
Abstract
Background
Extracorporeal life support is influenced by membrane integrity and nutrition. The impact of parenteral nutrition with lipid injectable emulsions on membrane function remains uncertain.
Methods
This retrospective cohort included 151 critically ill adults receiving extracorporeal life support at a single center. Patients were classified based on exposure to parenteral nutrition: those exposed to parenteral nutrition (n = 38, 25%) and those not exposed to parenteral nutrition (n = 113, 75%). Data included demographics, clinical, biochemical, and extracorporeal membrane oxygenation–related variables. Our primary outcome was time to first extracorporeal membrane oxygenation circuit dysfunction; secondary outcomes were extracorporeal life support duration, metabolic changes, and transfusion needs, analyzed with Kaplan-Meier curves and Cox models.
Results
Extracorporeal membrane oxygenation circuit life span was similar between groups (8.5 vs 7 days, P = 0.079). Replacement occurred in 18 patients (11.9%): 7 (18.4%) with parenteral nutrition and 11 (9.7%) without. Multivariate analysis showed a hazard ratio (HR) of 2.35 for membrane replacement in parenteral nutrition patients (CI 0.71–7.69, P = 0.157). The parenteral nutrition group had higher D-dimer (3470 vs 650 ng/ml, P = 0.06), lower fibrinogen (159 vs 308 mg/dl, P = 0.05), longer extracorporeal life support (10 vs 7 days, P = 0.047). Parenteral nutrition was also linked to elevated peak triglycerides (352 vs 261 mg/dl, P < 0.01) and lower HDL (14.5 vs 22 mg/dl, P < 0.01). However, these metabolic changes did not significantly impact membrane replacement (HR 0.99, CI 0.99–1.00, P = 0.817).
Conclusions
Although membrane dysfunction seemed more frequent in the parenteral nutrition group, evidence was insufficient to confirm an association, underscoring the need for further research.