Optimal Number of Blood Cultures and Volume of Blood Needed to Detect Bacteremia in Children

Kim, Jong Jae; Kim, Cheol Am; Baik, Sang Ho; Oh, Eui Tak; Kang, Hong Ja; Kim, Kil Seo

Abstract

PURPOSE: We compared pathogen recovery rates by obtaining two blood cultures instead of one blood culture containing 1ml and collecting a larger volume, 1 to 3ml.
METHODS
Total of 750 blood specimens from 250 patients with fever, a temperature higher than 39degrees C and suspected bacteremia were obtained. Each patient had two samples of blood, A (1ml) and B (4ml), obtained at 30-minute interval from separate sites of extremities and B was divided into B1 (1ml) and B2 (3ml). Each sample was inoculated into aerobic culture media. Patients were excluded if two samples of blood were not obtained or if the isolate represented a contaminant.
RESULTS
A pathogen was isolated in 19 (7.6%) of 250 patients and 37 (4.9%) of 750 specimens. In 7 patients, the pathogen was isolated with all the culture methods and in 12 patients, one or more of the cultures yielded no growth. The pathogen recovery rates were 53% (10/19) in A and B1, 89% (17/19) in B2 and 68% (13/19) in A+B1. No difference was detected between A or B1 and A+B1 (P>0.05) and the pathogen recovery rate for B2 was significantly greater than that for A or B1 (P<0.05), but no significant differences were found in pathogen recovery when B2 was compared with A+B1.
CONCLUSION
Increasing volume of blood from 1 to 3ml inoculated into blood culture bottles improves detection of bacteremia in pediatric patients and spares patients the cost and pain of an additional venipuncture.