Intravenous lipid dose and incidence of bacteremia and fungemia in patients undergoing bone marrow transplantation

Experimental data have implicated intravenous lipids as being immunosuppressive, yet evidence that lipids are associated with an increase in clinically documented infections is sparse. A prospective trial conducted in patients with hematologic malignancies who were undergoing bone marrow transplantation compared the incidence of bacteremia and fungemia during the first month after the transplant. Patients (n = 512) were randomly assigned to receive 6-8% (low dose) or 25-30% (standard dose) of total daily energy as a 20% lipid emulsion. An adaptive randomization scheme stratified for treatments that might influence infection outcome (hematopoietic growth factors, fluconazole, graft-versus-host disease prophylaxis with steroids, pentoxifylline, intravenous immunoglobulin, and total body irradiation). The transplant type (autologous, related family donor, or unrelated donor) did not differ in distribution between treatment groups. Of the evaluable patients (n = 482), 55 patients in the standard-dose lipid group developed bacteremia or fungemia compared with 54 in the low-dose lipid group. The log-rank test comparing the time to first infection found no association between the incidence of bacteremia or fungemia and intravenous lipid (P = 0.95). Similar results were found when analyzed as intent-to-treat (P = 0.98), when bacterial or fungal infections at all sites were included (P = 0.94), and when the observation period was extended to 60 d (P = 0.58 for blood infections, P = 0.77 for infections at all sites). These data indicate that moderate amounts of intravenous lipid rich in linoleic acid are not associated with an increased incidence of bacterial or fungal infections in patients undergoing bone marrow transplantation and receiving total parenteral nutrition. Am J Clin Nutr 1998;67:927-33.