The added role of bronchoalveolar lavage (BAL) in SARS-CoV-2 detection in hospitalized patients with suspected COVID-19 pneumonia and at least one negative nasopharyngeal swab (NPS) has yet to be definitively established. We aimed to provide a systematic review and meta-analysis to summarize data from the literature on the diagnostic yield of BAL in this context. We searched Medline and Embase for all studies reporting outcomes of interest published up to October 2021. Two authors reviewed all titles/abstracts and retrieved the selected full texts according to predefined selection criteria. The summary estimate was derived using the random-effects model. Thirteen original studies, involving 868 patients, were included. The summary estimate of proportions of SARS-CoV-2 positivity in BAL fluid in patients with at least one previous negative NPS was 20% (95% confidence interval [CI]; 11-30%). Moreover, microbiological tests of BAL fluid led to the identification of other pathogens, mainly bacteria, in up to two-thirds of cases. BAL plays a crucial role in the diagnostic work-up of patients with clinical suspicion of COVID-19 and previous negative NPS, as it allowed to detect the infection in a significant proportion of subjects, who would have been otherwise misclassified, with relevant implications in the prevention of disease spread, especially in hospital settings.
Keywords: SARS-CoV-2 pneumonia; bronchoalveolar lavage; diagnostic yield; negative upper respiratory tract swabs.