A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography.

In tuberculosis (TB), chest radiography (CXR) patterns are highly variable, mimicking pneumonia and many other diseases. This study aims to evaluate the efficacy of Google teachable machine, a deep neural network-based image classification tool, to develop algorithm for predicting TB probability of CXRs. The training dataset included 348 TB CXRs and 3806 normal CXRs for training TB detection. We also collected 1150 abnormal CXRs and 627 normal CXRs for training abnormality detection. For external validation, we collected 250 CXRs from our hospital. We also compared the accuracy of the algorithm to five pulmonologists and radiological reports. In external validation, the AI algorithm showed areas under the curve (AUC) of 0.951 and 0.975 in validation dataset 1 and 2. The accuracy of the pulmonologists on validation dataset 2 showed AUC range of 0.936-0.995. When abnormal CXRs other than TB were added, AUC decreased in both human readers (0.843-0.888) and AI algorithm (0.828). When combine human readers with AI algorithm, the AUC further increased to 0.862-0.885. The TB CXR AI algorithm developed by using Google teachable machine in this study is effective, with the accuracy close to experienced clinical physicians, and may be helpful for detecting tuberculosis by CXR.

Effects of Early Pulmonary Rehabilitation on Hospitalized Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease: A Systematic Review and Meta-Analysis.

Background: Pulmonary rehabilitation (PR) effectively improves symptoms and exercise ability in patients with stable chronic obstructive pulmonary disease (COPD). However, the effectiveness and timing of early PR on hospitalized patients with acute exacerbation of COPD (AECOPD) is still debated. Methods: This study conducted a meta-analysis to compare the outcome benefits between early PR and usual care for patient hospitalized due to AECOPD. A systematic search was performed for retrieving randomized control trials (RCTs) from the PubMed, Embase, and Cochrane library until November 2021. RCTs reporting early PR for AECOPD with hospitalization, either during admission or within four weeks of discharge, were enrolled for systematic review and meta-analysis. Results: Twenty RCTs (1274 participants) were included. Early PR showed significantly improved readmission rate (ten trials, risk ratio 0.68, 95% confidence interval (CI) 0.50-0.92), 6-minute walking distance (6MWD, twelve trials, MD 59.73, 95% CI 36.34-83.12), St George's Respiratory Questionnaire score (eight trials, MD -10.65, 95% CI -14.78 to -6.52), Borg score (eight trials, MD -0.79, 95% CI -1.26 to -0.32), and modified Medical Research Council dyspnea scale (eight trials, MD -0.38, 95% CI -0.5 to -0.25). However, the trend of mortality (six trials, risk ratio 0.72, 95% CI 0.39-1.34) benefit was not significant. The subgroup analysis showed non-significant trends of better effect in early PR during admission than those after discharge for outcomes of 6MWD, quality of life, and dyspnea. However, non-significant trends of less benefits on mortality and readmission rate were found in early PR during the admission. Conclusion: Overall, early PR is beneficial for AECOPD with hospitalization, and there was no significant outcome difference between PR initiated during admission or within 4 weeks of discharge.

Efficacy of HEPA Air Cleaner on Improving Indoor Particulate Matter 2.5 Concentration.

High-efficiency particulate air (HEPA) filters is a potential tool used to remove fine particles and improve indoor air quality. This study aims to analyze the real-world efficacy of portable HEPA air cleaners in a household environment. Laser light dispersion PM2.5 sensors are used to continuously monitor the indoor and outdoor PM2.5 level before and after HEPA air cleaner filtration. Overall, HEPA air cleaners significantly reduce the indoor PM2.5 level (33.5 ± 10.3 vs. 17.2 ± 10.7 µg/m3, mean difference (MD) = -16.3 µg/m3, p < 0.001) and indoor/outdoor PM2.5% (76.3 ± 16.8 vs. 38.6 ± 19.8%, MD = -37.7%, p < 0.001). The efficacy to reduce PM2.5 is strongest in three machines with medium-flow setting group (indoor PM2.5 MD: -26.5 µg/m3, indoor/outdoor PM2.5 percentage MD: -56.4%). Multiple linear regression demonstrates that outdoor PM2.5, machine number, airflow speed, and window ventilation are significant factors associated with indoor PM2.5 concentrations (R = 0.879) and percentage of the indoor/outdoor PM2.5 ratio (R = 0.808). HEPA air cleaners can effectively improve indoor PM2.5 air pollution. Adequate air cleaner machine numbers, appropriate airflow, and window ventilation limitations are important to achieve the best efficacy of the HEPA air cleaner.

