Berlin Definition Research Articles

Effects of different positive end-expiratory pressure strategies on treating overweight patients with acute respiratory distress syndrome: A retrospective study based on large intensive care unit databases.

Effects of different positive end-expiratory pressure strategies on treating overweight patients with acute respiratory distress syndrome: A retrospective study based on large intensive care unit databases.

Diagnostic biomarkers and miRNAs in prognosis of acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) is a disease of the lung and/or extrapulmonary system characterized by acute, progressive breathing difficulty and refractory hypoxemia. After years of revision, the 2012 International Expert Conference developed a new diagnostic standard for ARDS, known as the Berlin definition, which provides good guidance on how to define and judge the disease in clinical practice. Despite the establishment of diagnostic standards and treatment improvements, ARDS mortality rate still remains high. The primary reason is that the pathophysiology has not been fully elucidated. In patients with ARDS, damage to the alveolar capillary membrane may occur, leading to increased vascular permeability and the occurrence of pulmonary edema. Therefore, exploring the pathogenesis of ARDS from the perspective of microvascular permeability and identification of effective targets may be key factors in the diagnosis and treatment of ARDS. This review presents the current literature regarding the role of miRNAs (micro ribonucleic acids) in early detection and prediction of ARDS outcome.

Efficacy of solnatide to treat pulmonary permeability edema in SARS-CoV-2 positive patients with moderate to severe ARDS: A randomized controlled pilot-trial.

Efficacy of solnatide to treat pulmonary permeability edema in SARS-CoV-2 positive patients with moderate to severe ARDS: A randomized controlled pilot-trial.

The Norwegian national trauma registry: development process and essential data insights

BackgroundUnderstanding trauma epidemiology, patient demographics, injury characteristics, and outcomes is essential for optimising trauma systems. The Norwegian Trauma Registry (NTR) monitors and improves the Norwegian Trauma System, setting care standards and overseeing system development. The registry was officially recognised as a national register in 2013. This study outlines the establishment of the population-based national registry and provides an overview of selected data.MethodsNorway’s trauma system includes trauma centres, acute care hospitals, and prehospital services. The registry collects injury details, clinical outcomes, and patient experiences. Local NTR databases that are linked to a central database are maintained at each hospital, and only certified data registrars can enter and validate data. This enables data linkages across hospitals. The NTR includes potentially severely injured patients but also includes undertriaged patients (defined as severely injured patients who are not met by a trauma team activation upon hospital arrival). Descriptive statistics were used to analyse data from trauma patients registered between 2015 and 2023. Patient-Reported Outcome Measures (PROMs) from 2022 were also assessed.ResultsFrom 2015 to 2023, 78 275 trauma patients were recorded, with annual patient inclusion rising from 7586 in 2015 to 9759 in 2023. All 38 Norwegian hospitals contributed data in 2023. Median age was 41 years (IQR: 21–62), and 66.5% were men. The highest injury rate was among those aged 15–24 years. Penetrating injuries accounted for 4.6% of cases. Severely injured patients with New Injury Severity Score (NISS) ≥ 16 totalled 16 678 (21.3%), while 10 509 (13.4%) had an Injury Severity Score (ISS) ≥ 16. Polytrauma was identified in 3783 (4.9%) of patients using the Newcastle definition and in 2508 (3.2%) patients using the Berlin definition. In 2023, a trauma team was activated for 8731(89.4%) patients recorded in the registry. PROMs data from 2022 showed that 47.2% (1018/2157) of the patients reported anxiety or depression 12 months post-injury. Among those without physical injuries, 8.0% (11/138) were out of work or education. Of the severely injured patients (NISS ≥ 16) who were employed or in education prior to the injury, 26.4% (83/314) had not returned to work or education after 12 months.ConclusionsThe Norwegian Trauma Registry has been successfully implemented in all trauma hospitals in Norway, enabling comprehensive data collection to support trauma care improvements and research.

