ABSTRACT
BACKGROUND: Automated analysis of lung computed tomography (CT) scans may help characterize subphenotypes of acute respiratory illness. We integrated lung CT features measured via deep learning with clinical and laboratory data in spontaneously breathing subjects to enhance the identification of COVID-19 subphenotypes. METHODS: This is a multicenter observational cohort study in spontaneously breathing patients with COVID-19 respiratory failure exposed to early lung CT within 7 days of admission. We explored lung CT images using deep learning approaches to quantitative and qualitative analyses; latent class analysis (LCA) by using clinical, laboratory and lung CT variables; regional differences between subphenotypes following 3D spatial trajectories. RESULTS: Complete datasets were available in 559 patients. LCA identified two subphenotypes (subphenotype 1 and 2). As compared with subphenotype 2 (n = 403), subphenotype 1 patients (n = 156) were older, had higher inflammatory biomarkers, and were more hypoxemic. Lungs in subphenotype 1 had a higher density gravitational gradient with a greater proportion of consolidated lungs as compared with subphenotype 2. In contrast, subphenotype 2 had a higher density submantellar-hilar gradient with a greater proportion of ground glass opacities as compared with subphenotype 1. Subphenotype 1 showed higher prevalence of comorbidities associated with endothelial dysfunction and higher 90-day mortality than subphenotype 2, even after adjustment for clinically meaningful variables. CONCLUSIONS: Integrating lung-CT data in a LCA allowed us to identify two subphenotypes of COVID-19, with different clinical trajectories. These exploratory findings suggest a role of automated imaging characterization guided by machine learning in subphenotyping patients with respiratory failure. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04395482. Registration date: 19/05/2020.
Subject(s)
COVID-19 , Lung , Phenotype , Respiratory Insufficiency , Tomography, X-Ray Computed , Humans , COVID-19/diagnostic imaging , COVID-19/physiopathology , Tomography, X-Ray Computed/methods , Female , Male , Middle Aged , Lung/diagnostic imaging , Lung/physiopathology , Aged , Respiratory Insufficiency/diagnostic imaging , Respiratory Insufficiency/etiology , Respiratory Insufficiency/physiopathology , Cohort Studies , AdultABSTRACT
BACKGROUND: Lung perfusion defects, mainly due to endothelial and coagulation activation, are a key contributor to COVID-19 respiratory failure. COVID-19 patients may also develop acute kidney injury (AKI) because of renal perfusion deficit. We aimed to explore AKI-associated factors and the independent prediction of standardized minute ventilation (MV)-a proxy of alveolar dead space-on AKI onset and persistence in COVID-19 mechanically ventilated patients. METHODS: This is a multicenter observational cohort study. We enrolled 157 COVID-19 patients requiring mechanical ventilation and intensive care unit (ICU) admission. We collected clinical information, ventilation, and laboratory data. AKI was defined by the 2012 KDIGO guidelines and classified as transient or persistent according to serum creatinine criteria persistence within 48 h. Ordered univariate and multivariate logistic regression analyses were employed to identify variables associated with AKI onset and persistence. RESULTS: Among 157 COVID-19 patients on mechanical ventilation, 47% developed AKI: 10% had transient AKI, and 37% had persistent AKI. The degree of hypoxia was not associated with differences in AKI severity. Across increasing severity of AKI groups, despite similar levels of paCO2, we observed an increased MV and standardized MV, a robust proxy of alveolar dead space. After adjusting for other clinical and laboratory covariates, standardized MV remained an independent predictor of AKI development and persistence. D-dimer levels were higher in patients with persistent AKI. CONCLUSIONS: In critically ill COVID-19 patients with respiratory failure, increased wasted ventilation is independently associated with a greater risk of persistent AKI. These hypothesis-generating findings may suggest that perfusion derangements may link the pathophysiology of both wasted ventilation and acute kidney injury in our population.
