Predicting the Length of Mechanical Ventilation in Acute Respiratory Disease Syndrome Using Machine Learning: The PIONEER Study.

Background: The ability to predict a long duration of mechanical ventilation (MV) by clinicians is very limited. We assessed the value of machine learning (ML) for early prediction of the duration of MV > 14 days in patients with moderate-to-severe acute respiratory distress syndrome (ARDS). Methods: This is a development, testing, and external validation study using data from 1173 patients on MV ≥ 3 days with moderate-to-severe ARDS. We first developed and tested prediction models in 920 ARDS patients using relevant features captured at the time of moderate/severe ARDS diagnosis, at 24 h and 72 h after diagnosis with logistic regression, and Multilayer Perceptron, Support Vector Machine, and Random Forest ML techniques. For external validation, we used an independent cohort of 253 patients on MV ≥ 3 days with moderate/severe ARDS. Results: A total of 441 patients (48%) from the derivation cohort (n = 920) and 100 patients (40%) from the validation cohort (n = 253) were mechanically ventilated for >14 days [median 14 days (IQR 8-25) vs. 13 days (IQR 7-21), respectively]. The best early prediction model was obtained with data collected at 72 h after moderate/severe ARDS diagnosis. Multilayer Perceptron risk modeling identified major prognostic factors for the duration of MV > 14 days, including PaO2/FiO2, PaCO2, pH, and positive end-expiratory pressure. Predictions of the duration of MV > 14 days showed modest discrimination [AUC 0.71 (95%CI 0.65-0.76)]. Conclusions: Prolonged MV duration in moderate/severe ARDS patients remains difficult to predict early even with ML techniques such as Multilayer Perceptron and using data at 72 h of diagnosis. More research is needed to identify markers for predicting the length of MV. This study was registered on 14 August 2023 at ClinicalTrials.gov (NCT NCT05993377).

A genome-wide association study of survival in patients with sepsis.

BACKGROUND: Sepsis is a severe systemic inflammatory response to infections that is accompanied by organ dysfunction and has a high mortality rate in adult intensive care units. Most genetic studies have identified gene variants associated with development and outcomes of sepsis focusing on biological candidates. We conducted the first genome-wide association study (GWAS) of 28-day survival in adult patients with sepsis. METHODS: This study was conducted in two stages. The first stage was performed on 687 European sepsis patients from the GEN-SEP network and 7.5 million imputed variants. Association testing was conducted with Cox regression models, adjusting by sex, age, and the main principal components of genetic variation. A second stage focusing on the prioritized genetic variants was performed on 2,063 ICU sepsis patients (1362 European Americans and 701 African-Americans) from the MESSI study. A meta-analysis of results from the two stages was conducted and significance was established at p < 5.0 × 10-8. Whole-blood transcriptomic, functional annotations, and sensitivity analyses were evaluated on the identified genes and variants. FINDINGS: We identified three independent low-frequency variants associated with reduced 28-day sepsis survival, including a missense variant in SAMD9 (hazard ratio [95% confidence interval] = 1.64 [1.37-6.78], p = 4.92 × 10-8). SAMD9 encodes a possible mediator of the inflammatory response to tissue injury. INTERPRETATION: We performed the first GWAS of 28-day sepsis survival and identified novel variants associated with reduced survival. Larger sample size studies are needed to better assess the genetic effects in sepsis survival and to validate the findings.

Respiratory Subsets in Patients with Moderate to Severe Acute Respiratory Distress Syndrome for Early Prediction of Death.

