Severe Thrombocytopenia Does Not Increase Bleeding Complications in Patients Undergoing Bronchoscopy

Introduction

Prophylactic platelet transfusions are often performed prior to bronchoscopy or broncho-alveolar lavage (BAL) to prevent bleeding in thrombocytopenic patients. There is a paucity of data to validate this approach, with a platelet transfusion threshold of <50,000/mm³ largely based on expert opinion. We conducted a retrospective study on the incidence of bleeding complications in thrombocytopenic patients undergoing bronchoscopy.

Methods

We identified 150 consecutive patients with platelet counts <100,000/mm³ who underwent bronchoscopy and/or BAL from January 2009 to May 2014 at our institution. Bronchoscopies performed in patients with frank hemoptysis and trans-bronchial lung biopsy procedures were excluded. Patient characteristics, underlying diagnosis, platelet count prior to bronchoscopy, administration of platelet transfusions and bronchoscopy details were recorded. Factors affecting bleeding risk including presence of renal dysfunction (defined as BUN >30 and/or Cr>2.0) and coagulation studies (PT, PTT, INR) were identified. The British Thoracic Society guidelines¹ were used to categorize bleeding as a result of bronchoscopy. Data were analyzed using descriptive statistics.

Results

The median age was 59 years (range 27-90), with two-thirds of patients (63%) being male. One hundred and seventeen (78%) patients had underlying malignancy and 55 (37%) had thrombocytopenia related to malignancy. Fellows and residents under the supervision of a bronchoscopy certified attending performed all but 4 of the bronchoscopies. Infection (40%) was the primary indication for bronchoscopy with BAL performed in 127 (85%) patients. Fifty-eight of 89 (65%) patients with baseline platelet counts <50,000/mm³ received prophylactic transfusions compared to 8% of those with platelet counts >50,000/mm³. The platelet count did not rise to >50,000//mm³ in many transfused patients.

Seventy patients (47%) had counts <50,000/mm³ and eighty patients (53%) had counts >50,000/mm³ at the time of bronchoscopy. 49% were receiving immunosuppressive medications, 45% had renal dysfunction and 8% had INR >1.5. Bloody lavage that resolved spontaneously without continuous suctioning (Grade 0) was observed in 9 (6%) patients. Bleeding that required continuous suctioning but then resolved spontaneously (Grade 1) was noted in 1 patient with a platelet count of 61,000/mm³. Of 10 total bleeding events, 7 occurred in patients who were intubated. Two additional patients with platelet counts of 30,000/mm³ and 53,000/mm³ had diffuse alveolar hemorrhage, which was present before bronchoscopy. “Old” blood and blood clots were observed in 6 patients.

Discussion

The low incidence of bleeding complications from bronchoscopy +/- BAL even in patients with platelet counts <30,000/mm³ (3 episodes in 31 patients, all grade 0) demonstrates that bronchoscopy can be safely done in severely thrombocytopenic patients. Adopting a lower threshold for prophylactic transfusions could save a considerable number of platelet units and translate into significant cost savings and decreased risk of transfusion-related complications.

1.Du Rand IA, Blaikley J, Booton R, et al. British Thoracic Society guideline for diagnostic flexible bronchoscopy in adults: accredited by NICE. Thorax. 2013:68 Suppl 1:i1-i44