Clinical Characteristics and Predictors of Mortality in Critically Ill Adult Patients with Influenza Infection.

Patients with influenza infection may develop acute respiratory distress syndrome (ARDS), which is associated with high mortality. Some patients with ARDS receiving extracorporeal membrane oxygenation (ECMO) support die of infectious complications. We aimed to investigate the risk factors affecting the clinical outcomes in critically ill patients with influenza. We retrospectively reviewed the medical records of influenza patients between January 2006 and May 2016 at the Kaohsiung Veterans General Hospital in Taiwan. Patients aged below 20 years or without laboratory-confirmed influenza were excluded. Critically ill patients who presented with ARDS (P = 0.004, odds ratio (OR): 8.054, 95% confidence interval (CI): 1.975-32.855), a higher Acute Physiology and Chronic Health Evaluation (APACHE) II score (P = 0.008, OR: 1.102, 95% CI: 1.025-1.184), or higher positive end-expiratory pressure (P = 0.008, OR: 1.259, 95% CI: 1.061-1.493) may have a higher risk of receiving ECMO. Influenza A (P = 0.037, OR: 0.105, 95% CI: 0.013-0.876) and multiple organ failure (P = 0.007, OR: 0.056, 95% CI: 0.007-0.457) were significantly associated with higher mortality rates. In conclusion, our study showed critically ill influenza patients with ARDS, higher APACHE II scores, and higher positive end-expiratory pressure have a higher risk of receiving ECMO support. Influenza A and multiple organ failure are predictors of mortality.

The effect of tracheostomy delay time on outcome of patients with prolonged mechanical ventilation: A STROBE-compliant retrospective cohort study.

The tracheostomy timing for patients with prolonged mechanical ventilation (PMV) was usually delayed in our country. Both physician decision time and tracheostomy delay time (time from physician's suggestion of tracheostomy to procedure day) affect tracheostomy timing. The effect of tracheostomy delay time on outcome has not yet been evaluated before.Patients older than 18 years who underwent tracheostomy for PMV were retrospectively collected. The outcomes between different timing of tracheostomy (early: ≤14 days; late: >14 days of intubation) were compared. We also analyzed the effect of physician decision time, tracheostomy delay time, and procedure type on clinical outcomes.A total of 134 patients were included. There were 57 subjects in the early tracheostomy group and 77 in the late group. The early group had significantly shorter mechanical ventilation duration, shorter intensive care unit stays, and shorter hospital stays than late group. There was no difference in weaning rate, ventilator-associated pneumonia, and in-hospital mortality. The physician decision time (8.1 ±â€Š3.4 vs 18.2 ±â€Š8.1 days, P < .001) and tracheostomy delay time (2.1 ±â€Š1.9 vs 6.1 ±â€Š6.8 days, P < .001) were shorter in the early group than in the late group. The tracheostomy delay time [odds ratio (OR) = 0.908, 95% confidence interval (CI) = 0.832-0.991, P = .031) and procedure type (percutaneous dilatation, OR = 2.489, 95% CI = 1.057-5.864, P = .037) affected successful weaning. Platelet count of >150 × 10/µL (OR = 0.217, 95% CI = 0.051-0.933, P = .043) and procedure type (percutaneous dilatation, OR = 0.252, 95% CI = 0.069-0.912, P = .036) were associated with in-hospital mortality.Shorter tracheostomy delay time is associated with higher weaning success. Percutaneous dilatation tracheostomy is associated with both higher weaning success and lower in-hospital mortality.

Outcomes of unplanned extubation in ordinary ward are similar to those in intensive care unit: A STROBE-compliant case-control study.