Mapping Berlin in the Global Definition of ARDS: Overlap and Progression Between Berlin and Global Definitions

Mapping Berlin in the Global Definition of ARDS: Overlap and Progression Between Berlin and Global Definitions

51 Utilizing Indirect Intrapleural Pressure to Guide Mechanical Ventilation in Burn Patients with ARDS

Abstract Introduction Patients with severe burn injuries are at increased risk of developing acute respiratory distress syndrome (ARDS). There is a paucity of literature examining mechanical ventilation strategies such as an open lung approach in burn injured patients with ARDS. Use of esophageal manometry may provide guidance on optimal positive end expiratory pressure (PEEP) required to prevent alveolar collapse while ensuring safe plateau pressures to prevent barotrauma. The aim of this work was to report on a single institution’s use of esophageal manometry in burn injured patients with ARDS to examine outcomes in mortality, oxygenation, and resolution of ARDS. Methods Burn injured patients admitted to an ABA verified burn center from January 2017 to May 2024 were retrospectively reviewed for those who required mechanical ventilation for more than 48 hours, developed ARDS by the Berlin definition, and received esophageal manometry to determine optimal PEEP, safe plateau pressures, and airway pressures. Medical records were reviewed for demographics, injury characteristics, ventilator settings, intrapleural pressures obtained by manometry, and arterial partial pressure of oxygen. Oxygenation indices (OI) [(Fraction of inspired oxygen x mean airway pressure) / arterial partial pressure of oxygen] were calculated to compare hypoxic respiratory failure (HRF) severity prior to manometry use, and at 1-, 3- and 5-days following manometry initiation. Friedman’s test was used to compare OI levels among the four time points. Results Of 242 patients who required mechanical ventilation during this period, 32 developed ARDS. Overall mortality rate was 30.1%, with 24.6% in the non-ARDS cohort compared to 71.9% in the ARDS cohort (p< 0.0001). Esophageal manometry was utilized in 25 of the 32 patients with ARDS. Of these 25 patients, the median (IQR) TBSA burned was 41% (29-50). Oxygenation indices improved at all subsequent time points after utilization of esophageal manometry, with statistically significant improvements between post manometry days 1 and 5 [OI: 17.7 (8.9-21.4) vs OI: 11.0 (8.6-12.6); p=0.02] and days 3 and 5 [OI: 14.8 (9.3-19.6) vs OI: 11.0 (8.6-12.6); p=0.04]. Despite having median revised Baux scores of 83 (72.1-97.5), 10 out of the 25 (40%) patients with ARDS who utilized manometry had resolution of ARDS. Conclusions A strategy incorporating esophageal manometry to determine optimal PEEP and safe plateau pressure can be used to guide ventilator management in burn injured patients with ARDS. Manometry provides an additional tool to help rescue patients who have a high risk of mortality due to the severity of their injury. Applicability of Research to Practice Esophageal manometry can be used to provide an individualized approach to mechanical ventilation and improve oxygenation in burn patients with ARDS. Funding for the Study N/A

Effects of 'Head Up' Prone Position on Transcranial Color Doppler-Based Estimators of Intracranial Pressure in Moderate to Severe Acute Respiratory Distress Syndrome Without Brain Injury: A Cross-Over, Longitudinal, Physiological Study.

