番茄社区

An algorithm for testing and treatment of patients with staphylococcal bacteremia bloodstream infections was non-inferior to usual care, and the difference between the rate of serious adverse events was similar, a randomized trial found.

Rates of clinical success were comparable among patients assigned to algorithm-based therapy compared with those who received usual practice (82.0% vs 81.5%, respectively), reported Vance G. Fowler, Jr., MD, of Duke University in Durham, North Carolina, and colleagues.

While the rate of serious adverse events was higher among those randomized to the algorithm-based therapy, the difference was non-significant, the authors wrote in

"The big point of this study is doing the same with less," said Fowler in a statement. "If we are able to have the same outcomes but use less antibiotics, that has tremendous benefit at all levels of care."

Consensus guidelines for treatment of staphylococcal bacteremia differ depending on the complexity of the infection, the authors said, with prolonged courses of antibiotics recommended for complicated staphylococcal bacteremia and shorter courses for uncomplicated infections. However, they noted "considerable variation" in treatment practices, as the guidelines are primarily based on "low-quality evidence and expert opinion."

They sought to examine a standardized strategy for classifying patients with either complicated or uncomplicated infections. Patients were from 16 medical centers, mostly in the U.S., from 2011 to 2017. They were eligible if they had one or more blood cultures positive for Staphylococcus aureus (followed up for 42 days after therapy) or coagulase-negative staphylococci (followed up for 18 days after therapy).

In the algorithm group, duration of treatment was defined by the algorithm-defined category, which was based on specific clinical characteristics. The choice and duration of antibiotics in the usual practice group was determined by the physician, the authors said.

The primary outcome was defined as both success rate at test-of-cure evaluation and investigator-reported serious adverse events in the two treatment groups.

Overall, 509 patients were randomized to therapy -- 255 to the algorithm therapy group and 254 to the usual practice group. Patients were a mean age of about 57 and 44.4% were women. About three-quarters had coagulase-negative staphylococcal bacteremia.

An intention-to-treat analysis found 82.0% of patients in the algorithm-based group achieved clinical success vs 81.5% of patients in the usual practice group (difference 0.5%, 1-sided 97.5% CI -6.2% to ∞). The authors said that "since the lower limit of the confidence interval was greater than -15%, algorithm-based therapy was determined to be noninferior to usual practice."

There was also no significant difference in serious adverse events (32.5% in algorithm group vs 28.3% in usual care, difference 4.2%, 95% CI -3.8% to 12.2%), though the authors noted the wide confidence intervals. Adverse events leading to discontinuation of the study drug were reported in four patients in the algorithm group and one patient in the usual care group.

There were 16 patients who died prior to test-of-cure assessment in the algorithm group and 14 in the usual practice group. Two deaths in the algorithm group and three in the usual practice group were attributed to infection, and all patients were among patients with S. aureus infection, the authors said.

In a per-protocol analysis, mean duration of therapy among patients with simple or uncomplicated bacteremia was 4.4 days in the algorithm therapy group versus 6.2 days in the usual-care group.

In , Eli N. Perencevich, MD, of the University of Iowa Carver College of Medicine in Iowa City, and Preeti N. Malani, MD, of the University of Michigan in Ann Arbor, also noted that there was a 3-day reduction in therapy for uncomplicated coagulase-negative staphylococcus bacteremia.

"Given that vancomycin is the most commonly prescribed antibiotic in U.S. acute care hospitals, a 3-day reduction for a high-incidence condition such as uncomplicated coagulase-negative staphylococcus bacteremia could have a sizable public health effect," they wrote.

Perencevich and Malani also characterized this study as "an elegant addition to the evidence base of how to manage staphylococcal bacteremia" and that these results should "likely influence the next iteration of treatment guidelines."

Study limitations included the fact that it was not powered to compare subgroups, such as patients with uncomplicated S. aureus bacteremia; that the open-label design of the study could have introduced bias; and that "repeated exposure to the algorithm may have influenced management decisions in clinicians caring for patients receiving algorithm-based therapy as well as those receiving usual practice."