β-lactam infusion strategies in critical illness: is continuous superior to extended infusion? A systematic review and network meta-analysis

Background Prolonged β-lactam infusion is common in critical care, but whether continuous infusion (CI) improves outcomes more than extended infusion (EI) is unclear. Methods We conducted a systematic review and random-effects Bayesian network meta-analysis of randomized controlled trials in critically ill adults comparing intravenous β-lactam administration by continuous infusion (24-hour infusion), extended infusion (≥3 h per dose), and intermittent infusion (≤1 h per dose). We searched PubMed, EMBASE, Web of Science, and the Cochrane Library from January 1, 2000, to July 25, 2025. Two reviewers independently screened records, extracted data, and assessed risk of bias. Relative risks (RRs) and mean differences (MDs) were synthesized using random-effects models. Network meta-analysis used Markov chain Monte Carlo methods in R, with complementary pairwise analyses in Stata. Certainty of evidence was assessed using GRADE for network meta-analysis. Results Twenty-five randomized trials ( N  = 9,915) were included; no direct CI vs. EI trials were identified. Versus II, CI reduced mortality (RR 0.85, 95% CI 0.66–0.98; moderate certainty); EI had uncertain mortality effects (RR 0.80, 95% CI 0.60–1.10; low certainty). Clinical cure improved versus II for CI (RR 1.10, 95% CI 1.03–1.20) and EI (RR 1.28, 95% CI 1.07–1.55). ICU length of stay was uncertain (CI vs. II: MD −0.47 days, 95% CI −2.10–1.28; EI vs. II: MD −1.32 days, 95% CI −4.65–2.52). Indirect CI vs. EI comparisons showed no significant differences in mortality (RR 1.06, 95% CI 0.73–1.44), clinical cure (RR 0.86, 95% CI 0.70–1.05), or ICU length of stay (MD 0.86 days, 95% CI −3.32–4.54). Bayesian posterior probability analysis suggested a 93.2% probability that EI is superior to CI for clinical cure. Conclusions CI did not demonstrate greater clinical benefit than EI. CI can reduce patient mortality, and both CI and EI appear to improve clinical cure rates compared with II. CI conferred no additional benefit over EI, suggesting that EI may be sufficient as a prolonged-infusion strategy in clinical practice. However, this finding is based on indirect evidence. The apparent advantage of EI should therefore be regarded as hypothesis-generating and requires confirmation in adequately powered head-to-head randomized controlled trials. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD42023452814 , PROSPERO CRD42023452814.