Shorter duration of antibiotics, similar success rates with algorithm-based therapy for staphylococcal bacteremia

Référence

Holland TL, Raad I, Boucher HW, et al, for the Staphylococcal Bacteremia Investigators. Effect of algorithm-based therapy vs. usual care on clinical success and serious adverse events in patients with staphylococcal bacteremia. JAMA 2018;320(12):1249-1258.

Sommaire

Recommended treatment duration for staphylococcal bacteremia is based on low-quality evidence and expert opinion. In this study, patients with positive blood cultures for either Staphylococcus aureus or coagulase-negative staphylococci were randomized to algorithm-based therapy or usual care. The algorithm stated that patients with S aureus bacteremia who had uncomplicated infections (eg, defervescence within 72 hours, no indwelling devices) should be treated for 14 days, and those with complicated infections (eg, endocarditis, signs of metastatic infection) should be treated for 28 days to 42 days. For patients with coagulase-negative staphylococcal bacteremia, the algorithm stated that treatment should last from 0 to 28 days based on the complexity of the infection with 5 days of therapy recommended for uncomplicated bacteremia. Further, in the algorithm group, patients with methicillin-resistant staphylococci were treated with vancomycin or daptomycin while patients with methicillin-susceptible staphylococci were treated with an anti-staphylococcal penicillin or cefazolin. In the usual care group, the choice and duration of antibiotics were determined by the treating clinician. Patients in the 2 groups were similar: More than 80% of patients had simple or uncomplicated bacteremia, mostly with coagulase-negative staphylococci, and two-thirds were health-care–related infections. The duration of antibiotic therapy was shorter in the algorithm group (4.4 days vs 6.2 days; difference -1.8 days; 95% CI -3.1 to -0.6 days), driven by the shorter duration of treatment in patients with uncomplicated coagulase-negative staphylococcal bacteremia (5.3 days vs. 8.4 days; difference -3.1 days; -4.9 to -1.3 days). The primary outcome was clinical success rate, which was defined as clinical and microbiological resolution of bacteremia at 28 days. Clinical success occurred in 82% of the algorithm group and 81.5% of the usual care group, with a lower limit of the confidence interval of greater than -15%, which fulfills the noninferiority criteria. There was no significant difference detected in the rate of serious adverse events, but the confidence interval was wide (32.5% vs 28.3%; difference 4.2%; -3.8% to 12.2%).