Correlating Reiff scores with clinical, functional, and prognostic factors: characterizing noncystic fibrosis bronchiectasis severity: validation from a nationwide multicenter study in Taiwan.

BACKGROUND: Our study aimed to confirm a simplified radiological scoring system, derived from a modified Reiff score, to evaluate its relationship with clinical symptoms and predictive outcomes in Taiwanese patients with noncystic fibrosis bronchiectasis (NCFB). METHODS: This extensive multicenter retrospective study, performed in Taiwan, concentrated on patients diagnosed with NCFB verified through high-resolution computed tomography (HRCT) scans. We not only compared the clinical features of various types of bronchiectasis (cylindrical, varicose, and cystic). Furthermore, we established relationships between the severity of clinical factors, including symptom scores, pulmonary function, pseudomonas aeruginosa colonization, exacerbation and admission rates, and HRCT parameters using modified Reiff scores. RESULTS: Data from 2,753 patients were classified based on HRCT patterns (cylindrical, varicose, and cystic) and severity, assessed by modified Reiff scores (mild, moderate, and severe). With increasing HRCT severity, a significant correlation was found with decreased forced expiratory volume in the first second (FEV1) (p < 0.001), heightened clinical symptoms (p < 0.001), elevated pathogen colonization (pseudomonas aeruginosa) (p < 0.001), and an increased annual hospitalization rate (p < 0.001). In the following multivariate analysis, elderly age, pseudomonas aeruginosa pneumonia, and hospitalizations per year emerged as the only independent predictors of mortality. CONCLUSION: Based on our large cohort study, the simplified CT scoring system (Reiff score) can serve as a useful adjunct to clinical factors in predicting disease severity and prognosis among Taiwanese patients with NCFB.

Lower body mass index is an independent predictor of mortality in older patients with acute respiratory distress syndrome.

Background: Acute respiratory distress syndrome (ARDS) is associated with high mortality. The impacts of body mass index (BMI) on the morality of older patients with ARDS remain unclear. Methods: This is a single-center cohort study which was conducted at Taichung Veterans General Hospital, Taiwan. Adult patients admitted to the ICU needing mechanical ventilation with ARDS were included for analysis. We compared the data of older patients (age ≥65 years) with those of younger patients (Age <65 years). The factors associated with in-hospital mortality of older patients were investigated. Results: This study included a total of 728 (mean age: 66 years; men: 63%) patients, and 425 (58.4%) of them aged ≥65 years. Older patients exhibited lower body mass index (BMI) (23.8 vs 25.2), higher Acute Physiology and Chronic Health Evaluation (APACHE) II scores (28.9 vs 26.3), higher Charlson Comorbidity Index (CCI) (4.0 vs 3.4), and lower Sequential Organ Failure Assessment (SOFA) scores (10.0 vs 11.1) than younger patients. Furthermore, older patients had mortality rates similar to younger patients (40.5% vs 42.9%, P = 0.542), but had longer length of stay in the ICU (17.6 vs 15.6 days, P = 0.047). For older patients, BMI <18.5 (odds ratio [OR], 2.78; 95% confidence interval [CI], 1.45-5.34), high SOFA score (OR, 1.20; 95% CI, 1.12-1.28), and moderate (OR, 1.95; 95% CI 1.20-3.14) or severe ARDS (OR, 2.30; 95% CI 1.26-4.22) were independent risk factors for mortality. Conclusions: In this cohort, critical ill older patients with ARDS had lower BMI, more comorbidities, and higher APACHE II scores than younger patients. Mortality rate was similar between older and younger patients. Low BMI, high SOFA score, and moderate or severe ARDS were independently associated with mortality in older patients with ARDS.

Risk factors for prolonged mechanical ventilation in critically ill patients with influenza-related acute respiratory distress syndrome.

