The findings from a randomized clinical trial in Spain support de-escalation of antibiotic therapy in patients with bacteremia caused by Enterobacterales, researchers reported yesterday in The Lancet Infectious Diseases.
Although empiric treatment with antipseudomonal beta-lactams is often necessary with severe infections to ensure that the causative organism is covered, these drugs can exert selection pressure for resistance mechanisms and increase the risk of Clostridioides difficile, which is why de-escalation to a narrower-spectrum drug is recommended once the infecting pathogen has been identified. But the dearth of evidence on the safety and efficacy of de-escalation in specific clinical situations has been a barrier to wider acceptance of the practice.
To determine the safety and efficacy of de-escalation in patients with bacteremia caused by Enterobacterales, investigators randomly assigned trial participants to de-escalate to one of several antibiotics according to susceptibility (the de-escalation group) in the first 48 hours after the antimicrobial susceptibility report was available or to continue with empiric treatment (the control group). The primary outcome was clinical cure 3 to 5 days after end of treatment. The non-inferiority margin was –10%.
Potential ecological benefit
Of the 344 patients assigned to the treatment groups, 164 patients in the de-escalation group and 173 in the control group were included in the modified intention-to-treat (mITT) population. Analysis of the primary outcome showed that 90% of patients in the de-escalation group had clinical cure, compared with 89% in the control group, for a risk difference of 1.6 percentage points (95% confidence interval, –5.0 to 8.2).
Given the potential ecological benefit, these results provide evidence to implement active actions to promote de-escalation in this clinical setting.
The number of adverse events reported was 219 in the de-escalation group and 175 in the control group, with 24% and 32% considered severe, respectively. Five percent of patients in the de-escalation group and 6% in the control group died during the 60-day follow-up, but no deaths were treatment-related.
The investigators note that the most frequent sources of bacteremia were the urinary and biliary tracts, so the results apply mainly to those populations.
"Given the potential ecological benefit, these results provide evidence to implement active actions to promote de-escalation in this clinical setting," they wrote.