Written by Valeria Ilieva, MD, Infectious Disease Physician, Medical Director AMS subcommittee, South Puget Sound
SOUTH PUGET SOUND – Balance Trial: https://www.nejm.org/doi/abs/10.1056/NEJMoa2404991 was published in the New England Journal of Medicine (Nov. 2024).
A multicenter international noninferiority trial collected data over the course of almost 10 years (2014 to 2023). A total of 3608 patients across 74 hospitals were randomized to receive antibiotics for 7 vs 14 days.
This is the first large scale non-inferiority trial to address some of the weaknesses of previously published studies addressing the length of treatment of bacteremia that were criticized for lacking sufficient power, selection of non-critically ill patients, selecting patients with uncomplicated sources leading to simple bacteremia with predominantly Gram-negative pathogens. This could possibly explain why from more than 36,000 who were eligible to participate in the BALANCE trial 9,000 were excluded due to lack of consent, including attending physicians declining to enroll their patients. Possible explanation - the relative insufficiency in the data backing up the decisions for shorter antibiotic courses at the time of enrollment.
The researchers excluded: Staph aureus bacteremia, candidemia, severely immunocompromised patients with persistent neutropenia, hematopoietic cell transplant recipients, solid organ transplant on immunosuppressive therapy, cases when bacteremia originated from foci requiring prolonged treatment (e.g. osteomyelitis, septic joints, infected intravascular devices, etc.), and single positive cultures for pathogens which were considered contaminant.
Results: 1,814 patients were assigned to 7 days of antibiotic treatment, and 1,794 to 14 days. At enrollment, 55% of patients were in the ICU and 45% were in the hospital wards. Most of the infectious community was acquired (75.4%) vs hospital acquired (13.4%) and ICU acquired (11.2%). Most common sources include urinary tract, intra-abdominal or hepatobiliary, respiratory tract, vascular catheters, and skin and soft tissue. Bacterial pathogens were both Gram-negative and Gram-positive bacteria: Enterobacteriaceae (E, coli, Klebsiella, Proteus, Serratia, Pseudomonas), Enterococci, coagulase- negative staphylococci, and streptococci. Co-morbidities include diabetes, obesity, arrythmia, steroid use or some degree of immunosuppression, renal insufficiency (including HD), CHF, CAD, PAD, liver disease, leukemia or lymphoma, solid-organ cancers. Antibiotic selection, dosing, and route of administration were at the discretion of the treating team.
Primary outcome: 90-days all-cause mortality. Overall mortality by 90 days was slightly higher in the 14-days group (16.1%) vs (14.5%) in the one-week course group, showing non-inferiority in the shorter duration of treatment.
Secondary outcomes: Mortality in ICU vs floors, difference in bacteremia relapse, measures of length of stay, vasopressor use, mechanical ventilation use, were similar in both groups. The percentage of antimicrobial related adverse events, C. difficile infection, and secondary infection or colonization with antibiotic-resistant bacteria were similar in both groups.
Some non-adherence to the study protocol was noted. For example, some patients received a shorter or longer duration of treatment than 7 or 14 days, respectively. The median duration of treatment in the 7 days group was 8 days (7-11 days) vs. 14 days (14-15) in the 14 days group. Despite that, per protocol analysis showed that 7 days of treatment were non-inferior to 14 days of treatment. The non-inferiority was preserved in the modified intention -to- treat analysis that excluded patients who died before day 7 of antibiotics. The results were consistent across secondary clinical outcomes and across prespecified subgroups defined according to patient, pathogen, and syndrome characteristics.
The BALANCE trial showed that physician driven 7-day antibiotic treatment course is non-inferior to 14-days strategy among hospitalized patients. This strategy does not require new expensive diagnostic or treatment technologies and could lead to large savings in drug acquisition cost. However, the results of this trial can’t be applied to Staphylococcus aureus bacteremia, bacteremia due to Candida spp. and does not fully address the rate of recurrence of infections with the same pathogen in different anatomic location within the studied 90 days. In almost all cases, source control was achieved. Also, most infections were community-acquired, leading to a relatively low percentage of drug-resistant bacteria.
For simpler review of this trial please refer to this short video: https://www.youtube.com/shorts/0VBD2fh7r1o