BACKGROUND: Patients with influenza-related acute respiratory distress syndrome (ARDS) are critically ill and require mechanical ventilation (MV) support. Prolonged mechanical ventilation (PMV) is often seen in these cases and the optimal management strategy is not established. This study aimed to investigate risk factors for PMV and factors related to weaning failure in these patients. METHODS: This retrospective cohort study was conducted by eight medical centers in Taiwan. All patients in the intensive care unit with virology-proven influenza-related ARDS requiring invasive MV from January 1 to March 31, 2016, were included. Demographic data, critical illness data and clinical outcomes were collected and analyzed. PMV is defined as mechanical ventilation use for more than 21 days. RESULTS: There were 263 patients with influenza-related ARDS requiring invasive MV enrolled during the study period. Seventy-eight patients had PMV. The final weaning rate was 68.8% during 60 days of observation. The mortality rate in PMV group was 39.7%. Risk factors for PMV were body mass index (BMI) > 25 (kg/m2) [odds ratio (OR) 2.087; 95% confidence interval (CI) 1.006-4.329], extracorporeal membrane oxygenation (ECMO) use (OR 6.181; 95% CI 2.338-16.336), combined bacterial pneumonia (OR 4.115; 95% CI 2.002-8.456) and neuromuscular blockade use over 48 h (OR 2.8; 95% CI 1.334-5.879). In addition, risk factors for weaning failure in PMV patients were ECMO (OR 5.05; 95% CI 1.75-14.58) use and bacteremia (OR 3.91; 95% CI 1.20-12.69). CONCLUSIONS: Patients with influenza-related ARDS and PMV have a high mortality rate. Risk factors for PMV include BMI > 25, ECMO use, combined bacterial pneumonia and neuromuscular blockade use over 48 h. In addition, ECMO use and bacteremia predict unsuccessful weaning in PMV patients.

Efficacy of combination therapy with standard-dose carbapenem for treating nosocomial pneumonia caused by carbapenem-resistant Acinetobacter baumannii in intensive care units: A multicentre retrospective propensity score-matched study.

Carbapenem-resistant Acinetobacter baumannii (CRAB) infection is common worldwide. Despite carbapenem resistance, standard-dose carbapenems are still used in clinical practice. Hence in this study, we aimed to compare the efficacy and outcomes of a regimen containing standard-dose carbapenems with those of a regimen lacking carbapenems during the treatment of critically ill patients with CRAB nosocomial pneumonia in the intensive care unit (ICU). Initially, 735 patients were recruited for this multicentre retrospective cohort study. After exclusion, time-window bias adjustment, and propensity score matching, multiple clinical outcomes were compared between the carbapenem-containing (CC) (n = 166) and no carbapenem-containing (NCC) (n = 166) groups. The CC group showed a higher risk of clinical failure on day 7 than the NCC group (44.6% vs. 33.1%, P = 0.043). The lengths of ICU stay (21 and 16 days, P = 0.024) and hospital stay (61 and 44 days, P = 0.003) were longer in the CC group than in the NCC group. Multivariate analysis showed that the CC regimen was associated with higher clinical failure (adjusted odds ratio (aOR) = 1.64, 95% CI = 1.05-2.56, P = 0.031) and lower microbiological eradication (aOR = 0.48, 95% CI = 0.23-1.00, P = 0.049) at day 7 than the NCC group. Thus, a regimen containing a standard dose of carbapenem should be prescribed with caution for treating CRAB nosocomial pneumonia in the ICU.

Mechanical ventilation practices in Asian intensive care units: A multicenter cross-sectional study.

PURPOSE: This study investigated current practices of mechanical ventilation in Asian intensive care units, focusing on tidal volume, plateau pressure, and positive end-expiratory pressure (PEEP). MATERIALS AND METHODS: In this multicenter cross-sectional study, data on mechanical ventilation and clinical outcomes were collected. Predictors of mortality were analyzed by univariate and multivariable logistic regression. A scoring system was generated to predict 28-day mortality. RESULTS: A total of 1408 patients were enrolled. In 138 patients with acute respiratory distress syndrome (ARDS), 65.9% were on a tidal volume ≤ 8 ml/kg predicted body weight (PBW), and 71.3% were on sufficient PEEP. In 1270 patients without ARDS, 88.8% were on a tidal volume ≤ 10 ml/kg PBW. A plateau pressure < 30 cmH2O was measured in 92.2% of patients. Mortality rates increased from 13% to 74% as the generated predictive score increased from 5 to ≥8.5. Income classification, age, SOFA score, PaO2/FiO2 ratio, plateau pressure, number of vasopressors, and steroid use were associated with mortality. CONCLUSIONS: In Asia, low tidal volume ventilation and sufficient PEEP were underused in patients with ARDS. The majority of patients without ARDS were on intermediate tidal volumes. Country income, age, and severity of illness were associated with mortality.

A combination of oxygenation and driving pressure can provide valuable information in predicting the risk of mortality in ARDS patients.

