BRONCHOSCOPIC FINDINGS OF ALVEOLAR HEMORRHAGE: HOW BLOODY IS BLOODY?

Abstract

SESSION TITLE: Monday Electronic Posters 3 SESSION TYPE: Original Inv Poster Discussion PRESENTED ON: 10/21/2019 02:30 PM - 03:15 PM PURPOSE: Alveolar hemorrhage (AH) is a complicated presentation with many disparate etiologies including capillaritis (e.g., microscopic polyangiitis) and non-capillaritis (e.g., infection). Management of steroid-responsive etiologies may allow prompt treatment while awaiting biopsy or serological testing, but may be harmful in other diagnoses. Recognition of AH is difficult due to several methods for diagnosis and no gold standard. In general, serial aliquots of saline are used to observe persistently or progressively bloody return from bronchoalveolar lavage (BAL). However, description is subjective and relies on physician technique and index of suspicion, and the phrase AH may bias clinical judgement. The incidence of bloody return among all BALs is uncertain. We aim to characterize the frequency and description of bloody return from BAL and associated clinical diagnoses and management. METHODS: We searched all 2017 bronchoscopy records from the EMR for description of bloody return from BALs, with one procedure per included patient. RESULTS: Of 1792 total bronchoscopies in 2017, our search returned 686 procedures in 352 patients. Of these, 14 were described as progressively bloody return on BAL, versus 33 persistently bloody, and 83 blood-tinged; 222 noted blood unrelated to BAL. A total of 6 patients were felt to have capillaritis causing AH based on clinical/serological features - 1 with progressively bloody return, 4 persistently bloody, and 1 blood-tinged. Among the 47 patients with persistently or progressively bloody return, 39 had underlying infection, 6 ARDS, 5 pulmonary edema, 10 coagulopathy, 5 thrombocytopenia, 8 malignancy, and 1 BMT; 23 of 47 had >1 contributing diagnosis. Among the 6 patients with capillaritis, 3 had neutrophil predominant BAL and 3 macrophage predominant. Ten patients were treated empirically with corticosteroids after bronchoscopy while awaiting further data, only 5 of which had capillaritis. CONCLUSIONS: Progressively or persistently bloody return from BAL was more common than expected, however few patients had capillaritis. The majority had infection among many other contributing diagnoses. This suggests diagnosis of capillaritis based on bronchoscopic findings alone is not as useful as previously described. More data are needed regarding the utility of cell counts to differentiate etiologies of bloody return. Additionally, description of BAL return is subjective. Prospective studies and reporting standardization are needed. CLINICAL IMPLICATIONS: Numerous etiologies yield persistently or progressively bloody return on BAL with conflicting management strategies. Pulmonary capillaritis should not be diagnosed with BAL data alone. Given subjectivity of AH interpretation, prospective studies and reporting standardization are warranted. Additionally, biomarkers to rapidly distinguish etiologies of AH and direct early management of life-threatening entities are needed. DISCLOSURES: No relevant relationships by Lynn Fussner, source=Web Response No relevant relationships by Kevin Patterson, source=Web Response