Patients who received an appropriate initial antibiotic for bloodstream infections (BSIs) had a lower risk of death compared with those who received an inappropriate antibiotic, US researchers reported today in JAMA Network Open.
The multicenter cross-sectional study, which included more than 32,000 patients who had been hospitalized with BSIs, found that receipt of appropriate initial empiric antibiotic therapy was associated with lower risk of in-hospital death for three pathogen groups compared with those whose initial antibiotic was inappropriate. For all three groups, the risk of in-hospital death was more than or nearly cut in half for patients who received the right antibiotic.
"Given these findings, it is important for clinicians to carefully choose empirical antimicrobial agents to improve outcomes in patients with BSIs," the study authors wrote.
For the study, a team led by researchers with Duke University Medical Center analyzed data from the Premier Healthcare database from 2016 through 2020, including data from a subset of US hospitals that reported laboratory and microbiologic data. The study included adults who had positive results from the first blood cultures taken during hospitalization and received treatment with at least one systemic antibiotic.
BSIs are severe bacterial infections that can lead to sepsis and can be deadly in 10% to 30% of cases, which makes the choice of initial antibiotic therapy—before the results of antibiotic susceptibility tests have been obtained—a critical one. Broad-spectrum antibiotics are typically used when a patient is suspected of having a severe infection.
The study authors note that while there is concern about overuse of broad-spectrum antibiotics, because of adverse side effects and the potential for promoting antibiotic resistance, the emergence and rapid spread of drug-resistant pathogens means there a risk of choosing an antibiotic that doesn't work.
"Therefore, a delicate balance exists between overuse of broad-spectrum antibiotics and undertreatment of infections, both of which can lead to harm," they wrote.
The aim of the study was to examine the pathogens causing BSIs, their antibiotic resistance profile, and in-hospital mortality associated with appropriate versus inappropriate empirical antibiotic therapy. Previous studies that looked at associations between empirical antibiotic therapy and outcomes in BSI patients have shown conflicting results.
A delicate balance exists between overuse of broad-spectrum antibiotics and undertreatment of infections, both of which can lead to harm.
The pathogens were categorized into three groups: gram-negative rods (GNRs), gram-positive cocci (GPC), and Candida species. The study also assessed resistant organisms.
Appropriate empirical antibiotic therapy was defined as initiation of an antibiotic to which the pathogen isolated from blood culture was susceptible.
Lower rates of appropriate drugs for resistant bacteria
A total of 32,100 patients from 183 hospitals who had BSIs were included in the study (mean age 64 years, 54.8% male, 69.9% non-Hispanic White). Overall, 46.6% of patients had positive results for GNRs, 52.5% were infected with GPC, and 0.9% had BSIs caused by Candida species. The most frequent GNRs were Escherichia coli and Klebsiella pneumoniae, and the most frequent GPC were Staphylococcus aureus and Streptococcus species. The in-hospital mortality rate was 14.3%
The overall proportions of appropriate antibiotic therapy were high for GNRs and GPC (94.4% and 97.0%, respectively) but lower for Candida species (65.1%). They were lower still for resistant organisms: 55.3% for carbapenem-resistant Enterobacterales species and 60.4% for vancomycin-resistant Enterococcus. The proportion of appropriate empirical therapy for non-resistant pathogens was generally higher than it was for resistant pathogens.
The crude in-hospital mortality rates for the appropriate empirical antibiotic therapy group compared with the inappropriate empirical antibiotic therapy group were 12.7% versus 20.8% for GNR, 14.8% versus 21.1% for GPC, and 23.7% versus 34.3% for Candida species. In multivariable multilevel logistic regression models that adjusted for age, sex, race and ethnicity, comorbidities and other variables, receipt of appropriate empirical therapy was associated with lower risk of in-hospital deaths for GNR (adjusted odds ratio [aOR], 0.52; 95% confidence interval [CI], 0.42 to 0.64), GPC (aOR, 0.60; 95% CI, 0.47 to 0.78), and Candida species (aOR, 0.48; 95% CI, 0.23 to 0.99).
Additional analysis of specific types of pathogens found that the in-hospital risk of death associated with appropriate empirical therapy was lower for patients infected with E coli (aOR, 0.64; 95% CI, 0.44 to 0.93), Klebsiella species (aOR, 0.54; 95% CI, 0.31 to 0.93), ceftriaxone-resistant gram-negative organisms (aOR, 0.62; 95% CI, 0.46 to 0.83), S aureus (aOR, 0.40; 95% CI, 0.27 to 0.61), methicillin-resistant S aureus (aOR, 0.46; 95% CI, 0.30 to 0.71), and Enterococcus species (aOR, 0.48; 95% CI, 0.30 to 0.77).
Identifying risk factors for resistance
The authors say the rates of the appropriate empirical antibiotic therapy observed in the study are higher than have been reported in previous studies, which could be attributed to implementation of antimicrobial stewardship programs, better diagnostic tools, and development of better treatment guidelines.
It is important for clinicians to carefully choose empirical antimicrobial agents to improve outcomes in patients with BSIs.
But they note that the findings indicate that selecting the right antibiotic for patients with BSIs remains a challenge, particularly for resistant pathogens, and that finding ways to determine which patients may be at greater risk for drug-resistant BSIs could help improve both antibiotic selection and outcomes.
"Identifying potential risk factors for resistant pathogens might help improve the choice of correct empirical antibiotic therapy in patients with a high likelihood of infection with resistant pathogens," they wrote. "A judicious selection of broad empirical antimicrobial agents to treat BSIs is essential, but a comprehensive approach would also be warranted to further improve outcomes."