Bleeding risk of transbronchial cryobiopsy compared to transbronchial forceps biopsy in interstitial lung disease - A prospective, randomized, multicentre cross-over trial

Abstract

Background: Bronchoscopic cryobiopsy is a new method of bronchoscopic tissue sampling in interstitial lung disease. In case of transbronchial biopsies, the resultant tissue samples are of high quality, and the lung parenchyma seen in the samples is adequate for a histological diagnosis in most cases. Bleeding after transbronchial biopsy is the most important procedure- associated complication and may be life threatening. This study addresses the risk of bleeding of transbronchial cryobiopsy. Methods: In this prospective, randomized, controlled multicentre study 359 patients with interstitial lung disease requiring diagnostic bronchoscopic tissue sampling were included. Both conventional transbronchial forceps biopsy and transbronchial cryobiopsy were undertaken in each patient. The sequence of the procedures was randomized. Bleeding severity was evaluated semi-quantitatively as "no bleeding", "mild" (suction alone), "moderate" (additional intervention) or "severe" (prolonged monitoring necessary or fatal outcome), for each intervention. Results: In 359 patients atotal of 1160 cryobiopsies and 1302 forceps biopsies were performed. Bleeding was observed after forceps biopsy in 173 patients (48.2%) and after cryobiopsy in 261 patients (72.7%). Bleeding was significantly greater in the cryobiopsy group (cryobiopsy/forceps biopsy: no bleeding 27.3%/51.8%; mild 56.5%/44.0%; moderate 15.0%/4.2%; severe 1.2%/0%; p < 0.001). The rate of clinically relevant bleeding (moderate or severe) was higher after the cryobiopsy procedures compared to the forceps biopsies (16.2% vs. 4.2%, p < 0.05). No fatal bleeding complications occurred. Conclusions: Compared to transbronchial forceps biopsy, transbronchial cryobiopsy was associated with an increased risk of bleeding which is of clinical relevance. Therefore training and additional precautions for bleeding control should be considered. Trial registration: The study was registered with clinicaltrials.gov (NCT01894113).