Exacerbations of Chronic Obstructive Pulmonary Disease (ECOPD) are often triggered by bacterial and viral infections, yet current diagnostic methods are time-consuming and biomarkers (e.g. C-reactive protein (CRP)) show limited accuracy in differentiating between these triggers. Hemocytometric parameters derived from routine blood analysis may provide a rapid, accessible alternative. This study evaluated whether machine learning applied to routine hemocytometric parameters can accurately distinguish between bacterial and viral etiology in ECOPD.We performed a post hoc, exploratory analysis on 64 patients hospitalized for an ECOPD. Bacterial and viral pathogens were detected using sputum culture and/or polymerase chain reaction on nasopharyngeal swab samples. After preprocessing 53 hemocytometric parameters, Random Forest models were developed using selected hemocytometry features for bacterial (n = 12) and viral (n = 6) presence. Model performance was assessed using repeated stratified train-test splits with out-of-sample predictions.Hemocytometry-based models showed good performance for classification of bacterial presence (AUC 0.783; 95% CI 0.713–0.853), with accuracy, sensitivity, and specificity of 72%, 73%, and 70%, respectively, outperforming CRP (AUC 0.585, p < 0.001). For classification of viral presence, the model achieved an AUC of 0.717 (95% CI 0.636–0.798), with accuracy, sensitivity, and specificity of 72%, 59%, and 80%, respectively.In this exploratory analysis, hemocytometry-based machine learning models show promising performance for classifying the presence of bacterial and viral infections in ECOPD, outperforming CRP. The worldwide, routine availability of hemocytometry underscores the relevance of our findings as a practical tool for identifying bacterial and viral causes and supports potential for rapid clinical implementation. However, they require validation in larger cohorts.Clinical trial registration NCT04961060.
Bronchoscopic lung volume reduction (BLVR) is a minimally invasive option for advanced COPD, but carries bleeding risk. A significant proportion of COPD patients requiring BLVR are on chronic anticoagulation, yet data on procedural safety in this population remains scarce. We conducted a retrospective study using the National Inpatient Sample (NIS) database (2016–2023). Adult BLVR patients were stratified by chronic anticoagulation status. The primary outcomes were hemothorax and pulmonary bleeding. Secondary outcomes included any bleeding (gastrointestinal, pulmonary, intracranial, retroperitoneal, hematuria, hemothorax, or acute blood loss anemia), pneumothorax, in-hospital mortality, major adverse cardiac events (MACE), and prolonged hospitalization (≥7 days). Multivariable logistic regression yielded adjusted odds ratios (aORs) controlling demographics, comorbidities, and hospital characteristics. Of 16,340 weighted BLVR hospitalizations, 1,275 (7.8%) involved anticoagulated patients. Anticoagulated patients were older (68.6 vs. 65.7 years) and more frequently male (70.6% vs. 58.5%), with a higher burden of cardiovascular comorbidities. Hemothorax and pulmonary bleeding occurred in 2.7% and 7.1% of anticoagulated patients versus 1.1% and 4.7% of non-anticoagulated patients, respectively. Any bleeding occurred in 14.1% of anticoagulated versus 10.8% of non-anticoagulated patients. These differences were not statistically significant, and chronic anticoagulation was not independently associated with any increased bleeding on multivariable analysis. No significant differences were observed in mortality (aOR 0.88), pneumothorax (aOR 0.72), MACE (aOR 1.13), or prolonged hospitalization (aOR 1.09). Chronic anticoagulation was not independently associated with adverse outcomes following BLVR, suggesting comparable short-term outcomes in this population. Prospective studies examining the timing and types of anticoagulation in BLVR are warranted.
Long-term oxygen therapy (LTOT) for patients with severe chronic hypoxemia is a well-established treatment, but data on long-term outcomes are limited. This study aims to describe the characteristics and survival of patients using LTOT in Belgium. Data from patients receiving LTOT were extracted from the Belgian Social Security databank with 4 years of follow-up and compared with a more detailed tertiary care cohort from the University Hospital of Leuven. Kaplan-Meier curves for overall and transplant-free survival were constructed. We performed log-rank univariate analyses for diagnosis and starting treatment modality and applied a Cox proportional hazards model adjusting for age and underlying diagnosis. In Belgium, 4973 patients started LTOT in 2017. Overall 4-year survival was 22% and determined by age, baseline oxygen modality and diagnosis. LTOT was primarily prescribed for patients with COPD (75%) who had higher 4-year survival (24%) compared to patients with restrictive pulmonary disease (17%) (HR 1.23, p < 0.005). In the University hospital cohort (n = 533), 4-year transplant-free survival was 26% with significantly lower survival rates of patients with ILD (13%) (HR 1.98, p < 0.005) and other obstructive lung diseases (24%) (HR 1.56, p < 0.01) as compared to COPD (33%). Survival for patients receiving LTOT remains poor. Best prognosis is seen in patients with COPD, while patients with restrictive pulmonary disease and ILD in particular have the poorest outcome. Palliative care and end-of-life discussions should be an integral part of prescribing LTOT.
