Antoni Rosell

Background: Airway microbiome composition in bronchiectasis extends beyond conventional microbiology, but its clinical implications remain unclear. Aim: This study evaluated the airway bacteriome and mycobiome in bronchiectasis, examining their associations with clinical characteristics, inflammatory patterns and disease aetiology. Methods: This study reports the pre-specified baseline cross-sectional analysis from a prospective multicentre cohort of 153 adults with non-CF bronchiectasis studied during clinical stability. Airway bacterial and fungal communities were characterised using 16S rRNA (V3–V4) and ITS2 sequencing. Samples were stratified by dominant bacterial genus, associations with clinical features, inflammatory biomarkers and aetiological subgroups were assessed. LEfSe was used as an exploratory method to identify bacterial genera with differential relative abundance across dominance groups. Results: Pseudomonas -dominant bacteriomes showed the lowest α-diversity, and the strongest associations with disease severity, including poorer lung function, higher BSI scores and elevated sputum inflammatory markers. Haemophilus -dominant communities also correlated with reduced diversity and airway inflammation, although were associated with milder clinical severity compared with Pseudomonas . In contrast, Streptococcus -dominant bacteriomes demonstrated the highest bacterial diversity, enrichment of commensal genera, lower inflammatory marker levels and more favourable clinical features. The mycobiome also contributed to airway ecology: Ascomycota abundance correlated with higher bacterial diversity and lower inflammatory biomarker levels, whereas Basidiomycota dominance was associated with reduced diversity and a pro-inflammatory pattern. Distinct taxa were associated with specific aetiologies, including Rothia and Porphyromonas in immunodeficiencies and Staphylococcus in primary ciliary dyskinesia. Conclusion: Airway microbiome dominance is associated with distinct clinical characteristics and inflammatory patterns in bronchiectasis. In a well-characterised multicentre cohort, integration of bacteriome and mycobiome data with inflammatory and remodelling markers provides external validation of previous observations and may contribute to a better characterisation of disease heterogeneity.

Bacterial colonization and biofilm formation remain major limitations of airway stenting, contributing to device obstruction and infection-related complications. Here, we translated to silicone tracheal stents a previously developed plasma-assisted micro/nanostructured silver interface designed to provide hybrid anti-biofilm activity through reduced bacterial adhesion and controlled silver release. The coated stents were characterized to confirm the generation of the nanostructured silver coating and were evaluated against Gram-positive and Gram-negative bacteria in vitro, followed by in vivo assessment in a miniature pig airway model. In vitro, the coating reproduced the dual bacteriophobic and silver-mediated antimicrobial behavior previously observed on model silicone surfaces. In vivo, animals implanted with silver-coated stents showed lower endpoint bacterial loads in bronchial washes than those receiving commercial stents (1.0 × 108 vs 1.2 × 109 CFU/mL, p =0.0085). Bacterial adhesion on explanted stents was also markedly lower for coated devices (109 vs 1011 CFU /cm2, 99.8% reduction, p < 0.0001), and scanning electron microscopy showed a thinner and less mature biofilm structure. Potentially pathogenic bacteria, including Klebsiella pneumoniae and Pseudomonas aeruginosa, were more frequently detected on commercial stents, whereas coated stents were mainly associated with the native respiratory microbiota. Overall, these results indicate that nanostructured silver-coated silicone tracheal stents display hybrid anti-biofilm functionality in a challenging in vivo airway setting.

Background: Chronic respiratory diseases often require long-term ventilatory support, leading to a growing number of patients treated with home mechanical ventilation (HMV). Despite advancements in telemonitoring with real-time tracking of noninvasive mechanical ventilation enabled by integrated software in HMV devices, early signs of respiratory decompensation may go unnoticed, leading to emergency visits and hospitalizations, which burden both patients and health care systems. Objective: This study aims to develop and evaluate a machine learning–based model capable of predicting respiratory decompensation events, defined as emergency visits or hospitalizations due to acute deterioration in the patient’s underlying respiratory condition. These events reflect episodes of worsening respiratory status that require urgent medical attention. The model uses data from HMV telemonitoring platforms, with the aim of improving patient outcomes. Methods: This retrospective study analyzed data from 482 patients on HMV monitored via ResMed, Philips, and Breas platforms, collected between March 2021 and November 2024 at the Germans Trias i Pujol Hospital in Catalonia, Spain. Data included device use, compliance, mask leakage, and ventilator settings. Decompensation was defined as emergency department visits or hospitalizations. A windowing strategy captured the 5 weeks prior to events. Multiple machine learning models were trained using grid search to identify the optimal hyperparameters, prioritizing recall in order to minimize false negatives. Models were evaluated using 10-fold cross-validation. Finally, Shapley Additive Explanations (SHAP) analysis was used for model interpretability. Results: The final dataset included 157 data windows, balanced for positive and negative cases. Among the models tested, logistic regression achieved the highest recall (mean 0.94, SD 0.06, 95% CI 0.90‐0.98) though with moderate accuracy (mean 0.60, SD 0.05, 95% CI 0.56‐0.64). The random forest classifier achieved the best balance across metrics (accuracy: mean 0.66, SD 0.10, 95% CI 0.59‐0.73; recall: mean 0.78, SD 0.15, 95% CI 0.67‐0.89; F ₁ -score: mean 0.70, SD 0.10, 95% CI 0.63‐0.77). SHAP analysis revealed that higher use, leakage, and compliance in the week before a decompensation event were key predictors, suggesting compensatory behavior or early clinical deterioration. Overall performance remained moderate, reflecting limitations in sample size, incomplete longitudinal records for daily data, and the absence of key physiological measurements. Conclusions: This study demonstrates the feasibility of predicting respiratory decompensation using data from HMV telemonitoring systems. Tree-based ensemble models, particularly random forests, provided the most balanced performance, while SHAP analysis offered clinically relevant insights. Although performance was moderate, the findings support further development of predictive tools to enable timely telemedical interventions. Limitations include sample size, missing physiological parameters, and a single-center design. Future research should expand to multicenter datasets and incorporate additional clinical variables to enhance model robustness and generalizability.

