Artificial intelligence (AI) is revolutionizing interventional pulmonology (IP) by enhancing diagnostics, procedural precision, and patient outcomes. AI-powered tools improve lung nodule detection, radiomics-based risk stratification, and bronchoscopic navigation. Machine learning (ML) algorithms aid in lung cancer screening by analyzing imaging data, reducing false positives, and improving early diagnosis. AI-driven robotic-assisted bronchoscopy enhances navigation and biopsy accuracy, particularly for peripheral lung lesions. Endobronchial ultrasound (EBUS) and cytopathology benefit from AI’s ability to assess lymph node malignancy and optimize rapid on-site evaluation (ROSE). AI applications extend to phenotyping chronic obstructive pulmonary disease (COPD) and identifying candidates for bronchoscopic lung volume reduction (BLVR). Deep learning (DL) models analyze computed tomography (CT) imaging and spirometry data to optimize patient selection. AI-driven algorithms are also advancing pleural effusion detection, differentiation, and classification, supporting clinical decision-making. Education and research in IP are also transforming with AI-driven simulation, virtual reality, and automated assessment tools that enhance procedural training and competency evaluation. The integration of AI into clinical work and procedural training accelerates advancements while presenting challenges in ethical AI implementation, data security, and bias mitigation. As AI continues to evolve, its role in IP will expand, improving procedural efficiency, personalizing treatment plans, and optimizing patient selection for interventions. Future developments will focus on refining AI-driven predictive analytics, enhancing robotic-assisted procedures, and integrating AI seamlessly into clinical workflows. The responsible implementation of AI in IP holds the potential to transform patient care, reduce complications, and advance precision medicine.
AIM (Anticipate/Adjust, Image, Manage) to Prevent Bleeding During Robotically‐Assisted Bronchoscopic Ablation
ABSTRACTTherapeutic treatment of lung nodules by ablation is a new field. Even though not considered standard of care, lung nodule ablation can be appropriate for select cases. Even though ablation is a safe and well‐tolerated procedure, bleeding is a potential complication. We introduce our stepwise approach through two cases on how we AIM (Anticipate/Adjust, Image, Manage) to prevent uncontrolled bleeding during Robotically Assisted Bronchoscopic Ablation: 1. Anticipate structures with high risk for bleeding and adjust accordingly, accounting for the needle ablation field; 2. Image: utilise cone beam CT before the first and after each ablation session to screen for bleeding; 3. Manage bleeding by pre‐emptively utilising Fogarty balloon occlusion.
Successful Treatment of a Post-lobectomy Bronchopleural Fistula With an Endobronchial Valve
Abstract A 77-year-old female with history of a right lower lobe adenocarcinoma (pT1a, pN0) underwent right lower lobectomy in August 2023, which subsequently was complicated by a bronchopleural fistula at the level of the right lower lobe bronchial stump. This led to a chronic hydropneumothorax, including a fungal empyema requiring placement of a 14-French thoracostomy tube and prolonged antimicrobial treatment over the next year. In June 2024 she underwent bronchoscopy, with application of argon plasma coagulation (APC) to cauterize the immediate area abutting the bronchopleural fistula and Gore VBX stent placement from the bronchus intermedius to the right middle lobe bronchus, bypassing the right lower lobe bronchial stump. There was no air leak immediately after that procedure and she continued with a 14-French thoracostomy tube in place for empyema drainage. Unfortunately, within the next 10 days the stent migrated requiring replacement. About a month later in July, 2024 the second stent was removed and revealed a persistent defect at the level of the bronchial stump with no apparent regression in size relative to the images captured prior to the initial bronchial stent placement. Under the expanded access pathway, a size 5 Spiration endobronchial valve was placed within the fistula in the right lower lobe bronchial stump (Figure 1). At the conclusion of the bronchoscopy procedure, a recruitment maneuver was performed while the suction on the thoracostomy tube was briefly increased to -40 cm H2O suction and no air leak was evident. The next CT chest 2 weeks later showed substantial improvement of the right hydropneumothorax. The air leak never returned and the thoracostomy tube was removed once the fluid drainage became negligible. The patient's dyspnea gradually improved and she was able to wean off supplemental oxygen. A follow-up CT chest 3 months later demonstrated a very small residual right pleural effusion with no gas component and stable positioning of the endobronchial valve. The multidisciplinary team involved in the patient's care intend to maintain the endobronchial valve in position indefinitely.
