Résumé
Importance There is uncertainty about whether prolonged infusions of beta-lactam antibiotics improve clinically important outcomes in critically ill adults with sepsis or septic shock. Objective To determine whether prolonged beta-lactam antibiotic infusions are associated with a reduced risk of death in critically ill adults with sepsis or septic shock compared with intermittent infusions. Data Sources The primary search was conducted with MEDLINE (via PubMed), CINAHL, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov from inception to May 2, 2024. Study Selection Randomized clinical trials comparing prolonged (continuous or extended) and intermittent infusions of beta-lactam antibiotics in critically ill adults with sepsis or septic shock. Data Extraction and Synthesis Data extraction and risk of bias were assessed independently by 2 reviewers. Certainty of evidence was evaluated with the Grading of Recommendations Assessment, Development and Evaluation approach. A bayesian framework was used as the primary analysis approach and a frequentist framework as the secondary approach. Main Outcomes and Measures The primary outcome was all-cause 90-day mortality. Secondary outcomes included intensive care unit (ICU) mortality and clinical cure. Results From 18 eligible randomized clinical trials that included 9108 critically ill adults with sepsis or septic shock (median age, 54 years; IQR, 48-57; 5961 men [65%]), 17 trials (9014 participants) contributed data to the primary outcome. The pooled estimated risk ratio for all-cause 90-day mortality for prolonged infusions of beta-lactam antibiotics compared with intermittent infusions was 0.86 (95% credible interval, 0.72-0.98; I2 = 21.5%; high certainty), with a 99.1% posterior probability that prolonged infusions were associated with lower 90-day mortality. Prolonged infusion of beta-lactam antibiotics was associated with a reduced risk of intensive care unit mortality (risk ratio, 0.84; 95% credible interval, 0.70-0.97; high certainty) and an increase in clinical cure (risk ratio, 1.16; 95% credible interval, 1.07-1.31; moderate certainty). Conclusions and Relevance Among adults in the intensive care unit who had sepsis or septic shock, the use of prolonged beta-lactam antibiotic infusions was associated with a reduced risk of 90-day mortality compared with intermittent infusions. The current evidence presents a high degree of certainty for clinicians to consider prolonged infusions as a standard of care in the management of sepsis and septic shock. Trial RegistrationPROSPERO Identifier: CRD42023399434