Background/Objectives: This study evaluates the adherence to guidelines for the management of asthma exacerbations in the ED, recommendations at discharge, and impact at a 1-year of follow-up. Methods: An observational study of 87 asthma patients who attended the ED during 2022 and were discharged within 24 h was carried out. Data before the ED admission, care in the ED, and discharge reports, as well as the clinical characteristics at follow-up, were recorded. The relationship between complete ED discharge reports and outcome at 1 year, and factors associated with new exacerbations were analyzed. Results: The mean age was 51 years, 80% of the patients were women, and 50% had severe asthma. Prior to ED admission, 58.8% of patients used ICS-LABA, 26.2% triple therapy, 31.8% had not been treated, and 51.2% had presented at least one exacerbation. On ED admission, PEF was measured in 21% of patients only, decreasing to 6.8% at 3 h. In the ED discharge reports, the use of systemic corticosteroids was recommended in 76.5% of the cases and ICS-LABA in 46.9%. However, complete ED discharge reports were recorded for only 18.2% of patients. A total of 6.7% of patients were referred to a primary care physician and 29.9% to a pneumologist. Complete ED discharge forms did not improve asthma control at follow-up or reduce new exacerbations. Exacerbations before ED admission (OR 2.49, 95% CI 1.47–4.22, p = 0.001) and the use of any asthma controller treatment (OR 1.84, 95% CI 1.84–507, p = 0.017) were associated with ≥2 exacerbations at follow-up. Conclusions: Contact with ED did not improve disease control or reduce exacerbations. It is necessary to optimize care before, during, and after exacerbations by developing integrated programs with primary care to improve asthma management.

BACKGROUND Chronic respiratory diseases often require long-term ventilatory support, leading to a growing number of patients treated with home mechanical ventilation (HMV). Despite advancements in telemonitoring with real-time tracking of non-invasive mechanical ventilation (NIMV) enabled by integrated software in HMV devices, early signs of respiratory decompensation may go unnoticed, leading to emergency visits and hospitalizations, which burden both patients and healthcare systems. OBJECTIVE The objective of this study is to develop and evaluate a machine learning-based model capable of predicting respiratory decompensation events, defined as emergency visits and hospitalizations, using data from HMV telemonitoring platforms, with the aim of improving patient outcomes. METHODS This retrospective study analyzed data from 482 HMV patients monitored via ResMed, Philips, and Breas platforms, collected between March 2021 and November 2024 at Germans Trias i Pujol Hospital in Catalonia (HUGTiP), Spain. Data included device usage, compliance, mask leakage, and ventilator settings. Decompensation was defined as emergency department visits or hospitalizations. A windowing strategy captured the five weeks prior to events. Multiple machine learning models were trained using grid search to identify the optimal hyperparameters, prioritizing recall in order to minimize false negatives. Models were evaluated using 10-fold cross-validation. Finally, Shapley additive explanations (SHAP) were used for model interpretability. RESULTS The final dataset included 157 data windows, balanced for positive and negative cases. Among the models tested, logistic regression (LR) achieved the highest recall (0.94 ± 0.06) though with moderate accuracy (0.60 ± 0.05). The random forest classifier (RFC) achieved the best balance across metrics (accuracy: 0.66 ± 0.10; recall: 0.78 ± 0.15; F1 score: 0.70 ± 0.10). SHAP analysis revealed that higher usage, leakage, and compliance in the week before a decompensation event were key predictors, suggesting compensatory behavior or early clinical deterioration. Overall performance remained moderate, reflecting limitations in sample size, incomplete longitudinal records for daily data, and the absence of key physiological measurements. CONCLUSIONS This study demonstrates the feasibility of predicting respiratory decompensation using data from HMV telemonitoring systems. Tree-based ensemble models, particularly random forests, provided the most balanced performance, while SHAP analysis offered clinically relevant insights. Although performance was moderate, the findings support further development of predictive tools to enable timely telemedical interventions. Limitations include sample size, missing physiological parameters, and single-center design. Future research should expand to multicenter datasets and incorporate additional clinical variables to enhance model robustness and generalizability.

