CHEST critical care最新文献

Status asthmaticus is refractory bronchospasm that can result in hypercarbia, altered mental status, respiratory failure, hypoxemia, and death. The care of patients with status asthmaticus often requires care in the ICU and, in rare circumstances, consideration of extracorporeal membrane oxygenation (ECMO) or extracorporeal CO₂ removal support. Compared with ARDS, status asthmaticus is a relatively rare indication for venovenous ECMO and, to our knowledge, its use has not been examined in prospective multicenter studies. As ECMO becomes more widely available, it may be valuable for providers to understand better its role in the management of status asthmaticus. In this edition of “How I Do It,” we provide a case example of life-threatening status asthmaticus to discuss unique considerations in the care of these patients with this complex disease.

Background

Ventilation and sedation are used for the management of acute hypoxemic respiratory failure (AHRF), but their optimal combination to minimize the risks of ventilation is not well understood.

Research Question

What are the individual effects and interactions of inspiratory and positive end-expiratory pressure (PEEP), sedation, and venovenous extracorporeal membrane oxygenation (VV-ECMO) on respiratory drive, effort, and lung-distending pressure in patients with AHRF triggering the ventilator?

Study Design and Methods

In this secondary exploratory analysis of a trial of lung and diaphragm protection in AHRF, inspiratory pressure, sedation, PEEP, and VV-ECMO were titrated while respiratory drive (airway pressure in the first 100 ms [P_0.1]), effort (esophageal pressure swing [|ΔPes|]), and lung-distending pressure (dynamic transpulmonary driving pressure [ΔP_L,dyn]) were recorded. Associations were evaluated using linear mixed-effects regression models including prespecified terms for potential interactions.

Results

The study included 223 individual measurements of P_0.1 and 235 individual measurements of |ΔPes| and ΔP_L,dyn from 30 patients. Propofol-attenuated P_0.1 (–0.4 cm H₂O; 95% CI, –0.3 to –0.1 cm H₂O per 10-μm/kg/min increase), |ΔPes| (–2.5 cm H₂O; 95% CI, –3.4 to –1.7 cm H₂O per 10-μm/kg/min increase), and ΔP_L,dyn (–1.6 cm H₂O; 95% CI, –2.3 to –0.8 cm H₂O per 10-μm/kg/min increase). The effect of inspiratory pressure on |ΔPes| varied depending on propofol dose: with higher propofol dose, inspiratory pressure resulted in higher ΔP_L,dyn. With VV-ECMO, patients (n = 16) showed significantly lower |ΔPes| (–10 cm H₂O; 95% CI, –17.5 to –2.5 cm H₂O) and required less sedation to reduce |ΔPes| than without VV-ECMO (n = 14).

Interpretation

Mechanical ventilation, sedation, and VV-ECMO exert interdependent effects on respiratory drive, effort, and lung-distending pressure in AHRF. Patients receiving VV-ECMO require less sedation to control respiratory effort.

Background

Metabolic derangements in sepsis stem from mitochondrial injury and contribute to organ dysfunction and mortality. Thus, repair of mitochondrial damage seems pivotal for recovery and determining clinical outcome in sepsis. However, reliable biomarkers assessing mitochondrial repair noninvasively in peripheral blood are currently lacking.

Research Question

Are different gene transcripts related to mitochondrial repair (ie, biogenesis, fusion, fission, mitophagy) and the protein interaction assessing mitochondrial biogenesis, both measured in peripheral blood, associated with disease severity and clinical outcome?

Study Design and Methods

Healthy control patients (n = 22), uninfected critically ill control patients (n = 13), and patients with sepsis (n = 75) were included in this prospective multicentric observational study. Gene products of mitochondrial quality control and mitochondrial DNA were measured on day 1 and 4 in peripheral blood mononuclear cells. In addition, we assessed in the same samples the mitochondrial protein interaction of mitochondrial transcription factor A (TFAM)-mitochondrial transcription factor B2 (TFB2M) using a proximity ligation assay. Patients with sepsis were stratified in the outcome-related subgroups ICU-free within 1 week (n = 16), not ICU-free within 1 week (n = 36), and 30-day nonsurvivors (n = 23).

