CLINICAL MYTHS AND EVIDENCE-BASED MEDICINE Bronchoscopy in Uremic Patients Irtaza Khan, MD,* Christina Bellinger, MD,* Carla Lamb, MD,† Robert Chin, MD,* and John Conforti, DO* Abstract: Uremia is associated with an increased risk of bleeding by virtue of alteration in platelet adhesion. Pulmonologists are frequently called upon to perform flexible bronchoscopy in patients with chronic renal insufficiency. The high blood urea nitrogen levels may predispose these patients to a high risk of bleeding complications with bronchoscopic procedures. We carried out a literature review to evaluate the myth that bronchoscopy is unsafe in uremic patients. Key Words: flexible bronchoscopy, uremia, hemorrhage, safety (Clin Pulm Med 2010;17: 146 –148) MYTH Uremia increases the risk of bleeding in patients undergoing bronchoscopy. Flexible bronchoscopy is relatively a safe procedure but does carry a risk of potential complications including hemorrhage. This risk can potentially increase by coexisting medical conditions. Bleeding is a potentially serious complication in patients with renal failure and results from abnormalities in interactions between platelets themselves and with the vessel wall. 1 This risk increases with the type of procedure performed and the severity of renal failure. However, not all clinicians perform bleeding time before performing invasive procedures in this population nor has this test been shown to be helpful in surgical procedures. Though commonly used in patients with renal disease, the risks of bronchoscopy induced hemorrhage in these patients has rarely been investigated. In 1987, the American Thoracic Society published guide- lines 2 listing conditions that can result in an increase in complica- tions associated with bronchoscopy. In this article we review the available literature regarding the risk of bronchoscopy in uremic patients. DATA Wahidi et al 3 in 2005 published a survey of the beliefs and clinical practices among pulmonologists regarding the safety of bronchoscopy. Of the 158 pulmonologists surveyed at the ACCP meeting in Philadelphia in 2001, 55% of responders did not consider elevated creatinine a potential contraindication for transbronchial lung biopsy with 22.6% performing the procedure in patients with creatinine levels of more than 3.0. Only 37.6% pulmonologists gave 1-deamino-8-D-arginine-vasopressin (DDAVP) before transbron- chial lung biopsy in uremic patients with a majority (55.6%) choos- ing not to intervene. However, the rate of hemorrhage in these patients was not surveyed. In 1977, Cunningham et al 4 reviewed their experience of transbronchial forceps biopsy. Eleven of the 31 immunosuppressed patients undergoing trans- bronchial forceps biopsy had elevated blood urea nitrogen (BUN) ranging from 31 to 147 mg/100 mL. Bleeding in uremic patients was 3 times more than in the nonuremic population (45% vs. 15%). Although the authors defined hemorrhage as mild (20 mL) or explosive (100 mL), the quantity of bleeding and its conse- quences were not clear in this uremic population. Two of the above patients had initial platelet counts of 2000 and 38,000, respectively, and were treated with platelet infusions before the procedure. The platelet count at the time of biopsy ranged from 90,000 to 797,000. The authors made no conclusions regarding the safety of bronchoscopy. (Category C; Table 1). 5 Cordasco et al 6 in 1991 published a retrospective review on bronchoscopies performed at the Cleveland Clinic from January 1981 to December 1989. BUN of less than 25 mg/dL and creatinine of less than 1.5 mg/dL were considered normal. Bleeding was categorized as minimal (50 mL of blood mixed with lavage fluid), moderate (50 mL to 100 mL of bloody lavage fluid), and profuse (100 mL). Clinically significant bleeding was seen in 58 cases. Coagulation abnormalities were seen in 3 patients. Eleven episodes of bleeding occurred in immunosuppressed patients. Of the 58 patients who bled, 2 had renal insufficiency (BUN 29 and 40 mg/dL and serum creatinine 2.5 and 2.3 mg/dL, respectively) and 2 patients had end stage renal disease (ESRD). Minimal to moderate bleeding was seen in 2 of the 4 patients with renal insufficiency (one with chronic renal insufficiency and the other with ESRD). The other 2 (one patient with chronic renal insufficiency and one with ESRD) had profuse bleeding. Patients with renal insufficiency who bled had normal coagulation parameters and platelet counts, however, a bleeding time was not performed. The authors concluded that the degree of bleeding was related to the type of bronchoscopy performed. Transbronchial biopsies were associated with an increased incidence of bleeding (34 total with 25 minimal to moderate and 9 episodes of profuse bleeding). Bronchial brushings and endobronchial biopsies were associated with 17 episodes of bleeding respectively (14 episodes associated with mild to moderate and 3 with profuse bleeding in each case). There overall risk of bleeding was low with no deaths reported. (Category C; Table 1). 5 Diette et al in 1999 7 performed a prospective cohort study of all adult patients undergoing flexible bronchoscopy over a 1-year period. A total of 720 procedures were performed including 38 in lung transplant patients. Although the degree of renal dysfunction was not mentioned, 6 patients were reported to have renal failure. None of the broncho- scopies were done in the immediate postoperative period and were performed for surveillance for rejection, airway inspection, solitary nodule, pleural effusion, and evaluation of focal or diffuse infiltrates. None of the patients with renal failure had hemorrhagic complications in this study though the lung transplant patients were 5 times more likely to have the procedure terminated prematurely for bleeding and had significantly more blood loss (928 vs. 13 mL, P  0.0001) at the end of the procedure. However, severity of bleeding was not related to the degree of renal dysfunction, coagu- lation parameters, and platelet counts. Increase in bleeding tendencies was explained by possible increase in inflammation and altered local hemodynamic factors seen in lung transplant patients. (Category C; Table 1). 5 From the *Pulmonary and Critical Care Medicine, Wake Forest University, Baptist Medical Center, Winston Salem, NC; and †Tufts University School of Medicine, MA. Address correspondence to: John Conforti, DO, Department of Pulmonary and Critical Care Medicine, Wake Forest University, Baptist Medical Center, Medical Center Blvd, Winston Salen, NC 27157. E-mail: jconfort@wfubmc.edu. Copyright © 2010 by Lippincott Williams & Wilkins ISSN: 1068-0640/10/1703-0146 DOI: 10.1097/CPM.0b013e3181da8a0e Clinical Pulmonary Medicine • Volume 17, Number 3, May 2010 146 | www.clinpulm.com