Institutional members access full text with Ovid®

Share this article on:

Prospective Aspergillus Galactomannan Antigen Testing in Pediatric Hematopoietic Stem Cell Transplant Recipients

Steinbach, William J. MD*; Addison, Rachel M.; McLaughlin, Lisa; Gerrald, Quincy; Martin, Paul L. MD, PhD*; Driscoll, Timothy MD*; Bentsen, Christopher; Perfect, John R. MD; Alexander, Barbara D. MD

The Pediatric Infectious Disease Journal: July 2007 - Volume 26 - Issue 7 - p 558-564
doi: 10.1097/INF.0b013e3180616cbb
Original Studies

Background: The galactomannan (GM) assay is an approved noninvasive test for detection of invasive aspergillosis (IA) that has been validated in adult patients with hematologic malignancies who are undergoing bone marrow transplantation. There have been few studies with this assay in pediatric patients, but early reports suggest that there may be differences in the performance such that false-positive GM tests in pediatric patients are more common than in adult patients.

Methods: We performed a prospective study in pediatric hematopoietic stem cell transplant recipients with twice-weekly sampling for GM detection during the highest risk periods of neutropenia and graft-versus-host disease. We analyzed 826 serum samples from 64 patients, including 15 serum samples from one patient diagnosed with probable IA according to defined criteria.

Results: Twenty of 811 samples tested positive on repeat testing (specificity, 97.5%; 95% CI: 96.2–98.4%) including samples from 8 of 63 patients without clinical evidence of IA according to study criteria (specificity, 87.3%; 95% CI: 76.9–93.4%). Eleven patients received piperacillin/tazobactam therapy, and 4 of the 11 patients had a positive assay result coinciding with the dates of piperacillin/tazobactam administration. When samples from these patients were excluded, specificity increased to 98.4% (95% CI: 97.2–99.1%) by sample and to 91.5% (95% CI: 81.6–96.3%) by patient.

Conclusions: The GM assay holds promise for early, noninvasive diagnosis of IA in high-risk children and false-positive results were not common or unexplainable. This study supports further validation of this assay in a large-scale, pediatric-dedicated format.

From the Departments of *Pediatrics and †Medicine, Duke University Medical Center, Durham, NC; and ‡Bio-Rad Laboratories, Redmond, WA.

Accepted for publication March 22, 2007.

Address for correspondence: William J. Steinbach, MD, Division of Pediatric Infectious Diseases, Box 3499, Duke University Medical Center, Durham, NC 27710. E-mail:

© 2007 Lippincott Williams & Wilkins, Inc.