Frequency of Intravenous Administration Set Changes and Bacteremia: Defining the Risk

Abstract

Despite numerous technological advances in the deliv-ery of intravenous (IV) therapy over the past 3 decades, infection remains a major cause of morbidity and mortality among its recipients. Furthermore, IV-therapy-related infec-tion is the single leading cause of nosocomial bacteremia. Approximately 90% of IV-therapy-related bacteremias are associated with catheter colonization. Most infections of this type are caused by gram-positive cocci, in particular staphylococci. Appropriately, most recent clinical investiga-tions aimed at reducing the risk of IV-related infection are centered on reducing colonization of the catheter or sur-rounding skin by these organisms. 1 Bacteremia caused by the infusion of contaminated fluids is less common, but nev-ertheless is a potentially serious problem that can affect single patients or dozens of patients, depending on the cir-cumstances of contamination. Unlike catheter-related bacteremia, bacteremia due to contaminated infusate is often due to gram-negative bacilli such as Klebsiella, Enterobacter, and Serratia species, which, unlike staphylo-cocci, can multiply in glucose-containing IV solutions. Staphylococci can, however, multiply in blood products and lipid emulsions. Contamination of the infusate can occur through sev-eral routes. Intrinsic contamination occurs at the time of manufacture of the fluid. This type of contamination fortu-nately is rare, but has the potential to cause a large number of episodes of infection. In what is perhaps the best studied outbreak due to intrinsic contamination, 378 cases of bac-teremia due to Enterobacter aerogenes or Enterobacter agglomerans were documented in 25 different US hospi-tals. 2 The cases were traced to contamination of the cap used in IV-fluid bottles; bacteria were readily transferred from the cap to the fluid during manipulation of the bottle. The low numbers of bacteria present were able to multiply within the fluid; there was an association between the occurrence of bacteremia and duration of time that the bot-tle and administration set were in use. The clinical features of bacteremia in this outbreak were notable in that patients often had no conditions predisposing to gram-negative bac-teremia; the mortality rate was 13.4%. The outbreak was terminated by discontinuing use of the implicated product. More common, and perhaps more insidious, is extrinsic contamination, which occurs during subsequent admixture of IV fluids or during manipulation of the infu-sion apparatus. Breaks in sterile technique can introduce organisms into the infusate, where they may multiply to high concentrations. Repeated manipulation of infusion sets is a potential means of introduction of bacteria into IV fluids. 3 Prolonged "hang time" of the fluid or components of the administration set may permit the multiplication of these organisms to relatively high concentration. Infusion of such fluids can lead to the abrupt onset of fever and hypotension in a previously stable patient. 4 Because single patients or a very limited number of patients may be affect-ed, such contamination may be difficult to recognize. The Centers for Disease Control and Prevention has developed a comprehensive set of guidelines for the preven-tion of intravascular-device-related infections. 5 With regard to replacement of administration sets, the current recom-mendation is that replacement should occur "... no more fre-quently than at 72-hour intervals "An exception is made for administration of blood or lipid emulsion, which requires change of the administration set at 24 hours. The current recommendations evolved over a 25-year period, beginning with an initial recommendation of change at 24 hours that was based on observations made during the nationwide epidemic of Enterobacter septicemia. Subsequent studies showed that changes at 48 hours