BACKGROUND: Acute respiratory distress syndrome (ARDS) is a common life-threatening condition in critically ill patients. Itis also an important public health issue because it can cause substantial mortality and health care burden worldwide. The objective of this study was to investigate therisk factors that impact ARDS mortality in a medical center in Taiwan. METHODS: This was a single center, observational study thatretrospectively analyzed data from adults in 6 intensive care units (ICUs) at Taichung Veterans General Hospital in Taiwan from 1st October, 2018to30th September, 2019. Patients needing invasive mechanical ventilation and meeting the Berlin definition criteria were included for analysis. RESULTS: A total of 1,778 subjects were screened in 6 adult ICUs and 370 patients fulfilled the criteria of ARDS in the first 24 hours of the ICU admission. Among these patients, the prevalenceof ARDS was 20.8% and the overall hospital mortality rate was 42.2%. The mortality rates of mild, moderate and severe ARDS were 35.9%, 43.9% and 46.5%, respectively. In a multivariate logistic regression model, combination of driving pressure (DP) > 14cmH2O and oxygenation (P/F ratio)≤150 was an independent predictor of mortality (OR2.497, 95% CI 1.201-5.191, p = 0.014). Patients with worse oxygenation and a higher driving pressure had the highest hospital mortality rate(p<0.0001). CONCLUSIONS: ARDS is common in ICUs and the mortality rate remains high. Combining oxygenation and respiratory mechanics may better predict the outcomes of these ARDS patients.

Prolonged use of neuromuscular blocking agents is associated with increased long-term mortality in mechanically ventilated medical ICU patients: a retrospective cohort study.

BACKGROUND: Neuromuscular blockade agents (NMBAs) can be used to facilitate mechanical ventilation in critically ill patients. Accumulating evidence has shown that NMBAs may be associated with intensive care unit (ICU)-acquired weakness and poor outcomes. However, the long-term impact of NMBAs on mortality is still unclear. METHODS: We conducted a retrospective analysis using the 2015-2019 critical care databases at Taichung Veterans General Hospital, a referral center in central Taiwan, as well as the Taiwan nationwide death registry profile. RESULTS: A total of 5709 ventilated patients were eligible for further analysis, with 63.8% of them were male. The mean age of enrolled subjects was 67.8 ± 15.8 years, and the one-year mortality was 48.3% (2755/5709). Compared with the survivors, the non-survivors had a higher age (70.4 ± 14.9 vs 65.4 ± 16.3, p < 0.001), Acute Physiology and Chronic Health Evaluation II score (28.0 ± 6.2 vs 24.7 ± 6.5, p < 0.001), a longer duration of ventilator use (12.6 ± 10.6 days vs 7.8 ± 8.5 days, p < 0.001), and were more likely to receive NMBAs for longer than 48 h (11.1% vs 7.8%, p < 0.001). After adjusting for age, sex, and relevant covariates, the use of NMBAs for longer than 48 h was found to be independently associated with an increased risk of mortality (adjusted HR: 1.261; 95% CI: 1.07-1.486). The analysis of effect modification revealed that this association was tended to be strong in patients with a Charlson Comorbidity Index of 3 or higher. CONCLUSIONS: Our study demonstrated that prolonged use of NMBAs was associated with an increased risk of long-term mortality in critically ill patients requiring mechanical ventilation. Further studies are needed to validate our findings.

The Association Between Absolute Lymphocyte Count and Long-Term Mortality in Critically Ill Medical Patients: Propensity Score-Based Analyses.

Objective: Absolute lymphocyte count (ALC) has been implicated with short-term outcomes in a number of diseases, and we aimed to investigate the association between week-one ALC and long-term mortality in patients who were admitted to the medical intensive care units (ICUs). Methods: We enrolled patients who were admitted to the medical ICUs at the Taichung Veterans General Hospital, a referral centre located in central Taiwan, between 2015 and 2020 to conduct this retrospective cohort study. The outcome of interest was long-term all-cause mortality, and hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated to determine the association. Furthermore, we employed propensity score-matching (PSM) and weighting techniques, consisting of inverse probability of treatment weighting (IPTW) and covariate balancing propensity score (CBPS), to confirm the association between ALC and mortality. Results: A total of 5722 critically ill patients were enrolled, and the one-year mortality was 44.8%. The non-survivor group had a lower ALC (1549, 1027-2388 vs 1948, 1373-2743 counts/µL, p<0.01) compared with those in the survivor group. Cox regression showed that low ALC was independently associated with mortality (adjHR 1.091, 95% CI 1.050-1.134). Propensity score-based analyses demonstrated the robust association, with adjHRs in the original, PSM, IPTW, and CBPS populations of 1.327 (95% CI 1.224-1.438), 1.301 (95% CI 1.188-1.424), 1.292 (95% CI 1.186-1.407), and 1.297 (95% CI 1.191-1.412), respectively. Sensitivity analyses further showed that the association between low ALC and mortality existed in a dose-response manner. Conclusion: We found that low ALC was associated with long-term mortality in critically ill patients; further studies are warranted to validate and translate these findings into clinical utility.

