Bacterial infections among COVID-19 patients are rare, but antibiotic use is high

Analysis of antimicrobial use among COVID-19 patients in the UK, published in The Lancet Microbe journal, highlights a need to include guidance on optimal usage of the medicines in patient care.

The most comprehensive analysis of its kind to date confirms that bacterial co-infections among COVID-19 patients are rare, yet antimicrobial use was high during the country’s first wave between February 6 and June 8, 2020.

COVID-19 and antimicrobial use

The findings suggest that adopting measures to support optimal antimicrobial use, known as antimicrobial stewardship, in COVID-19 patients will help to slow the emergence of antimicrobial resistance. In particular, the authors say emphasis should be placed on restricting empirical antimicrobial prescribing – when a medicine is administered before tests confirm a bacterial infection – especially when patients are first admitted to hospital.

Dr Antonia Ho, of the MRC-University of Glasgow Centre for Virus Research, UK, said: “In making any assessment of the use of antimicrobials in the treatment of COVID-19 patients it is essential to acknowledge that clinicians in the UK – and worldwide – have been battling a global medical emergency. Given the unprecedented challenges posed by the pandemic, particularly during its early stages when admitted patients were very sick, effective treatments were limited, and the role of possible co-infections unknown, it is unsurprising that doctors would prescribe antimicrobials. However, we now know that bacterial co-infection is uncommon in patients with community-acquired COVID-19. Since antimicrobial resistance remains one of the biggest public health challenges of our time, measures to combat it are essential to help ensure that these life-saving medicines remain an effective treatment for infection in years to come.”

Bacterial respiratory infections among patients hospitalised with COVID-19

A previous systematic review and meta-analysis of 38 studies had suggested that bacterial respiratory and bloodstream infections are rare among patients hospitalised with COVID-19, but the studies included were small and provided limited information on the timing of infection, specific bacteria causing the infections, and antimicrobial drug classes used.

The authors of the new study report the numbers of COVID-19-related bacterial infections, types of bacteria identified, the number of people prescribed antimicrobials, and the types of antimicrobials prescribed among 48,902 patients admitted to 260 hospitals in England, Scotland, and Wales due to COVID-19 between February 6 and June 8, 2020. The timing of infection was recorded either as occurring prior to hospital admission (co-infection) or acquired after admission (secondary infection). The average patient age was 74 years, and 43% were female.

Microbiological tests – including blood tests and analysis of sputum and deep lung samples – were recorded for 8,649 patients, with COVID-19-related respiratory or bloodstream bacterial infections detected in 1,107 patients. By contrast, bacterial co-infections are far more common with severe influenza, occurring in 23% of patients.

Secondary infections in prevalence

When bacterial infections were detected in COVID-19 patients, 71% were secondary infections, acquired more than 2 days after patients were admitted to hospital. Staphylococcus aureus and Haemophilus influenzae were the most common causes of respiratory co-infections, while Enterobacteriaceae and S aureus were the most common in secondary respiratory infections. Bloodstream infections were most often caused by Escherichia coli and S aureus.

Among patients with available data, 37% (13,390/36,145 patients) had been prescribed antimicrobials for their illness by a doctor or pharmacist before being admitted to hospital, while 85% (39,258/46,061 patients) received one or more antimicrobials at some point during their hospital stay. Antimicrobial use was highest during March and April 2020, but fell during May, so further assessment of any changes to patterns of prescribing is essential.

Broad-spectrum agents such as carbapenems – a class of antimicrobials reserved for the treatment of severe or high-risk bacterial infections – were used frequently, accounting for 3.8% of all prescriptions. In contrast, carbapenem-sparing alternatives were used less often, accounting for between 0.2% and 1.5% of all prescriptions.

Based on their findings, the authors recommend a range of existing antimicrobial stewardship interventions that should be prioritised for incorporation into COVID-19 patient care. As well as restricting prescribing without a confirmed diagnosis, these include tailoring the choice of antimicrobials (when required) to likely pathogens and local resistance patterns and encouraging clinicians to discontinue antimicrobials if co-infection is deemed unlikely and tests confirm that patients do not have a bacterial infection.

Antimicrobial stewardship principles

Dr. Clark Russell, of the University of Edinburgh Centre for Inflammation Research, said: “Our findings add much-needed depth to our understanding of how antimicrobials have been used in the treatment of patients with COVID-19, and how antimicrobial usage could be optimised. Prioritising and incorporating existing antimicrobial stewardship principles into care plans could help to prevent a rise of drug-resistant infections becoming a longer-term sequela of the pandemic. Our study looked at only the first pandemic wave in the UK, so it is important that future studies assess antimicrobial usage later in the pandemic, both in the UK and other parts of the world.”

The authors acknowledge some limitations to their study. Microbiological diagnosis of co-infection is challenging, particularly during a pandemic, therefore, as fewer than 20% of participants had microbiological investigations recorded, the true rate of bacterial infections may differ from what the study reported. Clinical findings that would help inform a diagnosis of bacterial infection were not collected at the time of microbiological sampling, contributing to the existing challenges of using test results to retrospectively distinguish between a true infection and other plausible explanations.

 

 

 

 

http://www.thelancet.com/journals/lanmic/article/PIIS2666-5247(21)00090-2/fulltext

 

This post is also available in: Chinese (Simplified)

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *