Bacterial co-infections and antibiotic prescribing practice in adults with COVID-19: experience from a single hospital cluster

Cheng, Lily Shui-kuen; Chau, Sandy Ka-yee; Tso, Eugene Yuk-keung; Tsang, Steven Woon-choy; Li, Issac Yuk-fai; Wong, Barry Kin-chung; Fung, Kitty Sau-chun

Background: Coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has affected millions of individuals since December 2019, resulting in significant morbidity and mortality globally. During the 1918 Influenza Pandemic, it was observed that influenza was associated with bacterial co-infections. However, empirical or prophylactic antibiotic use during viral pandemics should be balanced against the associated adverse drug events. Methods: In this retrospective cohort study, we investigated bacterial co-infections in adults with COVID-19 in Hong Kong. Notably, at the time of writing this report, patients with varying disease severities were isolated in hospitals until confirmatory evidence of virological clearance or immunity was available. The study included adults with laboratory-confirmed COVID-19 admitted to a single hospital cluster between 8 January 2020 and 1 May 2020. We obtained data regarding patient demographics, clinical presentations, blood test results, treatment, and outcomes. Bacteriological profiles and risk factors for co-infections were investigated. Antibiotic prescription practices were also reviewed. Results: Of the 147 patients recruited, clinical disease was suspected in 42% (n = 62) of patients who underwent testing for other respiratory infections. Notably, 35% (n = 52) of the patients were prescribed empirical antibiotics, predominantly penicillins or cephalosporins. Of these, 35% (n = 18) received more than one class of antibiotics and 37% (n = 19) received empirical antibiotics for over 1 week. Overall, 8.2% (n = 12) of patients developed bacterial co-infections since the detection of COVID-19 until discharge. Methicillin-susceptible Staphylococcus aureus was the most common causative pathogen identified. Although 8.2% (n = 12) of patients developed hypoxia and required oxygen therapy, no mortality was observed. Multivariate analysis showed that pneumonic changes on chest radiography at the time of admission predicted bacterial co-infections. Conclusion: These findings emphasise the importance of judicious administration of antibiotics throughout the disease course of COVID-19 and highlight the role of antimicrobial stewardship during a pandemic.