Association between early blood urea nitrogen-to-albumin ratio and one-year post-hospital mortality in critically ill surgical patients: a propensity score-matched study.

BACKGROUND: Blood urea nitrogen to albumin ratio (BAR) is increasingly recognized as an early predictor for short-term outcomes in critically ill patients, but the association of BAR with long-term outcomes in critically ill surgical patients remains underexplored. METHODS: We enrolled consecutive patients who were admitted to surgical intensive care units (ICUs) at Taichung Veterans General Hospital between 2015 and 2020, and the dates of death were retrieved from Taiwan's National Health Insurance Research Database. In addition to Cox regression, we also used propensity score matching to determine the hazard ratios (HRs) and 95% confidence intervals (CIs) for one-year post-hospital mortality of the variables. RESULTS: A total of 8,073 eligible subjects were included for analyses. We found that age, male gender, high Charlson Comorbidity Index, high Acute Physiology and Chronic Health Evaluation II score, positive microbial culture, and leukocytosis were predictors for mortality, whereas high body mass index, scheduled surgery, and high platelet counts were protective factors against long-term mortality. The high BAR was independently associated with increased post-hospital mortality after adjustment for the aforementioned covariates (adjHR 1.258, 95% CI, 1.127-1.405). Notably, the association tended to be stronger in females and patients with fewer comorbidities and lower disease severity of critical illness. The propensity score matching, dividing subjects by BAR higher or lower than 6, showed a consistent association between week-one BAR and post-hospital mortality (adjHR 1.503, 95% CI 1.247-1.811). CONCLUSIONS: BAR is a newly identified predictor of short-term outcome, and we identified long-term outcome-relevant factors, including BAR, and the identified factors may be useful for risk stratification of long-term outcomes in patients discharged from surgical ICUs.

Mortality and ventilator dependence in critically ill patients with ventilator-associated pneumonia caused by carbapenem-resistant Acinetobacter baumannii.

BACKGROUND: Carbapenem-resistant Acinetobacter baumannii (CRAB) is a key pathogen associated with ventilator-associated pneumonia (VAP). Research on treatment outcomes, especially ventilator dependence, in patients with VAP caused by CRAB remains limited. METHODS: This retrospective multicenter study included ICU-admitted patients with VAP caused by CRAB. The original cohort was included as the mortality evaluation cohort. The ventilator dependence evaluation cohort included cases that survived more than 21 days after VAP and without prolonged ventilation before VAP onset. The mortality rate, ventilator dependence rate, clinical factors associated with treatment outcomes, and treatment outcome differences with various VAP onset times were investigated. RESULTS: In total, 401 patients with VAP caused by CRAB were analyzed. The 21-day all-cause mortality rate was 25.2%, and the 21-day ventilator dependence rate was 48.8%. Clinical factors associated with 21-day mortality included lower body mass index, higher sequential organ failure assessment score, vasopressors usage, CRAB persistence, and VAP onset time > seven days. Clinical factors associated with 21-day ventilator dependence included older age, vasopressors usage, and VAP onset time > seven days. CONCLUSIONS: ICU-admitted patients with CRAB-related VAP had high mortality and ventilator dependence rates. Older age, vasopressor usage, and longer VAP onset time were independent factors associated with ventilator dependence.

Anaemia in the first week may be associated with long-term mortality among critically ill patients: propensity score-based analyses.

