The clinical benefits of an antimicrobial de-escalation strategy were compared with those of a no-switch strategy in bacteremic patients. Adults with community-onset monomicrobial Escherichia coli, Klebsiella species and Proteus mirabilis bacteremia treated empirically using broad-spectrum beta-lactams, including third-generation cephalosporins (GCs), fourth-GC or carbapenems, were treated definitively with first- or second-GCs (de-escalation group), the same regimens as empirical antibiotics (no-switch group), or antibiotics with a broader-spectrum than empirical antibiotics (escalation group). The eligible 454 adults were categorized as the de-escalation (231 patients, 50.9%), no-switch (177, 39.0%), and escalation (46, 10.1%) groups. Patients with de-escalation therapy were more often female, had less critical illness and fatal comorbidity, and had a higher survival rate than patients in the other two groups. After propensity score matching in the de-escalation and no-switch groups, critical illness at onset (Pitt bacteremia score ≥ 4; 16.5% vs. 12.7%; P = 0.34) or day 3 (2.5% vs. 2.5%; P = 1.00), fatal comorbidity (16.5% vs. 21.5%; P = 0.25), time to defervescence (4.6 vs. 4.7 days; P = 0.89), hospital stays (11.5 vs. 10.3 days; P = 0.13) and 4-week crude mortality rate (4.4% vs. 4.4%; P = 1.00) were similar. However, lower antibiotic cost (mean: 212.1 vs. 395.6 US$, P < 0.001) and fewer complications of bloodstream infections due to resistant pathogens (0% vs. 5.1%, P = 0.004) were observed in the de-escalation group. De-escalation to narrower-spectrum cephalosporins is safe and cost-effective for adults with community-onset EKP bacteremia stabilized by empirical broad-spectrum beta-lactams.
All Science Journal Classification (ASJC) codes
- Microbiology (medical)
- Infectious Diseases
- Pharmacology (medical)