction model performed well in discriminating AAS and could be clinically useful in prehospital triage of patients with suspected AAS.

ANTECEDENTES: Los síndromes aórticos agudos (aortic acute syndromes, AAS) constituyen un grupo complejo y potencialmente letal de entidades que requieren un tratamiento especializado en emergencias. El objetivo de este estudio fue construir un algoritmo de predicción para ayudar a la selección prehospitalaria de los AAS. MÉTODOS: Se recogieron prospectivamente una serie de pacientes consecutivos inscritos en una red regional especializada en patología aórtica. Se desarrollaron dos algoritmos de predicción para AAS basados en una regresión logística y en un método de aprendizaje automático denominado Super Learner (SL). Undertriage (infra-selección) se definió como la proporción de pacientes con AAS no transportados al centro especializado en patología aórtica y el overtriage (sobre-selección) como la proporción de pacientes con diagnósticos alternativos al AAS pero transportados al centro especializado en patología aórtica. RESULTADOS: Se incluyeron los datos de 976 ingresos hospitalarios entre febrero de 2010 y junio de 2017, con 609 (62,4%) AAS. Las tasas de overtriage y undertriage fueron del 52,3% y del 16,1%, respectivamente. La población se dividió en una cohorte de entrenamiento (n = 743) y en una cohorte de validación (n = 233). El área bajo la curva ROC para la puntuación de regresión logística y el SL fueron de 0,68 (0,64, 0,72) y de 0,87 (0,84, 0,89), respectivamente (P < 0,001) en la cohorte de entrenamiento, y de 0,67 (0,60, 0,74) y de 0,73 (0,66, 0,79) en la cohorte de validación (P = 0,038). La puntuación de regresión logística se asoció con tasas de undertriage y overtriage de 33,7% (i.c. del 95% bootstrapped 29,3%, 38,3%) y de 7,2% (4,8%, 9,8%), respectivamente, mientras que el SL presentó tasas de undertriage y overtriage de 1,0% (0,3%, 2,0%) y de 30,2% (25,8%, 34,8%), respectivamente. CONCLUSIÓN: El modelo de predicción de aprendizaje automático funcionó bien para discriminar AAS y podría ser clínicamente útil en la selección prehospitalaria de pacientes con sospecha de síndrome aórtico agudo.

Impact of sleep deprivation on anaesthesia residents' non-technical skills: a pilot simulation-based prospective randomized trial.

BACKGROUND: Sleep deprivation is common in anaesthesia residents, but its impact on performance remains uncertain. Non-technical skills (team working, situation awareness, decision making, and task management) are key components of quality of care in anaesthesia, particularly in crisis situations occurring in the operating room. The impact of sleep deprivation on non-technical skills is unknown. We tested the hypothesis that in anaesthesia residents sleep deprivation is associated with impaired non-technical skills. METHODS: Twenty anaesthesia residents were randomly allocated to undergo a simulation session after a night shift [sleep-deprived (SLD) group, n =10] or after a night of rest [rested (R) group, n =10] from January to March 2015. The simulated scenario was a situation of crisis management in the operating room. The primary end point was a composite score of anaesthetists' non-technical skills (ANTS) assessed by two blinded evaluators. RESULTS: Non-technical skills were significantly impaired in the SLD group [ANTS score 12.2 (interquartile range 10.5-13)] compared with the R group [14.5 (14-15), P <0.02]. This difference was mainly accounted for by a difference in the team working item. On the day of simulation, the SLD group showed increased sleepiness and decreased confidence in anaesthesia skills. CONCLUSIONS: In this randomized pilot trial, sleep deprivation was associated with impaired non-technical skills of anaesthesia residents in a simulated anaesthesia intraoperative crisis scenario. TRIAL REGISTRATION: NCT02622217.

Noninvasive ventilation during acute respiratory distress syndrome in patients with cancer: Trends in use and outcome.

PURPOSE: The objectives of our study were to describe the outcome of patients with malignancies treated for acute respiratory distress syndrome (ARDS) with noninvasive ventilation (NIV) and to evaluate factors associated with NIV failure. METHODS: Post hoc analysis of a multicenter database within 20 years was performed. All patients with malignancies and Berlin ARDS definition were included. Noninvasive ventilation use was defined as NIV lasting more than 1 hour, whereas failure was defined as a subsequent requirement of invasive ventilation. Conditional backward logistic regression analyses were conducted. RESULTS: A total of 1004 met the Berlin definition of ARDS. Noninvasive ventilation was used in 387 patients (38.6%) and NIV failure occurred in 71%, with an in-hospital mortality of 62.7%. Severity of ARDS defined by the partial pressure arterial oxygen and fraction of inspired oxygen ratio (odds ratio [OR], 2.20; 95% confidence interval [CI], 1.15-4.19), pulmonary infection (OR, 1.81; 95% CI, 1.08-3.03), and modified Sequential Organ Failure Assessment (SOFA) score (OR, 1.13; 95% CI, 1.06-1.21) were associated with NIV failure. Factors associated with hospital mortality were NIV failure (OR, 2.52; 95% CI, 1.56-4.07), severe ARDS as compared with mild ARDS (OR, 1.89; 95% CI, 1.05-1.19), and modified SOFA score (OR, 1.12; 95% CI, 1.05-1.19). CONCLUSION: Noninvasive ventilation failure in ARDS patients with malignancies is frequent and related to ARDS severity, SOFA score, and pulmonary infection-related ARDS. Noninvasive ventilation failure is associated with in-hospital mortality.

Early detection with the ß-LACTA™ test of extended-spectrum ß-lactamase-producing Enterobacteriaceae in blood cultures.

Effective antibiotic therapy is crucial for the outcome of septic patients and requires early diagnosis of ß-lactam resistance in cases of Gram-negative bacteremia. Here, we report high sensitivity of the ß-LACTA™ test in rapid detection of extended-spectrum ß-lactamase-producing Enterobacteriaceae in blood cultures positive for Gram-negative bacilli.