BACKGROUND: Anaemia is highly prevalent in critically ill patients; however, the long-term effect on mortality remains unclear. METHODS: We retrospectively included patients admitted to the medical intensive care units (ICUs) during 2015-2020 at the Taichung Veterans General Hospital. The primary outcome of interest was one-year mortality, and hazard ratios (HRs) with 95% confidence intervals (CIs) were determined to assess the association. We used propensity score matching (PSM) and propensity score matching methods, including inverse probability of treatment weighting (IPTW) as well as covariate balancing propensity score (CBPS), in the present study. RESULTS: A total of 7,089 patients were eligible for analyses, and 45.0% (3,189/7,089) of them had anaemia, defined by mean levels of haemoglobin being less than 10 g/dL. The standardised difference of covariates in this study were lower than 0.20 after matching and weighting. The application of CBPS further reduced the imbalance among covariates. We demonstrated a similar association, and adjusted HRs in original, PSM, IPTW and CBPS populations were 1.345 (95% CI 1.227-1.474), 1.265 (95% CI 1.145-1.397), 1.276 (95% CI 1.142-1.427) and 1.260 (95% CI 1.125-1.411), respectively. CONCLUSIONS: We used propensity score-based analyses to identify that anaemia within the first week was associated with increased one-year mortality in critically ill patients.

The prevalence, presentation and outcome of colistin susceptible-only Acinetobacter Baumannii-associated pneumonia in intensive care unit: a multicenter observational study.

Hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP) caused by carbapenem-resistant Acinetobacter baumannii (CRAB) are both associated with significant morbidity and mortality in daily clinical practice, as well as in a critical care setting. It is unclear whether colistin susceptible-only Acinetobacter baumannii (CSO AB) is a unique phenotype separate from or a subset of CRAB-associated pneumonia. The aim of this study is to investigate the prevalence of CSO AB pneumonia and compare the presentation and outcome between CSO AB and CRAB-associated pneumonia in critically ill patients. This multicenter retrospective cohort study initially recruited 955 patients with CR-GNB pneumonia. After exclusion, 575 patients left who were ICU-admitted and had CRAB nosocomial pneumonia remained. Among them, 79 patients had CSO AB pneumonia, classified as the CSO AB group. The other 496 patients were classified as the CRAB group. We compared demographic characteristics, disease severity, and treatment outcomes between the two groups. The prevalence of CSO AB among all cases of CRAB pneumonia was 13.74% (79/575). The CSO AB and CRAB groups had similar demographic characteristics and disease severities at initial presentation. The in-hospital mortality rate was 45.6% and 46.4% for CSO AB and CRAB groups, respectively (p = 0.991). The CSO AB group had significantly better clinical outcomes at day 7 (65.8% vs 52.4%, p = 0.036) but longer length of ICU stay (27 days vs 19 days, p = 0.043) compared to the CRAB group. However, other treatment outcomes, including clinical outcomes at day 14 and 28, mortality, microbiological eradication, ventilator weaning, and newly onset dialysis, were similar. In conclusion, CSO AB accounted for 13.74% of all cases of CRAB pneumonia, and the clinical presentation and treatment outcomes of CSO AB and CRAB pneumonia were similar.

Clinical effectiveness of tigecycline in combination therapy against nosocomial pneumonia caused by CR-GNB in intensive care units: a retrospective multi-centre observational study.

BACKGROUND: Tigecycline has in vitro bacteriostatic activity against a broad spectrum of bacteria, including carbapenem-resistant Gram-negative bacteria (CR-GNB). However, the role of tigecycline in treatment of nosocomial pneumonia caused by CR-GNB remains controversial and clinical evidences are limited. We aimed to investigate the clinical benefits of tigecycline as part of the combination treatment of nosocomial CR-GNB pneumonia in intensive care unit (ICU). METHODS: This multi-centre cohort study retrospectively enrolled ICU-admitted patients with nosocomial pneumonia caused by CR-GNB. Patients were categorized based on whether add-on tigecycline was used in combination with at least one anti-CR-GNB antibiotic. Clinical outcomes and all-cause mortality between patients with and without tigecycline were compared in the original and propensity score (PS)-matched cohorts. A subgroup analysis was also performed to explore the differences of clinical efficacies of add-on tigecycline treatment when combined with various anti-CR-GNB agents. RESULTS: We analysed 395 patients with CR-GNB nosocomial pneumonia, of whom 148 received tigecycline and 247 did not. More than 80% of the enrolled patients were infected by CR-Acinetobacter baumannii (CRAB). A trend of lower all-cause mortality on day 28 was noted in tigecycline group in the original cohort (27.7% vs. 36.0%, p = 0.088). In PS-matched cohort (102 patient pairs), patients with tigecycline had significantly lower clinical failure (46.1% vs. 62.7%, p = 0.017) and mortality rates (28.4% vs. 52.9%, p < 0.001) on day 28. In multivariate analysis, tigecycline treatment was a protective factor against clinical failure (PS-matched cohort: aOR 0.52, 95% CI 0.28-0.95) and all-cause mortality (original cohort: aHR 0.69, 95% CI 0.47-0.99; PS-matched cohort: aHR 0.47, 95% CI 0.30-0.74) at 28 days. Kaplan-Meier survival analysis in subgroups of patients suggested significant clinical benefits of tigecycline when added to a colistin-included (log rank p value 0.005) and carbapenem-included (log rank p value 0.007) combination regimen. CONCLUSIONS: In this retrospective observational study that included ICU-admitted patients with nosocomial pneumonia caused by tigecycline-susceptible CR-GNB, mostly CRAB, tigecycline as part of a combination treatment regimen was associated with lower clinical failure and all-cause mortality rates.

