Flow starvation during square-flow assisted ventilation detected by supervised deep learning techniques.

BACKGROUND: Flow starvation is a type of patient-ventilator asynchrony that occurs when gas delivery does not fully meet the patients' ventilatory demand due to an insufficient airflow and/or a high inspiratory effort, and it is usually identified by visual inspection of airway pressure waveform. Clinical diagnosis is cumbersome and prone to underdiagnosis, being an opportunity for artificial intelligence. Our objective is to develop a supervised artificial intelligence algorithm for identifying airway pressure deformation during square-flow assisted ventilation and patient-triggered breaths. METHODS: Multicenter, observational study. Adult critically ill patients under mechanical ventilation > 24 h on square-flow assisted ventilation were included. As the reference, 5 intensive care experts classified airway pressure deformation severity. Convolutional neural network and recurrent neural network models were trained and evaluated using accuracy, precision, recall and F1 score. In a subgroup of patients with esophageal pressure measurement (ΔPes), we analyzed the association between the intensity of the inspiratory effort and the airway pressure deformation. RESULTS: 6428 breaths from 28 patients were analyzed, 42% were classified as having normal-mild, 23% moderate, and 34% severe airway pressure deformation. The accuracy of recurrent neural network algorithm and convolutional neural network were 87.9% [87.6-88.3], and 86.8% [86.6-87.4], respectively. Double triggering appeared in 8.8% of breaths, always in the presence of severe airway pressure deformation. The subgroup analysis demonstrated that 74.4% of breaths classified as severe airway pressure deformation had a ΔPes > 10 cmH2O and 37.2% a ΔPes > 15 cmH2O. CONCLUSIONS: Recurrent neural network model appears excellent to identify airway pressure deformation due to flow starvation. It could be used as a real-time, 24-h bedside monitoring tool to minimize unrecognized periods of inappropriate patient-ventilator interaction.

Specific Training Improves the Detection and Management of Patient-Ventilator Asynchrony.

BACKGROUND: Patient-ventilator asynchrony is common in patients undergoing mechanical ventilation. The proportion of health-care professionals capable of identifying and effectively managing different types of patient-ventilator asynchronies is limited. A few studies have developed specific training programs, but they mainly focused on improving patient-ventilator asynchrony detection without assessing the ability of health-care professionals to determine the possible causes. METHODS: We conducted a 36-h training program focused on patient-ventilator asynchrony detection and management for health-care professionals from 20 hospitals in Latin America and Spain. The training program included 6 h of a live online lesson during which 120 patient-ventilator asynchrony cases were presented. After the 6-h training lesson, health-care professionals were required to complete a 1-h training session per day for the subsequent 30 d. A 30-question assessment tool was developed and used to assess health-care professionals before training, immediately after the 6-h training lecture, and after the 30 d of training (1-month follow-up). RESULTS: One hundred sixteen health-care professionals participated in the study. The median (interquartile range) of the total number of correct answers in the pre-training, post-training, and 1-month follow-up were significantly different (12 [8.75-15], 18 [13.75-22], and 18.5 [14-23], respectively). The percentages of correct answers also differed significantly between the time assessments. Study participants significantly improved their performance between pre-training and post-training (P < .001). This performance was maintained after a 1-month follow-up (P = .95) for the questions related to the detection, determination of cause, and management of patient-ventilator asynchrony. CONCLUSIONS: A specific 36-h training program significantly improved the ability of health-care professionals to detect patient-ventilator asynchrony, determine the possible causes of patient-ventilator asynchrony, and properly manage different types of patient-ventilator asynchrony.

Occurrence, co-occurrence and persistence of symptoms of depression and post-traumatic stress disorder in survivors of COVID-19 critical illness.

Background: Intensive care unit (ICU) admission and invasive mechanical ventilation (IMV) are associated with psychological distress and trauma. The COVID-19 pandemic brought with it a series of additional long-lasting stressful and traumatic experiences. However, little is known about comorbid depression and post-traumatic stress disorder (PTSD).Objective: To examine the occurrence, co-occurrence, and persistence of clinically significant symptoms of depression and PTSD, and their predictive factors, in COVID-19 critical illness survivors.Method: Single-centre prospective observational study in adult survivors of COVID-19 with ≥24 h of ICU admission. Patients were assessed one and 12 months after ICU discharge using the depression subscale of the Hospital Anxiety and Depression Scale and the Davidson Trauma Scale. Differences in isolated and comorbid symptoms of depression and PTSD between patients with and without IMV and predictors of the occurrence and persistence of symptoms of these mental disorders were analysed.Results: Eighty-nine patients (42 with IMV) completed the 1-month follow-up and 71 (34 with IMV) completed the 12-month follow-up. One month after discharge, 29.2% of patients had symptoms of depression and 36% had symptoms of PTSD; after one year, the respective figures were 32.4% and 31%. Coexistence of depressive and PTSD symptoms accounted for approximately half of all symptomatic cases. Isolated PTSD symptoms were more frequent in patients with IMV (p≤.014). The need for IMV was associated with the occurrence at one month (OR = 6.098, p = .005) and persistence at 12 months (OR = 3.271, p = .030) of symptoms of either of these two mental disorders.Conclusions: Comorbid depressive and PTSD symptoms were highly frequent in our cohort of COVID-19 critical illness survivors. The need for IMV predicted short-term occurrence and long-term persistence of symptoms of these mental disorders, especially PTSD symptoms. The specific role of dyspnea in the association between IMV and post-ICU mental disorders deserves further investigation.Trial registration: ClinicalTrials.gov identifier: NCT04422444.

