Evaluation of the impact of artificial intelligence-assisted image interpretation on the diagnostic performance of clinicians in identifying pneumothoraces on plain chest X-ray: a multi-case multi-reader study.

BACKGROUND: Artificial intelligence (AI)-assisted image interpretation is a fast-developing area of clinical innovation. Most research to date has focused on the performance of AI-assisted algorithms in comparison with that of radiologists rather than evaluating the algorithms' impact on the clinicians who often undertake initial image interpretation in routine clinical practice. This study assessed the impact of AI-assisted image interpretation on the diagnostic performance of frontline acute care clinicians for the detection of pneumothoraces (PTX). METHODS: A multicentre blinded multi-case multi-reader study was conducted between October 2021 and January 2022. The online study recruited 18 clinician readers from six different clinical specialties, with differing levels of seniority, across four English hospitals. The study included 395 plain CXR images, 189 positive for PTX and 206 negative. The reference standard was the consensus opinion of two thoracic radiologists with a third acting as arbitrator. General Electric Healthcare Critical Care Suite (GEHC CCS) PTX algorithm was applied to the final dataset. Readers individually interpreted the dataset without AI assistance, recording the presence or absence of a PTX and a confidence rating. Following a 'washout' period, this process was repeated including the AI output. RESULTS: Analysis of the performance of the algorithm for detecting or ruling out a PTX revealed an overall AUROC of 0.939. Overall reader sensitivity increased by 11.4% (95% CI 4.8, 18.0, p=0.002) from 66.8% (95% CI 57.3, 76.2) unaided to 78.1% aided (95% CI 72.2, 84.0, p=0.002), specificity 93.9% (95% CI 90.9, 97.0) without AI to 95.8% (95% CI 93.7, 97.9, p=0.247). The junior reader subgroup showed the largest improvement at 21.7% (95% CI 10.9, 32.6), increasing from 56.0% (95% CI 37.7, 74.3) to 77.7% (95% CI 65.8, 89.7, p<0.01). CONCLUSION: The study indicates that AI-assisted image interpretation significantly enhances the diagnostic accuracy of clinicians in detecting PTX, particularly benefiting less experienced practitioners. While overall interpretation time remained unchanged, the use of AI improved diagnostic confidence and sensitivity, especially among junior clinicians. These findings underscore the potential of AI to support less skilled clinicians in acute care settings.

National survey on management of spontaneous pneumothorax from emergency department to specialised treatment: room for improvement.

Introduction: Spontaneous pneumothorax (SP) affects both young, otherwise healthy individuals and older persons with known underlying pulmonary disease. Initial management possibilities are evolving and range from observation to chest tube insertion. SP guidelines suggest an individualized approach based on multiple factors such as symptoms, size of pneumothorax, comorbidity and patient preference. Aim: With this Danish national survey we aimed to map organization of care including involved specialties, treatment choice, training, and follow-up plans to identify aspects, and optimization of spontaneous pneumothorax management. Method: A survey developed by the national interest group for pleural medicine was sent to all departments of emergency medicine, thoracic surgery, respiratory medicine, and to relevant departments of abdominal or orthopaedic surgery. Results: The response rate was 75 % (47 of 65). Overall, 21% of responding departments had no guideline for SP management, which was provided by multiple specialties with marked heterogeneity in choice of treatment including tube size, management during admission, and referral procedure to follow-up. Few departments required procedure training, and nearly all of the responders called for improvements in management of pneumothorax. Conclusion: This survey suggests that SP management and care is delivered heterogeneously across Danish hospitals with marked difference between respiratory physicians, emergency physicians, general surgeons and thoracic surgeons. It is therefore likely that management is sub-optimal. There is a need for a common Danish SP guideline to ensure optimal treatment across involved specialties.

COVID-19-Related Pleural Diseases.

Coronavirus disease 2019 (COVID-19)-related pleural diseases are now well recognized. Since the beginning of the pandemic, increasing cases of pleural diseases including pneumothorax, pneumomediastinum, and pleural effusion with severe COVID-19 infection have attracted the attention of physicians and are not incidental or due to barotrauma. The complicated course of COVID-19 illness highlights the complex pathophysiological underpinnings of pleural complications. The management of patients with pneumothorax and pneumomediastinum is challenging as the majority require assisted ventilation; physicians therefore appear to have a low threshold to intervene. Conversely, pleural effusion cases, although sharing some similar patient characteristics with pneumothorax and pneumomediastinum, are in general managed more conservatively. The evidence suggests that patients with COVID-19-related pleural diseases, either due to air leak or effusion, have more severe disease with a worse prognosis. This implies that prompt recognition of these complications and targeted management are key to improve outcomes.

