ABSTRACT
BACKGROUND: This study compared overall bacterial and bloodstream infection rates in patients receiving premixed parenteral nutrition (PN) with vs without lipid emulsion. METHODS: Data from hospitalized patients who were ≥18 years of age and receiving premixed PN between 2005 and 2007 were extracted from the Premier Perspective database. Data were categorized into 2 groups: patients who received premixed PN only and those receiving premixed PN with lipids. Multiple logistic regression was used to adjust for risk factors and potential confounders, reporting the probability of risk for an infection. RESULTS: The group without lipids was observed to have lower rates of both overall bacterial infection (43.5% vs 53.5%) and bloodstream infection (14.5% vs 18.9%). However, after adjusting for baseline characteristics, there were no significant differences in overall risk of bacterial infections (51.4% vs 53.5%; odds ratio [OR] = 1.11; 95% confidence interval [CI], 0.96-1.27) or bloodstream infections (19.6% vs 19.2%; 0.97; 0.81-1.16). In a subset of patients in the intensive care unit for ≥3 days, lower overall bacterial infection rates (58.3% vs 67.3%) and bloodstream infection rates (31.0% vs 37.0%) were observed in the group without lipids. After adjustment, there were no significant differences in risk of overall bacterial infection (OR = 0.95; 95% CI, 0.75-1.22) or bloodstream infection (0.92; 0.71-1.19) between the 2 groups. CONCLUSIONS: When administered with premixed PN, lipid emulsion was not significantly associated with an increase in the risk of infectious morbidity when compared to omitting lipids from therapy.