Results

Transcript levels of the assessed messenger RNA markers of patients with sepsis were not associated with disease severity nor did they predict clinical outcome. Strikingly, the mitochondrial protein interaction of TFAM-TFB2M on day 4 (P < .05) and the difference between day 1 and 4 (P < .001) allowed stratification in the three clinical outcome subgroups. In addition, a decline in TFAM-TFB2M protein interactions between day 1 and 4 was an independent predicator for 30-day mortality (adjusted hazard ratio, 8.34; 95% CI, 2.73-25.45; P < .001).

Interpretation

Patients with sepsis with an early activation of mitochondrial biogenesis were more likely to be ICU-free within 1 week. A mitochondrial and clinical recovery can be assessed via the protein interaction of TFAM-TFB2M in peripheral blood. Thus, mitochondrial protein interactions targeting mitochondrial biogenesis provide a promising dimension of novel biomarkers assessing mitochondrial dysfunction in sepsis.

Case Presentation

A 48-year-old man with history of recent travel to central Mexico and immunosuppression sought treatment with a 1-month-long history of progressive headache, fatigue, word-finding difficulties, and night sweats. The patient had a history of end-stage renal disease; he had undergone a kidney transplantation 7 years prior with good graft function with immunosuppression with tacrolimus, everolimus, and low-dose prednisone. At an outside hospital, he recently had been treated with empiric antibiotics for meningitis, but these were discontinued given the low suspicion for a bacterial cause. After discharge, he continued to have headaches, limited oral intake, persistent nausea, urinary frequency, and falls, prompting him to seek treatment at the ED. Physical examination findings were benign aside from disorientation. Laboratory workup was significant for hyponatremia of 122 mM, creatinine of 1.4 mg/dL (baseline, 1.4-1.5 mg/dL), WBC count of 7.2 10⁹/L, hemoglobin of 13 g/dL, and platelet count of 349 10⁹/L. Neither tacrolimus nor everolimus levels were supratherapeutic.

Background

The exploration of subphenotypes in hospitalized patients with COVID-19 has garnered substantial attention. Most existing studies operate under the assumption of heterogeneity in COVID-19 patient populations, and this assumption can lead to erroneous conclusions.

Research Question

Do plasma biomarker profiles reflective of various pathophysiologic pathways provide evidence for heterogeneity in hospitalized patients with COVID-19?

Study Design and Methods

This is a secondary analysis of two prospective observational studies of adult patients hospitalized with COVID-19-related respiratory failure in the general ward and ICU of two medical centers and with 44 host response biomarkers available. Parsimonious models were used to allocate and validate ARDS inflammatory subphenotypes. Novel biological subphenotypes were identified using latent profile analysis (LPA) and hierarchical clustering. Heterogeneity of treatment effect for corticosteroids was assessed using an interaction term in a logistic regression model.

Results

The cohort consisted of 162 patients admitted to the ICU and 464 patients admitted to the ward. Using the parsimonious models in ICU patients, only 3.1% to 13% of patients were classified as hyperinflammatory subphenotype. Using de novo subphenotyping techniques, neither clustering nor LPA revealed significant evidence for heterogeneity in the ward (P = .11-.13), ICU (P = .23-.88), or combined cohort (P = .05-.88). Adding clinical variables did not alter results in the ICU or combined cohort. Using the combined approach in the ward cohort, indices provided borderline significance for two subphenotypes, and there was good agreement between clustering and LPA (87.9%), but no heterogeneity of treatment effect for corticosteroids was observed between these two classes (P = .198).

Interpretation

Systemic inflammatory subphenotypes derived from patients with ARDS did not reflect the variation in severity of COVID-19 in this study. Empirical evidence, derived from cluster analysis or LPA, offers limited support for biological heterogeneity in COVID-19.

Background

Post-COVID-19 clinics were implemented to improve postacute care for patients with COVID-19, including survivors of critical illness, many of whom experience post-intensive care syndrome (PICS). Whether postacute care changed over the course of the pandemic and if inequities in utilization exist remain unclear.