Prone positioning is recommended in acute respiratory distress syndrome (ARDS) to ensure adequate gas exchange. However, it may lead to an increase in intracranial pressure (ICP), mostly due to a reduction of venous return from the brain. ICP can be noninvasively estimated with transcranial color-coded Doppler (TCCD) using methods based on the relationships between the pulsatility index (PI) and ICP or methods based on the estimate of cerebral perfusion pressure (eCPP) and estimate of ICP (eICP). This study was aimed at assessing the effects of a 30° reverse Trendelenburg ('head up') prone position on two noninvasive estimators of ICP (eICP and PI). This is a cross-over, longitudinal, physiological study conducted on a cohort of adult patients fulfilling Berlin definition criteria for moderate to severe ARDS without brain injury but with clinical indication to prone positioning. We registered TCCD parameters of cerebral hemodynamic and systemic hemodynamic parameters, blood gas exchange data, and respiratory mechanics parameters in a horizonal supine position, in a 30° semirecumbent supine position, in the standard prone position, and, finally, in the 30° 'head up' prone position, obtained by tilting the entire bed to a reverse Trendelenburg position. One-way repeated measures analysis of variance was used to analyze data. In 20 patients included, switching from a supine position to the standard prone position resulted in a significant increase in mean ± SD PI (from 0.99 ± 0.22 to 1.29 ± 0.25, p < 0.01) and eICP (from 12.5 ± 3.8 to 17.5 ± 4.1, p < 0.01), whereas moving from this latter position to the 'head up' prone position resulted in a decrease in the mean ± SD PI (from 1.29 ± 0.25 to 1.0 ± 0.23, p < 0.01). Hemodynamic and respiratory mechanics parameters did not differ. The 30° 'head up' prone position may limit the increase in PI in moderate to severe ARDS without brain injury. As a noninvasive estimator of ICP, PI may allow detection of changes in ICP when moving from the 'head up' semirecumbent supine position to the standard prone position and from this latter position to the 'head up' prone position.

Open-source computational pipeline automatically flags instances of acute respiratory distress syndrome from electronic health records.

Physicians, particularly intensivists, face information overload and decision fatigue, underscoring the need for automated diagnostic tools. Acute Respiratory Distress Syndrome (ARDS) affects over 10% of critical care patients, with over 40% mortality rate, yet is only recognized in 30-70% of cases in clinical settings. We present a reproducible computational pipeline that automates ARDS adjudication in retrospective datasets of mechanically ventilated adults, implementing the Berlin Definition via natural language processing and classification algorithms. We used labeled chest imaging reports from two hospitals to train an XGBoost model to detect bilateral infiltrates, and a labeled subset of attending physician notes from one hospital to train another XGBoost model to detect a pneumonia diagnosis. Both models achieve high discriminative performance on test sets-an area under the receiver operating characteristic curve (AUROC) of 0.88 for adjudicating bilateral infiltrates on chest imaging reports, and an AUROC of 0.87 for detecting pneumonia on attending physician notes. We integrated these models with rule-based components and validated the entire pipeline on a subset of healthcare encounters from a third hospital (MIMIC-III). We find a sensitivity of 93.5% in adjudicating ARDS - far surpassing the 22.6% ARDS documentation rate we found for this cohort - along with a false positive rate of 17.4%. We conclude that our reproducible, automated pipeline holds promise for improving ARDS recognition and could aid clinical practice through real-time EHR integration.

The UTAMI score: a chest x-ray-based tool for predicting ICU admission in ARDS of pneumonia patients.

This study proposes and evaluates the Universal Thorax ARDS Modification Index (UTAMI), a new method based on chest x-ray findings, for rapid ICU admission prediction in pneumonia with ARDS. Clinical and laboratory variables are analyzed to find potential predictors. A cross-sectional study at Fatmawati Central General Hospital (2022-2023) compared the diagnostic accuracy of UTAMI method against the gold standard for ARDS diagnosis; Berlin Definition. We analyzed 318 patients' data that were hospitalized for pneumonia. Clinical and laboratory predictors of ARDS were also analyzed. Neutrophil levels, CRP, D-dimer, oxygen saturation, and respiratory rate can predict ARDS diagnosis according to the Berlin Definition. The patient cohort showed that those with moderate-severe ARDS were admitted to the ICU. With ARDS categorized as ARDS requiring ICU admission (ARDS ICU) and ARDS not requiring ICU admission, the UTAMI method requires only history of coronary artery disease (CAD), CRP, and oxygen saturation as key predictors. CRP was a predictor in both the Berlin Definition (PR 1.28) and the UTAMI method (PR 1.71). In the AUROC test, the Berlin Definition distinguished moderate-severe ARDS with 81.2% accuracy using chest radiographs, clinical and laboratory values. The UTAMI method, based solely on chest radiographs achieved 79.6% accuracy, showing fair discrimination against the gold standard. UTAMI Score is a viable tool for predicting the risk of ARDS in pneumonia. Utilizing UTAMI method, ARDS can be predicted using only chest radiograph, making it easier for clinicians to be alerted earlier. Predicting ARDS ICU from UTAMI method requires only 3 variables; CAD comorbid, laboratory CRP and peripheral oxygen saturation.

