A prospective observational study to evaluate the clinical utility of staged Broncho alveolar fluid (BAL) analysis in adolescent and adult leukemia and lymphoma patients with febrile neutropenia and lung infiltrates.

Febrile neutropenia (FN) is an important complication while on chemotherapy in patients with hematological malignancies. Lung infiltrates (LI) are seen in 30% of patients with FN. The prevalence of LI depends on the duration of neutropenia, underlying malignancy and the imaging modality used (CT chest being more sensitive as compared to Chest X rays). Patients of FN with LI have a higher mortality. A wide spectrum of organisms can cause LI in this setting and includes bacterial, fungal, viral or poly-microbial infections. Treatment is mostly empirical and usually consists of a combination of antibiotics that cover Enterobacteriaceae and Pseudomonas (Such as Cefoperazone-sulbactam or carbapenems) with an aminoglycoside and an agent that act against staphylococcal aureus (Teicoplanin or vancomycin). The addition of antifungal is considered if fever persists beyond 96 hours as per the IDSA guidelines. The antibiotics are continued for at least 10-14 days and in the case of an antifungal for throughout the duration of myelosuppression and resolution of radiological findings. The radiological findings are non-specific. Other non-invasive tests such as blood cultures and serum galactomannan help in guiding therapy in <30% patients. An accurate etiological diagnosis is essential and could help in narrowing down the use of antimicrobials, thereby improving the efficacy, preventing toxicity and emergence of drug resistance. Bronchoscopy is an important diagnostic tool that has a diagnostic yield reported in the range of 25-50%. These studies have been mainly retrospective in nature and used a varied microbiologic algorithm. The studied population has also been very heterogeneous precluding any firm conclusion on its clinical utility. This also has to be looked at in the context that most of the FN patients have a higher risk of complications such as bleeding (due to thrombocytopenia and coagulopathy) due to the procedure. These patients are also sick and could be dependent on high flow oxygen or hemodynamically unstable thereby making them unsuitable for this procedure. We routinely perform BAL analysis on patients of FN with LI and use a microbiologic algorithm tapered to the clinical presentation and radiological findings and have found it very useful. The exact clinical utility and feasibility of this test need to be studied in a prospective manner, we propose to do in our study.