Secondary infection rates and antibiotic prescribing in a COVID-19 hdu population

Introduction Secondary infection in COVID-19 has been associated with adverse outcomes and high mortality The prevalence of secondary infection in COVID-19 and optimal antimicrobial strategies remain unclear Methods Retrospective case-note review of patients with COVID-19 admitted to our institution's high dependency unit (HDU) from March to June 2020 Patients were PCR-positive for SARS-CoV-2 or had classical CT appearances and a compatible clinical presentation for COVID-19 Microbiological tests, antimicrobial prescriptions and clinical outcomes were recorded Results 84 patients were identified Median age was 68 5 years and 29/84 (34 5%) were female Respiratory support included HFNO (n=39), CPAP (n=56), non-invasive ventilation (n=3) and invasive ventilation (n=14) Overall mortality was 36/84 (42 9%) 6/84 patients (7 1%) had evidence of secondary infection (>105 CFUs on bronchoalveolar lavage (BAL);positive sputum culture or positive blood culture excluding skin contaminants) 28/84 (33 3%) had a respiratory sample sent: BAL n=10;sputum culture n=2;Legionella antigen n=15;throat swab multiplex PCR n=3;Biofire respiratory viral panel n=7 BAL was positive in 3/10 cases (Enterococcus faecium;Serratia marcescens and Escherichia coli;Pseudomonas aeruginosa) One sputum culture was positive for M abscessus 71/84 (84 5%) had blood cultures 8 (11 2%) were positive, of which 6 were considered skin contaminants and not deemed true secondary infection (coagulase negative Staphylococci n=5;Lysinibacillus sp n=1;Proteus mirabilis n=1;Staphylococcus epidermidis and Serratia marcescens n=1) All 84 patients received antimicrobials 32 (38 1%) received a macrolide, predominantly azithromycin Macrolide usage was not associated with mortality or admission length, but was associated with increased intubation rate (28 1% vs 9 6%, p=0 027) Initial antibiotic treatment was monotherapy in 45 (53 6%) cases and dual therapy in 39 (46 4%) Initial treatment with two antibiotics versus monotherapy was not associated with mortality but was associated with increased intubation rate (25 6% vs 8 9%, p=0 040) and increased mean admission length (16 5 vs 11 6 days, p= 036) Discussion Robust evidence of secondary infection in patients with COVID-19 was uncommon in our cohort Increased intubation rates in patients prescribed a macrolide and those initially prescribed dual antibiotic therapy is likely to reflect more severe disease There is considerable potential for enhanced antimicrobial stewardship in further waves of COVID-19