Impact of the Kigali Modifications on ARDS Diagnosis.

Background: The Berlin definition of ARDS has recently been expanded to include the Kigali criteria, which allows for the inclusion of more non-intubated patients. However, there is concern that the expanded criteria may be overly inclusive, potentially representing different pathologies. This study evaluates the characteristics of patients with sepsis who meet the Kigali criteria compared with those who meet the original Berlin criteria. Methods: This retrospective cohort study was conducted at a single institution and included subjects aged 18 and older who were admitted for sepsis. Berlin-ARDS was defined by a PaO2/FIO2 ratio of ≤300 mm Hg within the first 7 days of admission, the use of intubation or noninvasive ventilation with PEEP of at least 5 cm H2O, bilateral opacities on chest imaging, and respiratory failure not attributed to heart failure. Kigali-ARDS was defined for subjects who did not meet the Berlin criteria owing to unavailable arterial blood gas values, a PaO2/FIO2 ratio >300 mm Hg, or who had a PaO2/FIO2 ratio ≤300 mm Hg but were not ventilated with a PEEP of 5 cm H2O. Results: Of 427 subjects, 73 developed ARDS according to the Berlin criteria, and 94 met the ARDS criteria only by the Kigali definition, whereas 260 did not meet either definition. Smoking was significantly associated with meeting the Kigali criteria (P = .045) but not the Berlin criteria (P = .49). Higher lactate and white blood cell levels were linked to the Berlin criteria (P = .02 and P = .01, respectively) but not to the Kigali criteria. Conclusions: Smoking was a risk factor for meeting the Kigali criteria but not the Berlin criteria, suggesting that the Kigali criteria might include patients with chronic lung conditions rather than true acute lung injury. Additionally, higher lactate and white blood cell levels were associated with Berlin-ARDS, indicating more severe sepsis in these subjects compared with those meeting the Kigali criteria.

SRAGE as a Prognostic Biomarker in ARDS: Insights from a Clinical Cohort Study.

Background and Objectives: Acute respiratory distress syndrome (ARDS) is a severe form of acute lung injury with high mortality, characterized by hypoxemic respiratory failure and diffuse lung damage. Despite advancements in care, no definitive biomarkers have been established for ARDS diagnosis and prognostic stratification. Soluble receptor for advanced glycation end-products (sRAGE), a marker of alveolar epithelial injury, has shown promise as a prognostic indicator in ARDS. This study evaluates sRAGE's utility in predicting 28-day mortality. Materials and Methods: A retrospective cohort study was conducted at a tertiary care ICU in Serbia from January 2021 to June 2023. Adult patients meeting the Berlin definition of ARDS were included. Exclusion criteria included pre-existing chronic respiratory diseases and prolonged mechanical ventilation before diagnosis. Serum sRAGE levels were measured within 48 h of ARDS diagnosis using enzyme-linked immunosorbent assay (ELISA). Clinical severity scores, laboratory markers, and ventilatory parameters were recorded. Logistic regression and survival analyses were used to assess the prognostic value of sRAGE for 28-day mortality. Results: A cohort of 121 patients (mean age 55.5 years; 63.6% male) was analyzed. Non-survivors exhibited higher median sRAGE levels than survivors (5852 vs. 4479 pg/mL, p = 0.084). The optimal sRAGE cut-off for predicting mortality was >16,500 pg/mL (sensitivity 30.4%, specificity 86.9%). Elevated sRAGE levels were associated with greater disease severity and an increased risk of 28-day mortality in ARDS patients, highlighting its potential as a prognostic biomarker. The main findings, while indicative of a trend toward higher sRAGE levels in non-survivors, did not reach statistical significance. Conclusions: The main findings, while indicative of a trend toward higher sRAGE levels in non-survivors, did not reach statistical significance (p = 0.084). sRAGE demonstrates potential as a prognostic biomarker in ARDS and has moderate correlation with 28-day mortality. Integrating sRAGE with other biomarkers could enhance risk stratification and guide therapeutic decisions. The retrospective design limits the ability to establish causation, underscoring the need for multicenter prospective studies.