ABSTRACT
BACKGROUND: Patients with high spinal cord injury (SCI) are unable to breathe on their own and require mechanical ventilation (MV). The long-term use of MV is associated with increased morbidity and mortality. In patients with intact phrenic nerve function, patients can be partially or completely removed from MV by directly stimulating the diaphragm motor points with a diaphragm pacing system (DPS). OBJECTIVES: We describe our multicenter European experience using DPS in SCI patients who required MV. METHODS: We conducted a retrospective study of patients who were evaluated for the implantation of DPS. Patients evaluated for DPS who met the prospectively defined criteria of being at least 1 year of age, and having cervical injury resulting in a complete or partial dependency on MV were included. Patients who received DPS implants were followed for up to 1 year for device usage and safety. RESULTS: Across 3 centers, 47 patients with high SCI were evaluated for DPS, and 34 were implanted. Twenty-one patients had 12 months of follow-up data with a median DPS use of 15 h/day (interquartile range 4, 24). Eight patients (38.1%) achieved complete MV weaning using DPS 24 h/day. Two DPS-related complications were surgical device revision and a wire eruption. No other major complications were associated with DPS use. CONCLUSIONS: Diaphragm pacing represents an attractive alternative stand-alone treatment or adjunctive therapy compared to MV in patients with high SCI. After a period of acclimation, the patients were able to reduce the daily use of MV, and many could be completely removed from MV.
Subject(s)
Electric Stimulation Therapy , Spinal Cord Injuries , Diaphragm , Humans , Respiration, Artificial/adverse effects , Retrospective Studies , Spinal Cord Injuries/complications , Spinal Cord Injuries/surgery , Ventilator Weaning/methodsABSTRACT
BACKGROUND: The performances of 2 noninvasive CPAP systems (high flow and low flow air-entrainment masks) were compared to the Boussignac valve in 3 different scenarios. METHODS: Scenario 1: pneumatic lung simulator with a tachypnea pattern (tidal volume 800 mL at 40 breaths/min). Scenario 2: Ten healthy subjects studied during tidal breaths and tachypnea. Scenario 3: Twenty ICU subjects enrolled for a noninvasive CPAP session. Differences between set and effective CPAP level and F(IO(2)), as well as the lowest airway pressure and the pressure swing around the imposed CPAP level, were analyzed. The lowest airway pressure and swing were correlated to the pressure-time product (area of the airway pressure curve below the CPAP level) measured with the simulator. P(aO(2)) was a subject's further performance index. RESULTS: Lung simulator: Boussignac F(IO(2)) was 0.54, even if supplied with pure oxygen. The air-entrainment masks had higher swing than the Boussignac (P = .007). Pressure-time product correlated better with pressure swing (Spearman correlation coefficient [ρ] = 0.97) than with lowest airway pressure (ρ = 0.92). In healthy subjects, the high-flow air-entrainment mask showed lower difference between set and effective F(IO(2)) (P < .001), and lowest airway pressure (P < .001), compared to the Boussignac valve. In all measurements the Boussignac valve showed higher than imposed CPAP level (P < .001). In ICU subjects the high-flow mask had lower swing than the Boussignac valve (P = .03) with similar P(aO(2)) increase. CONCLUSIONS: High-flow air-entrainment mask showed the best performance in human subjects. During high flow demand, the Boussignac valve delivered lower than expected F(IO(2)) and showed higher dynamic hyper-pressurization than the air-entrainment masks.
Subject(s)
Continuous Positive Airway Pressure/instrumentation , Inhalation/physiology , Masks , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Oxygen/blood , Pressure , Respiratory Rate , Tidal Volume , Ventilator Weaning , Young AdultABSTRACT
Non invasive ventilation (NIV) has been shown to be an effective therapy in selected patients with acute respiratory failure. Due to its benefit and relative ease of use, NIV is frequently used. In addition, the shortage and high cost of intensive care beds have prompted the use of NIV outside the intensive care unit. Choosing the right time and type of patient with acute respiratory failure to improve the chances of success with NIV requires an appropriate environment and monitoring. This review presents and discusses the currently available data regarding NIV success outside the intensive care unit, the optimal ventilatory strategy, possible solutions to the mechanical problems and the minimum monitoring required.