Achieving Molecular Profiling in Pleural Biopsies: A Multicenter, Retrospective Cohort Study.

BACKGROUND: Pleural biopsy findings offer greater diagnostic sensitivity in malignant pleural effusions compared with pleural fluid. The adequacy of pleural biopsy techniques in achieving molecular marker status has not been studied, and such information (termed "actionable" histology) is critical in providing a rational, efficient, and evidence-based approach to diagnostic investigation. RESEARCH QUESTION: What is the adequacy of various pleural biopsy techniques at providing adequate molecular diagnostic information to guide treatment in malignant pleural effusions? STUDY DESIGN AND METHODS: This study analyzed anonymized data on 183 patients from four sites across three countries in whom pleural biopsy results had confirmed a malignant diagnosis and molecular profiling was relevant for the diagnosed cancer type. The primary outcome measure was adequacy of pleural biopsy for achieving molecular marker status. Secondary outcomes included clinical factors predictive of achieving a molecular diagnosis. RESULTS: The median age of patients was 71 years (interquartile range, 63-78 years), with 92 of 183 (50%) male. Of the 183 procedures, 105 (57%) were local anesthetic thoracoscopies (LAT), 12 (7%) were CT scan guided, and 66 (36%) were ultrasound guided. Successful molecular marker analysis was associated with mode of biopsy, with LAT having the highest yield and ultrasound-guided biopsy the lowest (LAT vs CT scan guided vs ultrasound guided: LAT yield, 95%; CT scan guided, 86%; and ultrasound guided, 77% [P = .004]). Biopsy technique and size of biopsy sample were independently associated with successful molecular marker analysis. LAT had an adjusted OR for successful diagnosis of 30.16 (95% CI, 3.15-288.56; P = .003) and biopsy sample size an OR of 1.18 (95% CI, 1.02-1.37) per millimeter increase in tissue sample size (P < .03). INTERPRETATION: Although previous studies have shown comparable overall diagnostic yields, in the modern era of targeted therapies, this study found that LAT offers far superior results to image-guided techniques at achieving molecular profiling and remains the optimal diagnostic tool.

The bacteriology of pleural infection (TORPIDS): an exploratory metagenomics analysis through next generation sequencing.

BACKGROUND: Pleural infection is a common and severe disease with high morbidity and mortality worldwide. The knowledge of pleural infection bacteriology remains incomplete, as pathogen detection methods based on culture have insufficient sensitivity and are biased to selected microbes. We designed a study with the aim to discover and investigate the total microbiome of pleural infection and assess the correlation between bacterial patterns and 1-year survival of patients. METHODS: We assessed 243 pleural fluid samples from the PILOT study, a prospective observational study on pleural infection, with 16S rRNA next generation sequencing. 20 pleural fluid samples from patients with pleural effusion due to a non-infectious cause and ten PCR-grade water samples were used as controls. Downstream analysis was done with the DADA2 pipeline. We applied multivariate Cox regression analyses to investigate the association between bacterial patterns and 1-year survival of patients with pleural infection. FINDINGS: Pleural infection was predominately polymicrobial (192 [79%] of 243 samples), with diverse bacterial frequencies observed in monomicrobial and polymicrobial disease and in both community-acquired and hospital-acquired infection. Mixed anaerobes and other Gram-negative bacteria predominated in community-acquired polymicrobial infection whereas Streptococcus pneumoniae prevailed in monomicrobial cases. The presence of anaerobes (hazard ratio 0·46, 95% CI 0·24-0·86, p=0·015) or bacteria of the Streptococcus anginosus group (0·43, 0·19-0·97, p=0·043) was associated with better patient survival, whereas the presence (5·80, 2·37-14·21, p<0·0001) or dominance (3·97, 1·20-13·08, p=0·024) of Staphylococcus aureus was linked with lower survival. Moreover, dominance of Enterobacteriaceae was associated with higher risk of death (2·26, 1·03-4·93, p=0·041). INTERPRETATION: Pleural infection is a predominantly polymicrobial infection, explaining the requirement for broad spectrum antibiotic cover in most individuals. High mortality infection associated with S aureus and Enterobacteriaceae favours more aggressive, with a narrower spectrum, antibiotic strategies. FUNDING: UK Medical Research Council, National Institute for Health Research Oxford Biomedical Research Centre, Wellcome Trust, Oxfordshire Health Services Research Committee, Chinese Academy of Medical Sciences, and John Fell Fund.

Cost-effectiveness of ambulatory care management of primary spontaneous pneumothorax: an open-label, randomised controlled trial.