Evaluating the impact of ESICM 2023 guidelines and the new global definition of ARDS on clinical outcomes: insights from MIMIC-IV cohort data

BackgroundIn 2023, the European Society of Intensive Care Medicine (ESICM) recommended updated criteria for acute respiratory distress syndrome (ARDS). In 2024, Matthay et al. updated the global ARDS definition in AJRCCM, titled “A New Global Definition of Acute Respiratory Distress Syndrome.” However, the impact of this new definition on ARDS treatments is currently unknown.ObjectiveThis study aims to determine the effect of the new ARDS definition on patients with hypoxemic respiratory failure and study the heterogeneity of patients in the new definition to guide treatment.MethodsClinical consultation data from the Medical Information Mart for Intensive Care IV database were extracted using Structured Query Language based on the PostgreSQL tool (version 10.0). Data were analyzed using Python (version 3.9) and the deep learning framework Pytorch. Kaplan–Meier survival analysis was used to compare survival between the old and new definitions. A hierarchical clustering approach was applied to identify potential ARDS clinical subtypes.ResultsThe new definition diagnosed ARDS earlier and included individuals with lower mortality rates compared with the Berlin definition. Patients meeting the new definition but not the Berlin criteria exhibited a favorable response to non-invasive ventilation strategies (p = 0.009). The XGBoost classifier, trained to predict subphenotypes, achieved an AUC of 0.88 ± 0.02 on the training set. Additionally, mortality was significantly associated with patients with hypoxemia compared with survivors, particularly regarding respiratory parameters. Easily accessible metrics, such as respiratory rate and urea nitrogen (BUN), can help diagnose ARDS in high-risk populations in resource-limited settings.ConclusionsThe new ARDS definition offers advantages in earlier detection, more accurate grading, and more precise diagnosis in resource-limited settings compared with the Berlin definition. This study also established a robust prediction model for early ARDS identification, improving the patient prognosis and reducing the mortality rate.

Current Improvements to the Acute Respiratory Distress Syndrome Definition

ARDS was first described with cases in 1967 and it became a broad agenda with 4 basic criteria determined at the AmericanEuropean Consensus conference in 1994, and then it reached its peak with the Berlin Definition in 2012, with effects that continue to this day. However, later on, the definition was moved to a better level with objections to the Berlin definition and remarkable new recommendations. However, remarkable new recommendations after Berlin brought its definition to a better level. An attempt was made to better define it with many biomarkers in blood and broncho alveolar lavage. It was recommended to define ARDS Specific Marker (ASM) and ARDS Severity Score (ASS), which could help determine its severity and mortality. Although there is no definitive treatment, many recommendations have been made for its management. Low tidal volume ventilation, prone position and High-Flow Nasal Oxygen (HFNO) application to non-intube patients have remained important in management over a long period of time. PEEP titration, negative fluid balance, non-invasive ventilation, use of muscle relaxants and Extracorporeal Membrane Oxygenation (ECMO) applications were also discussed on a patient-by-patient basis. Despite improvements in definition and management, the mortality rate of it still remains high. There continues to be a need for new studies and methods regarding the definition and management of ARDS.

Operationalizing the New Global Definition of ARDS: A Retrospective Cohort Study From South Africa

Operationalizing the New Global Definition of ARDS: A Retrospective Cohort Study From South Africa

Oxygenation Index, Oxygen Saturation Index vs PaO2/FiO2 *PEEP: A Secondary Analysis of OXIVA-CARDS Study.