Introduction: In patients with acute respiratory distress syndrome (ARDS), the PaO2/FiO2 ratio at the time of ARDS diagnosis is weakly associated with mortality. We hypothesized that setting a PaO2/FiO2 threshold in 150 mm Hg at 24 h from moderate/severe ARDS diagnosis would improve predictions of death in the intensive care unit (ICU). Methods: We conducted an ancillary study in 1303 patients with moderate to severe ARDS managed with lung-protective ventilation enrolled consecutively in four prospective multicenter cohorts in a network of ICUs. The first three cohorts were pooled (n = 1000) as a testing cohort; the fourth cohort (n = 303) served as a confirmatory cohort. Based on the thresholds for PaO2/FiO2 (150 mm Hg) and positive end-expiratory pressure (PEEP) (10 cm H2O), the patients were classified into four possible subsets at baseline and at 24 h using a standardized PEEP-FiO2 approach: (I) PaO2/FiO2 ≥ 150 at PEEP < 10, (II) PaO2/FiO2 ≥ 150 at PEEP ≥ 10, (III) PaO2/FiO2 < 150 at PEEP < 10, and (IV) PaO2/FiO2 < 150 at PEEP ≥ 10. Primary outcome was death in the ICU. Results: ICU mortalities were similar in the testing and confirmatory cohorts (375/1000, 37.5% vs. 112/303, 37.0%, respectively). At baseline, most patients from the testing cohort (n = 792/1000, 79.2%) had a PaO2/FiO2 < 150, with similar mortality among the four subsets (p = 0.23). When assessed at 24 h, ICU mortality increased with an advance in the subset: 17.9%, 22.8%, 40.0%, and 49.3% (p < 0.0001). The findings were replicated in the confirmatory cohort (p < 0.0001). However, independent of the PEEP levels, patients with PaO2/FiO2 < 150 at 24 h followed a distinct 30-day ICU survival compared with patients with PaO2/FiO2 ≥ 150 (hazard ratio 2.8, 95% CI 2.2−3.5, p < 0.0001). Conclusions: Subsets based on PaO2/FiO2 thresholds of 150 mm Hg assessed after 24 h of moderate/severe ARDS diagnosis are clinically relevant for establishing prognosis, and are helpful for selecting adjunctive therapies for hypoxemia and for enrolling patients into therapeutic trials.

Clinical and biological markers for predicting ARDS and outcome in septic patients.

Sepsis is a common cause of acute respiratory distress syndrome (ARDS) associated with a high mortality. A panel of biomarkers (BMs) to identify septic patients at risk for developing ARDS, or at high risk of death, would be of interest for selecting patients for therapeutic trials, which could improve ARDS diagnosis and treatment, and survival chances in sepsis and ARDS. We measured nine protein BMs by ELISA in serum from 232 adult septic patients at diagnosis (152 required invasive mechanical ventilation and 72 had ARDS). A panel including the BMs RAGE, CXCL16 and Ang-2, plus PaO2/FiO2, was good in predicting ARDS (area under the curve = 0.88 in total septic patients). Best performing panels for ICU death are related to the presence of ARDS, need for invasive mechanical ventilation, and pulmonary/extrapulmonary origin of sepsis. In all cases, the use of BMs improved the prediction by clinical markers. Our study confirms the relevance of RAGE, Ang-2, IL-1RA and SP-D, and is novel supporting the inclusion of CXCL16, in BMs panels for predicting ARDS diagnosis and ARDS and sepsis outcome.

Whole-Blood Mitochondrial DNA Copies Are Associated With the Prognosis of Acute Respiratory Distress Syndrome After Sepsis.

Acute respiratory distress syndrome (ARDS) is an inflammatory process of the lungs that develops primarily in response to pulmonary or systemic sepsis, resulting in a disproportionate death toll in intensive care units (ICUs). Given its role as a critical activator of the inflammatory and innate immune responses, previous studies have reported that an increase of circulating cell-free mitochondrial DNA (mtDNA) is a biomarker for fatal outcome in the ICU. Here we analyzed the association of whole-blood mtDNA (wb-mtDNA) copies with 28-day survival from sepsis and sepsis-associated ARDS. We analyzed mtDNA data from 687 peripheral whole-blood samples within 24 h of sepsis diagnosis from unrelated Spanish patients with sepsis (264 with ARDS) included in the GEN-SEP study. The wb-mtDNA copies were obtained from the array intensities of selected probes, with 100% identity with mtDNA and with the largest number of mismatches with the nuclear sequences, and normalized across the individual-probe intensities. We used Cox regression models for testing the association with 28-day survival. We observed that wb-mtDNA copies were significantly associated with 28-day survival in ARDS patients (hazard ratio = 3.65, 95% confidence interval = 1.39-9.59, p = 0.009) but not in non-ARDS patients. Our findings support that wb-mtDNA copies at sepsis diagnosis could be considered an early prognostic biomarker in sepsis-associated ARDS patients. Future studies will be needed to evaluate the mechanistic links of this observation with the pathogenesis of ARDS.

