This Week in Pulmonary — Sep 29, 2026

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Pulmonary. This week we're covering 10 notable papers spanning imaging biomarkers in obstructive and vascular lung disease, targeted therapeutics in cystic fibrosis and severe asthma, and risk stratification in the critically ill. Let's dive in.

We'll start with diagnosis, because the most immediately usable paper this week is a meta-analysis in Chest on adaptive D-dimer thresholds in suspected pulmonary embolism [1]. The worry has been that age-adjusted and clinical-probability-adjusted cut-offs, which let patients with a D-dimer above the traditional five hundred go home without imaging, might be trading safety for efficiency. Graziani and colleagues pooled six prospective diagnostic management studies covering just over twelve thousand patients with suspected embolism, where prevalence ranged from about seven to nineteen percent, and asked specifically about failures inside the so-called adaptive window, that band above five hundred but below the adjusted threshold. Across everyone ruled out without CT pulmonary angiography, the three-month failure rate was about one in a thousand. Within the adaptive window itself, where risk should concentrate, it was about one in two hundred, and roughly six in ten patients avoided imaging altogether. Heterogeneity for the safety outcome was essentially absent. The authors conclude the failure rate is acceptably low, including in that adaptive window, which is reassuring for the strategies already embedded in several guidelines and undercuts the post-hoc analyses that had raised the alarm.

Turning to imaging as prognosis rather than diagnosis, two papers in the American Journal of Respiratory and Critical Care Medicine make the case that a routine chest CT carries information we are largely throwing away. Wu and colleagues followed nearly two thousand community-based participants in China for three years, about eleven hundred of them with preserved spirometry, and scored airway mucus plugs by the number of lung segments involved [2]. Among people with entirely normal spirometry, those with one or two plugged segments had roughly a twenty percent higher risk of respiratory exacerbations, and those with three or more had a risk about forty percent higher. More striking, participants with three or more plugs went on to develop spirometry-defined chronic obstructive pulmonary disease about eighteen percent of the time, compared with under eight percent of those with no plugs, with a correspondingly faster fall in the ratio of forced expiratory volume to forced vital capacity. The same threshold predicted exacerbations in those with established, including asymptomatic, chronic obstructive pulmonary disease. This is observational and from a single national cohort, so it identifies a marker of risk rather than a treatment target, but it extends mucus plugging well upstream of what we currently call disease.

Also in the American Journal of Respiratory and Critical Care Medicine, Hermann and colleagues used SPIROMICS to ask whether one year of change on quantitative CT forecasts the next several years of lung function loss [3]. Among eight hundred and eighty participants, average decline in forced expiratory volume in one second was about thirty millilitres a year. A one standard deviation increase over a single year in machine-learned combined bronchitic-apical emphysema added roughly four millilitres a year of extra decline, and the same increase in diffuse emphysema added about six. Conventional density-based emphysema measures also predicted decline, but when everything was modelled together, only the two machine-learned subtypes held up independently. The effect sizes are modest at the individual level; the authors frame them as candidate intermediate endpoints that could shorten chronic obstructive pulmonary disease trials, not as a clinic-ready prognostic tool.

The pulmonary vascular theme produced two papers that speak to each other. In Chest, Couch and colleagues measured pulmonary artery diameter and the pulmonary artery to aorta ratio on CT in two hundred and fifty-two PVDOMICS participants with chronic lung disease and right heart catheterisation [8]. Larger arteries tracked with higher mean pulmonary artery pressure and with a diagnosis of pulmonary hypertension, with moderate overall discrimination, around an area under the curve of point seven nine for diameter. The important nuance is that performance varied sharply by substrate: the ratio performed well in obstructive and non-parenchymal disease but noticeably worse in interstitial lung disease, where the artery dilated less for the same rise in pressure. The evidence suggests a normal-looking pulmonary artery is considerably less reassuring in fibrotic lung disease than in emphysema. Complementing that, Komori and colleagues in Respiratory Medicine retrospectively reviewed one hundred and fifty-eight patients and found that nearly half of patients with pulmonary arterial hypertension had coexisting lung disease on baseline CT [9]. Those patients were older, walked shorter distances, and had survival that resembled group three pulmonary hypertension rather than classical pulmonary arterial hypertension, with an age-adjusted mortality roughly five-fold higher, though with wide statistical uncertainty in a small cohort. Respiratory failure accounted for about forty percent of deaths in that overlap group. It is retrospective and single-country, but it argues the overlap phenotype is common and prognostically distinct.