The classification of Berlin definition is based on the PaO2/FiO2 ratio, which has been found to have a poor association with mortality. Airway pressures reflect lung compliance and the settings of mechanical ventilators. In this study, we aimed to investigate the change in the severity of COVID-19-associated acute respiratory distress syndrome (ARDS) classification using [PaO2/FiO2 × PEEP] (P/FP) ratio compared to the traditional P/F ratio, and whether the P/FP ratio improves the predictive validity of in-hospital mortality. Our study sample included patients from the OXIVA-CARDS study. In this secondary analysis, we examined the oxygenation index and oxygen saturation index in relation to the P/FP ratio, as well as the risk of P/FP in mortality. We used Pearson's correlation to assess the relationships between various parameters. Receiver operating characteristic analysis with Youden's index was used to compare the prognostic value of the oxygenation index (OI), oxygen saturation index (OSI), P/F ratio, P/FP ratio, and SaO2/FiO2 ratio for predicting overall mortality. Multiple logistic regression was also performed to determine the impact of mean airway pressure (Pmean), S/F ratio, OI, and P/FP ratio on mortality. A total of 201 patients (with 1543 measurements) were included in the analysis. Overall, 522 (34%) were reclassified into either more or less severe categories. Patients who were classified as having severe ARDS based on the P/FP ratio had significantly lower P/FP ratio, oxygenation index, and A-a O2 gradient as compared to those classified as having severe ARDS based on the P/F ratio (p < 0.05) at all levels of ARDS severity. On multivariate regression analysis, only the OI significantly impacted mortality (p < 0.05). We observed that the oxygen index and oxygen saturation index were more sensitive than the PaO2/FiO2 ratio and P/FP ratio. Additionally, only the oxygenation index had a significant impact on mortality. By including airway pressures in the calculation of the OI, its predictive ability is enhanced compared to using the S/F ratio, P/F ratio, or P/FP ratio. The sensitivity of mortality by including Pmean is higher as compared to when only PEEP is taken into consideration. P/FP is a weak predictor of mortality as compared to OI and OSI. Vadi SMR, Sanwalka N, Suthar D. Oxygenation Index, Oxygen Saturation Index vs PaO2/FiO2 *PEEP: A Secondary Analysis of OXIVA-CARDS Study. Indian J Crit Care Med 2024;28(10):917-922.

Pressure Control Surrogate Formula for Estimating Mechanical Power in ARDS Is Associated With Mortality.

Background: Mechanical power (MP) applied to the respiratory system (MPRS) is associated with ventilator-induced lung injury (VILI) and ARDS mortality. Absent automated ventilator MPRS measurements, the alternative is clinically unwieldy equations. However, simplified surrogate formulas are now available and accurately reflect values produced by airway pressure-volume curves. This retrospective, observational study examined whether the surrogate pressure -control equation alone could accurately assess mortality risk in subjects with ARDS managed almost exclusively with volume control (VC) ventilation. Methods: Nine hundred and forty-eight subjects were studied in whom invasive mechanical ventilation and implementation of ARDS Network ventilator protocols commenced ≤ 24 h after ARDS onset and who survived > 24 h. MPRS was calculated as 0.098 x breathing frequency x tidal volume x (PEEP + driving pressure). MPRS was assessed as a risk factor for hospital mortality and compared between non-survivors and survivors across Berlin definition classifications. In addition, mortality was compared across 4 MPRS thresholds associated with VILI or mortality (ie, 15, 20, 25, and 30 J/min). Results: MPRS was associated with increased mortality risk: odds ratio (95% CI) of 1.06 (1.04-1.07) J/min (P < .001). Median MPRS differentiated non-survivors from survivors in mild (24.7 J/min vs 18.5 J/min, respectively, P = .034), moderate (25.7 J/min vs 21.3 J/min, respectively, P < .001), and severe ARDS (28.7 J/min vs 23.5 J/min, respectively, P < .001). Across 4 MPRS thresholds, mortality increased from 23-29% when MPRS was ≤ threshold versus 38-51% when MPRS was > threshold (P < .001). In the > cohort, the odds ratio (95% CI) increased from 2.03 (1.34-3.12) to 2.51 (1.87-3.33). Conclusion: The pressure control surrogate formula is sufficiently accurate to assess mortality in ARDS, even when using VC ventilation. In our subjects when MPRS exceeds established cutoff values for VILI or mortality risk, we found mortality risk consistently increased by a factor of > 2.0.