C61. FINDING HIDDEN GEMS AT SUNSET CLIFFS: NOT JUST PE AND PH! PULMONARY VEINS, AVMS, AND OTHER PULMONARY VASCULAR CASES | April 30, 2024 | Volume: No volume available | Issue: No issue available | Pages: A6087-A6087
AbstractHumidified rapid‐insufflation ventilatory exchange (HRIVE) is an option for maintenance of oxygenation. This technique allows for oxygenation while the patient is apnoeic due to continuous positive airway pressure and gas exchange through flow‐dependent dead space flushing. There is no study about the usage of HRIVE during rigid bronchoscopy. This retrospective study looked at rigid bronchoscopy cases utilizing HRIVE. Data points assessing adequacy of oxygenation and ventilation were recorded at time points: oxygen saturation (SpO2), partial pressure of oxygen (PaO2) and partial pressure of carbon dioxide (PaCO2). Our nine cases had an average baseline SpO2 of 99.26%, 95.56% at 10 min into HRIVE and 95.27% at the end of HRIVE. The average baseline PaO2 was 309.01 mmHg, 124.99 mmHg at 10 min into HRIVE and 128.17 mmHg at the end of HRIVE. The average baseline PaCO2 was 43.26 mmHg, 68.76 mmHg at 10 min into HRIVE and 75.52 mmHg at the end of HRIVE. The average pre‐HRIVE end‐tidal CO2 (ETCO2) was 38.56 mmHg and the average post‐HRIVE ETCO2 was 61.22 mmHg. The average baseline pH was 7.36, 7.22 at 10 min into HRIVE and 7.19 at the end of HRIVE. In this small cohort study, HRIVE was able to maintain adequate oxygenation via the rigid bronchoscope in a select group of patients. Hypercapnia and respiratory acidosis did result after 10 min, which may predispose certain patient populations to complications. HRIVE potentially offers an additional option of oxygenation via the rigid bronchoscope.
Risk of complications rise with coronary artery disease and diabetes mellitus after endobronchial valve placement in severe heterogeneous emphysema
AbstractBackgroundPatients with chronic obstructive pulmonary disease (COPD) commonly have comorbidities which may impact both symptoms and prognosis. Endoscopic lung volume reduction (ELVR) with endobronchial valves (EBV) is an innovative, effective and safe treatment for patients with severe emphysema who remain symptomatic despite optimal medical therapy.ObjectivesTo evaluate medical comorbidities associated with increased risk of complications after EBV procedure.MethodThis was a retrospective cohort review of patients with severe, heterogeneous emphysema who underwent Zephyr® EBV installation for ELVR. Demographics, baseline comorbidities, dyspnea scores, data regarding procedure, pulmonary function test, minimal clinically important difference and post‐EBV complications were recorded.ResultsOf a total 82 participants, 24 were identified as having a post procedure complication. There was a significant difference in the incidence of type 2 diabetes mellitus (T2DM) at baseline (25%) in complication group compared with non‐complication group (5.1%) (p value = 0.01). There was also a significant difference between baseline coronary artery disease (CAD) in those with (58.3%) or without (29.3%) complication (p = 0.01). Multivariate analysis of other covariables, identified associations of both CAD and T2DM with a heightened risk of complication (adjusted OR 4.19 CI: 1.23–14.2, p value = 0.02 and adjusted OR of 6.1 CI: 1.3–26.1, p value = 0.02 respectively).ConclusionsOur study found that patients with severe, heterogeneous emphysema who undergo ELVR utilizing Zephyr® EBV and suffer complications during post‐procedure period are more likely to have CAD or T2DM as baseline comorbidity. These data suggest that baseline CAD and T2DM might be predictive of increased risk of complications following ELVR with EBV.
Lung Transplantation in a Patient with COPA Syndrome
COPA syndrome is a newly discovered, rare genetic autoimmune disorder, which can affect the lungs, joints, and kidneys. It is difficult to recognize, and the survival benefit of lung transplantation for these patients is not yet known. We present a case of a 24-year-old woman who received bilateral lung transplant for COPA syndrome. At 15 months posttransplant, her pulmonary function is stable with no episodes of acute cellular- or antibody-mediated rejection and no evidence of disease recurrence.
Purpose of review Anaphylaxis is a systemic allergic reaction and without the appropriate treatment can progress fast and may be fatal. Information about allergy and immunology topics including anaphylaxis can be easily found online and are widely searched by patients as well as medical personnel. Social media and search engines not only bring a number of benefits to users but also carry risks for breaches of confidentiality and privacy, unprofessional behaviour, malpractice and litigation. Social media use by medical healthcare workers has liability risks and can result in malpractice. Recent findings We used PubMed with the terms anaphylaxis; social network; mobile apps that revealed a very small number of publications addressing the use of these platforms and anaphylaxis. Google Trends and KWFinder were used to analyse the relative search interest and absolute search volumes on ‘anaphylaxis’. Summary Multiple sources exist on the Internet for health-related searches for both healthcare providers and patients. We analysed different websites, mobile apps, Facebook, YouTube and Twitter. To avoid any posting of patient identifiers, it is imperative to be aware of the 18 HIPAA identifiers. We list websites with accurate information about anaphylaxis for the use of both patients and healthcare professionals for purposes of patient education. Further research on health information-seeking behaviour is clearly needed.