Unplanned extubation (UE) may cause considerable adverse effects in patients receiving mechanical ventilation (MV). Previous literature showed inconsistent prognosis in patients with UE. This study aimed to evaluate the clinical implications and outcomes of UE.The intubated adult patients with MV support in our hospital were enrolled, and they were divided into the UE and non-UE groups. Demographic data, admission unit, MV duration, overall weaning rate, and mortality rates were compared. The outcomes of UE in ordinary ward and intensive care unit (ICU) were also assessed.Totally 9245 intubated adult patients were included. UE occurred in 303 (3.5%) patients, and the UE events were 0.27 times/100 MV days. Old age, nonoperation related MV cause, and admission out of the ICU were significant factors associated with UE events. UE patients showed a trend of better overall weaning rate (71.9% vs 66.7%, P = .054) than non-UE. However, the in-hospital mortality rate (25.7% vs 24.8%, P = .713) were similar between the UE and non-UE patients. The reintubation rate of UE patients was 44.1% (142/322). Successful UEs were associated with patients in weaning process (52.8% vs 38.7%, P = .012), and patients received non-invasive positive pressure ventilation (NIPPV) support after UE (19.4% vs 3.5%, P < .001). Patients with successful UE had significantly shorter MV days, higher overall weaning rate, and lower mortality than those with unsuccessful UE. Outcomes of UE in ordinary ward and in ICU had similar MV duration, reintubation rate, overall weaning rate, and in-hospital mortality rate.The overall weaning rate and in-hospital mortality rates of the UE and non-UE patients were similar. UE occurred in ordinary ward had similar outcomes to those in ICU. Patients receiving MV should be assessed daily for weaning indications to reduce delayed extubation, and therefore, may decrease UE occurrence. Once the UE happened, NIPPV support may reduce the reintubation rate.

Monitoring sedation for bronchoscopy in mechanically ventilated patients by using the Ramsay sedation scale versus auditory-evoked potentials.

BACKGROUND: Appropriate sedation benefits patients by reducing the stress response, but it requires an appropriate method of assessment to adjust the dosage of sedatives. The aim of this study was to compare the difference in the sedation of mechanically ventilated patients undergoing flexible bronchoscopy (FB) monitored by auditory-evoked potentials (AEPs) or the Ramsay sedation scale (RSS). METHODS: In a prospective, randomized, controlled study, all patients who underwent FB with propofol sedation were monitored and their sedation adjusted. During FB, one group was monitored by AEP and another group was monitored by RSS. The propofol dosage was adjusted by the nursing staff during examination to maintain the Alaris AEP index (AAI) value between 25 and 40 in the AEP group and the RSS at 5 or 6 in the RSS group. Before FB and during FB, the AAI, heart rate (HR), and mean arterial pressure (MAP) were recorded every 5 min. The percentages of time at the sedation target and the propofol dosages were calculated. RESULTS: Nineteen patients received AEP monitoring and 18 patients received RSS monitoring. The percentage of time at the sedation target during FB was significantly higher in the AEP monitoring group (51.3%; interquartile range [IQR], 47.0-63.5%) than in the RSS group (15.4%; IQR, 9.5-23.4%), (P < 0.001). During FB, the RSS group had a significantly higher AAI (P = 0.011), HR (P < 0.001), and MAP (P < 0.001) than the AEP group. CONCLUSIONS: In mechanically ventilated patients undergoing FB, AEP monitoring resulted in less variation in AAI, HR, and MAP, and a higher percentage of time at the sedation target than RSS monitoring.

Impact of disease severity on gastric residual volume in critical patients.

