Objective: Anaerobic bacteremia rarely occurs in children, Therefore w
e assessed the usefulness of routinely obtaining anaerobic blood cultu
res in our pediatric patients. Study design: Records of 9360 paired ae
robic and anaerobic blood culture bottles (Bactec NR660 System) contai
ning blood specimens from pediatric inpatients and outpatients at Duke
University Medical Center, Durham, N.C., were reviewed retrospectivel
y. Yield and speed of detection were calculated for each bottle and co
mpared for statistical significance by the McNemar test. Results: A to
tal of 723 clinically important microorganisms were isolated; only 15
(2.1%) were strict anaerobes. Significantly more microorganisms (p <0.
001), especially staphylococci, nonfermenting gram-negative rods, ente
ric gramnegative rods, and yeasts, were detected by use of the aerobic
bottle, The anaerobic bottle was important in identifying an anaerobi
c microorganism as the cause of sepsis in only five patients, all of w
hom were at increased risk of having anaerobic infection. Conclusions:
Anaerobic blood cultures are rarely helpful in the majority of pediat
ric patients and usually show positive results only in clinical settin
gs associated with anaerobic infection. Microorganisms that prefer an
aerobic environment, such as Pseudomonas aeruginosa and yeasts, are no
w far more common than anaerobes in children; aerobic culturing of the
entire volume of blood collected might increase the yield from pediat
ric blood cultures.