Daily Respiratory Research Analysis
Three high-impact respiratory studies emerged today: a multicenter lung CT vision foundation model (LCTfound) that advances diagnostic and prognostic imaging across eight tasks; a large, pragmatic cluster RCT in Vietnam showing a stepped anti‑inflammatory reliever strategy with budesonide–formoterol reduces exacerbations and hospitalizations in undifferentiated chronic respiratory disease; and real‑world evidence from Argentina demonstrating strong effectiveness of RSV maternal immunization in p
Summary
Three high-impact respiratory studies emerged today: a multicenter lung CT vision foundation model (LCTfound) that advances diagnostic and prognostic imaging across eight tasks; a large, pragmatic cluster RCT in Vietnam showing a stepped anti‑inflammatory reliever strategy with budesonide–formoterol reduces exacerbations and hospitalizations in undifferentiated chronic respiratory disease; and real‑world evidence from Argentina demonstrating strong effectiveness of RSV maternal immunization in preventing infant hospitalizations.
Research Themes
- AI-enabled lung imaging foundation models for multi-task clinical utility
- Stepped anti-inflammatory reliever strategies for chronic respiratory disease in resource-limited settings
- Maternal immunization for RSV and infant hospitalization prevention
Selected Articles
1. A lung CT vision foundation model facilitating disease diagnosis and medical imaging.
LCTfound is a large-scale lung CT foundation model trained on 105,184 scans that jointly encodes imaging and clinical data to support eight clinical tasks, outperforming strong baselines across centers. It provides a unified, deployable framework for diagnosis, prognosis, reconstruction, and navigation, potentially standardizing AI-assisted thoracic imaging.
Impact: A scalable, multicenter-validated foundation model for lung CT is likely to reshape AI-enabled imaging workflows and catalyze research across diagnosis, prognosis, and therapy planning.
Clinical Implications: If prospectively validated and integrated, this model could improve consistency and speed of thoracic imaging interpretation, enhance lesion detection/segmentation, support treatment planning, and enable advanced reconstruction in low-dose or sparse-view settings.
Key Findings
- Trained on 105,184 multicenter lung CT scans with diffusion-based pretraining and joint imaging–clinical encoding.
- Supports eight tasks (enhancement, virtual CTA, sparse-view reconstruction, segmentation, diagnosis, prognosis, response prediction, 3D navigation).
- Consistently outperformed leading baselines across multiple centers, providing a unified deployable framework.
Methodological Strengths
- Large multicenter dataset (105,184 CT scans) with comprehensive, multi-task evaluations.
- Advanced training (diffusion-based pretraining) and joint encoding of imaging and clinical data.
Limitations
- Lack of prospective, randomized clinical deployment studies demonstrating impact on patient outcomes.
- Generalizability and fairness across diverse scanners, populations, and institutions require further auditing.
Future Directions: Prospective clinical trials, regulatory-grade validation, fairness auditing across demographics and scanners, and integration studies assessing workflow efficiency, safety, and outcome benefits.
The concomitant development and evolution of lung computed tomography (CT) and artificial intelligence (AI) have made non-invasive lung imaging a key component of clinical care of patients. However, the scarcity of labeled CT data and the limited generative capacity of existing models have constrained their clinical utility. Here, we present LCTfound, a large-scale vision foundation model designed to overcome these limitations. Trained on a multi-center dataset comprising 105,184 CT scans, LCTfound leverages diffusion-based pretraining and joint encoding of imaging and clinical information to support 8 tasks, including CT enhancement, virtual computed tomography angiography (CTA), sparse-view reconstruction, lesion segmentation, diagnosis, prognosis, cancer pathological response prediction, and three-dimensional surgical navigation. In comprehensive multicenter evaluations, LCTfound consistently outperforms leading baseline models, delivering a unified, broadly deployable solution that both augments clinical decision-making and elevates CT image quality across diverse practice settings. LCTfound establishes a scalable foundation for next-generation clinical imaging intelligence, uniting large AI model with precision healthcare.
2. Implementation of a stepped anti-inflammatory reliever therapy intervention with budesonide-formoterol 160/4·5 mcg by Turbuhaler versus usual care for adults presenting during exacerbations of obstructive lung disease suggestive of asthma or chronic obstructive pulmonary disease in a resource-limited setting: an open-label, cluster randomised trial.