AIM: To investigate whether illness severity has an impact on gastric residual volume (GRV) in medical critically ill patients. METHODS: Medical intensive care unit (ICU) patients requiring nasogastric feeding were enrolled. Sequential Organ Failure Assessment (SOFA) score was assessed immediately preceding the start of the study. Acute Physiology and Chronic Health Evaluation (APACHE) II scores were recorded on the first, fourth, seventh, and fourteenth day of the study period. GRV was measured every 4 h during enteral feeding. The relationship between mean daily GRV and SOFA scores and the correlation between mean daily GRV and mean APACHE II score of all patients were evaluated and compared. RESULTS: Of the 61 patients, 43 patients were survivors and 18 patients were non-survivors. The mean daily GRV increased as SOFA scores increased (P < 0.001, analysis of variance). Mean APACHE II scores of all patients correlated with mean daily GRV (P = 0.011, Pearson correlation) during the study period. Patients with decreasing GRV in the first 2 d had better survival than patients without decreasing GRV (P = 0.017, log rank test). CONCLUSION: GRV is higher in more severely ill medical ICU patients. Patients with decreasing GRV had lower ICU mortality than patients without decreasing GRV.

The initial response to inhaled nitric oxide treatment for intensive care unit patients with acute respiratory distress syndrome.

BACKGROUND: Inhaled nitric oxide (INO) can improve hypoxemia and reduce pulmonary hypertension, but there is a wide range of response to INO. OBJECTIVES: The aim of this study was to evaluate the effect of different INO concentrations in acute respiratory distress syndrome (ARDS) patients. METHODS: Thirty-two adult ARDS patients who were supported by mechanical ventilator were included. INO was given at a concentration of 1, 5, 10, 20 and 40 ppm, sequentially. Arterial blood gas and mean pulmonary artery pressure (MPAP) were measured 30 min after INO concentrations changed. RESULTS: There was a significant increase in PaO(2)/FiO(2) (p < 0.0001) and a decrease in pulmonary vascular resistance index and MPAP (p < 0.0001) after INO use. INO concentrations required for improving oxygenation were in the range of 1-20 ppm, whereas concentrations required for decreasing MPAP were in the range of 1-40 ppm. PaO(2)/FiO(2) worsened when the INO dose was adjusted above 20 ppm. Sixty-nine percent of ARDS patients were responders; 31% of them were nonresponders. Among responders, 64% of patients responded at 1 ppm, 36% at 5 ppm. CONCLUSIONS: The optimal doses for improving oxygenation and reducing PAP differ. The maximum PaO(2)/FiO(2) was observed at a lower INO concentration than that required for the minimal MPAP. There was no further improvement in PaO(2)/FiO(2) when the INO dose was adjusted above 20 ppm. Higher doses of INO treatment worsened oxygenation.

Beneficial effects of albuterol therapy driven by heliox versus by oxygen in severe asthma exacerbation.

OBJECTIVES: To determine and define the beneficial effects of heliox-driven albuterol therapy on severe asthma exacerbation and clinical factors that affect greater response. METHODS: The authors conducted two randomized, double-blinded, controlled trials in patients with severe asthma exacerbation. The first trial recruited 80 patients in the emergency department (ED). They received three consecutive doses of albuterol delivered by a nebulizer powered by either O(2) (O(2) group) or heliox (He/O(2) = 80:20; heliox group). Changes in peak expiratory flow rate (PEF) were compared, and factors influencing the response to heliox-driven albuterol therapy were identified. The second trial of 80 patients was conducted in older patients, a subpopulation associated with greater response in the first trial. RESULTS: In the first trial, the heliox group had more rapid and greater improvement in PEF compared with the O(2) group. There tended to be more patients in the heliox group reaching the predetermined dischargeable PEF (>60% predicted) after three albuterol treatments (odds ratio, 2.58; 95% confidence interval = 1.03 to 6.46; p = 0.069). For patients eventually discharged from the ED, the ED stay was shorter by 10 minutes per patient in the heliox group compared with the O(2) group (p = 0.007). Logistic regression showed older age and lower pretreatment PEF to be associated with favorable heliox responses. The second trial, which recruited older patients (older than 40 years), showed greater improvement in PEF and dyspnea score with heliox-driven albuterol therapy in patients with lower pretreatment PEF. CONCLUSIONS: Heliox-driven albuterol may be a useful adjunct therapy for older asthmatic patients with severe asthma exacerbation.

Heliox improves hemodynamics in mechanically ventilated patients with chronic obstructive pulmonary disease with systolic pressure variations.

