Reply to the comment on “Prolonged versus short-term infusion of meropenem for the treatment of sepsis: a systematic review and meta-analysis”

Dear Editor, In accordance with the TITAN Guidelines 2025[1], we declare that no generative artificial intelligence tools were used at any stage of the research process or during the preparation of this manuscript. We are pleased to note the significant interest and insightful comments[2–4] from readers regarding our article, “Prolonged versus short-term infusion of meropenem for the treatment of sepsis: a systematic review and meta-analysis,” published in the International Journal of Surgery[5]. We sincerely appreciate and value the perspectives raised, which carry important clinical and methodological implications. Our specific responses to the points raised are detailed below. Commentary on dosage heterogeneity and confounding effects We fully agree with the reviewers that heterogeneity in the total daily meropenem dosage across the included studies is a critical methodological concern, as it could confound the assessment of the independent effect of infusion duration. To address this, we conducted an additional subgroup analysis examining the impact of total daily dosage on mortality (Fig. 1). Given that some studies used a loading dose in the prolonged-infusion arms, we categorized studies based on the total daily dose in the intermittent-infusion arm: 3 g/day, 6 g/day, and weight-based dosing. The subgroup analysis showed no heterogeneity (I2 = 0%) and indicated that a total daily dose of 3 g was associated with a statistically significant reduction in mortality. Figure 1.: Forest plot of mortality among patients treated with prolonged versus short-term infusion of meropenem stratified by total daily dose. The points and the bars represent the relative risk (RR) and 95% confidence intervals (CIs). Commentary on the lack of stratified analysis for microbiological eradication We agree entirely that for beta-lactam antibiotics like meropenem, whose efficacy is closely tied to pharmacodynamic/pharmacokinetic parameters (%fT > MIC), a stratified analysis by pathogen type and resistance profile is crucial for clinical guidance. Following the reviewer’s suggestion, we sought pathogen-specific data from the original studies. However, most included studies reported only the overall microbiological eradication rate. Only three studies provided eradication data stratified by specific resistant bacteria[6–8]. Consequently, we performed a subgroup analysis only for pathogen species with sufficient data for meta-analysis: Acinetobacter spp., Enterobacter spp., Escherichia coli, Klebsiella spp., Pseudomonas spp., and Serratia spp. (Supplemental Digital Content eFigure 1, available at: https://links.lww.com/JS9/H445). Unfortunately, likely due to the limited sample sizes, a significant advantage of prolonged over intermittent infusion for microbiological eradication was observed only in the Klebsiella spp. subgroup. Commentary on the unclear classification and attribution of adverse events We appreciate the reviewers’ attention to the safety details. The lack of detailed information on adverse event (AE) types and attribution in the primary literature indeed limited our ability to draw definitive clinical conclusions regarding the safety profile of prolonged infusion. AE reporting was sparse. Among the included studies, only three[6–8] reported AE types in a stratified manner, listing the following: elevated liver enzymes, diarrhea, acute kidney injury following meropenem therapy, rash, vomiting, and seizures. Due to sample size constraints, we could only perform subgroup analyses for elevated liver enzymes and diarrhea, neither of which showed a statistically significant difference between infusion strategies (Supplemental Digital Content eFigure 2, available at: https://links.lww.com/JS9/H445). We strongly encourage thorough safety profiling in future studies. We believe that addressing these important concerns enhances the rigor and clinical relevance of our study’s conclusions. We again thank the readers for their valuable input and contributions to our work.