Abstract
Post-infusion hyperalimentation surveillance cultures (HC) are commonly obtained in many hospitals and have been routinely performed in our Special Care Unit (neonatal intensive care). We reviewed the microbiology records for the 1 year period from October 1983 through September 1984, to identify infants who were receiving hyperalimentation, and who had blood cultures (BC) for suspected sepsis. Thirty four patients who had been hyperalimented in that period had 101 BC. The incidence of positive BC was 15%. Two patients each had 2 organisms isolated from one BC. S. epidermidis (6/17) was the most common isolate. The same 34 patients had 1213 HC, 84 (0.7%) of which were positive: 3 for 2 organisms and 1 for 3 organisms. The most common organisms isolated were S. epidermidis 40/89, yeast 10/89 and enterococcus 8/89. The ability of routine HC to predict a subsequent episode of sepsis (positive BC) was studied. Two of 15 episodes of positive BC occurred subsequent to the isolation of the same organism from HC: sensitivity 0.13, specificity 0.94, predictive value of a positive HC 0.03, predictive value of a negative HC 0.98. The ability of concurrent (±24 hrs) HC to predict a positive BC was also studied. Five of 15 positive BC were accompanied by a positive HC: sensitivity 0.33, specificity 0.84, predictive value of a positive HC 0.26, predictive value of a negative HC 0.88. The annual cost of routine HC was $55,529. HC's are not effective predictors of positive BC and their routine use could substantially misdirect antimicrobial therapy. Concurrent HC and BC may be useful in establishing the pathogenesis of an infection.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Krasinski, K., Beagan, E., Holzman, R. et al. 1125 NON-UTILITY OF HYPERALIMENTATION SURVEILLANCE CULTURES. Pediatr Res 19, 298 (1985). https://doi.org/10.1203/00006450-198504000-01155
Issue Date:
DOI: https://doi.org/10.1203/00006450-198504000-01155