Platelet count as a prognostic marker for acute respiratory distress syndrome

BackgroundThis study aimed to evaluate the role of platelet count (PLT) in the prognosis of patients with acute respiratory distress syndrome (ARDS).MethodsThe data were extracted from the Medical Information Mart for Intensive Care database (version 2.2). Patients diagnosed with ARDS according to criteria from Berlin Definition and had the platelet count (PLT) measured within the first day after intensive care unit admission were analyzed. Based on PLT, ARDS patients were divided into four groups: PLT ≤ 100 × 109/L, PLT 101–200 × 109/L, PLT 201–300 × 109/L, and PLT > 300 × 109/L. The primary outcome was 28-day mortality. Survival probabilities were analyzed using Kaplan–Meier. Furthermore, the association between PLT and mortality in ARDS patients was assessed using a univariate and multivariable Cox proportional hazards model.ResultsOverall, the final analysis included 3,207 eligible participants with ARDS. According to the Kaplan–Meier curves for 28-day mortality of PLT, PLT ≤ 100 × 109/L was associated with a higher incidence of mortality (P = 0.001), the same trends were observed in the 60-day (P = 0.001) and 90‐day mortality (P = 0.001). In the multivariate model adjusted for the potential factors, the adjusted hazard ratio at PLT 101–200 × 109/L group, PLT 201–300 × 109/L, and PLT > 300 × 109/L was 0.681 [95% confidence interval (CI): 0.576–0.805, P < 0.001], 0.733 (95% CI: 0.604–0.889, P = 0.002), and 0.787 (95% CI: 0.624–0.994, P = 0.044) compared to the reference group (PLT ≤ 100 × 109/L), respectively. Similar relationships between the PLT ≤ 100 × 109/L group and 28-day mortality were obtained in most subgroups.ConclusionPLT appeared to be an independent predictor of mortality in critically ill patients with ARDS.

Risk factors for in-hospital mortality in recipients of allogeneic hematopoietic stem cell transplantation with acute respiratory distress syndrome: a retrospective study based on the 2023 new definition of acute respiratory distress syndrome

IntroductionARDS (acute respiratory distress syndrome) is the most severe form of acute hypoxic respiratory failure. Most studies related to ARDS have excluded patients with hematologic diseases, let alone allogeneic hematopoietic stem cell transplantation (allo-HSCT) recipients. Numerous patients experiencing severe hypoxic respiratory failure do not meet the Berlin definition due to the limitations of diagnosis and treatment. A new definition of ARDS, remove some diagnosis restrictions, was proposed in 2023. Based on the 2023 new definition of ARDS, we investigated the clinical features of ARDS in allo-HSCT recipients and reported risk factors for in-hospital mortality in allo-HSCT recipients defined by the Berlin definition and the new definition of ARDS respectively.MethodsFrom Jan 2016 to Dec 2020, 135 allo-HSCT recipients identified with the new definition and 87 identified with the Berlin definition at three teaching hospitals were retrospectively included in this study. Variables (demographic information, characteristics of hematologic disease and ARDS episode, laboratory tests and SOFA score) with P < 0.05 in univariate logistic regression analysis were included in multivariate stepwise logistic regression analysis. Adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) were reported.ResultsUnder the new definition, SOFA score (OR = 1.351, 95% CI: 1.146–1.593, P < 0.01) were found as an independent risk factor for in-hospital mortality in ARDS after allo-HSCT, while SpO2/FiO2 (OR = 0.984, 95% CI: 0.972–0.996, P < 0.01) was a protective factor. The infusion of peripheral-derived stem cells was found to be a protective factor against in-hospital mortality in post-transplantation ARDS compared with the infusion of bone marrow-derived stem cells (OR = 0.726, 95% CI: 0.164–3.221, P = 0.04). Under the Berlin definition, PaO2/FiO2 (OR = 0.977, 95% CI: 0.961–0.993, P = 0.01, lactate (OR = 7.337, 95% CI: 1.313–40.989, P < 0.01) and AST (OR = 1.165, 95% CI: 1.072–1.265, P < 0.01) were independently associated with in-hospital mortality.ConclusionThese prognostic risk factors we found in allo-HSCT recipients may contribute to closer monitoring and ARDS prevention strategies. These findings require confirmation in prospective, large sample size studies.

