Evaluation of persistent pulmonary infiltrates in pediatric oncology patients

Purpose. To evaluate the role of bronchoalveolar lavage (BAL), computed-tomography-guided biopsy (CTB), and open lung biopsy (OLB) in the management of persistent pulmonary infiltrates in pediatric oncology patients. Methods. Retrospective review of clinical records of pediatric oncology patients who underwent BAL, CTB, and OLB over a 7-year period. Data was compared across the three procedures using chi-square analysis. Logistic regression was used to adjust potential confounding variables for diagnostic yield. Results. There were 113 consecutive patients who underwent 140 separate procedures during their hospitalization. Thirty (26%) patients had a previous BMT. BALs were more likely to occur as the first line of investigation (98% vs. 47%, 45%; P < 0.01) and in patients with diffuse infiltrates (64% vs. 61%, 26%; P < 0.01) when compared to CTB and OLB, respectively. OLBs were performed less frequently in neutropenic patients (26% vs. 53%, 54%; P < 0.05), more often led to change in management directly because of procedure (61% vs. 12%, 33%; P < 0.01), and had higher diagnostic yield (61% vs. 24%, 36%; P < 0.01) when compared to CTB and BAL, respectively. Diagnostic yield of OLB was significantly higher regardless of diffuse or focal nature of infiltrate. Major adverse events after a procedure were not significantly different across the three procedures. Logistic regression demonstrated that having an OLB was independently associated with identifying the cause of pulmonary infiltrate. Conclusion. OLB appears to be safe, has the best diagnostic yield, and leads to change in management more often than CTB or BAL in pediatric oncology patients with persistent pulmonary infiltrates.