A 7-day course of antibiotics for patients hospitalized with bloodstream infections was just as effective as a 14-day course, according to results of the randomized, open-label BALANCE trial.
Among patients with bloodstream infections who received 7 days of antibiotics, 14.5% had died at 90 days, versus 16.1% of those who received 14 days of antibiotics, for an absolute difference of -1.6% (95% CI -4.0 to 0.8), reported Nick Daneman, MD, MSc, of the University of Toronto, at the IDWeek annual meeting in Los Angeles.
The finding was consistent in the per-protocol analysis (-2.0% absolute difference, 95% CI -4.5 to 0.6) and in the modified intention-to-treat (ITT) analysis.
Moreover, the findings were also generally consistent for a wide range of secondary outcomes, including:
"We were not surprised by our findings, because a growing body of research has shown that shorter treatment durations are sufficient for many other infections," Daneman told MedPage Today. "But optimal antibiotic treatment durations have not been determined for patients with bloodstream infection, particularly for patients that are critically ill with these infections."
Subanalyses showed that the shorter antibiotic regimen was also non-inferior across patient location (ie, ICU or hospital ward) severity of illness, source of infection, different types of infectious syndromes, as well as gram-negative, gram-positive, or polymicrobial infections.
"Seven-day treatment should be the general strategy for most patients with bloodstream infection," he said. More than half of enrolled patients were critically ill, he noted. "So we know this treatment is good enough for our sickest of patients."
Of note, however, the study did not demonstrate a significant reduction in Clostridioides difficile infections or antibiotic resistance with the shorter antibiotic regimen, Daneman said.
He also pointed out that the trial excluded patients who were neutropenic, those who had undergone transplantation, and those with infections caused by Staphylococcus aureus.
"You have to decide whether you think you can generalize [these findings] to your transplant patients or your neutropenic patients," he told attendees. "And I would advise against extrapolating it to your Staph aureus patients until someone has the time to do that trial."
The BALANCE trial enrolled 3,608 patients across 74 hospitals in seven countries. In the ITT analysis, 1,814 participants were randomized to the 7-day antibiotic treatment arm and 1,794 to the 14-day arm. At enrollment, 55% of patients were in the ICU.
About 75% of infections were community-acquired, 13% were hospital-acquired, and 11% were ICU-acquired. The most common sources of bacteremia were the urinary tract (42.2%), abdomen (18.8%), lung (13.0%), vascular catheters (6.3%), and skin or soft tissue (5.2%).
Patient characteristics were balanced between the two groups, with a mean age of 70. About 32% had diabetes and 22% had solid organ malignancies. Approximately 12% were immunosuppressed and the same percentage had renal insufficiency.
More than 70 pathogens were isolated from index blood cultures and were representative of bacteria seen in surveillance studies, including "all the usual culprits and a great mix of gram-negative and gram-positive organisms," Daneman said.
The top three pathogens were Escherichia coli (44%), Klebsiella spp (15%) and Enterococcus spp ( 7%). Other common pathogens included Streptococcus spp, Pseudomonas, coagulase-negative Staph, Enterobacter spp, Proteus spp, and Serratia spp.
"As a community, I think we've all started to realize in the last several years that we don't want to treat patients for too long because that's more selective pressure on our bystander flora, greater risk of C. difficile, more risk of adverse events, and higher costs," Daneman told attendees. Excessive durations of treatment are the number one contributor to unnecessary days of antibiotic use, he said.
Study limitations included its open-label design, the exclusion of some patients, and the potential for nonadherence to the treatment duration assignment.