Response to Letter to the Editor: Association between mixed fatty acid emulsion and parenteral nutrition‐associated cholestasis in extremely low‐birth‐weight infants: A retrospective cohort study

We appreciate the authors’ careful reading of our work and their thoughtful comments, which provide an opportunity to further clarify our study design and findings. In this letter, we respond to each of the points raised. We agree that temporal changes in clinical practice may introduce confounding factors. In our study, aside from the introduction of donated breast milk in 2021, the nutrition protocol remained consistent throughout the study period, as described in the Methods section. Human milk is indeed the optimal nutrition for preterm infants and has been shown to reduce the risk of complications such as necrotizing enterocolitis (NEC).1 Our center introduced donor milk in 2021 for cases in which maternal milk was unavailable or insufficient. However, donor milk constituted a relatively small proportion of total enteral intake in our cohort, and its availability varied. We acknowledge that changes in feeding composition could potentially influence outcomes, including parenteral nutrition–associated cholestasis (PNAC), although the exact role of donor milk vs formula in PNAC pathogenesis remains unclear and warrants further study. Importantly, all infants received the same parenteral nutrition initiation strategy, lipid dosing escalation, and enteral feeding advancement protocol. Although sepsis management and other clinical practices may have evolved over time, we adjusted for sepsis and other known confounders (eg, NEC, duration of parenteral nutrition) in our multivariable analysis. This approach helps mitigate the potential influence of temporal trends on our primary outcome. We thank the authors for this suggestion. In our study, we reported both average daily lipid intake and actual total lipid exposure and found no significant differences between the two groups (Table 2). Although cumulative lipid exposure is indeed relevant, our data showed similar durations of lipid administration and comparable lipid intakes between groups. Therefore, differential lipid exposure is unlikely to explain the lack of significant difference in PNAC incidence. Future prospective studies with detailed longitudinal lipid dosing data may further elucidate dose-response relationships. We acknowledge that our subgroup analyses were limited by small sample sizes, particularly for infants with NEC stage ≥IIa (n = 13). This was clearly stated as a limitation in the Discussion. Although these subgroups are clinically important, our study was not powered to detect differences within them. We agree that larger, multicenter studies are needed to evaluate the effect of lipid emulsions in high-risk subpopulations. We observed significantly lower peak gamma-glutamyl transferase (gamma-GT) levels in the group receiving soybean oil, medium-chain triglycerides (MCTs), olive oil, and fish oil (SMOF), as noted in Table 4. Although gamma-GT is a marker of cholestasis and oxidative stress, our primary outcome was based on direct bilirubin, which remains the most widely accepted diagnostic criterion for PNAC in neonates.2 We agree that the lower gamma-GT may suggest a milder hepatic insult in the SMOF group, and this finding merits further investigation in future studies focused on biomarkers of hepatic injury beyond bilirubin. We thank the authors for their constructive comments, which highlight important considerations for interpreting our study and designing future research. Our findings suggest that, in our cohort of infants with extremely low birth weight, SMOF lipid emulsion did not significantly reduce the incidence of PNAC compared with MCT/long-chain triglyceride emulsion. We agree that further large-scale, randomized trials are needed to clarify the role of mixed lipid emulsions in preventing PNAC, especially in high-risk subgroups. None declared.