Explainable machine learning approach to predict extubation in critically ill ventilated patients: a retrospective study in central Taiwan.

BACKGROUND: Weaning from mechanical ventilation (MV) is an essential issue in critically ill patients, and we used an explainable machine learning (ML) approach to establish an extubation prediction model. METHODS: We enrolled patients who were admitted to intensive care units during 2015-2019 at Taichung Veterans General Hospital, a referral hospital in central Taiwan. We used five ML models, including extreme gradient boosting (XGBoost), categorical boosting (CatBoost), light gradient boosting machine (LightGBM), random forest (RF) and logistic regression (LR), to establish the extubation prediction model, and the feature window as well as prediction window was 48 h and 24 h, respectively. We further employed feature importance, Shapley additive explanations (SHAP) plot, partial dependence plot (PDP) and local interpretable model-agnostic explanations (LIME) for interpretation of the model at the domain, feature, and individual levels. RESULTS: We enrolled 5,940 patients and found the accuracy was comparable among XGBoost, LightGBM, CatBoost and RF, with the area under the receiver operating characteristic curve using XGBoost to predict extubation was 0.921. The calibration and decision curve analysis showed well applicability of models. We also used the SHAP summary plot and PDP plot to demonstrate discriminative points of six key features in predicting extubation. Moreover, we employed LIME and SHAP force plots to show predicted probabilities of extubation and the rationale of the prediction at the individual level. CONCLUSIONS: We developed an extubation prediction model with high accuracy and visualised explanations aligned with clinical workflow, and the model may serve as an autonomous screen tool for timely weaning.

Artificial intelligence-aided diagnosis model for acute respiratory distress syndrome combining clinical data and chest radiographs.

Objective: The aim of this study was to develop an artificial intelligence-based model to detect the presence of acute respiratory distress syndrome (ARDS) using clinical data and chest X-ray (CXR) data. Method: The transfer learning method was used to train a convolutional neural network (CNN) model with an external image dataset to extract the image features. Then, the last layer of the model was fine-tuned to determine the probability of ARDS. The clinical data were trained using three machine learning algorithms-eXtreme Gradient Boosting (XGB), random forest (RF), and logistic regression (LR)-to estimate the probability of ARDS. Finally, ensemble-weighted methods were proposed that combined the image model and the clinical data model to estimate the probability of ARDS. An analysis of the importance of clinical features was performed to explore the most important features in detecting ARDS. A gradient-weighted class activation mapping (Grad-CAM) model was used to explain what our CNN sees and understands when making a decision. Results: The proposed ensemble-weighted methods improved the performances of the ARDS classifiers (XGB + CNN, area under the curve [AUC] = 0.916; RF + CNN, AUC = 0.920; LR + CNN, AUC = 0.920; XGB + RF + LR + CNN, AUC = 0.925). In addition, the ML model using clinical data to present the top 15 important features to identify the risk factors of ARDS. Conclusion: This study developed combined machine learning models with clinical data and CXR images to detect ARDS. According to the results of the Shapley Additive exPlanations values and the Grad-CAM techniques, an explicable ARDS diagnosis model is suitable for a real-life scenario.

Comparing Clinical Outcomes of Tiotropium/Olodaterol, Umeclidinium/Vilanterol, and Indacaterol/Glycopyrronium Fixed-Dose Combination Therapy in Patients with Chronic Obstructive Pulmonary Disease in Taiwan: A Multicenter Cohort Study.

