The diagnosis of pulmonary invasive fungal infection (IFI) in the pediatric oncology patient is challenging. Consensus criteria developed in 2008 state that bronchioalveolar lavage (BAL) results cannot confirm this diagnosis. A video-assisted thoracoscopic biopsy (VATS-biopsy) of lungs has been increasingly used to assist in evaluating these children for IFI. Our goal was to evaluate the impact of BAL and VATS-biopsy results on the management of IFI among pediatric oncology patients.
A retrospective review of all oncology patients evaluated for IFI with VATS-biopsy and/or BAL over 9 years was carried out at a single free-standing children's hospital. The primary outcome was management changes in the use of antifungal therapy on the basis of diagnostic procedure, fungal culture results, lung imaging, and serological markers.
A total of 102 patients underwent 122 diagnostic evaluations for IFI. Ninety-one workups included only BAL, 17 evaluations involved only VATS-biopsy, and 14 cases involved both BAL and VATS-biopsy. The diagnostic yield of VATS-biopsy (38.7%) was superior to that of BAL (27.6%). There was poor concordance between VATS-biopsy and BAL results in the 14 cases where both were performed. Upon workup completion, IFI was proven in 12 children, probable in 29, and possible in 52. The odds of continuing antifungals increased 3-fold for patients with probable IFI and 12.7 times for those with the proven disease.
On the basis of the inferior diagnostic yield of BAL, we believe that VATS-biopsy may be a more useful diagnostic adjuvant in the diagnosis of IFI in the immunocompromised pediatric oncologic patient population.