Bronchoscopic Blood Patch Delivery for Treatment of Persistent Bronchopleural Fistula

Abstract

INTRODUCTION: Bronchopleural Fistula (BPF) may result from a primary pathological process or iatrogenic causes such as accidental alveolar overdistention during mechanical ventilation. One study estimated BPF formation associated with mechanical ventilation at approximately 23%, 67% of which was associated with mortality. CASE PRESENTATION: A 56 y/o white female with a history of Crohn's disease and fibromyalgia was transferred to our institution from a long term acute care facility where she was treated for bilateral pneumonia. Her pneumonia also resulted in a tracheostomy and she required continuous mechanical ventilation. During her hospitalization at our institution, she had a left chest tube inserted for an iatrogenic pneumothorax related to mechanical ventilation. A communication persisted between her lung parenchyma and pleural space preventing chest tube removal. After five days of chest tube drainage, a large (5 chamber) air leak persisted. Bronchoscopy was performed to isolate the bronchopleural fistula (BPF). A 5 french Fogarty balloon was passed through the working channel of a bronchoscope into the left lower lobe and inflated to occlude all of the left basilar segments. This resulted in cessation of her air leak in the water chamber of the plural drain. The location of the BPF was further isolated to the left antero-medial basal segment, which had closed her air leak when occluded with the Fogarty balloon. Several alternating layers of an absorbable hemostat (knitted fabric prepared by controlled oxidation of cellulose-Surgicel; Ethicon; Piscataway, NJ) was placed within the antero-medial basal segment using bronchoscopy forceps. Through a cut Fogarty balloon, 3 ml of the patient's blood was delivered onto the absorbable hemostat to create an occluding blood patch of the left antero-medial basal segment. At the conclusion of the procedure, there was no air leak in the left chest tube during positive pressure ventilation. While on mechanical ventilation, the chest tube was removed 2 days later without radiographic recurrence of her pneumothorax. DISCUSSION: Mortality of BPF increased if the fistula occurred late in disease course and larger air leaks were associated with a higher mortality when compared to smaller leaks. Consequences of persistent air leaks included incomplete lung expansion leading to atelectasis and V/Q mismatch, hypoxia, and infection of pleural space. Once the underlying lung injury improves and the need for positive pressure ventilation decreases, the BPF may resolve. Unfortunately, our patient required continuous positive pressure mechanical ventilation which prevented healing of the underlying injury and BPF closure. Additional approaches to BPF closure include surgical management with direct closure of the fistula by using intercostal muscle support, omental or thoracic flaps. Other bronchoscopic techniques include unidirectional valve placement and other occlusive techniques such as a Watanabe Spigot and fibrin glue. Although there is a lack of large randomized trials supporting bronchoscopic methods for BPF management, successful fistula closure has been documented as in our case. To our knowledge, this is the first case report of an autologous blood patch with oxidized regenerated cellulose used for BPF closure. The use of these agents has been described for pleural closure of BPF but not within the airway. We believe this technique provides effective and minimally invasive BPF closure. CONCLUSION(S): Bronchoscopic options may provide a minimally invasive management strategy of BPF which otherwise could result in morbidity and mortality.