Background: Long-acting beta-agonists (LABA) and long-acting muscarinic antagonists (LAMA) combination therapy improved lung function and health-related quality-of-life and reduced exacerbation rates and dyspnea in symptomatic chronic obstructive pulmonary disease (COPD) patients. We compared the real-world effects of three fixed-dose LABA/LAMA combinations for COPD in Taiwan. Methods: This multicenter, retrospective study evaluated 1-year outcomes after LABA/LAMA combination therapy in patients with symptomatic COPD. Exacerbations and symptoms of COPD, lung functions, and therapy escalation were compared among patients using tiotropium/olodaterol, umeclidinium/vilanterol and indacaterol/glycopyrronium. Propensity score matching (PSM) was applied to balance the baseline characteristics. Results: Data of 1,617 patients were collected. After PSM, time to first moderate-to-severe COPD exacerbation was comparable among three groups, while the annualized rates of the exacerbation (episodes/patient/year) in patients receiving tiotropium/olodaterol (0.19) or umeclidinium/vilanterol (0.17) were significantly lower than those receiving indacaterol/glycopyrronium (0.38). COPD-related symptoms were stable over the treatment period, and there was no significant difference in the changes of symptom scores including CAT and mMRC among three groups at the end of the study period. Conclusion: This study presented valuable real-world outcome in terms of exacerbation and treatment response of COPD patients treated with fixed-dose LABA/LAMA regimens in Taiwan. The annualized rates of moderate-to-severe exacerbation in patients receiving tiotropium/olodaterol or umeclidinium/vilanterol were significantly lower than those receiving indacaterol/glycopyrronium, though the time to first moderate-to-severe exacerbation was similar among different fixed-dose LABA/LAMA combinations.

Clinical outcome of nosocomial pneumonia caused by Carbapenem-resistant gram-negative bacteria in critically ill patients: a multicenter retrospective observational study.

Nosocomial pneumonia caused by carbapenem-resistant gram-negative bacteria (CRGNB) is a growing threat due to the limited therapeutic choices and high mortality rate. The aim of this study was to evaluate the prognostic factors for mortality in patients with nosocomial pneumonia caused by CRGNB and the impact of colistin-based therapy on the outcomes of intensive care unit (ICU) patients. We conducted a retrospective study of the ICUs in five tertiary teaching hospitals in Taiwan. Patients with nosocomial pneumonia caused by CRGNB from January 2016 to December 2016 were included. Prognostic factors for mortality were analyzed using multivariate logistic regression. The influence of colistin-based therapy on mortality and clinical and microbiological outcomes were evaluated in subgroups using different severity stratification criteria. A total of 690 patients were enrolled in the study, with an in-hospital mortality of 46.1%. The most common CRGNB pathogens were Acinetobacter baumannii (78.7%) and Pseudomonas aeruginosa (13.0%). Significant predictors (odds ratio and 95% confidence interval) of mortality from multivariate analysis were a length of hospital stay (LOS) prior to pneumonia of longer than 9 days (2.18, 1.53-3.10), a sequential organ failure assessment (SOFA) score of more than 7 (2.36, 1.65-3.37), supportive care with vasopressor therapy (3.21, 2.26-4.56), and escalation of antimicrobial therapy (0.71, 0.50-0.99). There were no significant differences between the colistin-based therapy in the deceased and survival groups (42.1% vs. 42.7%, p = 0.873). In the subgroup analysis, patients with multiple organ involvement (> 2 organs) or higher SOFA score (> 7) receiving colistin-based therapy had better survival outcomes. Prolonged LOS prior to pneumonia onset, high SOFA score, vasopressor requirement, and timely escalation of antimicrobial therapy were predictors for mortality in critically ill patients with nosocomial CRGNB pneumonia. Colistin-based therapy was associated with better survival outcomes in subgroups of patients with a SOFA score of more than 7 and multiple organ involvement.

The necessity of a loading dose when prescribing intravenous colistin in critically ill patients with CRGNB-associated pneumonia: a multi-center observational study.

