RESUMEN
BACKGROUND: De-escalation from broad-spectrum to narrow-spectrum antibiotics is considered an important measure to reduce the selective pressure of antibiotics, but a scarcity of adequate evidence is a barrier to its implementation. We aimed to determine whether de-escalation from an antipseudomonal ß-lactam to a narrower-spectrum drug was non-inferior to continuing the antipseudomonal drug in patients with Enterobacterales bacteraemia. METHODS: An open-label, pragmatic, randomised trial was performed in 21 Spanish hospitals. Patients with bacteraemia caused by Enterobacterales susceptible to one of the de-escalation options and treated empirically with an antipseudomonal ß-lactam were eligible. Patients were randomly assigned (1:1; stratified by urinary source) to de-escalate to ampicillin, trimethoprim-sulfamethoxazole (urinary tract infections only), cefuroxime, cefotaxime or ceftriaxone, amoxicillin-clavulanic acid, ciprofloxacin, or ertapenem in that order according to susceptibility (de-escalation group), or to continue with the empiric antipseudomonal ß-lactam (control group). Oral switching was allowed in both groups. The primary outcome was clinical cure 3-5 days after end of treatment in the modified intention-to-treat (mITT) population, formed of patients who received at least one dose of study drug. Safety was assessed in all participants. Non-inferiority was declared when the lower bound of the 95% CI of the absolute difference in cure rate was above the -10% non-inferiority margin. This trial is registered with EudraCT (2015-004219-19) and ClinicalTrials.gov (NCT02795949) and is complete. FINDINGS: 2030 patients were screened between Oct 5, 2016, and Jan 23, 2020, of whom 171 were randomly assigned to the de-escalation group and 173 to the control group. 164 (50%) patients in the de-escalation group and 167 (50%) in the control group were included in the mITT population. 148 (90%) patients in the de-escalation group and 148 (89%) in the control group had clinical cure (risk difference 1·6 percentage points, 95% CI -5·0 to 8·2). The number of adverse events reported was 219 in the de-escalation group and 175 in the control group, of these, 53 (24%) in the de-escalation group and 56 (32%) in the control group were considered severe. Seven (5%) of 164 patients in the de-escalation group and nine (6%) of 167 patients in the control group died during the 60-day follow-up. There were no treatment-related deaths. INTERPRETATION: De-escalation from an antipseudomonal ß-lactam in Enterobacterales bacteraemia following a predefined rule was non-inferior to continuing the empiric antipseudomonal drug. These results support de-escalation in this setting. FUNDING: Plan Nacional de I+D+i 2013-2016 and Instituto de Salud Carlos III, Subdirección General de Redes y Centros de Investigación Cooperativa, Ministerio de Ciencia, Innovación y Universidades, Spanish Network for Research in Infectious Diseases; Spanish Clinical Research and Clinical Trials Platform, co-financed by the EU; European Development Regional Fund "A way to achieve Europe", Operative Program Intelligence Growth 2014-2020.