Neurally adjusted ventilatory assist in acute respiratory failure: a randomized controlled trial.

PURPOSE: We hypothesized that neurally adjusted ventilatory assist (NAVA) compared to conventional lung-protective mechanical ventilation (MV) decreases duration of MV and mortality in patients with acute respiratory failure (ARF). METHODS: We carried out a multicenter, randomized, controlled trial in patients with ARF from several etiologies. Intubated patients ventilated for ≤ 5 days expected to require MV for ≥ 72 h and able to breathe spontaneously were eligible for enrollment. Eligible patients were randomly assigned based on balanced treatment assignments with a computerized randomization allocation sequence to two ventilatory strategies: (1) lung-protective MV (control group), and (2) lung-protective MV with NAVA (NAVA group). Allocation concealment was maintained at all sites during the trial. Primary outcome was the number of ventilator-free days (VFDs) at 28 days. Secondary outcome was all-cause hospital mortality. All analyses were done according to the intention-to-treat principle. RESULTS: Between March 2014 and October 2019, we enrolled 306 patients and randomly assigned 153 patients to the NAVA group and 153 to the control group. Median VFDs were higher in the NAVA than in the control group (22 vs. 18 days; between-group difference 4 days; 95% confidence interval [CI] 0 to 8 days; p = 0.016). At hospital discharge, 39 (25.5%) patients in the NAVA group and 47 (30.7%) patients in the control group had died (between-group difference - 5.2%, 95% CI - 15.2 to 4.8, p = 0.31). Other clinical, physiological or safety outcomes did not differ significantly between the trial groups. CONCLUSION: NAVA decreased duration of MV although it did not improve survival in ventilated patients with ARF.

Dexamethasone treatment for the acute respiratory distress syndrome: a multicentre, randomised controlled trial.

BACKGROUND: There is no proven specific pharmacological treatment for patients with the acute respiratory distress syndrome (ARDS). The efficacy of corticosteroids in ARDS remains controversial. We aimed to assess the effects of dexamethasone in ARDS, which might change pulmonary and systemic inflammation and result in a decrease in duration of mechanical ventilation and mortality. METHODS: We did a multicentre, randomised controlled trial in a network of 17 intensive care units (ICUs) in teaching hospitals across Spain in patients with established moderate-to-severe ARDS (defined by a ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen of 200 mm Hg or less assessed with a positive end-expiratory pressure of 10 cm H2O or more and FiO2 of 0·5 or more at 24 h after ARDS onset). Patients with brain death, terminal-stage disease, or receiving corticosteroids or immunosuppressive drugs were excluded. Eligible patients were randomly assigned based on balanced treatment assignments with a computerised randomisation allocation sequence using blocks of 10 opaque, sealed envelopes to receive immediate treatment with dexamethasone or continued routine intensive care (control group). Patients in the dexamethasone group received an intravenous dose of 20 mg once daily from day 1 to day 5, which was reduced to 10 mg once daily from day 6 to day 10. Patients in both groups were ventilated with lung-protective mechanical ventilation. Allocation concealment was maintained at all sites during the trial. Primary outcome was the number of ventilator-free days at 28 days, defined as the number of days alive and free from mechanical ventilation from day of randomisation to day 28. Secondary outcome was all-cause mortality 60 days after randomisation. All analyses were done according to the intention-to-treat principle. This study is registered with ClinicalTrials.gov, NCT01731795. FINDINGS: Between March 28, 2013, and Dec 31, 2018, we enrolled 277 patients and randomly assigned 139 patients to the dexamethasone group and 138 to the control group. The trial was stopped by the data safety monitoring board due to low enrolment rate after enrolling more than 88% (277/314) of the planned sample size. The mean number of ventilator-free days was higher in the dexamethasone group than in the control group (between-group difference 4·8 days [95% CI 2·57 to 7·03]; p<0·0001). At 60 days, 29 (21%) patients in the dexamethasone group and 50 (36%) patients in the control group had died (between-group difference -15·3% [-25·9 to -4·9]; p=0·0047). The proportion of adverse events did not differ significantly between the dexamethasone group and control group. The most common adverse events were hyperglycaemia in the ICU (105 [76%] patients in the dexamethasone group vs 97 [70%] patients in the control group), new infections in the ICU (eg, pneumonia or sepsis; 33 [24%] vs 35 [25%]), and barotrauma (14 [10%] vs 10 [7%]). INTERPRETATION: Early administration of dexamethasone could reduce duration of mechanical ventilation and overall mortality in patients with established moderate-to-severe ARDS. FUNDING: Fundación Mutua Madrileña, Instituto de Salud Carlos III, The European Regional Development's Funds, Asociación Científica Pulmón y Ventilación Mecánica.

