LAUSR

BACKGROUND Respiratory infections cause a significant amount of morbidity and mortality in pediatric and young adult patients with malignancy. Bronchoscopy with bronchoalveolar lavage (BAL) is frequently utilized in the diagnostic process, but which patients would most benefit is poorly understood. METHODS A retrospective study from 2013-2022 examined patients with active malignancy who underwent bronchoscopy with BAL. Positive and negative clinical impact were assessed by three independent reviewers according to predetermined criteria. Mixed-effects logistic regression was performed to identify factors associated with positive and negative clinical impact. RESULTS 145 bronchoscopies met inclusion criteria with a median patient age of 12 years (IQR 5-17). 30.3% of bronchoscopies had a positive clinical impact with 17.2% leading to a new diagnosis, most commonly Pneumocystis jirovecii pneumonia (PJP) (7.6%). Comparatively, 18.6% had a negative clinical impact, most commonly from a procedural complication (13.1%). Trimethoprim-sulfamethoxazole initiation for treatment of suspected PJP prior to BAL (aOR 11.20, 95% CI 1.32-95.29) was associated with positive clinical impact. Requirement for ICU-level care (aOR 18.85, 95% CI 3.60-98.69) or oxygen supplementation by nasal canula prior to BAL (aOR 18.41, 95% CI 4.78-70.95) were associated with negative clinical impact while prior invasive ventilation (aOR 0.09, 95% CI 0.01-0.58) was associated with the absence of negative clinical impact. CONCLUSIONS Patients with potential respiratory infections with high clinical suspicion for PJP who had been started on treatment trimethoprim-sulfamethoxazole prior to BAL benefit most from bronchoscopy with BAL. Patients previously intubated deemed clinically safe for bronchoscopy tolerated the procedure better than patients in the ICU on nasal cannula.