Format

Send to:

Choose Destination
See comment in PubMed Commons below
J Hosp Med. 2012 Nov-Dec;7(9):702-5. doi: 10.1002/jhm.1978. Epub 2012 Sep 28.

Predicting bacteremia based on nurse-assessed food consumption at the time of blood culture.

Author information

  • 1Department of Emergency and Critical Care Medicine, Juntendo University Nerima Hospital, Tokyo, Japan.

Abstract

BACKGROUND:

Bacteremia and its complications are important causes of morbidity and mortality in hospitalized patients. However, the yield of blood cultures is relatively low, with many false-positive results from bacterial contamination.

METHODS:

We investigated the relationship between patient food consumption and the presence of bacteremia. This was an observational analysis of a cohort of 1179 patients who underwent blood culture analysis between January 2005 and December 2009. Patients with anorexia-inducing conditions, such as gastrointestinal illness and malignant disease treated with chemotherapy, were excluded. Food consumption was rated by nurses as the percentage of food consumed during the meal preceding the blood culture. Groupings were as follows: low consumption (<50%), moderate (>50% to <80%), and high (>80%).

RESULTS:

Low consumption was observed in 39.8% of patients, moderate in 17.8%, and high in 41.6%. The average body temperature was 38.1 ± 1.1°C. Bacteremia was present in 18.5%, 3.9%, and 1.4% of patients in the low, moderate, and high food consumption groups, respectively. The negative predictive value was 98.3%, suggesting that bacteremia is very unlikely in the setting of good food intake.

CONCLUSION:

Bacteremia is an unlikely occurrence in hospitalized patients who maintain adequate food consumption at the time of blood culture.

Copyright © 2012 Society of Hospital Medicine.

PMID:
23024032
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Wiley
    Loading ...
    Write to the Help Desk