A Prognostic Enrichment Strategy for Selection of Patients With Acute Respiratory Distress Syndrome in Clinical Trials.

OBJECTIVES: Incomplete or ambiguous evidence for identifying high-risk patients with acute respiratory distress syndrome for enrollment into randomized controlled trials has come at the cost of an unreasonable number of negative trials. We examined a set of selected variables early in acute respiratory distress syndrome to determine accurate prognostic predictors for selecting high-risk patients for randomized controlled trials. DESIGN: A training and testing study using a secondary analysis of data from four prospective, multicenter, observational studies. SETTING: A network of multidisciplinary ICUs. PATIENTS: We studied 1,200 patients with moderate-to-severe acute respiratory distress syndrome managed with lung-protective ventilation. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We evaluated different thresholds for patient's age, PaO2/FIO2, plateau pressure, and number of extrapulmonary organ failures to predict ICU outcome at 24 hours of acute respiratory distress syndrome diagnosis. We generated 1,000 random scenarios as training (n = 900, 75% of population) and testing (n = 300, 25% of population) datasets and averaged the logistic coefficients for each scenario. Thresholds for age (< 50, 50-70, > 70 yr), PaO2/FIO2 (≤ 100, 101-150, > 150 mm Hg), plateau pressure (< 29, 29-30, > 30 cm H2O), and number of extrapulmonary organ failure (< 2, 2, > 2) stratified accurately acute respiratory distress syndrome patients into categories of risk. The model that included all four variables proved best to identify patients with the highest or lowest risk of death (area under the receiver operating characteristic curve, 0.86; 95% CI, 0.84-0.88). Decision tree analyses confirmed the accuracy and robustness of this enrichment model. CONCLUSIONS: Combined thresholds for patient's age, PaO2/FIO2, plateau pressure, and extrapulmonary organ failure provides prognostic enrichment accuracy for stratifying and selecting acute respiratory distress syndrome patients for randomized controlled trials.

A vascular endothelial growth factor receptor gene variant is associated with susceptibility to acute respiratory distress syndrome.

