Secondary Logo

Journal Logo

Brief Reports

Epidemiology of Bloodstream Infections in Children with Sickle Cell Disease

Ellison, Angela M. MD, MSc*; Ota, Kaede V. MD, MSc; McGowan, Karin L. PhD, MS; Smith-Whitley, Kim MD

Author Information
The Pediatric Infectious Disease Journal: May 2013 - Volume 32 - Issue 5 - p 560-563
doi: 10.1097/INF.0b013e318286c75b
  • Free


Bloodstream infections (BSIs) are a major cause of morbidity and mortality in children with sickle cell disease (SCD).1–4 Previous studies identified Streptococcus pneumoniae and Haemophilus influenzae as the leading etiologic pathogens responsible for invasive bacterial infections in this high-risk population.1–4Escherichia coli, Salmonella spp and Staphylococcus aureus followed in frequency.2,4 In the years following these studies, the incidence of invasive bacterial infections in the SCD population significantly decreased with the introduction of the penicillin prophylaxis and conjugated H. influenzae type b and pneumococcal vaccines.5–8 Given the reported decline in the incidence of invasive S. pneumoniae and H. influenzae type b infections in the United States,5–8 new data are needed to determine the organisms responsible for BSIs in this population. This information would help to fine tune existing recommendations pertaining to empiric antibiotic administration in patients with SCD presenting with fever and reassess the effectiveness of current prophylactic strategies and to plan for future preventive strategies.

The primary objective of this study was to determine the organisms responsible for BSIs in a large pediatric SCD population in the era following heptavalent conjugate vaccine (PCV7). The secondary objectives were to examine changes in the annual incidence of BSI and to investigate foci of BSIs in this population.


We reviewed the medical records of 815 children, age 2 months to 22 years, who were followed at the Comprehensive Sickle Cell Center at the Children’s Hospital of Philadelphia from January 1, 2000, through December 31, 2010, to identify episodes of BSI. The Sickle Cell Center maintains a list of all patients who have been managed at Children’s Hospital of Philadelphia. Each patient’s chart was reviewed. The results of every blood culture obtained during the study period were reviewed and data pertaining to each BSI episode recorded.

SCD was classified as SS-type (sickle cell anemia), SC-type (hemoglobin-SC disease), Sβ0-thalassemia and Sβ+-thalassemia (sickle cell thalassemia). Penicillin prophylaxis is initiated in our center at 2 months of age and continued until at least 6 years of age regardless of sickle cell genotype. We continue penicillin prophylaxis in children who have had a splenectomy or in those with a prior history of invasive pneumococcal infection. Approximately 50% of our patients without a history of invasive pneumococcal disease are continued on penicillin prophylaxis from 6 years of age until transitioning to adult care, at the discretion of the primary provider. The 23-valent pneumococcal polysaccharide vaccine (PPV23) is routinely administered at 2 years of age with a booster dose administered at 5 years of age. Until recently, the primary care provider administered PCV7, following the schedule recommended by the American Academy of Pediatrics.9 In 2010, PCV7 was replaced by the 13-valent conjugate vaccine (PCV13).

BSIs were detected using the BacT/Alert blood culture system with Pediatric Fastidious Antibiotic Neutralization bottles throughout the study period. Antibiotic susceptibility testing for Staphylococcus spp, Enterobacteriaceae and nonglucose fermenting organisms (eg, Pseudomonas aeruginosa, Stenotrophomonas maltophilia) was performed using the Vitek 2 system (Biomerieux, Durham, NC). Streptococcus spp were tested using E-Test (Biomerieux, Durham, NC). Candida spp was tested using the Yeast One susceptibility microtiter system (TREK Diagnostic Systems, Cleveland, OH). Repeatedly positive blood cultures from the same BSI episode were removed from the dataset to avoid duplication.

Organisms isolated from blood cultures were considered “contaminants” if (1) the organism is commonly associated with contaminated blood cultures and (2) a repeat blood culture without effective antibiotic exposure during the interval between the 2 blood cultures was negative or the patient recovered without receiving effective antibiotic therapy. “Effective” antibiotic therapy was deemed to have been provided if the isolated organism was susceptible to the antibiotics received, as determined by susceptibility testing performed by the microbiology laboratory. If effective antibiotics were given before the drawing of a repeat blood culture and the organism could be categorized as a contaminant, the blood culture was considered “indeterminate.” Positive blood cultures that were deemed “indeterminate” or containing “contaminants” only were excluded from the study.

The annual incidence of true BSI caused by all pathogens was calculated for each year of the study from 2001 through 2010, inclusive. In calculating the incidence, the numerator was defined as the number of patients with a true BSI who had at least 1 visit to the Sickle Cell Center clinic during the year of interest. The denominator was defined as the total number of patients who had at least 1 visit to the Sickle Cell Center clinic during the year of interest. The incidence was reported in units of BSI episodes/1000 clinic patients/year.