Clinically significant depressive and post-traumatic stress disorder symptoms in survivors of COVID-19 critical illness, especially in patients who had undergone invasive mechanical ventilation, were highly frequent, occurred soon after discharge, and persisted over the long term.

Tratamento prévio com adalimumabe reduz lesão pulmonar induzida por ventilação mecânica em um modelo experimental / Pretreatment with adalimumab reduces ventilator-induced lung injury in an experimental model

RESUMO Objetivo: Determinar se a administração de adalimumabe previamente à ventilação mecânica reduz a lesão pulmonar induzida por ventilação mecânica. Métodos: Randomizaram-se 18 ratos em três grupos submetidos à ventilação mecânica por 3 horas com uma fração inspirada de oxigênio de 0,40%. Os três grupos foram assim caracterizados: um grupo com baixo volume corrente (n = 6), no qual se utilizaram volume corrente de 8mL/kg e pressão expiratória final positiva de 5cmH2O; um grupo com alto volume corrente (n = 6), no qual se utilizaram volume corrente de 35mL/kg e pressão expiratória final positiva de zero; e um grupo pré-tratado com alto volume corrente (n = 6), no qual se administraram adalimumabe (100µg/kg) por via intraperitoneal 24 horas antes do início da ventilação mecânica, volume corrente de 35mL/kg e pressão expiratória final positiva de zero. Realizou-se ANOVA para comparação de dano histológico (com utilização de escores segundo o ATS 2010 Lung Injury Scoring System), edema pulmonar, complacência pulmonar, pressão parcial de oxigênio arterial e pressão arterial média entre os grupos. Resultados: Após 3 horas de ventilação, o escore médio de lesão histológica pulmonar foi mais elevado no grupo com alto volume corrente do que no grupo com baixo volume corrente (0,030 versus 0,0051; p = 0,003). O grupo com alto volume corrente demonstrou complacência pulmonar diminuída após 3 horas (p = 0,04) e hipoxemia (p = 0,018 versus controle). O grupo alto volume corrente tratado previamente teve melhora do escore histológico, principalmente devido à redução significante da infiltração leucocitária (p = 0,003). Conclusão: O exame histológico após 3 horas de ventilação lesiva revelou lesão pulmonar induzida por ventilação mecânica na ausência de modificações mensuráveis na mecânica pulmonar e na oxigenação; a administração de adalimumabe antes da ventilação mecânica diminuiu o edema pulmonar e o dano histológico.

ABSTRACT Objective: To determine whether adalimumab administration before mechanical ventilation reduces ventilator-induced lung injury (VILI). Methods: Eighteen rats randomized into 3 groups underwent mechanical ventilation for 3 hours with a fraction of inspired oxygen = 0.40% including a low tidal volume group (n = 6), where tidal volume = 8mL/kg and positive end-expiratory pressure = 5cmH2O; a high tidal volume group (n = 6), where tidal volume = 35mL/kg and positive end-expiratory pressure = 0; and a pretreated + high tidal volume group (n = 6) where adalimumab (100ug/kg) was administered intraperitoneally 24 hours before mechanical ventilation + tidal volume = 35mL/kg and positive end-expiratory pressure = 0. ANOVA was used to compare histological damage (ATS 2010 Lung Injury Scoring System), pulmonary edema, lung compliance, arterial partial pressure of oxygen, and mean arterial pressure among the groups. Results: After 3 hours of ventilation, the mean histological lung injury score was higher in the high tidal volume group than in the low tidal volume group (0.030 versus 0.0051, respectively, p = 0.003). The high tidal volume group showed diminished lung compliance at 3 hours (p = 0.04) and hypoxemia (p = 0,018 versus control). Pretreated HVt group had an improved histological score, mainly due to a significant reduction in leukocyte infiltration (p = 0.003). Conclusion: Histological examination after 3 hours of injurious ventilation revealed ventilator-induced lung injury in the absence of measurable changes in lung mechanics or oxygenation; administering adalimumab before mechanical ventilation reduced lung edema and histological damage.

Volumetric capnography in acute respiratory distress syndrome / Capnografia volumétrica na síndrome do desconforto respiratório agudo

Volumetric capnography is especially sensitive to disturbances affecting the efficiency of ventilation for gas exchange. Because lung homogeneity is a very fragile property, it is endangered in the majority of diseases that affect the airways, lung parenchyma, or alveolar microcirculation. Acute lung injury and acute respiratory distress syndrome can be conveniently monitored with volumetric capnography. The combination of two advanced technologies—airway flow monitoring and mainstream capnography—allows breath-by-breath bedside computerized determination of the physiological dead space, alveolar heterogeneity, and CO2 elimination. The use of volumetric capnography at the bedside can provide clinicians with important physiological and prognostic data, as well as allowing the effects of therapeutic interventions to be evaluated in critical ill patients receiving mechanical ventilation.

A capnografia volumétrica é especialmente sensível aos problemas que afetam a eficiência da ventilação para a troca gasosa. Uma vez que a homogeneidade do pulmão é uma propriedade muito frágil, a medida da capnografia é um desafio na maioria das doenças que comprometem as vias aéreas, o parênquima pulmonar e a microcirculação alveolar. A lesão pulmonar aguda e síndrome do desconforto respiratório agudo são situações que devem ser monitoradas com a capnografia volumétrica. Essa tecnologia avançada é uma combinação da medida do fluxo aéreo e a capnografia convencional, fazendo com que seja possível computar, à beira do leito, parâmetros como espaço morto, heterogeneidade alveolar e eliminação do CO2. O uso da capnografia volumétrica à beira do leito pode fornecer aos clínicos importantes informações fisiológicas e sobre o prognóstico, assim como seguir o efeito de intervenções terapêuticas nos doentes críticos ventilados mecanicamente.