BACKGROUND: The acute respiratory distress syndrome (ARDS) is one of the main causes of mortality in adults admitted to intensive care units. Previous studies have demonstrated the existence of genetic variants involved in the susceptibility and outcomes of this syndrome. We aimed to identify novel genes implicated in sepsis-induced ARDS susceptibility. METHODS: We first performed a prioritization of candidate genes by integrating our own genomic data from a transcriptomic study in an animal model of ARDS and from the only published genome-wide association study of ARDS study in humans. Then, we selected single nucleotide polymorphisms (SNPs) from prioritized genes to conduct a case-control discovery association study in patients with sepsis-induced ARDS (n = 225) and population-based controls (n = 899). Finally, we validated our findings in an independent sample of 661 sepsis-induced ARDS cases and 234 at-risk controls. RESULTS: Three candidate genes were prioritized: dynein cytoplasmic-2 heavy chain-1, fms-related tyrosine kinase 1 (FLT1), and integrin alpha-1. Of those, a SNP from FLT1 gene (rs9513106) was associated with ARDS in the discovery study, with an odds ratio (OR) for the C allele of 0.76, 95% confidence interval (CI) 0.58-0.98 (p = 0.037). This result was replicated in an independent study (OR = 0.78, 95% CI = 0.62-0.98, p = 0.039), showing consistent direction of effects in a meta-analysis (OR = 0.77, 95% CI = 0.65-0.92, p = 0.003). CONCLUSIONS: We identified FLT1 as a novel ARDS susceptibility gene and demonstrated that integration of genomic data can be a valid procedure to identify novel susceptibility genes. These results contribute to previous firm associations and functional evidences implicating FLT1 gene in other complex traits that are mechanistically linked, through the key role of endothelium, to the pathophysiology of ARDS.

A Quantile Analysis of Plateau and Driving Pressures: Effects on Mortality in Patients With Acute Respiratory Distress Syndrome Receiving Lung-Protective Ventilation.

OBJECTIVES: The driving pressure (plateau pressure minus positive end-expiratory pressure) has been suggested as the major determinant for the beneficial effects of lung-protective ventilation. We tested whether driving pressure was superior to the variables that define it in predicting outcome in patients with acute respiratory distress syndrome. DESIGN: A secondary analysis of existing data from previously reported observational studies. SETTING: A network of ICUs. PATIENTS: We studied 778 patients with moderate to severe acute respiratory distress syndrome. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We assessed the risk of hospital death based on quantiles of tidal volume, positive end-expiratory pressure, plateau pressure, and driving pressure evaluated at 24 hours after acute respiratory distress syndrome diagnosis while ventilated with standardized lung-protective ventilation. We derived our model using individual data from 478 acute respiratory distress syndrome patients and assessed its replicability in a separate cohort of 300 acute respiratory distress syndrome patients. Tidal volume and positive end-expiratory pressure had no impact on mortality. We identified a plateau pressure cut-off value of 29 cm H2O, above which an ordinal increment was accompanied by an increment of risk of death. We identified a driving pressure cut-off value of 19 cm H2O where an ordinal increment was accompanied by an increment of risk of death. When we cross tabulated patients with plateau pressure less than 30 and plateau pressure greater than or equal to 30 with those with driving pressure less than 19 and driving pressure greater than or equal to 19, plateau pressure provided a slightly better prediction of outcome than driving pressure in both the derivation and validation cohorts (p < 0.0000001). CONCLUSIONS: Plateau pressure was slightly better than driving pressure in predicting hospital death in patients managed with lung-protective ventilation evaluated on standardized ventilator settings 24 hours after acute respiratory distress syndrome onset.

Neurally adjusted ventilatory assist in patients with acute respiratory failure: study protocol for a randomized controlled trial.

BACKGROUND: Patient-ventilator asynchrony is a common problem in mechanically ventilated patients with acute respiratory failure. It is assumed that asynchronies worsen lung function and prolong the duration of mechanical ventilation (MV). Neurally Adjusted Ventilatory Assist (NAVA) is a novel approach to MV based on neural respiratory center output that is able to trigger, cycle, and regulate the ventilatory cycle. We hypothesized that the use of NAVA compared to conventional lung-protective MV will result in a reduction of the duration of MV. It is further hypothesized that NAVA compared to conventional lung-protective MV will result in a decrease in the length of ICU and hospital stay, and mortality. METHODS/DESIGN: This is a prospective, multicenter, randomized controlled trial in 306 mechanically ventilated patients with acute respiratory failure from several etiologies. Only patients ventilated for less than 5 days, and who are expected to require prolonged MV for an additional 72 h or more and are able to breathe spontaneously, will be considered for enrollment. Eligible patients will be randomly allocated to two ventilatory arms: (1) conventional lung-protective MV (n = 153) and conventional lung-protective MV with NAVA (n = 153). Primary outcome is the number of ventilator-free days, defined as days alive and free from MV at day 28 after endotracheal intubation. Secondary outcomes are total length of MV, and ICU and hospital mortality. DISCUSSION: This is the first randomized clinical trial examining, on a multicenter scale, the beneficial effects of NAVA in reducing the dependency on MV of patients with acute respiratory failure. TRIAL REGISTRATION: ClinicalTrials.gov website ( NCT01730794 ). Registered on 15 November 2012.

