Prospective monitoring for invasive aspergillosis using galactomannan and polymerase chain reaction in high risk pediatric patients

The diagnosis of invasive aspergillus remains a challenge in the care of high-risk patients. Outcomes are improved when invasive aspergillus is diagnosed early, prompting the initiation of appropriate antifungal therapy. We evaluated the utility of prospective monitoring for invasive aspergillosis (IA) using biomarkers such as serum galactomannan (GM) and/or blood polymerase chain reaction (PCR) in high-risk pediatric patients. Patients with high-risk leukemia (HRL) or allogenic hematopoietic cell transplant (HCT) recipients were prospectively monitored twice weekly for IA using GM and PCR for Aspergillus species. Sixty-eight patients had collected >or=2 specimens. The 1086 specimens were collected; 627 from HRL (58%) and 459 (42%) from HCT recipients. Median specimens/patient was 11.0 (2 to 58), and median follow-up/patient was 98.5 days (14 to 437). Fifty-six percent of samples were obtained from patients receiving mold-active agents; 32% HRL and 89% HCT. There were no proven, 3 probable, and 20 possible episodes of IA. Thirteen specimens (1.2%) from 4 patients (5%) were GM+. None were positive by PCR. The prospective use of GM and PCR in this high-risk pediatric population did not identify cases of proven IA. A high false positive rate was not detected. It is speculated that changes in clinical practice, such as early use of empiric and/or prophylactic mold-active agent and frequent imaging studies have impacted the epidemiology of IA. In a population with low incidence of IA, the use of these assays as a screening device on blood may not further enhance current outcomes.