Microbial contamination of enteral feeding solutions in a community hospital

Decanting—a source of contamination of enteral feeds?

The techniques of opening and decanting ready-to-use enteral feeds packaged in bottles (crown-cap and screw-cap), cans and tetrapaks were evaluated as potential routes for the contamination of these feeds. It was found that the outsides of the feed containers, bottle openers, scissors and the experimenters' hands all acted as sources of contamination during the transfer of feeds to the nutrient container. The main source of contamination appeared to be the experimenters' hands with counts up to 10(2) cfu ml(-1) being recorded for feeds that had been decanted from screw-cap bottles, cans and tetrapaks by experimenters with either unprotected bare hands or hands experimentally contaminated with K. aerogenes. Levels of contamination and the number of samples contaminated after opening and decanting were consistently higher for cans and tetrapaks than for crown-cap or screw-cap bottles. Disinfection of feed containers followed by the use of sterile gloves and/or disinfected openers yielded bacteria-free feed from all the types of feed container studied.

Comparison of isomolar vs hyperosmolar enteral diets in experimental ileus

The diarrhea observed after infusing hypertonic enteral alimentation solutions may be due to the high osmolality. We compared Vivonex HN (810 mOsm) to Osmolite (300 mOsm) in two canine ileus models. After having bipolar electrodes implanted in the stomach, duodenum, jejunum, and colon, four dogs sequentially underwent operations to produce intestinal obstruction (SBO) or perforation/peritonitis (PER). The SBO was released and the perforation closed 24 hr later. GI myoelectric activity (MEA) was monitored during the first 4 postoperative days and again on the 10th day to determine steady-state MEA. Fasting MEA was recorded for 1 hr, followed by 1-hr recordings after intragastric cannula infusion of either Vivonex HN or Osmolite. There was no significant difference in MEA produced by Osmolite vs Vivonex at any recording site. The 24-hr postop gastric and small bowel MEA was significantly decreased (p less than 0.05) vs postoperative days 2 to 4 and 10 under both fasting and fed conditions. The ileus operations had no effect on colonic MEA. These data show that SBO and PER cause significant decreases in gastrointestinal MEA for 24 hr. There was no difference in MEA response of Osmolite vs Vivonex HN. These results suggest that infusion of hypertonic enteral alimentation solutions does not produce increases in GI myoelectric responsiveness vs isotonic solutions.

Enterobacter sakazakii infections in neonates associated with intrinsic contamination of a powdered infant formula

We report an outbreak of Enterobacter sakazakii infection and colonization in neonates related to an infant formula contaminated during the manufacturing process. The outbreak occurred in a 20-bed neonatal intensive care unit during a six-week period in 1988, and involved a total of four infants. Three infants had sepsis and three had bloody diarrhea; all patients responded to intravenous antibiotics and recovered without complications. The E sakazakii isolated from the formula had the same plasmid and multilocus enzyme profile as those isolated from patients. This outbreak demonstrates the significance of commercially contaminated formulas and emphasizes the need to limit contamination and multiplication of bacteria in enteral formulas.

Contaminated enteral nutrition solutions as a cause of nosocomial bloodstream infection: a study using plasmid fingerprinting

In July 1984, two patients fed enteral nutrition solutions contaminated with Enterobacter cloacae developed nosocomial bacteremia. Despite careful review of the preparation procedures as well as repeated microbiological surveys, 83 (27%) of the 309 formula bottles tested over a 1-yr period were contaminated and the source of contamination remained unknown. E. cloacae was the most frequent organism isolated (34%). The plasmid profiles of E. cloacae recovered from enteral nutrition solutions remained identical for several months. Blood culture isolates from 10 of the 40 patients who had developed E. cloacae nosocomial sepsis over a 7-yr period (1979-1985) had plasmid profiles linking them to contaminated enteral nutrition solutions. Epidemiological data from a case control study revealed that these 10 patients were indeed more likely to be exposed to enteral nutrition than the 30 others: 9/10 vs 10/30 (odds ratio 18, p = 0.002). Similarly, two of seven nosocomial Klebsiella pneumoniae bacteremias over a 6-month period in 1986 could be ascribed to administration of contaminated enteral liquid feeds prompting a general policy for using sterile commercially prepared solutions. Our results suggest that contaminated enteral nutrition solutions represent a significant cause of nosocomial sepsis.