In a 52-week, open-label cluster RCT (n=3095) across 40 district facilities in Vietnam, a stepped budesonide–formoterol anti‑inflammatory reliever strategy reduced moderate/severe exacerbations (RR 0.79) and hospitalizations (RR 0.74) versus usual care, with similar mortality and serious adverse events. Effects were consistent irrespective of baseline eosinophils.
Impact: This pragmatic, scalable intervention can be rapidly adopted in resource-limited settings to reduce exacerbations and hospitalizations in undifferentiated chronic respiratory disease.
Clinical Implications: Adopting a simple stepped budesonide–formoterol reliever algorithm with periodic review can standardize care, reduce exacerbations, and lower hospital burden where diagnostic and medication access is limited.
Key Findings
- Reduced moderate/severe exacerbations: 28.6% vs 36.0% (RR 0.794; p=0.03) over 12 months.
- Lower respiratory hospitalizations: 17.4% vs 24.1% (RR 0.737; p=0.05); mortality and grade 3–4 adverse events were similar.
- No significant interaction by baseline blood eosinophils, supporting broad applicability.
Methodological Strengths
- Large, multicenter, cluster randomized pragmatic design with 52-week follow-up.
- Clinically relevant outcomes (exacerbations, hospitalizations) and subgroup analysis by eosinophils.
Limitations
- Open-label design with potential performance bias and cluster-level contamination.
- Undifferentiated disease population; limited diagnostic phenotyping and lack of spirometric confirmation.
Future Directions: Cost-effectiveness analyses, implementation research across other LMICs, phenotyping to refine step assignment, and head-to-head comparisons with alternative simplified strategies.
BACKGROUND: In resource-limited countries, chronic respiratory disease is common, and access to diagnosis and to the multiple medications in asthma and chronic obstructive pulmonary disease guidelines is limited. Building on a previous pilot study, we hypothesised that implementation of a simple stepped anti-inflammatory reliever (AIR) approach with budesonide-formoterol in patients presenting with obstructive airways disease and recurrent exacerbations in resource-limited settings would reduce exacerbations compared to usual care. METHODS: We conducted a 52-week open-label, cluster randomised controlled trial among adults presenting with recurrent acute respiratory symptoms in Vietnam. 40 district health facilities were assigned to usual care or intervention, which comprised budesonide-formoterol 160/4·5 mcg Turbuhaler: Step 1, one inhalation as-needed for symptoms; Step 2, one inhalation twice-daily plus one inhalation as-needed; Step 3, referral for higher-level care. This was accompanied by quarterly clinical review. Primary outcome was the proportion of participants with at least one moderate or severe exacerbation over 12-months. Secondary outcomes included respiratory hospitalisations, grade 3-4 adverse events, and all-cause mortality. The trial was registered on ANZCTR.org.au (identifier ACTRN12620000649910). FINDINGS: Analysis included 3095 participants recruited between 21 July 2020 and 11 January 2022 (control:1421 vs. intervention: 1674; 75·4% [2334/3095] males; median age 63 years [IQR 55-70]). For the primary outcome, 36·0% (511/1421) control participants had at least one exacerbation during follow-up compared to 28·6% (478/1674) intervention participants (relative risk (RR) 0·794; 95% CI 0·649-0·971; p = 0·03); there was no significant interaction with baseline blood eosinophil count (β = -0·133; 95% CI -0·343 to 0·076; p = 0·21). For the secondary outcomes, fewer participants were hospitalised in the intervention group (control: 24·1% [342/1421] vs. intervention: 17·4% [292/1674]; RR 0·737, 95% CI 0·544-0·998; p = 0·05), and all-cause mortality was similar between groups (control: 3·7% [53/1421] vs. intervention: 3·4% [57/1674]; RR 0·887, 95% CI 0·548-1·435; p = 0·62). There was no significant difference in grade 3-4 adverse events (control: 6·3% [90/1421] vs. intervention: 6·4% [107/1674]; RR 1·004, 95% CI 0·747-1·349). Pneumonia risk was similar (control: 0·8% [11/1421] vs. intervention: 0·4% [7/1674]; RR 0·417, 95% CI 0·135-1·286). INTERPRETATION: Among participants presenting with recurrent exacerbations of undifferentiated chronic respiratory disease, a stepped anti-inflammatory reliever approach was associated with fewer exacerbations and hospitalisations than usual care. This approach may represent a feasible population-level strategy to reduce the global burden of chronic respiratory disease. FUNDING: This trial was funded by the National Health and Medical Research Council of Australia. The study sponsor was the Woolcock Institute of Medical Research. Budesonide-formoterol was provided by AstraZeneca.