OBJECTIVE: To test the hypothesis that, compared with air-oxygen, heliox would improve cardiac performance in mechanically ventilated patients with severe chronic obstructive pulmonary disease and systolic pressure variations >15 mm Hg and to determine clinical variables associated with favorable hemodynamic responses to heliox. DESIGN: A prospective interventional study. SETTING: Medical and respiratory intensive care units at a university-affiliated tertiary medical center. PATIENTS: Twenty-five consecutive mechanically ventilated patients with severe chronic obstructive pulmonary disease and acute respiratory failure who had systolic pressure variations >15 mm Hg. INTERVENTIONS: Respiratory and hemodynamic measurements were taken at the following time with the same ventilator setting: a) baseline; b) after 30 mins with heliox; and c) 30 mins after return to air-oxygen. MEASUREMENTS AND MAIN RESULTS: Heliox ventilation decreased intrinsic positive end-expiratory pressure (air-oxygen vs. heliox [mean +/- sd] 13 +/- 4 cm H2O vs. 5 +/- 2 cm H2O, p < .05), trapped lung volume (air-oxygen vs. heliox 362 +/- 67 mL vs. 174 +/- 86 mL, p < .05), and respiratory changes in systolic pressure variations (DeltaPP) (air-oxygen vs. heliox 29 +/- 5% vs. 13 +/- 7%, p < .05). In the ten patients with pulmonary arterial catheters, heliox decreased mean pulmonary arterial pressure, right atrial pressure, and pulmonary arterial occlusion pressure and increased cardiac index. Preheliox DeltaPP correlated with the magnitude of reduction in intrinsic positive end-expiratory pressure during heliox ventilation. Age, preheliox Paco2, and ratio of forced expiratory volume at first second to forced vital capacity correlated inversely, whereas preheliox DeltaPP correlated positively with increases in cardiac index. CONCLUSIONS: Heliox may be a useful adjunct therapy in patients with severe chronic obstructive pulmonary disease during acute respiratory failure who have persistent intrinsic positive end-expiratory pressure-induced hemodynamic changes despite ventilator management.

Prone-position ventilation induces sustained improvement in oxygenation in patients with acute respiratory distress syndrome who have a large shunt.

OBJECTIVES: Prone-position ventilation (PPV) induces acute improvement in oxygenation in many patients with acute respiratory distress syndrome (ARDS), with some maintaining their oxygenation even after they were returned to the supine position, but it is unclear what clinical factors determine the sustained oxygenation benefit. We hypothesized that patients with ARDS who have a larger shunt would have a better acute and sustained oxygenation response to PPV. DESIGN: Prospective, nonrandomized interventional study. SETTING: Medical and surgical intensive care units, university tertiary care center. PATIENTS: Twenty-two consecutive patients, with ARDS with an average PaO2/FiO2 of 94, were administered PPV for 12 hrs followed by supine-position ventilation for 2 hrs. MEASUREMENTS: Hemodynamic and gas exchange variables were monitored. The shunt was measured as venous admixture at an FiO2 of 1.0, and compliances of the respiratory system, lung, and chest wall were measured by the esophageal balloon technique before PPV, during PPV, and during subsequent supine-position ventilation. MAIN RESULTS: Fourteen patients (64%) responded to PPV, with PaO2/FiO2 increasing by > or =20. These changes were associated with a decrease in chest wall compliance. Responders had significantly shorter time from ARDS to PPV, a lower baseline PaO2/FiO2, and a higher venous admixture. All responders maintained the improvement in oxygenation and had a greater respiratory system compliance after returning to the supine position. Time from ARDS to PPV and baseline lung injury score were negatively associated, whereas chest wall compliance, heart rate, and PaCO2 were positively associated with sustained improvement in oxygenation. CONCLUSIONS: PPV induced acute and sustained improvement in oxygenation in many patients with ARDS. The sustained improvement is more significant if PPV is administered early to patients with a larger shunt and a more compliant chest wall. Measuring venous admixture and chest wall compliance before PPV may help select a subgroup of patients with ARDS who may benefit the most from PPV.