Mediastinal lymph node (LN) staging is routinely performed using PET/CT and EBUS-TBNA. Promising predictive algorithms for lymph nodes have been reported for each technique, both individually and in combination. This study aims to develop a predictive algorithm that combines EBUS, PET/CT and clinical data to provide a probability of malignancy. A retrospective study was conducted on consecutive patients with non-small cell lung carcinoma staged using PET/CT and EBUS-TBNA. Lymph nodes were identified by level (N1, N2, and N3) and anatomical region (AR) (subcarinal, paratracheal, and hilar). A Standardized Uptake Value (SUV) was determined for each sampled LN. The ultrasound features collected included diameter in the short axis (DSA), morphology, border, echogenicity and the presence of the vascular hilum. A robust logistic regression model was used to construct an algorithm to estimate the probability of malignancy of the lymph node. A total of 116 patients with a mean age of 66, 93% of whom were men, were included. 358 lymph nodes were evaluated, 51% of which exhibited adenocarcinoma and 35% were squamous, while 14% were classified as non-small-cell lung carcinoma. The model estimated the probability of malignancy for each lymph node using age, DSA, SUVmax, and AR. The Area Under the ROC curve, was 0.89. A user-friendly application was also developed ( https://ubidi.shinyapps.io/lymma/ .) The integration of patient clinical characteristics, EBUS features, and PET/CT findings may generate a pre-sampling malignancy probability map for each lymph node. The model requires prospective and external validation. A predictive algorithm for lymph metastasis in NSCLC combining the patient clinical characteristics, EBUS features, and PET/CT findings could facilitate the estimation of the specific probability of malignancy of a single lymph node in the lymph node probability map.

Objective: To describe the cohort of COPD patients monitored by Long-Term Home Non-Invasive mechanical ventilation (LTH-NIV), to evaluate the economic impact of the intervention and to determine the existence of clinical or ventilatory variables that are related to the success of the therapy.Methodology: Retrospective observational descriptive study of COPD patients with LTH-NIV with telemonitoring between 2019 and 2022.Results: 68 patients were included, 73% male. Mean age 71 years. Emphysema 60.3% with a mean obstruction of 42.3%; comorbidity prevalence of OSA was 27.7% and AHT 19.5%. Mean PaCO2 before starting LTH-NIV was 57 mmHg; after starting therapy it was 46 mmHg. ST mode with the Resmed® Lumis 150 device. The mean IPAP was 17.04 cmH2O in 98.5%; the mean EPAP was 8.91 cmH2O. Compliance was 7.66 hours and mean leak was 26.65 L/min. In more than half of the patients the PaCO2 was reduced by 20% once the therapy was started, which was significantly associated with a good prognosis. Compliant patients have a lower economic cost. The mean cost of the patients is €8,548 (SD €14,662). AHT was the only comorbidity significantly associated with the failure of LTH-NIV.Conclusions: PaCO2 is not the only variable to be taken into account and specific clinical manifestations, obstruction severity, and certain comorbidities may be critical in determining LTH-NIV indication for COPD patients.

Background and Objective: Interstitial lung diseases (ILDs) encompass over 200 entities. Among them, fibrosing lung diseases, have recently generated special interest due to the emerging therapies for their management. However, it is important to deepen our knowledge of other less prevalent ILD, since many of them are associated with a poor prognosis. This narrative review aims to provide a practical and up-to-date description of some poorly recognized ILD. It covers rare idiopathic interstitial pneumonias and their histologic patterns, genetic disorders with interstitial lung involvement (Hermansky-Pudlak syndrome), and ILD associated with benign proliferation of pulmonary lymphoid tissue, namely follicular bronchiolitis and granulomatous-lymphocytic interstitial lung disease. Methods: Electronic searches of PubMed and Google Scholar using specific keywords were conducted. Articles underwent screening for relevance, covering review articles, meta-analyses, systematic reviews, case series, prospective studies, society guidelines, editorials in peer-reviewed journals; scientific books on the subject. The data included was limited to English and Spanish publications. Key Content and Findings: Despite the low prevalence of these diseases, the increased recognition of radiological patterns, pathological features, and diagnostic procedures, have permitted their better characterization. This review highlights epidemiology, clinical presentation, diagnosis, natural history, and treatment. Conclusions: Lesser-studied ILD represent a diagnostic and therapeutic challenge and can be frequently misdiagnosed. Also, due to the lack of randomized controlled trials, there are no well-established therapeutic options. Further studies or registries are needed to improve accurate diagnosis and management.