3. Impact and effectiveness of RSV maternal immunization on infant hospitalizations in Buenos Aires: a hospital-based, multicentre, retrospective surveillance cohort study.
In a multicenter hospital-based surveillance with a nested test‑negative analysis, maternal RSVpreF immunization showed high effectiveness against infant RSV‑ALRTI hospitalization (VE 80.8% <3 months; 66.1% <6 months), with marked reductions in PICU admissions and prolonged stays, and a one‑third population‑level reduction in infant hospitalizations.
Impact: Provides timely real-world effectiveness data supporting maternal RSV vaccination policies and resource allocation to reduce severe infant respiratory disease.
Clinical Implications: Health systems can expect substantial reductions in RSV-related infant hospitalizations, PICU admissions, and lengthier stays by implementing maternal RSV immunization, especially targeting the third trimester.
Key Findings
- Vaccine effectiveness against RSV‑ALRTI hospitalization: 80.8% (<3 months) and 66.1% (<6 months).
- VE against PICU admission 87.2% and against prolonged hospitalization 88.6% (<6 months).
- Population-level impact: 33.6% reduction in RSV‑ALRTI hospitalizations (infants <6 months) in 2024 vs expected from prior years; number needed to immunize ≈84.
Methodological Strengths
- Hospital-based multicenter surveillance over seven years with nested test‑negative design.
- Adjusted analyses for key confounders (age, sex, comorbidities, epidemiological week).
Limitations
- Retrospective design with potential residual confounding and data completeness limits.
- Generalizability limited to participating hospitals and early implementation period; VE sample relatively small.
Future Directions: Expand surveillance to diverse regions, evaluate durability across seasons, assess co‑administration and coverage strategies, and perform cost‑effectiveness analyses.
BACKGROUND: Respiratory syncytial virus (RSV) is a major cause of hospitalizations and mortality in young infants worldwide. The RSVpreF maternal immunization (MI) was recently introduced in Argentina. METHODS: This study assessed the impact of RSVpreF MI on RSV-related acute lower respiratory tract infections (ALRTI) hospitalizations through a hospital-based, multicentre, retrospective surveillance cohort study, and measured vaccine effectiveness (VE) using a nested test-negative case-control study. Data of hospitalized infants under 18 months of age was collected and analysed within seven years from three Argentine tertiary hospitals. VE analysis included ALRTI-hospitalized infants who were born between March 1 and November 9, 2024, were under 6 months of age when tested for RSV, and whose mothers were eligible for prenatal RSV immunization. Expected RSV-ALRTI hospitalizations were compared with observed cases using a Poisson model. We estimated the VE of RSVpreF MI against RSV-ALRTI hospitalizations, paediatric intensive care unit (PICU) admissions, and extended hospital stays by comparing these rates in vaccinated and unvaccinated under 3 and 6 months. FINDINGS: A total of 3373 participants were included in the impact analysis, fromof whom 323 were born during the vaccination period and were eligible for the VE analysis. The VE of RSVpreF MI was 80·8% (95% CI: 62·8-90·5%), and 66·1% (95% CI 30·1-83·8) for infants under 3 and 6 months, respectively, adjusted for age, sex, comorbidities, and epidemiological weeks. VE for PICU admission was 87·2% (95% CI 52·6-97·0) and 88·6% (95% CI 62·3-97·1) for extended hospital stays in infants under 6 months. The vaccine reduced RSV-ALRTI hospitalizations in infants under 6 months by 33·6% (95% CI 29·5-37·2) in 2024 compared to expected cases from previous years. The number needed to immunize to prevent one RSV-related hospitalization was 83·9 (95% CI 65·9-185·4). INTERPRETATION: RSVpreF MI significantly reduced RSV-ALRTI hospitalizations, averting one-third of such hospitalizations in infants under 6 months. These findings provide valuable evidence for policymakers and health authorities. FUNDING: Gates Foundation and Thrasher Research Fund.