On the therapeutic side, the European Respiratory Journal published a mechanistic study that helps explain why starting modulators early matters. Berger and colleagues performed intestinal current measurements on fresh rectal biopsies in twenty-six children aged two to eleven with cystic fibrosis and at least one F508del allele, before and four months after starting elexacaftor, tezacaftor and ivacaftor [4]. In adolescents and adults, this therapy restores CFTR function to around forty percent of non-cystic-fibrosis controls. In these children, the cyclic AMP–induced chloride response reached a median of about ninety percent of normal, and total chloride secretion essentially reached the non-cystic-fibrosis median. Response correlated strongly and inversely with age. It is a small, observational, multicentre study using a surrogate tissue rather than airway outcomes, but near-normalisation of CFTR function in young children is a meaningful biological signal.

In Respiratory Medicine, Betancor and colleagues report twelve-month real-world outcomes for tezepelumab in one hundred and fifty-seven patients with severe asthma, of whom twenty-three had aspirin-exacerbated respiratory disease [5]. In that subgroup, severe exacerbations fell by about seventy percent, emergency visits by about three quarters, and cumulative oral corticosteroid exposure by more than half, with significant improvement in asthma control. Lung function changes were modest and did not reach statistical significance. Patients with aspirin-exacerbated disease trended toward greater benefit than those without, but the authors are explicit that penalized sensitivity analyses were imprecise and preclude any firm conclusion about a differential treatment effect. This is hypothesis-generating support for upstream TSLP blockade in this phenotype, not confirmation of it.

Finally, three papers on stratification and classification. In Chest, Martínez de Victoria Carazo and colleagues studied seventy-eight adults with non-HIV Pneumocystis jirovecii pneumonia across three Spanish hospitals, where in-hospital mortality was fifty percent [7]. Using two universally available admission labs, lactate dehydrogenase at or above four hundred and lymphocyte count below seven hundred, they defined four phenotypes, and thirty-day mortality rose in an orderly gradient from about twelve percent in the lowest-risk group to nearly two thirds in the group with both abnormalities. That convergent cytolytic and immunodepressed phenotype carried roughly a six-fold higher adjusted mortality risk, and the model discriminated well on internal bootstrap validation. It is retrospective, small, and has not been externally validated, so it is best read as a framework for stratified investigation rather than a validated score. In the Journal of Thoracic Oncology, Rekhtman and colleagues assembled twenty-eight lung tumours with carcinoid morphology but more than ten mitoses, from nine institutions across six countries [6]. These tumours had wild-type Rb and p53, low tumour mutational burden, and carcinoid-typical genomic alterations, yet survival significantly worse than either typical or atypical carcinoids. The IASLC Pathology Committee proposes classifying them as grade three carcinoids rather than large cell neuroendocrine carcinoma, which would align management with their biology. And in Chest, Bonny and colleagues offer a narrative review of cough function during mechanical ventilation, laying out the mechanisms of cough failure in the critically ill and the available quantification methods, while making clear that assessment remains unstandardised despite ineffective cough being an established risk factor for extubation failure and post-extubation pneumonia [10].

If you only have time for one paper this week, make it the Chest meta-analysis of adaptive D-dimer thresholds [1]. It directly addresses the safety objection that has been raised against age- and probability-adjusted cut-offs in everyday emergency practice, and it does so with pooled prospective management data rather than post-hoc reanalysis.

Here is what this week's evidence adds up to in Pulmonary. Adaptive D-dimer strategies now have pooled prospective evidence that their failure rate is low even in the window that generated the concern, which strengthens rather than overturns current guidance. Routine chest CT appears to carry prognostic information well before spirometry becomes abnormal, with mucus plugs marking future exacerbations and incident airflow obstruction, and one-year change in emphysema subtypes predicting subsequent lung function loss, though both remain observational and the emphysema work is framed as a trial endpoint rather than a bedside test. In pulmonary vascular disease, the evidence suggests pulmonary artery dilation is a less sensitive marker of pulmonary hypertension in interstitial disease than in obstructive disease, and that coexisting lung disease in pulmonary arterial hypertension identifies a group whose prognosis looks like group three disease. Among therapeutics, modulator therapy restored CFTR function to near-normal in young children in a small observational study, while tezepelumab's apparently greater benefit in aspirin-exacerbated respiratory disease remains statistically unsettled. And the proposal for a grade three carcinoid category reopens a classification question the World Health Organization has not yet resolved.

That's your roundup for This Week in Pulmonary. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.