Early parenteral nutrition comparing to enteral nutrition cannot reduce 28-day mortality in critically ill patients: a retrospective comparative cohort study based on the MIMIC‑IV database

BACKGROUND: Enteral nutrition (EN) is recommended as the first choice by guidelines for critical ill patients. But the timing of safe and effective delivery of parenteral nutrition (PN) is unclear and the results of previous studies are controversial. There is insufficient evidence for the use of early PN, so we designed this cohort study to compared the clinical outcomes of critical ill patients who received early PN with those who did not. METHODS: This retrospective study conducted using the Medical Information Mart for Intensive Care (MIMIC)-IV database. Patients who received nutrition therapy within 3 days of ICU admission were included and we categorized them as patients who received any kind of PN (PN group) or only enteral nutrition (EN group). Confounding factors were adjusted by propensity-score matching (PSM). The primary outcome was the 28-day mortality rate, and secondary outcomes included length of stay (LOS) in the hospital and ICU, hospital infection, and mechanical ventilation time. RESULTS: A total of 5,019 patients (PN group, 357; EN group, 4,662) were included in the analyses. The 28-day mortality rates showed no significant intergroup difference (EN, 22.3% vs. PN, 20.2%; P=0.378). The PN group showed a shorter median ICU LOS (EN, 8.14 vs. PN, 6.89 days, P=0.00955), and a longer median hospital LOS (PN, 21.55 vs. EN, 15.1 days, P<0.001). After PSM, each group included 355 patients, with no significant intergroup difference in the 28-day mortality rate (EN, 18.9% vs. PN, 20.3%; P=0.705). The PN group still showed a longer hospital LOS (median LOS: PN, 21.45 vs. EN, 14.81 days, P<0.001), but the other outcomes showed no differences. CONCLUSIONS: PN within 3 days of ICU admission did not reduce the 28-day mortality rate and could extend hospital LOS. This study supports further fundamental and clinical research to ascertain the effect of PN for ICU patients.