Association of trauma classifications to long-term outcome in blunt trauma patients

PurposeThe impact of major trauma is long lasting. Although polytrauma patients are currently identified with the Berlin polytrauma criteria, data on long-term outcomes are not available. In this study, we evaluated the association of trauma classification with long-term outcome in blunt-trauma patients.MethodsA trauma registry of a level I trauma centre was used for patient identification from 1.1.2006 to 31.12.2015. Patients were grouped as follows: (1) all severely injured trauma patients; (2) all severely injured polytrauma patients; 2a) severely injured patients with AIS ≥ 3 on two different body regions (Berlin-); 2b) severely injured patients with polytrauma and a physiological criterion (Berlin+); and (3) a non-polytrauma group. Kaplan-Meier survival analysis was performed to estimate differences in mortality between different groups.ResultsWe identified 3359 trauma patients for this study. Non-polytrauma was the largest group (2380 [70.9%] patients). A total of 500 (14.9%) patients fulfilled the criteria for Berlin + definition, leaving 479 (14.3%) polytrauma patients in Berlin- group. Berlin + patients had the highest short-term mortality compared with other groups, although the difference in cumulative mortality gradually plateaued compared with the non-polytrauma patient group; at the end of the 10-year follow up, the non-polytrauma group had the greatest mortality due to the high number of patients with traumatic brain injury (TBI).ConclusionExcess mortality of polytrauma patients by Berlin definition occurs in the early phase (30-day mortality) and late deaths are rare. TBI causes high early mortality followed by increased long-term mortality.

ARDSFlag: an NLP/machine learning algorithm to visualize and detect high-probability ARDS admissions independent of provider recognition and billing codes

BackgroundDespite the significance and prevalence of acute respiratory distress syndrome (ARDS), its detection remains highly variable and inconsistent. In this work, we aim to develop an algorithm (ARDSFlag) to automate the diagnosis of ARDS based on the Berlin definition. We also aim to develop a visualization tool that helps clinicians efficiently assess ARDS criteria.MethodsARDSFlag applies machine learning (ML) and natural language processing (NLP) techniques to evaluate Berlin criteria by incorporating structured and unstructured data in an electronic health record (EHR) system. The study cohort includes 19,534 ICU admissions in the Medical Information Mart for Intensive Care III (MIMIC-III) database. The output is the ARDS diagnosis, onset time, and severity.ResultsARDSFlag includes separate text classifiers trained using large training sets to find evidence of bilateral infiltrates in radiology reports (accuracy of 91.9%±0.5%) and heart failure/fluid overload in radiology reports (accuracy 86.1%±0.5%) and echocardiogram notes (accuracy 98.4%±0.3%). A test set of 300 cases, which was blindly and independently labeled for ARDS by two groups of clinicians, shows that ARDSFlag generates an overall accuracy of 89.0% (specificity = 91.7%, recall = 80.3%, and precision = 75.0%) in detecting ARDS cases.ConclusionTo our best knowledge, this is the first study to focus on developing a method to automate the detection of ARDS. Some studies have developed and used other methods to answer other research questions. Expectedly, ARDSFlag generates a significantly higher performance in all accuracy measures compared to those methods.