An analysis of 15 years' of data from the Veterans Affairs (VA) Healthcare System found significant reductions in several antibiotic-resistant healthcare-associated infections (HAIs) prior to the COVID-19 pandemic, US researchers reported yesterday in The Lancet Infectious Diseases.
The study authors say the observed reductions in pathogens like methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecium indicate that hospital infection control measures within the VA system have been successful and should be continued.
But the study also found worrisome increases in HAIs across other pathogen-drug combinations over the study period, along with a nationwide surge in some drug-resistant HAIs during the first 2 years of COVID-19 pandemic. That combination, the authors say, highlights the need for ongoing and improved antimicrobial resistance surveillance.
A 'holistic view' of the changing AMR landscape
To characterize the landscape of infection incidence and antimicrobial resistance within the VA Healthcare System, a team led by researchers from the Harvard T.H. Chan School of Public Health used electronic health records to analyze clinical microbiology data from all patients admitted to 138 VA medical centers from February 2007 through March 2022.
The main outcomes were the resistance proportion (the proportion of incident bacterial isolates identified as resistant) and the phenotypic incidence rate (incidence of infections per 1,000 admissions identified as resistant or susceptible).
The analysis focused on nine pathogens that are most commonly associated with HAIs, community-acquired infections, and multidrug-resistance, and 28 drug-pathogen combinations. The authors note that few studies to date have provided estimates for multiple pathogen-drug combinations for more than 5 years. Moreover, most of those studies have focused on the percentage of isolates with a resistant test result, which doesn’t reveal how the incidence of infections caused by resistant pathogens has changed.
"Therefore, reporting both antimicrobial resistance proportions and the phenotypic incidence is crucial for a holistic view of the changing antimicrobial resistance landscape," they wrote.
Over the study period, the researchers collected 991,527 incident isolates from 507,760 patients. The most commonly isolated organisms were S aureus, Escherichia coli, and Klebsiella pneumoniae, mostly collected from non–blood culture sites.
Through December 2019, the overall infection incidence across all nine pathogens declined, with the largest declines seen in Enterococcus faecium and S aureus. In addition, the proportion of S aureus infections that were MRSA declined from 57.7% to 44.6% (average annual percentage change [AAPC], –1.8%; 95% confidence interval [CI], –2.4% to –1.2%) and the proportion of E faecium infections resistant to vancomycin fell from 77.8% to 65.1% (AAPC, –1.2%; 95% CI, –2.5% to 0%). Fluoroquinolone resistance also declined, in both proportion and incidence, for most pathogens.
...reporting both antimicrobial resistance proportions and the phenotypic incidence is crucial for a holistic view of the changing antimicrobial resistance landscape.
But resistance didn't decline in all pathogens during the pre-COVID period. Third-generation cephalosporin resistance in E coli infections climbed from 6.7% in 2007 to 15.3% in 2019 (AAPC, 8.5%; 95% CI, 6.2% to 10.7%), caused primarily by an increase in the incidence of extended-spectrum beta-lactamase (ESBL)–producing E coli. Carbapenem resistance in Enterobacter cloacae infections increased from 1.1% to 7.3% (AAPC, 19.8%; 95% CI, 13.7% to 26.2%).
From January 2020 through March 2022, infection incidence and resistance trends rose for several pathogen-drug combinations, most notably carbapenem-resistant Acinetobacter baumannii (CRAB). The AACP for incidence of CRAB for the COVID period was 16.2% (95% CI, –12.8% to 54.9%), and the AACP for resistance proportion was 18.5% (95% CI, –16.2% to 67.7%).
These trends mirror reporting by the Centers for Disease Control and Prevention, which has documented sharp increases in CRAB and other drug-resistant HAIs in USA hospitals during the pandemic.
A new reporting standard
The study authors say the pre-pandemic declines in HAI incidence and resistance proportions could be explained by trends and policies observed in the VA Healthcare System over the study period.
For example, they note the decline in fluoroquinolone resistance correlated with substantial reductions in fluoroquinolone use in VA medical centers (from 125 days of therapy [DOT] per 1,000 patient-days in 2007 to 20 DOT in 2019). Those reductions, in turn, may be related to several fluoroquinolone safety warnings issued by the Food and Drug Administration.
The Veterans Health Administration also launched a number of infection control and stewardship initiatives from 2007 to 2014, including a MRSA prevention bundle.
The authors conclude that the findings have implications for policy, clinical practice, and antimicrobial stewardship, not just for the VA healthcare system but for all US hospitals. They also sound a note of caution.
"The reductions in methicillin-resistant S aureus, vancomycin-resistant E faecium, and HAI underscore the importance of system-wide infection prevention and antimicrobial stewardship programmes, such as those implemented in VA Medical Centers," they wrote. "Among these promising developments, the rising trends in ESBL-producing and carbapenem-resistant Enterobacterales, along with increased resistance of HAI during the COVID-19 pandemic necessitate enhanced surveillance and action."
In an accompanying commentary, experts from the National Institutes of Health Clinical Center call the study a "tremendous effort" and say that presenting both resistance proportion and phenotypic incidence is "crucial for better interpretation and could be considered a new reporting standard."
