1216. Limited Usefulness of Bronchoalveolar Lavage (BAL) in the Diagnosis of Pulmonary Infiltrates after Allogeneic Hematopoietic Stem Cell Transplantation (allo-HCT)
Session: Poster Abstract Session: Transplant: Epidemiology of Infections in Transplant Patients and Other Patients with Impaired Immunity
Friday, October 9, 2015
Room: Poster Hall
Posters
  • Slide1.png (306.0 kB)
  • Background: Pulmonary infiltrates are common after allo-HCT. BAL is often performed in these patients, although empirical treatment frequently is initiated and may decrease its yield. Sometimes less invasive diagnostic procedures may provide a potential explanation, but it is still not clear whether proceeding with a BAL is appropriate.

    Methods: We prospectively studied pulmonary infiltrates (PI) after allo-HCT in 76 patients with immunodeficiency or hematologic malignancies transplanted at our center between Aug 2013 and April 2015. Thirty-three patients developed 39 episodes of pulmonary infiltrates: 14 before-engraftment (PRE), 11 between engraftment and day 100 (EARLY) and 14 after day 100 (LATE).

    Results: Bronchoscopy with bronchoalveolar lavage (BAL) was performed in 19 PI. BAL was diagnostic in only 8 cases (42%): 2 cases of bacterial pneumonia, 1 case of Pneumocystis (PCP), 1 case of toxoplasmosis and 4 cases of respiratory viruses. Except the case of PCP, all  others were also identified by a less invasive diagnostic sample including blood culture, nasal bacterial culture from ENT exam, NP wash with PCR for respiratory virus as well as blood PCR for toxoplasmosis in one case and HHV6 in another). Of the 20 patients with PI who did not receive a BAL, 15 were considered infectious (fungal or bacterial) and treated empirically with good clinical outcome.  Five of the 20 had sputum sent (all non-diagnostic) and one patient had FNA performed later which revealed aspergillosis. 

    Infectious etiologies were considered the cause of 11, 6 and 11 of the PRE, EARLY and LATE PI, respectively. Non-infectious complications included progressive disease, heart failure, diffuse alveolar hemorrhage. Several remained undiagnosed, two of which responded to corticosteroids. 

    There were 8 deaths in patients who developed pulmonary infiltrates.  Three of the 4 patients with infection died of progressive disease, and the fourth died of HHV6 encephalitis (HHV6 was also found in the BAL).  Four patients with non-infectious PI died of progressive disease.

    Conclusion: Pulmonary infiltrates following allo-HCT are common (43% of patients) with most cases caused by infection that is frequently amenable to diagnosis by blood tests and NP wash. Non-invasive samples often provided sufficient diagnostic information to guide therapy.

    Juan Gea-Banacloche, MD1, Theresa Jerussi, PA-C2, Dier Hu, MS3, Mark Parta, Doctor of Medicine4, Sameer Kadri, MD5, Nirali Shah, MD6, Harry Malech, MD7, Ronald Gress, MD8, Steven Pavletic, MD9 and Kirsten Williams, MD9, (1)Experimental Transplantation and Immunology Branch, NCI, National Institutes of Health, Bethesda, MD, (2)NIH, Bethesda, MD, (3)NCI, Bethesda, MD, (4)LHD/ICMOB/NIAID (Transplant), Bethesda, MD, (5)Critical Care Medicine, National Institutes of Health, Bethesda, MD, (6)NCI, Pediatric Oncology Branch, Bethesda, MD, (7)National Institutes of Health, Bethesda, TX, (8)NCI, Experimental Transplantation and Immunology Branch, Bethesda, MD, (9)National Institutes of Health, Bethesda, MD

    Disclosures:

    J. Gea-Banacloche, None

    T. Jerussi, None

    D. Hu, None

    M. Parta, None

    S. Kadri, None

    N. Shah, None

    H. Malech, None

    R. Gress, None

    S. Pavletic, None

    K. Williams, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.