OBJECTIVE:To examine the seasonal distribution of AECOPD hospitalizations in Shijiazhuang (2020-2024), the lag associations of air pollutants (PM2.5, PM10, NO2, O3, SO2, CO) and meteorological factors (mean temperature, diurnal temperature range) across seasons, and to explore the season-specific models for high-incidence admission days as an exploratory environmental-surveillance analysis rather than a ready-to-deploy clinical early-warning system. METHODS:This city-wide retrospective time-series study analyzed 22,468 AECOPD admissions in Shijiazhuang from January 1, 2020 to December 31, 2024, using hospitalization front-page records covering all tertiary, secondary, and community hospitals in the city, obtained from the municipal medical insurance database (inpatient admissions only; ED discharges without admission were not included). Each year was divided by solar terms into spring (3/7-6/7), summer (6/8-9/7), autumn (9/8-12/7), and winter (12/8-3/6), with daily air pollution and meteorological data matched to admissions. Univariate and multivariate Poisson generalized linear regression (GLM) assessed single-day lag associations (lags 0-7 as eight separate lags); multivariable models and prediction features used the 7-day mean of lags 0-6 (admission day + preceding 6 days), not an 8-day lag0-7 average. Using each factor's 7-day cumulative window as input, we built logistic regression, random forest, and gradient boosting models and evaluated high-incidence-day prediction via fivefold time-series cross-validation, treating the many lag × pollutant × season tests as exploratory. Sensitivity analyses added during revision included collinearity diagnostics (VIF/correlation matrix), quasi-Poisson models with smooth time, a distributed-lag nonlinear temperature model (DLNM), COVID-period analyses, humidity/heating checks, alternative lag windows, and a near-term exposure prediction re-run. RESULTS:AECOPD admissions peaked in winter and were the lowest in summer (winter 6,009; spring 5,854; autumn 5,554; summer 5,051); the male-to-female ratio was ∼3.7:1 and mean age 71.7 ± 9.1 years. After multivariate adjustment, each 10 µg/m³ rise in winter PM10 and NO2 was associated with a 6.9% (IRR 1.069, 95%CI 1.052-1.085) and 8.9% (1.089, 1.050-1.129) higher admission rate, respectively. A 1 °C fall in mean temperature corresponded to a 4.4% increase (IRR 0.958, 0.947-0.969) and a 1 °C wider diurnal range to a 5.5% increase (1.055, 1.034-1.077). O3 was positively associated in summer and autumn (1.035, 1.015-1.056; 1.022, 1.005-1.039). Apparent protective multipollutant IRRs (e.g., winter PM2.5) were interpreted cautiously given collinearity. Prediction discrimination was the modest even when above chance: winter random forest fivefold CV AUC of 0.653 ± 0.112; other seasons ranged 0.50-0.62. Overall, environmental exposures were weak predictors of high-incidence days. Diagnostic re-analyses (Section 3.4), after reconciling the panel to the locked 22,468-admission cohort, showed acceptable collinearity (all VIF < 5, addressed with single-pollutant models), overdispersion handled by quasi-Poisson, and nonlinear temperature effects by DLNM (cold-induced excess, heat-related reduction), alongside a 2022 admission trough (∼50-54% of 2020/2024) and a near-term exposure re-run (same-day and lags 1-3 features) that reproduced the modest winter discrimination (AUC ≈0.66); against a calendar-only baseline the incremental value of the environmental block was small (overall ΔAUC ≈ +0.02 to +0.045), indicating the exposures are primarily explanatory rather than operationally predictive. CONCLUSION:AECOPD admissions show high winter incidence, with PM10, NO2, CO, low temperature, and a wide diurnal range as the main winter associated factors, and summer-autumn O3 exposure warranting attention. The 7-day cumulative-exposure model showed only the modest discrimination in winter (AUC ≈0.65) and is best viewed as an exploratory, hypothesis-generating surveillance analysis rather than a preliminary operational early-warning tool.
BACKGROUND:People with COPD are more likely to be sedentary, a behaviour associated with reduced life expectancy. COPD is also associated with social isolation and socioeconomic deprivation. Therefore, this study aimed to examine the relationship between these determinants and sedentary time. METHODS:This study analysed 23,559 UK Biobank participants with COPD (recruited 2006-2010), with a follow-up accelerometer assessment (2013-2015). Primary analyses were performed on accelerometer participants (n = 1,558) with complete covariate and accelerometer data. Associations between sedentary time, social isolation, and deprivation were explored using multivariable linear regression, stratified by COPD severity (GOLD stages). RESULTS:Sedentary time rose stepwise with disease severity (p < 0.001); with GOLD 4 individuals spending 19% more of their day sedentary than GOLD 1. Social isolation interacted with severity (p = 0.033), predicting an extra 1.5 h of daily sedentary time in GOLD 3-4 (95% CI, 0.49-2.47; p = 0.004). Deprivation had the greatest impact on sedentary time in mild disease. CONCLUSIONS:This large observational study suggests sedentary behaviour is associated with social isolation in severe COPD and socioeconomic deprivation in mild disease severity. This highlights the need to consider individual social and environmental factors alongside a clinical focus to reduce sedentary behaviour.
