Dose of intravenous lipids and rate of bacterial clearance in preterm infants with blood stream infections

Intravenous lipid emulsion (IVLE) is an integral part of the total parenteral nutrition (TPN) regimen in neonates. The use of IVLE during sepsis is the subject of controversy because it may interfere with phagocytosis of microbes by macrophages and may lead to significant hypertriglyceridemia. This paper aims to study the rate of clearance of bacteria in relation to dose of IVLE administered to preterm infants with blood stream infections (BSIs). Preterm infants (mean gestational age +/- SD, 32.0 +/- 2.5 weeks) with culture-proven BSI and receiving TPN were randomized to two groups. The first group (n = 22) was given the usual dose of IVLE according to a standard protocol (starting from 0.5 g kg(-1) day(-1) and gradually increased by 1 g kg(-1) day(-1) to a maximum of 3.5 g kg(-1) day(-1)); in the second group (n = 20), IVLE were restricted at a dose of 1 g kg(-1) day(-1). Samples for blood cultures were withdrawn every 24 h until a negative culture was obtained. CRP was measured daily until its normalization. Serum triglycerides were monitored daily. The rate of bacterial clearance was significantly more rapid in the restricted-dose IVLE group compared to the standard-dose group [72 (48-120) versus 144 (72-168) h, p = 0.001]. Daily weight increment was significantly greater in the standard-dose IVLE group compared to the restricted-dose IVLE group [25 (6.9-31.9) versus 0.9 (-3.3-11.7) g, p = 0.0001]. The duration of antibiotic use was significantly reduced in the restricted-dose IVLE group compared with the standard-dose IVLE group (10.0 +/- 4.5 vs 14.9 +/- 5.1 days; p = 0.003). The durations of TPN, mechanical ventilation, and hospitalization were not significantly different between groups. Restriction of the dose of IVLE to 1 g kg(-1) day(-1) in preterm infants with BSI is associated with earlier negative blood cultures and reduced duration of antibiotic therapy but was associated with a lower daily weight increments.