Assessment of PaO2/FiO2 for stratification of patients with moderate and severe acute respiratory distress syndrome.

OBJECTIVES: A recent update of the definition of acute respiratory distress syndrome (ARDS) proposed an empirical classification based on ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) at ARDS onset. Since the proposal did not mandate PaO2/FiO2 calculation under standardised ventilator settings (SVS), we hypothesised that a stratification based on baseline PaO2/FiOv would not provide accurate assessment of lung injury severity. DESIGN: A prospective, multicentre, observational study. SETTING: A network of teaching hospitals. PARTICIPANTS: 478 patients with eligible criteria for moderate (100300). PRIMARY AND SECONDARY OUTCOMES: Group severity and hospital mortality. RESULTS: At ARDS onset, 173 patients had a PaO2/FiO2≤100 but only 38.7% met criteria for severe ARDS at 24 h under SVS. When assessed under SVS, 61.3% of patients with severe ARDS were reclassified as moderate, mild and non-ARDS, while lung severity and hospital mortality changed markedly with every PaO2/FiO2 category (p<0.000001). Our model of risk stratification outperformed the stratification using baseline PaO2/FiO2 and non-standardised PaO2/FiO2 at 24 h, when analysed by the predictive receiver operating characteristic (ROC) curve: area under the ROC curve for stratification at baseline was 0.583 (95% CI 0.525 to 0.636), 0.605 (95% CI 0.552 to 0.658) at 24 h without SVS and 0.693 (95% CI 0.645 to 0.742) at 24 h under SVS (p<0.000001). CONCLUSIONS: Our findings support the need for patient assessment under SVS at 24 h after ARDS onset to assess disease severity, and have implications for the diagnosis and management of ARDS patients. TRIAL REGISTRATION NUMBERS: NCT00435110 and NCT00736892.

Clinical outcome of patients with major bleeding after venous thromboembolism. Findings from the RIETE Registry.

The natural history of patients with venous thromboembolism (VTE) who develop a major bleeding complication while on anticoagulant therapy is not well known. RIETE is a prospective registry of consecutive patients with symptomatic, objectively confirmed, acute VTE. The clinical characteristics, treatment decisions and outcome of all VTE patients who had major bleeding during the first three months of anticoagulant therapy were retrospectively studied. As of January 2007, 17, 368 patients were included in RIETE. Of these, 407 (2.3%) had major bleeding during the study period: 144 gastrointestinal, 119 haematoma, 51 intracranial, 43 genitourinary, 50 other. In 286 (69%) patients anticoagulant therapy was discontinued, in 74 (18%) not modified, in 38 (9.1%) a vena cava filter was inserted. During the first 30 days after bleeding, 24 (5.9%) patients re-bled, 20 (4.9%) had recurrent VTE, 133 (33%) died. Of these, 75 died of bleeding, 12 of recurrent pulmonary embolism. Most deaths occurred shortly after the bleeding episode (median: 1 day). On multivariate analysis, insertion of a vena cava filter was the only variable independently associated with a lower incidence of fatal bleeding (odds ratio [OR]: 0.10; 95% confidence interval [CI]: 0.01-0.79) and all-cause mortality (OR: 0.21; 95% CI: 0.07-0.63). In conclusion, the occurrence of major bleeding in patients with VTE is outstanding in terms of overall mortality (33% within 30 days), fatal bleeding (18%) or re-bleeding (5.9%). However, these patients also have an increased incidence of recurrent VTE (4.9%) and fatal pulmonary embolism (1.2%).