BACKGROUND: The importance or necessity of a loading dose when prescribing intravenous colistin has not been well established in clinical practice, and approximate one-third to half of patients with carbapenem-resistant gram-negative bacteria (CRGNB) infection did not receive the administration of a loading dose. The aim of this study is to investigate the efficacy and risk of acute kidney injury when prescribing intravenous colistin for critically ill patients with nosocomial pneumonia caused by CRGNB. METHODS: This was a multicenter, retrospective study that recruited ICU-admitted patients who had CRGNB-associated nosocomial pneumonia and were treated with intravenous colistin. Then, we classified the patients into colistin loading dose (N = 85) and nonloading dose groups (N = 127). After propensity-score matching for important covariates, we compared the mortality rate, clinical outcome and microbiological eradication rates between the groups (N = 67). RESULTS: The loading group had higher percentages of patients with favorable clinical outcomes (55.2% and 35.8%, p = 0.037) and microbiological eradication rates (50% and 27.3%, p = 0.042) at day 14 than the nonloading group. The mortality rates at days 7, 14 and 28 and overall in-hospital mortality were not different between the two groups, but the Kaplan-Meier analysis showed that the loading group had a longer survival time than the nonloading group. Furthermore, the loading group had a shorter length of hospital stay than the nonloading group (52 and 60, p = 0.037). Regarding nephrotoxicity, there was no significant difference in the risk of developing acute kidney injury between the groups. CONCLUSIONS: The administration of a loading dose is recommended when prescribing intravenous colistin for critically ill patients with nosocomial pneumonia caused by CRGNB.

Explainable machine learning to predict long-term mortality in critically ill ventilated patients: a retrospective study in central Taiwan.

BACKGROUND: Machine learning (ML) model is increasingly used to predict short-term outcome in critically ill patients, but the study for long-term outcome is sparse. We used explainable ML approach to establish 30-day, 90-day and 1-year mortality prediction model in critically ill ventilated patients. METHODS: We retrospectively included patients who were admitted to intensive care units during 2015-2018 at a tertiary hospital in central Taiwan and linked with the Taiwanese nationwide death registration data. Three ML models, including extreme gradient boosting (XGBoost), random forest (RF) and logistic regression (LR), were used to establish mortality prediction model. Furthermore, we used feature importance, Shapley Additive exPlanations (SHAP) plot, partial dependence plot (PDP), and local interpretable model-agnostic explanations (LIME) to explain the established model. RESULTS: We enrolled 6994 patients and found the accuracy was similar among the three ML models, and the area under the curve value of using XGBoost to predict 30-day, 90-day and 1-year mortality were 0.858, 0.839 and 0.816, respectively. The calibration curve and decision curve analysis further demonstrated accuracy and applicability of models. SHAP summary plot and PDP plot illustrated the discriminative point of APACHE (acute physiology and chronic health exam) II score, haemoglobin and albumin to predict 1-year mortality. The application of LIME and SHAP force plots quantified the probability of 1-year mortality and algorithm of key features at individual patient level. CONCLUSIONS: We used an explainable ML approach, mainly XGBoost, SHAP and LIME plots to establish an explainable 1-year mortality prediction ML model in critically ill ventilated patients.

Comparison of prone positioning and extracorporeal membrane oxygenation in acute respiratory distress syndrome: A multicenter cohort study and propensity-matched analysis.

BACKGROUND/PURPOSE: Both prone positioning and extracorporeal membrane oxygenation (ECMO) are used as rescue therapies for severe hypoxemia in patients with acute respiratory distress syndrome (ARDS). This study compared outcomes between patients with severe influenza pneumonia-related ARDS who received prone positioning and those who received ECMO. METHODS: This retrospective cohort study included eight tertiary referral centers in Taiwan. All patients who were diagnosed as having influenza pneumonia-related severe ARDS were enrolled between January and March 2016. We collected their demographic data and prone positioning and ECMO outcomes from medical records. RESULTS: In total, 263 patients diagnosed as having ARDS were included, and 65 and 53 of them received prone positioning and ECMO, respectively. The baseline PaO2/FiO2 ratio, Acute Physiology and Chronic Health Evaluation II score and Sequential Organ Failure Assessment score did not significantly differ between the two groups. The 60-day mortality rate was significantly higher in the ECMO group than in the prone positioning group (60% vs. 28%, p = 0.001). A significantly higher mortality rate was still observed in the ECMO group after propensity score matching (59% vs. 36%, p = 0.033). In the multivariate Cox regression analysis, usage of prone positioning or ECMO was the single independent predictor for 60-day mortality (hazard ratio: 2.177, p = 0.034). CONCLUSION: While the patients receiving prone positioning had better outcome, the causality between prone positioning and the prognosis is unknown. However, the current data suggested that patients with influenza-related ARDS may receive prone positioning before ECMO support.