Risk of bleeding associated with transbronchial biopsy using flexible bronchoscopy in patients with echocardiographic or chest CT evidence of pulmonary hypertension.

Background: Endobronchial ultrasound (EBUS)-guided transbronchial biopsy (TBB) facilitates the diagnosis of various respiratory diseases. The safety of performing EBUS-guided TBB in patients with a finding of pulmonary hypertension (PH) is controversial. Little is known about the relationship between the risk of bleeding associated with EBUS-guided TBB in the presence of PH suspected on echocardiography or chest CT.

Methods: To assess the risk of bleeding associated with EBUS-guided TBB in patients with presumed PH per echocardiography or chest CT, we retrospectively reviewed the medical records of 314 consecutive patients who underwent EBUS-guided TBB using a guide sheath (GS), as well as echocardiography and chest CT. Bleeding complication was defined as over one minute of suctioning; repeated wedging of the bronchoscope; instillation of cold saline, diluted vasoactive substances, or thrombin due to persistent bleeding. Findings of suspected PH were defined as peak tricuspid regurgitation velocity (TRV) > 2.8 m/s on echocardiography or pulmonary artery to aorta ratio (PA:A ratio) > 0.9 on chest CT.

Results: In total, 35 (11.1%) patients developed bleeding, and all cases were managed safely. Furthermore, 17 (5.4%) and 59 (18.8%) patients were suspected to have PH based on echocardiography and chest CT, respectively. Among the patients suspected to have PH on echocardiography, five (5/17 = 29.4%) patients developed bleeding. Among the patients suspected to have PH on chest CT, 11 (11/59 = 18.6%) patients developed bleeding. Univariate analysis revealed that long diameter (≥ 30 mm) of the lesion, lesion location (the biopsy site was inner than the segmental bronchus), bronchoscopic diagnosis of malignancy, and additional biopsy were potential predictive factors for bleeding. The finding of suspected PH on echocardiography correlated significantly with bleeding (p = 0.03). On multivariate analysis, long diameter (≥ 30 mm) of the lesion (p = .021) and findings of suspected PH on echocardiography (p = .049) were significantly associated with bleeding.

Conclusion: All cases of bleeding in the present study were managed safely. The risk of bleeding is moderately elevated when PH is suspected by echocardiography in patients undergoing EBUS-guided TBB using a GS.