We prospectively examined 298 sets (298 aerobic, 299 anaerobic, and 73 resin cultures) of blood cultures from 161 critically ill newborns. The attending physicians were unaware of the study. The mean blood volume per patient (aerobic and anaerobic) was 1.05 (range, 0.11 to 3.04) ml. The mean blood volume per aerobic bottle was 0.53 (range, 0.01 to 1.90) ml. Among aerobic samples 2.7% were .0.1 ml, 16% were s0.3 ml, 33% were s0.4 ml, and 55% were s0.5 ml. For anaerobic cultures the mean blood volume was 0.52 (range, 0.01 to 1.79) ml. Among anaerobic samples 2.7% were s0.1 ml, 15% were s0.3 ml, 35% were .0.4 ml, and 58% were .0.5 ml. Blood volume did not correlate with gestational age, chronologic age, or weight. The mean volume of blood submitted in positive cultures was not significantly greater than that in negative cultures. The blood volume used for culture from ill newborns may be inadequate for detecting sepsis, and the adequacy of currently available culture methods needs to be assessed for the small samples submitted from critically ill newborns.

No abstract

The BACTEC high-blood-volume fungal medium (HBV-FM) (Becton Dickinson Diagnostic Instrument Systems, Sparks, Md.) was compared with the Isolator (IS) tube and the BACTEC Plus 26 (BP26) blood culture bottle for the ability to recover fungi from the blood of adult patients suspected of having fungemia. A total of 6,836 blood culture sets that fulfilled criteria for inclusion in the study were received. Three separate comparisons were performed: 4,907 HBV-FM versus IS, 4,886 BP26 versus HBV-FM, and 4,949 BP26 versus IS. For the HBV-FM versus IS comparison, 218 isolates were recovered: 125 (57.3%) were bacteria and 93 (42.7%) were fungi. HBV-FM was comparable to IS for recovery of yeasts, but IS was superior for recovery of Histoplasma capsulatum (25 versus 0 isolates recovered [P < 0.001]). Growth of Torulopsis glabrata was detected earlier (P < 0.05) in HBV-FM bottles. For the BP26 versus HBV-FM comparison, 229 isolates were recovered: 161 (70.3%) were bacteria, and 68 (29.7%) were fungi. HBV-FM was superior for recovery of T. glabrata (P < 0.025) and all fungi combined (P < 0.025). There were no statistically significant differences in the speed of detection of microbial growth. For the BP26 versus IS comparison, 251 isolates were recovered: 165 (65.7%) were bacteria, and 86 (34.2%) were fungi. IS was superior for recovery of H. capsulatum (P < 0.001), T. glabrata (P < 0.05), and fungi other than H. capsulatum (P < 0.025).

A necrotizing meningoencephalitis complicated by ventricular compartmentalization and abscess formation caused by Enterobacter sakazakii in a previously healthy 5-week-old female is described. A detailed description of the isolate is presented. This communication firmly establishes the pathogenicity of E. sakazakii. We are grateful to M. A. Asbury of the Enteric Section, Centers for Disease Control, for confirming the identification of the isolate, to John J. Farmer III of the Centers for Disease Control and James W. Smith of Indiana University Medical Center for their helpful suggestions, and to Karen Kijovsky and Barbara Kauchak for secretarial assistance.