Enteral feeding contamination: comparison of diluents and feeding bag usage

This quasi-experimental study compared bacterial growth in enteral feeding solution in six Kangaroo feeding bags filled with enteral formula diluted with sterile water vs six of the same diluted with tap water. Feeding bags, which were on continuous pumps in patient-occupied rooms, were rinsed and refilled every 8 hours, and formula cultures were obtained at 0, 24, and 48 hr. Acceptability of cultures was based on published recommended standards. Low levels of nonpathogenic staphylococci were found in sterile water bags; high levels of bacilli and Gram negative rods were found in tap water bags. No significant differences were found in the relationship between type of formula diluent and outcomes. Type of water diluent and acceptability were not related. There was a systematic relationship between length of use and outcome; with both diluents, bags used over 24 hr became unacceptably contaminated. Further study with a larger sample is needed to determine the effects of using tap water rather than sterile water as diluent.

Microbial contamination of continuous drip feedings

We evaluated the extent and effects of bacterial contamination of an open continuous enteral feeding system. Eighty-two quantitative enteral feeding cultures and clinical data were obtained during 8 days of observation on each of 33 patients. Cultures of appropriate sites were obtained on febrile patients and compared to the enteral feeding culture. Gram negative bacilli (GNB) in the enteral feeding correlated with abdominal distension in the patients (10 of 12 patients with GNB compared to 5 of 21 without GNB; p less than 0.01). Nine of the 10 patients with GNB and distension were receiving systemic antimicrobics to which the organism was resistant. Contamination of feeding with Serratia marcescens correlated with cultures for the same organism in patients' other body sites (p less than 0.01). The feeding contaminant may have been the source of sepsis in one patient who expired from septic shock. No relationship was demonstrated between contamination and liquid stools or fever. Undiluted, canned feedings were significantly less contaminated at 24 hr (15%) than those requiring mixing of powder (94%) (p less than 0.0001). The canned feedings grew primarily enteric organisms, whereas the powder feedings grew flora typically resident on the skin. Mixing or diluting feedings appears to represent an increased risk of contamination. Growth of GNB may produce adverse effects. Further investigation into methods to limit contamination and growth is warranted.

Contamination of enteral nutrition systems during prolonged intermittent use

Two related studies were done to determine the incidence of bacterial contamination in enteral delivery systems that were used for 15 and 7.5 hr, rinsed after each use, and reused daily in vitro for 7 and 5 days, respectively. In the first study, systems infusing either a premixed formula (Ensure) or a hand-mixed formula (Vivonex) did not show bacterial growth until the 4th day, 1.0-2.0 X 10 colony-forming units per milliliter (CFU/ml) of Staphylococcus epidermidis. Thereafter there was sporadic growth of different organisms but never increasing growth during the 7 days of infusion. In the second study, systems with Ensure were initially contaminated with Staphylococcus aureus and Escherichia coli and reused for 5 days. S. aureus was eliminated by rinsing, but E. coli persisted in the delivery system at concentrations of 10(3)-10(6) CFU/ml. We conclude that clean enteral nutrition systems can be rinsed after short-infusion periods and reused up to 7 days in vitro without significant contamination; however, once a bag has become heavily contaminated some bacteria cannot be eradicated from the system by rinsing.

The effect of handling procedures on microbial contamination of enteral feeds

Three of the most commonly used delivery systems in enteral feeding were evaluated for potential routes of contamination during assembly and delivery of feeds. Assembly of systems wearing sterile gloves gave no contamination in the feeds but all systems were contaminated when assembled either with bare unprotected hands or with hands experimentally contaminated with bacterial cells. Delivery of contamination-free feed was only possible with the use of sterile gloves.

The role of the nutritional support team in preventing and identifying complications of parenteral and enteral nutrition

As the technology of parenteral and enteral nutrition advances, a multidisciplinary nutritional support team (NST) can be used to assure quality care. NST functions include developing and using standard protocols and solutions, monitoring patients, and preventing complications. Interventions by NSTs can save money for the hospital as well. Examples from Truman Medical Center (Kansas City, Missouri) illustrate how the NST can perform these functions.