Subject(s)
Respiration, Artificial , Respiratory Insufficiency/therapy , Acute Disease , Humans , Monitoring, PhysiologicABSTRACT
Descending necrotising mediastinitis is a severe infection spreading from the cervical region to the mediastinal connective tissue. It represents a virulent form of mediastinal infection, requiring prompt diagnosis and treatment to reduce the high mortality associated. An optimal debridement and drainage through an open thoracotomy access are the keys for a successful outcome. Two patients, males, 70 and 75-years-old with descending necrotising mediastinitis were treated in our Institution in April '05. One had an odontogenic abscess and the other had a retropharyngeal abscess. Operative procedures included thoracotomy with radical surgical debridement of the mediastinum and excision of necrotic tissue associated with transcervical surgical debridement and drainage. Postoperatively mediastinum-pleural and cervical irrigation with iodopovidone 2 per thousand was performed until a culture of pleural effusion become negative. Postoperatively both patients suffered from severe complication including septic shock and acute respiratory distress syndrome. The 70-years-old patient had an acute renal failure too. Postoperatively the length of the intensive care unit stay was 40 and 42 days, respectively. The outcome was favorable in both patients. Early detection and immediate open surgical treatment could be the best way to reduce morbidity and mortality rate. Descending Necrotising Mediastinitis cannot be adequately treated without mediastinal and cervical excision of necrotic tissue and drainage including an open thoracic and cervical approach.
Subject(s)
Mediastinitis/pathology , Mediastinitis/surgery , Thoracotomy/methods , Aged , Debridement , Drainage , Humans , Male , Mediastinitis/microbiology , Necrosis , Periodontal Abscess/complications , Retropharyngeal Abscess/complications , Therapeutic Irrigation , Treatment OutcomeABSTRACT
STUDY OBJECTIVES: Noninvasive ventilation, although effective as treatment for patients with acute cardiogenic pulmonary edema when prolonged for hours, is of limited use in the emergency department (ED). The aim of the study was to determine whether a short attempt at noninvasive pressure support ventilation avoids ICU admittance and to identify lack of response prediction variables. DESIGN: Prospective inception cohort study. SETTING: ED of a university hospital. PATIENTS: Fifty-eight consecutive patients with cardiogenic pulmonary edema who had been unresponsive to medical treatment and were admitted between January 1999 and December 2000. INTERVENTIONS: Pressure support ventilation was instituted through a full-face mask until the resolution of respiratory failure. A 15-min "weaning test" was performed to evaluate clinical stability. Responder patients were transferred to a medical ward. Nonresponding patients were intubated and were admitted to the ICU. MAIN OUTCOME MEASURES: The included optimal length of intervention, the avoidance of ICU admittance, the incidence of myocardial infarction, and predictive lack of response criteria. RESULTS: Patients completed the trial (mean [+/- SD] duration, 96 +/- 40 min). None of the responders (43 patients; 74%) was subsequently ventilated or was admitted to the ICU. Two new episodes of myocardial infarction were observed. Thirteen of 58 patients died. A mean arterial pressure of < 95 mm Hg (odds ratio [OR], 10.6; 95% confidence interval [CI], 1.8 to 60.8; p < 0.01) and COPD (OR, 9.4; 95% CI, 1.6 to 54.0; p < 0.05) at baseline predicted the lack of response to noninvasive ventilation. CONCLUSIONS: A short attempt at noninvasive ventilation is effective in preventing invasive assistance. A 15-min weaning test can identify patients who will not need further invasive ventilatory support. COPD and hypotension at baseline are negative predictive criteria.