Is bronchoscopic view a reliable method in diagnosis of tracheobronchial aspergillosis in critically ill non‐neutropenic patients?

• Published in: The clinical respiratory journal 2020-10, Vol.14 (10), p.956-964
• Main Authors: Youssif, Sahar Farghly, Hassan, Elham Ahmed, Moharram, Ahmed Mohamed, Farhan, Mohammed Ameen, Badary, Dalia M., Hasan, Ali Adel Azeem
• Format: Article
• Language: English
• Subjects: Aspergillosis; Bronchoscopy; Fungal infections; Infections; intensive care unit; Neutropenia; non‐neutropenic patients; tracheobronchial aspergillosis; Ulcers
• ISSN: 1752-6981; 1752-699X
• DOI: 10.1111/crj.13229

Introduction Pulmonary aspergillosis is the main respiratory fungal infection however; its diagnosis is missed or delayed in critically ill non‐neutropenic patients. Despite the utility of fiberoptic bronchoscopy for the evaluation of tracheobronchial aspergillosis (TBA) in immunocompromised patients has been extensively studied, however its utility in critically ill non‐neutropenic patients is underestimated. Objectives To assess the bronchoscopic changes suspected TBA relative to the microbiological and histopathological aspects in critically ill non‐neutropenic patients admitted to respiratory intensive care unit (RICU). Methods We prospectively studied 139 critically ill non‐neutropenic patients admitted to RICU and had a clinical suspicion of broncho‐pulmonary Aspergillus infection. Those patients were subjected to clinical and bronchoscopic assessment for the evaluation of suspected TBA. Microbiological culture of bronchoalveolar lavage (BAL) and histopathological examination of tracheobronchial biopsies were done. Results Bronchoscopic changes suspected TBA were found in 48.2% of patients (67/139), where Aspergillus infection was confirmed microbiologically in 59.7% (40/67) and histopathologically in 56.7% (38/67). Of these changes, whitish plaques ± ulcers, pseudomembrane and/or sticky secretion with hyperemic mucosa were detected in 68.7%, 26.9% and 16.4% respectively. These changes were mostly seen in the main bronchi (54/67; 80.6%). The sensitivity, specificity, positive, negative predictive values and overall accuracy of these bronchoscopic changes as compared with BAL fluid results were 83.3%, 70.3%, 53.2%, 91.2% and 74.1% respectively. Conclusions Bronchoscopy could be a reliable procedure for TBA diagnosis in critically ill non‐neutropenic patients. Whitish plaques ± ulcers were the prominent bronchoscopic changes with reasonable diagnostic accuracy for prediction of TBA.