Title: Evaluating the implementation of prolonged beta-lactam infusions for sepsis patients in the ICU
Authors: Nathaniel Park, Jeremy Frens, Alex Chappell, and Dustin Zeigler
Background/Purpose:
Beta-lactams are generally the preferred and most frequently prescribed antibiotics to treat patients with sepsis due to their broad spectrum of activity and favorable safety profile. Their efficacy is dependent on time above minimum inhibitory concentration (MIC) with requirements depending on beta-lactam class. Therefore, patients are at risk of treatment failure if beta-lactam concentrations at the site(s) of infection fall below the MIC for extended periods of time. Recently, data from the BLING-III trial and subsequent meta-analyses have suggested that there may be benefits in terms of clinical cure and reduced 90-day mortality with prolonged beta-lactam infusions in comparison with conventional, intermittent infusions. As a result, Cone Health adopted prolonged or continuous infusion protocols for piperacillin/tazobactam, penicillin G, and nafcillin. More recently, additional prolonged infusion protocols were implemented for cefepime, ceftazidime, and meropenem in critically ill patients with sepsis or septic shock. The aim of this study was to evaluate the efficacy and safety associated with the implementation of a prolonged infusion regimen protocol for cefepime, ceftazidime, and meropenem in ICU patients with sepsis or septic shock.
Methodology:
This Institutional Review Board (IRB) reviewed, determined exempt, retrospective, multicenter, single-health system, pre-post study was conducted from October 2024 to December 2025. Patients were included if they were > 18 years old; ICU admission for sepsis or septic shock; had documented site of infection or strong suspicion for bacterial infection; initiated on meropenem, cefepime, or ceftazidime < 24 hours from sepsis or septic shock diagnosis; and > 1 sign of organ dysfunction [mean arterial pressure (MAP) < 60 mm Hg for > 1 hour, vasopressor required > 4 hours, respiratory support required for > 1 hour, or serum creatinine > 2.49 mg/dL]. Exclusion criteria included renal replacement requirement at the time of antibiotic initiation, received antibiotics < 48 hours, and antibiotics not initiated within 24 hours of sepsis diagnosis. Baseline demographic data, safety data, and data relevant to the primary and secondary endpoints were collected among eligible patients. The primary endpoint was ICU length-of-stay (LOS) defined as days on ICU unit. Secondary endpoints were clinical cure (defined as completion of antibiotics within 14 days without resumption within 48 hours of cessation), vasopressor-free days, and 30-day all-cause mortality. Continuous data was analyzed with Student’s t-test or Mann-Whitney U test, and categorical data was analyzed with chi-square test or Fisher’s exact test. Two-sided alpha was set at 0.05.
Results:
In total, 200 patients were screened for inclusion (100 in pre- and 100 in post-intervention cohort). Ultimately, 38 patients were analyzed – 18 in the pre-intervention cohort, and 20 in the post-intervention cohort. Less than half of the patients were female (n=17, 44.7%) with an average age of 71 years old. Most patients were started on cefepime (n=29, 76.3%), with pulmonary and urinary sources accounting for most of the suspected sepsis sources (n=28, 73.7%). Of the 20 patients in the post-intervention cohort, 14 patients met criteria for prolonged beta-lactam infusion. Among patients meeting criteria (n=14), 1 patient (7.1%) received prolonged beta-lactam infusion. For the primary endpoint, the post-intervention group had a shorter median ICU LOS (2.7 vs 5.0 days, p=0.285). For secondary endpoints, the post-intervention group had less vasopressor-free days (0.7 vs 1.1 days, p=0.055), reduced 30-day mortality (15.0% vs 33.3%), and increased clinical cure (90.0% vs 77.8%). One occurrence of a new C. difficile infection occurred, which was in the pre-intervention cohort, and beta-lactams were appropriately renally adjusted in the 27 patients meeting criteria for renal dosing.
Conclusions:
Although numeric differences were observed in favor of the prolonged beta-lactam infusion protocol cohort with decreased ICU LOS, increased vasopressor-free days, 30-day mortality, and clinical cure, there were no statistical differences among any of the primary or secondary endpoints. Additionally, the observed low implementation rate of prolonged beta-lactam infusion among eligible patients further complicates interpretation. Most likely, the cohort identification methods did not have the sensitivity to identify more patients administered prolonged beta-lactam infusions. The results of this study should not dissuade use of the prolonged beta-lactam infusion protocol.
