Suspected sepsis driving US broad-spectrum antibiotic use, study finds

Sepsis illustration

Dr_Microbe / iStock

A new study of clinical data from US hospitals indicates that suspected community-onset sepsis accounts for half of broad-spectrum antibiotic use, but much of it may be unnecessary.

The study, published yesterday in JAMA Network Open, found that patients with suspected community-onset sepsis accounted for half of total inpatient antipseudomonal beta-lactam antibiotic and anti–methicillin-resistant Staphylococcus aureus (MRSA) antibiotic days. But antibiotic-resistant organisms were isolated from less than 10% of sepsis patients, and the proportion with resistant infections declined over time, while the proportion who received broad-spectrum antibiotics increased.

Sepsis occurs when the immune system overreacts to an infection, triggering a chain of events that can lead to tissue damage, organ failure, and death. The authors of the study say that while early, empiric administration of broad-spectrum antibiotics is strongly encouraged for sepsis patients to cover a wide range of potential pathogens, the "increasing mismatch between empiric prescribing and identified pathogens" observed in their study could promote antibiotic resistance and pose a risk to patient safety without any benefit.

"These results suggest that more attention is needed toward balancing early broad-spectrum antibiotic prescribing for patients with sepsis with limiting overuse for the majority who do not have antibiotic-resistant infections," they wrote.

Unnecessarily broad treatment

To assess the extent to which community-onset sepsis is driving broad-spectrum antibiotic use, a team led by researchers from Harvard Medical School analyzed data on nearly 6.3 million adults admitted to 241US hospitals from 2017 through 2021. The main outcomes were annual rates of empiric anti-MRSA and/or antipseudomonal beta-lactam antibiotic use—the two primary types of broad-spectrum antibiotics used for sepsis—and the proportion of use that was likely unnecessary based on the absence of beta-lactam–resistant gram-positive pathogens or ceftriaxone-resistant gram-negative pathogens.

Of the 6,272,538 hospitalizations (median patient age, 66; 49.6% male; 73.1% White) during the study period, 894,724 (14.3%) had suspected community-onset sepsis. Those patients accounted for 50.1% of total inpatient anti-MRSA antibiotic days and 49.3% of total antipseudomonal beta-lactam days. Patients with suspected sepsis who received anti-MRSA or antipseudomonal therapy tended to be more severely ill than those who did not.

The most common empiric antibiotic was vancomycin (41.2%), followed by ceftriaxone (37.8%), piperacillin-tazobactam (31.1%), and cefepime (24.8%). 

Over the study period, the proportion of patients with suspected sepsis who received anti-MRSA or antipseudomonal agents rose from 63.0% to 66.7% (adjusted odds ratio [aOR] per year, 1.03; 95% confidence interval [CI], 1.03 to 1.04). The increase was driven primarily by antipseudomonal beta-lactam use, which increased from 54.4% in 2017 to 59.6% in 2021.

These results suggest that more attention is needed toward balancing early broad-spectrum antibiotic prescribing for patients with sepsis with limiting overuse for the majority who do not have antibiotic-resistant infections.

But resistant organisms were isolated in only 65,434 (7.3%) of all suspected sepsis cases, and the proportion of patients who had any resistant organism decreased from 9.6% in 2017 to 7.3% in 2021 (aOR per year, 0.87; 95% CI, 0.87 to 0.88). Most patients with suspected sepsis who were treated with anti-MRSA and/or antipseudomonal agents—90.5%—had no resistant organisms, with the proportion increasing from 88.0% in 2017 to 91.6% in 2021 (aOR per year, 1.12; 95% CI, 1.11 to 1.13).

The fraction of patients who received inappropriately narrow-spectrum antibiotics was stable over time for those with resistant gram-positive organisms (aOR per year, 0.98; 95% CI, 0.96 to 1.00) but declined for those with resistant gram-negative organisms (aOR per year, 0.89; 95% CI, 0.87 to 0.90).

"There was a decrease in patients receiving inappropriately narrow therapy, but the absolute number who received unnecessarily broad treatment was nearly 50-fold higher than the number who were undertreated," the authors noted, adding that the increase in broad-spectrum antibiotic use was not associated with improved mortality.

Balancing the risks of undertreatment with stewardship

The authors say the results of the study should not be interpreted as critical of the prescribing standards of front-line providers, because broad-spectrum antibiotics are required by the Centers for Medicare & Medicaid Services' Severe Sepsis and Septic Shock Bundle within 3 hours of sepsis onset, while the Surviving Sepsis Campaign's 1-Hour Bundle calls for administration of broad-spectrum agents within 1 hour. These initiatives, they note, may be among the reasons why broad-spectrum antibiotic use for suspected sepsis is increasing.

"Additionally, front-line clinicians often must make treatment decisions with incomplete information," they wrote. "This leads many to err on the side of broader coverage, particularly when patients are severely ill."

They say the study was designed to gather epidemiologic data that could be used in efforts to strike the right balance between undertreatment and antimicrobial stewardship.

"These findings help elucidate the contribution of suspected sepsis to antibiotic prescribing in US hospitals and inform ongoing deliberations on finding the balance between assuring early, appropriate therapy for patients with severe infections vs limiting overuse for the majority of patients who do not have antibiotic-resistant infections," they concluded.

This week's top reads

Our underwriters