Microbial growth in clinically used enteral delivery systems

Potential contamination of enteral formulas has led to the development of policies limiting formula hangtimes. However, enteral administration bags can easily become contaminated during formula refilling. We prospectively studied enteral formula contamination when the hangtime of a prefilled 1000 ml pouch was compared with the standard 4-hour hangtime of a refilled enteral administration bag. Samples of formula collected from different locations along the enteral delivery system were cultured during 57 days of enteral hyperalimentation in 19 patients. The overall enteral formula contamination rate was 61%, where the greatest microbial growth occurred in reconstituted enteral formulas. The presence of microbial growth did not differ between canned formulas administered according to a 4-hour hangtime and the prefilled pouch. Greatest growth in all cases was at the distal tubing hub, where contamination during system manipulation or from the patient probably occurred. Use of prefilled enteral administration bags may delay formula contamination in the administration reservoir. A change in equipment design that would decrease the need to manipulate feeding sets or feeding tube connections should be further investigated.

Bacterial contamination of enteral diets

Enteral feeding solutions can be contaminated by bacterial micro-organisms already present in the ingredients, or introduced during preparation or transport, or in the hospital ward. During jejunostomy feeding without pump or filter, ascending bacterial invasion of the feeding bag is possible. In patients with lowered immune response contaminated feedings can cause serious septic clinical problems. The progressive loss of the nutritional value of the enteral feeding solution by bacterial contamination has to be considered for all patients.

Bacterial contamination of enteral nutrient solutions: intestinal colonization and sepsis in mice after ingestion

Contaminated enteral nutrient solution (ENS) was used to assess the risks of intestinal colonization and invasive sepsis using normal mice. Dilute Osmolite containing 10(6) Group B Streptococci/ml given for 2 to 10 consecutive days resulted in the detection of organisms by rectal washing sampling and in the recovery of 10(3) to 10(4) organisms from the cecum, large intestine, and rectum. Only 10(1) organisms survived in the small intestine. Colonization, or the persistence of organisms for 10 days after exposure, was produced after 2 days of ingestion in 44% of animals and in 100% after exposure for 5 days. There were 39 septic deaths. During the 7-day ingestion interval 15 died and 24 died within 10 days after exposure. The risk of sepsis was highest for young mice. Ways to minimize bacterial contamination of ENS intended for administration to patients are discussed.

Ascending contamination of a jejunostomy feeding reservoir

In general bacterial contamination of enteral feeding solutions is considered to be exogenous in origin. However, during feeding through catheter jejunostomy another source of contamination seems possible. We describe a patient in whom bacterial contamination of the feeding solution given through a catheter jejunostomy occurred frequently. Microbiological evaluation of several feeding samples showed that with this particular type of catheter enterostomy contamination was dual in origin. A not aseptic manipulation of the feed prior to administration led only to slight contamination. Most of the contamination however was endogenous--ie. ascending from the gut of the patient. In contrast to some previous studies, no clinical or laboratory deterioration could be observed in relation to the contamination. The food value of the feed expressed as the glucose concentration was significantly reduced after contamination with the causative organisms.

Bacterial contamination of tube-feeding formulas

Bacterial cultures were made on tube-feeding formulas provided to 35 unselected adult patients on termination of the infusion. Bacteria were counted and identified using routine procedures. Formulas were classified as nonmanipulated, manipulated, or locally prepared. Medical records were reviewed to determine if diarrhea was present during the period that included 2 days on either side of the sampling day. A significant association was observed between the extent of bacterial contamination and the presence of diarrhea (p = 0.027). Locally prepared and manipulated formulas contained a significantly greater number of organisms when compared to nonmanipulated formulas (x2 = 17.81, p less than 0.001). Counts for two baseline and four termination cultures on locally prepared formulas exceeded the acceptable limit for coliform organisms in pasteurized milk according to public health standards. Eight additional cultures on commercial feeding formulas which were presumably sterile initially, exceeded these standards at the end of the administration period. Thus 12 of 35 formulas (34%) would not meet public health standards applicable to pasteurized milk. Use of sterile nonmanipulated formulas in a closed administration set is recommended, along with routine, periodic bacteriologic surveillance of enteral feeding programs.