Article Text

  1. J. Schaffner1,
  2. M. Barragán1,
  3. S. Chochua1,2,
  4. Y. S. Wang1,
  5. H. M. Blumberg1,
  6. C. del Rio1,
  7. M. Leonard1
  1. 1Emory University, Atlanta, GA
  2. 2Children's Central Hospital, Tbilisi, Republic of Georgia


Background H. influenzae and S. pneumoniae are leading cause of childhood bloodstream infections (BSIs) in areas where vaccination is not available for these organisms. The prevalence of H. influenzae and S. pneumoniae BSI has not been established in Tbilisi, Republic of Georgia. This study attempts to determine the prevalence and outcomes of BSI in children at Children's Central Hospital (CCH), a referral pediatric hospital.

Methods We carried out a retrospective study of laboratory and medical records from January of 2004 to June of 2006. Patients with positive blood cultures were initially chosen for the study. Community-acquired (CA) BSIs were within the first 48 hours of admission and hospital acquired (HA) were 48 hours or greater.

Results Of 1,693 cultures performed in the study period, 339 (20%) were positive. Twenty-nine of these were omitted from the analysis due to incomplete medical records. Among the 310 children with positive cultures, the following organisms were identified: gram-negative rods (GNRs) (135, 43.6%); coagulase-negative S. aureus (CNS) (111, 35.8%); Klebsiella (17, 5.5%); S. aureus (12, 4.9%); Pseudomonas (10, 3.2%); yeast (9, 2.9%); Streptococcus spp. (5, 1.6%); Enterococcus (4, 1.3%); gram-positive rods (GPRs) (2, 0.7%); Listeria (2, 0.7%); Salmonella spp. (2, 0.7%); and Mixed (S. aureus and Klebsiella; 1, 0.3%). One hundred eighty-four (59%) patients with BSI were male; mean age was 0.4 (0-14) years. Ninety-three (30%) children with a BSI died, 90 of whom were under 1 year of age. CA BSI was present in 87 of 93 patients who died (p = .005). Death was primarily attributed to GNRs (59, 63%), followed by CNS (18, 19%), Pseudomonas (5, 5%), Klebsiella (4, 4%), and Enterococcus and yeast with 2 each (2%), and Listeria, S. aureus, and GPRs were responsible for one death each (1%). Any GNR BSI was significantly associated with mortality versus all none GNR BSI (OR = 3.4, 95% CI = 2.0, 5.7). No positive identifications were made for either H. influenzae or S. pneumoniae.

Conclusions Microbiologic identification of common organisms known to cause BSI in children is difficult in the Republic of Georgia due to limited resources available in the microbiology laboratory in the leading children's hospital. It is likely that causes of BSI due to vaccine-preventable etiologies are underdiagnosed due to laboratory capacity. Improving the infrastructure of diagnostic microbiology laboratories in resource-limited areas is critical to conduct appropriate surveillance and improve patient care.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.