BACKGROUND: Ambulatory management of primary spontaneous pneumothorax has been shown to reduce initial hospitalisation, but at the expense of increase adverse events. As a result, questions remain about the cost-effectiveness of this option. OBJECTIVES: A within-trial economic evaluation alongside a randomised controlled trial was performed to assess the cost-effectiveness of ambulatory care when compared with standard guideline-based management. METHODS: Patients were randomly assigned to treatment with either an ambulatory device or standard guideline-based management (aspiration, standard chest tube insertion or both). Follow-up was 12 months. Outcomes included healthcare resource use and costs, quality of life, quality-adjusted life-years (QALYs) and cost-effectiveness. RESULTS: 236 patients were recruited and randomly assigned to ambulatory care (n=117) and standard care (n=119). After multiple imputation for missing data, patients in the ambulatory care group had significantly lower National Health Service healthcare costs (-£788, 95% CI difference: -1527 to -50; p=0.037) than those in the standard care group. There were no differences in the number of QALYs gained (mean difference: -0.001, 95% CI difference: -0.032 to 0.030; p=0.95). When standard care was compared with ambulatory care, the incremental cost-effectiveness ratio was £799 066 per QALY gained, well above current thresholds of cost-effectiveness. As a result, the probability of ambulatory care being cost-effective was 0.93. CONCLUSION: Outpatient ambulatory management is highly likely to be a cost-effective option in the management of primary pneumothorax. TRIAL REGISTRATION NUMBER: ISRCTN79151659.

Pneumothorax management: current state of practice in the UK.

BACKGROUND AND OBJECTIVE: Spontaneous pneumothorax is a common pathology but optimal initial treatment regime is not well defined. Treatment options including conservative management, needle aspiration (NA) or insertion of a small-bore chest drain. Recent large randomised controlled trials may change the treatment paradigm: comparing conservative and ambulatory management to standard care, but current guidelines need to be updated. The aim of this study was to assess the current "state of play" in the management of pneumothorax in the UK. METHODS: Physicians and respiratory healthcare staff were invited to complete an online survey on the initial and subsequent management of pneumothorax. RESULTS: This study is the first survey of pneumothorax practice across the UK, which highlights variation in practice: 50% would manage a large primary pneumothorax with minimal symptoms conservatively, compared to only 3% if there were significant symptoms; 64% use suction if the pneumothorax had not resolved after > 2 days, 15% always clamp the chest drain prior to removal; whereas 30% never do. NICE guidance recommends the use of digital suction but this has not translated into widespread usage: only 23% use digital suction to check for resolution of air leak). CONCLUSION: Whilst there has always been allowance for individual clinician preference in guidelines, there needs to be consensus on the optimum management strategy. The challenge the new guidelines face is to design a simple and pragmatic approach, using this new evidence base.

Aetiology of Primary Spontaneous Pneumothorax.

Air in the pleural cavity is termed pneumothorax. When this occurs in the absence of trauma or medical intervention, it is called spontaneous pneumothorax. Primary spontaneous pneumothorax typically occurs in young patients without known lung disease. However, the idea that these patients have "normal" lungs is outdated. This article will review evidence of inflammation and respiratory bronchiolitis on surgical specimens, discuss the identification of emphysema-like change (i.e., blebs and bullae), the concept of pleural porosity and review recent data on the overexpression of matrix metalloproteinases in the lungs of patients who have had pneumothorax.

Role of thoracic ultrasonography in pleurodesis pathways for malignant pleural effusions (SIMPLE): an open-label, randomised controlled trial.

