Earlier radiographic detection of malignant airway obstruction by interventional pulmonologists highlights importance of dedicated radiographic review of the airways

TOPIC: Procedures TYPE: Original Investigations PURPOSE: Malignant central airway obstruction (CAO) is defined as neoplastic obstruction involving the trachea, mainstem bronchi, or bronchus intermedius, and portends worse prognosis1. Malignant airway obstruction of segmental airways (SAO) can produce respiratory symptoms requiring palliation. Radiographic identification of obstruction and timely referral for intervention are crucial in the management of intervenable disease. We audited the radiographic identification of SAO, CAO, and time to interventional pulmonology (IP) referral to inform and optimize our center’s practice. METHODS: Medical records of 100 consecutive patients between 10/1/2017-10/1/2019 who underwent therapeutic bronchoscopy for malignant CAO or SAO at Memorial Sloan Kettering were manually identified. Chest imaging scans (computerized tomography (CT) & positron emission tomography (PET)) were reviewed by one of 5 IP attendings for determination of the first evidence of radiographic obstruction. Attendings were blinded to the formal radiologist interpretation. Time from first radiographic evidence of obstruction, to IP consultation, and to bronchoscopy, were measured. IP attending interpretation was compared with the formal radiologist interpretation. RESULTS: Eighty-one medical records were eligible for final analysis. Patients with outside radiographic reports or non-malignant obstruction were excluded. IP attendings detected first evidence of airway obstruction a mean of 220 days prior to consultation for airway intervention (median = 67, inter-quartile range 16-244). 48% of imaging scans with first evidence of obstruction were described formally in the report by the radiologist; 52% were not. Of the imaging scans with first evidence of obstruction that were not formally described, 47% were consistent with CAO, and 53% were SAO. Of the scans taken immediately prior to procedure, 85% had formally reported obstruction by the radiologist, 15% did not. Of those that were not formally reported, 36% were consistent with CAO, and 64% were consistent with SAO. CONCLUSIONS: Our study was aimed to retrospectively explore the correlation between radiographic detection of malignant airway obstruction by IP attendings and the formal radiologist read. In our cohort, malignant CAO & SAO was detectable on thoracic imaging earlier than both initial radiologist commentary regarding CAO/SAO, and the time of referral for airway intervention. Intervention has shown to improve quality of life2and facilitates de-escalation of the patient's level of care3. More dedicated review and increased multi-disciplinary awareness of the radiographic progression of malignant CAO & SAO are needed to optimize care. CLINICAL IMPLICATIONS: Earlier detection of obstruction will allow for timely referral to IP. The next phase of our study will involve the development of an educational module to increase awareness of CAO and SAO amongst radiologists. After the intervention is implemented, we will retrospectively analyze any change in radiologic detection and referral patterns.1Daneshvar C. et al. Prevalence and outcome of CAO in patients with lung cancer. BMJ Open Respiratory Research.2019.2Amjadi K. et al. Impact of interventional bronchoscopy on quality of life in MAO. Respiration. 2008.3Henri G. et al. Therapeutic Rigid Bronchoscopy Allows Level of Care Changes in Patients With Acute Respiratory Failure From CAO, Chest. 1997. DISCLOSURES: No relevant relationships by Jason Beattie, source=Web Response No relevant relationships by Mohit Chawla, source=Web Response No relevant relationships by Paige Fuentes, source=Web Response No relevant relationships by Saamia Hossain, source=Web Response No relevant relationships by Bryan Husta, source=Web Response No relevant relationships by Or Kalchiem-Dekel, source=Web Response No relevant relationships by Robert Lee, source=Web Response No relevant relationships by I-Hsin Lin, source=Web Response No relevant relationships by Svetlana Visotski, source=Web Response