Research Question

Among survivors of COVID-19 critical illness, what were the patterns of postdischarge care during different pandemic waves, and are there inequities in outpatient utilization?

Study Design and Methods

In this retrospective cohort study, we describe sociodemographics, illness severity, outpatient utilization, and PICS burden up to 18 months after discharge for patients with COVID-19 admitted to an ICU at three Boston, Massachusetts, area hospitals during two waves (wave 1 and wave 2) of hospitalizations during the pandemic. Multivariable logistic regression models identified variables associated with follow-up in post-COVID-19 clinics and adverse postdischarge health care outcomes, including readmissions, ED visits, and all-cause postdischarge mortality.

Results

A total of 319 of 478 wave 1 patients (66.7%) and 80 of 187 wave 2 patients (42.8%) survived to hospital discharge. During wave 1, there was a higher proportion of patients with limited English proficiency (LEP) admitted to the ICU (45.5% vs 30.0%, P = .012) and a lower severity of illness on admission (Sequential Organ Failure Assessment score 4; interquartile range, 2-8 vs 6; interquartile range, 4-8; P = .013). PICS symptoms were common across both waves (80.6% vs 78.8%, P = .72). In multivariable analyses, LEP was associated with decreased odds of post-COVID-19 clinic follow-up (adjusted OR, 0.80; 95% CI, 0.70-0.92; P < .01) and increased odds of adverse postdischarge health care outcomes (adjusted OR, 1.49; 95% CI, 1.11-2.0; P < .01).

Interpretation

The overall burden of PICS was high across waves. LEP was associated with inequities in post-COVID-19 clinic follow-up and worse postdischarge outcomes, suggesting that language is an important target for further interventions to support equitable recovery after critical illness.

Background

Psychological distress symptoms are present and persistent among many patients who survive a critical illness like COVID-19.

Research Question

Could a self-directed mobile app-delivered mindfulness intervention be feasibly and rapidly implemented within a clinical trials network to reduce distress symptoms?

Study Design and Methods

A randomized clinical trial was conducted between January 2021 and May 2022 at 29 US sites and included survivors of hospitalization due to COVID-19-related illness with elevated symptoms of depression at discharge. Participants were randomized to intervention or usual care control. The intervention consisted of four themed weeks of daily audio, video, and text content. All study procedures were virtual. The primary outcome was depression symptoms assessed with the Patient Health Questionnaire 9 at 3 months. Secondary outcomes included anxiety (Generalized Anxiety Disorder 7-item scale), quality of life (EQ-5D), and adherence. We used general linear models to estimate treatment arm differences in outcomes over time.

Results

Among 56 randomized participants (mean age ± SD, 51.0 ± 13.2 years; 38 female [67.9%]; 14 Black participants [25%]), 45 (intervention: n = 23 [79%]; control: n = 22 [81%]) were retained at 6 months. There was no difference in mean improvement between intervention and control participants at 3 months in Patient Health Questionnaire 9 (−0.5 vs 0.1), Generalized Anxiety Disorder 7-item scale (−0.3 vs 0.1), or EQ-5D (−0.03 vs 0.02) scores, respectively; 6-month results were similar. Only 15 participants (51.7%) initiated the intervention, whereas the mean number ± SD of the 56 prescribed intervention activities completed was 12.0 ± 15.2. Regulatory approvals delayed trial initiation by nearly a year.

Interpretation

Among survivors of COVID-19 hospitalization with elevated psychological distress symptoms, a self-directed mobile app-based mindfulness intervention had poor adherence. Future psychological distress interventions mobilized at broad scale should focus efforts on patient engagement and regulatory simplification to enhance success.

Trial Registration

ClinicalTrials.gov; No.: NCT04581200; URL: www.clinicaltrials.gov

ICU-acquired weakness (ICU-AW) impacts up to 40% of patients admitted to the ICU and can have long-lasting effects on those who survive an ICU stay. In the last decade, early mobilization (EM) has emerged as an intervention to help prevent or to mitigate ICU-AW, or both, and to improve functional outcomes for patients. Despite its feasibility, safety, and potential benefits, a large gap in implementation of EM in ICUs globally remains. The purpose of this article is to review ICU-AW, to discuss the evidence base and current guidelines about EM, and to offer a practical approach for EM implementation with an emphasis on patient safety and common barriers.