BACKGROUND: Pleurodesis is done as an in-patient procedure to control symptomatic recurrent malignant pleural effusion (MPE) and has a success rate of 75-80%. Thoracic ultrasonography has been shown in a small study to predict pleurodesis success early by demonstrating cessation of lung sliding (a normal sign seen in healthy patients, lung sliding indicates normal movement of the lung inside the thorax). We aimed to investigate whether the use of thoracic ultrasonography in pleurodesis pathways could shorten hospital stay in patients with MPE undergoing pleurodesis. METHODS: The Efficacy of Sonographic and Biological Pleurodesis Indicators of Malignant Pleural Effusion (SIMPLE) trial was an open-label, randomised controlled trial done in ten respiratory centres in the UK and one respiratory centre in the Netherlands. Adult patients (aged ≥18 years) with confirmed MPE who required talc pleurodesis via either a chest tube or as poudrage during medical thorascopy were eligible. Patients were randomly assigned (1:1) to thoracic ultrasonography-guided care or standard care via an online platform using a minimisation algorithm. In the intervention group, daily thoracic ultrasonography examination for lung sliding in nine regions was done to derive an adherence score: present (1 point), questionable (2 points), or absent (3 points), with a lowest possible score of 9 (preserved sliding) and a highest possible score of 27 (complete absence of sliding); the chest tube was removed if the score was more than 20. In the standard care group, tube removal was based on daily output volume (per British Thoracic Society Guidelines). The primary outcome was length of hospital stay, and secondary outcomes were pleurodesis failure at 3 months, time to tube removal, all-cause mortality, symptoms and quality-of-life scores, and cost-effectiveness of thoracic ultrasonography-guided care. All outcomes were assessed in the modified intention-to-treat population (patients with missing data excluded), and a non-inferiority analysis of pleurodesis failure was done in the per-protocol population. This trial was registered with ISRCTN, ISRCTN16441661. FINDINGS: Between Dec 31, 2015, and Dec 17, 2019, 778 patients were assessed for eligibility and 313 participants (165 [53%] male) were recruited and randomly assigned to thoracic ultrasonography-guided care (n=159) or standard care (n=154). In the modified intention-to-treat population, the median length of hospital stay was significantly shorter in the intervention group (2 days [IQR 2-4]) than in the standard care group (3 days [2-5]; difference 1 day [95% CI 1-1]; p<0·0001). In the per-protocol analysis, thoracic ultrasonography-guided care was non-inferior to standard care in terms of pleurodesis failure at 3 months, which occurred in 27 (29·7%) of 91 patients in the intervention group versus 34 (31·2%) of 109 patients in the standard care group (risk difference -1·5% [95% CI -10·2% to 7·2%]; non-inferiority margin 15%). Mean time to chest tube removal in the intervention group was 2·4 days (SD 2·5) versus 3·1 days (2·0) in the standard care group (mean difference -0·72 days [95% CI -1·22 to -0·21]; p=0·0057). There were no significant between-group differences in all-cause mortality, symptom scores, or quality-of-life scores, except on the EQ-5D visual analogue scale, which was significantly lower in the standard care group at 3 months. Although costs were similar between the groups, thoracic ultrasonography-guided care was cost-effective compared with standard care. INTERPRETATION: Thoracic ultrasonography-guided care for pleurodesis in patients with MPE results in shorter hospital stay (compared with the British Thoracic Society recommendation for pleurodesis) without reducing the success rate of the procedure at 3 months. The data support consideration of standard use of thoracic ultrasonography in patients undergoing MPE-related pleurodesis. FUNDING: Marie Curie Cancer Care Committee.

Ambulatory management of secondary spontaneous pneumothorax: a randomised controlled trial.

Secondary spontaneous pneumothorax (SSP) is traditionally managed with an intercostal chest tube attached to an underwater seal. We investigated whether use of a one-way flutter valve shortened patients' length of stay (LoS).This open-label randomised controlled trial enrolled patients presenting with SSP and randomised to either a chest tube and underwater seal (standard care: SC) or ambulatory care (AC) with a flutter valve. The type of flutter valve used depended on whether at randomisation the patient already had a chest tube in place: in those without a chest tube a pleural vent (PV) was used; in those with a chest tube in situ, an Atrium Pneumostat (AP) valve was attached. The primary end-point was LoS.Between March 2017 and March 2020, 41 patients underwent randomisation: 20 to SC and 21 to AC (13=PV, 8=AP). There was no difference in LoS in the first 30 days following treatment intervention: AC (median=6 days, IQR 14.5) and SC (median=6 days, IQR 13.3). In patients treated with PV there was a high rate of early treatment failure (6/13; 46%), compared to patients receiving SC (3/20; 15%) (p=0.11) Patients treated with AP had no (0/8 0%) early treatment failures and a median LoS of 1.5 days (IQR 23.8).There was no difference in LoS between ambulatory and standard care. Pleural Vents had high rates of treatment failure and should not be used in SSP. Atrium Pneumostats are a safer alternative, with a trend towards lower LoS.

European Respiratory Society statement on thoracic ultrasound.

Thoracic ultrasound is increasingly considered to be an essential tool for the pulmonologist. It is used in diverse clinical scenarios, including as an adjunct to clinical decision making for diagnosis, a real-time guide to procedures and a predictor or measurement of treatment response. The aim of this European Respiratory Society task force was to produce a statement on thoracic ultrasound for pulmonologists using thoracic ultrasound within the field of respiratory medicine. The multidisciplinary panel performed a review of the literature, addressing major areas of thoracic ultrasound practice and application. The selected major areas include equipment and technique, assessment of the chest wall, parietal pleura, pleural effusion, pneumothorax, interstitial syndrome, lung consolidation, diaphragm assessment, intervention guidance, training and the patient perspective. Despite the growing evidence supporting the use of thoracic ultrasound, the published literature still contains a paucity of data in some important fields. Key research questions for each of the major areas were identified, which serve to facilitate future multicentre collaborations and research to further consolidate an evidence-based use of thoracic ultrasound, for the benefit of the many patients being exposed to clinicians using thoracic ultrasound.