Antibiotic Therapy for Staphylococcal Bloodstream Infections: Doing the Same with Less
Researchers found that the use of the treatment algorithm for staphylococcal bacteremia shortens therapy without compromising outcomes.
A treatment algorithm for staphylococcal bloodstream infections (BSIs) featuring markedly shorter antibiotic therapy than the conventional standard of care produces similar rates of success and serious adverse events. For complicated infections, the algorithm approach is better than standard care. The good news from the randomized, multinational, open-label, adjudicator-blinded trial was presented at the ID Week 2017 meeting in San Diego, California.
“The optimal duration of treatment for staphylococcal bacteremia is unknown. Long-course or short-course treatment may place the patient at risk. If treatment is too prolonged, there is a greater risk of the development of antibacterial resistance and antibiotic-associated adverse events. On the other hand, if the treatment duration is too short, there can be a greater chance of relapse due to inadequately treated infection,” said Thomas L. Holland, MD, Duke University Medical Center, Durham, North Carolina, on behalf of the Bacteremia Study Group.
With this in mind, the researchers sought to provide clarity concerning treatment time. “Our study rationale was that a strategy to identify patients with staphylococcal BSI who can safely be treated with shorter courses of therapy would improve care,” explained Dr. Holland.
The study explored whether an algorithm that defines treatment duration for staphylococcal BSI can provide similar efficacy and safety to the standard of care while reducing antibiotic duration—“doing the same with less.”
Standard of care included an unrestricted choice of antibiotics. The length of treatment was determined by the treating clinician. The algorithm-based therapy was more stringent. Antibiotics were limited to first-choice vancomycin or daptomycin for methicillin-resistant isolates, and cefazolin or nafcillin for methicillin-susceptible isolates. The duration of therapy was dictated by the severity of infection. For simple, uncomplicated, and complicated coagulase-negative staphylococci (CoNS) infections, antibiotics were given for 0-3, 5, and 7-28 days, respectively. For uncomplicated and complicated S. aureus infections, duration of antibiotic therapy was 14 and 28-42 days, respectively. The grading of infections involved various criteria set beforehand, and not by the treating clinician.
The study included adults with blood cultures that were positive for CoNS or Staphylococcus aureus. Intravenous catheters had been removed. Exclusion criteria included known or suspected complicated bacteremia and impaired kidney function evident as creatinine clearance <30 mL/min.
The co-primary endpoints were efficacy assessed as test of cure following treatment and safety. The pre-specified secondary endpoint was days of antibiotic therapy in patients able to be evaluated who did not have complicated BSI. An independent external adjudication committee established the primary efficacy outcome and assessed the significance of any potentially effective non-study antibiotic. An echocardiography lab conducted a blind review of all endocardiograms.
The 509 patients were randomized to receive standard of care (n=254) or the algorithm-based therapy (n=255). Four patients in each group were excluded when the bacteria were identified as being other than Staphylococcus. There were 190 CoNS patients in the standard of care group and 194 in the algorithm group. The respective numbers for S. aureus were 59 and 57. Of the CoNS infections in the standard of care group, 124 were simple, 52 were uncomplicated, and 15 were complicated. In those with S. aureus, 45 infections were uncomplicated and 14 were complicated. In the algorithm group, the CoNS infections were simple in 136 patients, uncomplicated in 39, and complicated in 19. For those with S. aureus infections, 34 were uncomplicated and 23 were complicated.
The 2 groups were balanced at baseline concerning age, sex, race, body mass index, risk factors (injection drug use, diabetes mellitus, immunosuppression, chronic renal insufficiency), and the setting of infection (nosocomial, community healthcare-associated, community-acquired).
Treatment was similarly successful in the 2 groups; 207 of the 254 (81.5%) of patients in the standard of care group, and 209 of 255 (82.0%) of patients in the algorithm-based therapy group (Absolute difference 0.5, 95% CI -5.2 to 6.1). “Algorithm-based therapy was as successful as the standard of care,” according to Dr. Holland.
Treatment success was similar for simple CoNS, uncomplicated CoNS, complicated CoNS, and uncomplicated S. aureus infections. For complicated S. aureus infections, treatment was successful for only 35.7% of standard care patients compared with 82.6% for the algorithm-based therapy group (Absolute difference 46.9, 95% CI 22.1 to 71.6). The treatment groups were similar in terms of treatment success for methicillin-susceptible S. aureus and CoNS, and methicillin-resistant S. aureus and CoNS.
The standard of care and algorithm-based groups displayed comparable rates of serious adverse events (28.3% vs 32.5%) and mortality (5.9% vs 6.7%). The comparability between the two groups extended to the duration of treatment.
“Use of the treatment algorithm for staphylococcal bacteremia shortens therapy without compromising outcomes. In cases of complicated S. aureus bacteremia, improved outcomes were evident with the algorithm-based therapy,” said Dr. Holland.
“This algorithm provides a means to accurately identify those patients with staphylococcal BSI for whom a short course therapy is appropriate,” he concluded.
Thomas L. Holland, MD: Consulting: The Medicines Company, Basilea Pharmaceuticals; Scientific Advisory Board: Motif Bio; Adjudication Committee, Achaogen; Grant support: NIH, FDA; Royalties: Up To Date; Employment: Duke University
Oral Abstract Session: Advances in Management of Bacteremia and Sepsis
Doing the Same with Less: A Randomized, Multinational, Open-Label, Adjudicator-Blinded Trial of an Algorithm vs. Standard of Care to Determine Treatment Duration for Staphylococcal Bacteremia
Brian Hoyle, PhD, is a medical and science writer and editor from Halifax, Nova Scotia, Canada. He has been a full-time freelance writer/editor for over 15 years. Prior to that, he was a research microbiologist and lab manager of a provincial government water testing lab. He can be reached at email@example.com.