Background

Prone positioning is a historically underused evidence-based practice for ARDS. Despite increased prone positioning during the COVID-19 pandemic, some patients may remain at risk of nonuse.

Research Question

What is the current evidence-based gap for prone positioning in ARDS, how is use changing over time, and what are patient-level barriers and facilitators to prone positioning?

Study Design and Methods

This retrospective cohort included invasively ventilated adults with ARDS and who met prone positioning criteria from six hospitals. The rate of prone positioning among eligible patients was summarized from January 2018 through December 2021. Segmented Poisson regression was used to describe temporal trends. Logistic regression was used to identify patient-level barriers and facilitators to prone positioning.

Results

Seven hundred ninety-nine patients fulfilled criteria for prone positioning. The mean age was 57 years, 125 patients (15.6%) had COVID-19, mean ICU stay was 19.5 days, and the mortality rate was 50.1%. Prone positioning was used in 297 of 799 patients (37.2%). Prone positioning was increasing before the pandemic with a relative rate (RR) of 1.12 per quarter (95% CI, 1.03-1.22). Prone positioning increased during the pandemic vs before the pandemic (RR, 1.62; 95% CI, 1.02-2.61), but not for patients with nonrespiratory diagnoses causing ARDS (RR, 0.74; 95% CI, 0.22-2.52). Barriers to prone positioning included vasopressor use (OR for withholding prone positioning, 1.15 per 0.1 μm/kg/min norepinephrine equivalent; 95% CI, 1.06-1.26), age (OR, 1.12 per 5 years; 95% CI, 1.03-1.22), and having undergone surgery (OR, 2.41; 95% CI, 1.00-5.81). Facilitators included having COVID-19 (OR for withholding prone positioning, 0.10; 95% CI, 0.04-0.24) or another respiratory illness (OR, 0.42; 95% CI, 0.23-0.79), and receiving neuromuscular blockade (OR, 0.22; 95% CI, 0.13-0.38).

Interpretation

Despite increased prone positioning during the COVID-19 pandemic, an evidence-based gap persists, especially for patients with nonrespiratory causes of ARDS. Multiple barriers and facilitators must be targeted to increase prone positioning.

Background

Systemic corticosteroid use in acute respiratory failure has yielded uncertain benefits, partially because of off-target side effects. Inhaled corticosteroids (ICSs) confer localized antiinflammatory benefits and may protect adults with direct lung injury (DLI) from developing respiratory failure. To our knowledge, this relationship has not been studied in children.

Research Question

Do children with DLI who are prescribed ICSs before hospitalization have lower odds of progressing to respiratory failure?

Study Design and Methods

This retrospective, single-center cohort identified children seeking treatment at the ED with DLI and medication records before hospitalization. The primary outcome was intubation; secondary outcomes included noninvasive respiratory support (NRS). We tested the association of ICSs with intubation and NRS, adjusting for confounders. We stratified analyses on history of asthma and performed a sensitivity analysis adjusting for systemic corticosteroid use to account for status asthmaticus.

Results

Of 35,220 patients, 17,649 patients (50%) were prescribed ICSs. Intubation occurred in 169 patients (73 patients receiving ICSs) and NRS was used in 3,582 patients (1,336 patients receiving ICS). ICS use was associated with lower intubation (adjusted OR, 0.46; 95% CI, 0.31-0.67) and NRS (aOR, 0.45; 95% CI, 0.40-0.49). The association between ICS and NRS differed according to history of asthma (P = .04 for interaction), with ICS exposure remaining protective only for patients with a history of asthma. Results held true in sensitivity analyses.

Interpretation

ICS use prior to hospitalization may protect children with DLI from progressing to respiratory failure, with possible differential efficacy according to history of asthma.