Thirty-seven (31%) of the 118 positive blood cultures identified were deemed contaminated and 33 (29%) were considered indeterminate. As a result, 48 positive blood cultures (41%) were analyzed as true BSI episodes. The 48 episodes of BSI occurred in 42 individuals. The mean age at the time of BSI was 8.2 years + 6.4 years. Seventy-six percent had sickle cell anemia (SCD-SS). There were no episodes of BSI among patients with Sβ0-thalassemia.

A total of 52 pathogens were recovered. The most commonly isolated pathogens were S. pneumoniae (23%), coagulase-negative Staphylococcus (15%) and Salmonella spp (11%). S. aureus and E. coli accounted for 9% and 6%, respectively. All S. pneumoniae isolates were susceptible to ceftriaxone and 10 of the 12 were susceptible to penicillin. Antibiotic susceptibility testing results were available for 7 of the 8 coagulase-negative Staphylococcus isolates, 1 of which was susceptible to oxacillin. One of the 5 S. aureus isolates was methicillin resistant. All Staphylococcus isolates were susceptible to vancomycin. The Salmonella spp isolates were all susceptible to ampicillin and ceftriaxone.

Eleven episodes (23%) of BSI occurred in patients with central venous access devices (CVADs), including 4 that were associated with an implantable central venous catheter. None of the excluded cases (“indeterminate” or “contaminant”) had a CVAD. There were an equal number of Gram-negative and Gram-positive organisms in this subgroup. The Gram-positive organisms included coagulase-negative Staphylococcus (4) and S. aureus (3). Acinetobacter baumanni complex, Klebsiella oxytoca, P. aeruginosa and S. maltophilia comprised the Gram-negative group. Mycobacterium fortuitum was isolated from 1 episode of catheter-associated BSI. Candida spp was the only nonbacterial pathogen isolated.

Four episodes of BSI were associated with acute chest syndrome at presentation (3 pneumococcal BSI and 1 Salmonella BSI), 2 with osteomyelitis (both Salmonella BSI) and 1 episode was associated with an epidural abscess (methicillin-resistant S. aureus).

The annual incidence of BSI remained stable during the 10 years of the study (2001 to 2010), (Fig. 1) and ranged from 4.9 to 9.9/1000 clinic patients/year with the exception of 2003.

Incidence of bacteremia in patients with SCD from 2003 to 2010 (95% confidence intervals).


This is the first study to investigate the bacterial pathogens responsible for BSI in a large cohort of children with SCD in the post-PCV7 era. Despite widespread availability of pneumococcal vaccines and penicillin prophylaxis, S. pneumoniae continues to account for largest proportion of bacterial pathogens isolated from BSI in our population. Serotype replacement has been a major concern in our population and others during the past decade.10–12 Nonadherence to penicillin prophylaxis may have been a factor in our pneumococcal cases given that 10 of the 12 isolates were susceptible to penicillin.10 In addition, we expect “breakthrough” cases of pneumococcal bacteremia to continue to occur, even among our most compliant patients, given that penicillin is not 100% effective. We had no cases of H. influenzae bacteremia, suggesting high efficacy of the conjugate H. influenzae type b vaccine.

Coagulase-negative Staphylococcus was recovered in 15% of cases. This group of organisms has not previously been recognized as potential pathogens among SCD patients without CVADs. Further research into the role of coagulase-negative Staphylococcus in BSIs in this population is needed.

Although difficult to make direct comparisons of our incidence data to published pre-PCV rates,1–4 our findings suggest that BSI incidence has dramatically decreased in the post-PCV7 era. Our overall annual incidence is approximately one-tenth of the published age-specific and organism-specific rates during the pre-PCV7 era.1–4 This is likely a result of the high efficacy of conjugate vaccines and penicillin prophylaxis. Although we did not determine the mortality in this study, we do expect that mortality in our population is lower given the overall decreased incidence of BSI.

Although S. pneumoniae continues to be the predominant organism isolated from BSIs in our population, there is a shift occurring favoring line-associated infections caused by a variety of different pathogens. The most common reason for placement of a CVAD in our population is to support chronic transfusion therapy for stroke prevention. Although several studies have reported infection to be a common complication of CVADs in both adult and pediatric SCD populations,13–16 1 recent study found a very low rate of infection associated with port-a-caths.17 Institutions should continue to enforce adherence to the Center for Disease Control guidelines for the prevention of intravascular catheter-related infections.18 Although an empiric combination of vancomycin and an antipseudomonal agent would be most appropriate in our SCD patients with CVAD-related BSIs, as recommended in national guidelines, we recommend that individual centers examine local antimicrobial susceptibility testing data when determining their choice of empiric antimicrobial coverage in this population, regardless of their CVAD status.19

The mean age of BSI in our population was 8.2 years, which is consistent with findings by McCavit et al.20 The increase in average age of BSIs is likely the result of the administration of conjugate pneumococcal and H. influenzae vaccines during infancy. Similar to McCavit, our data supports consideration of extending penicillin prophylaxis past 6 years of age. Clinicians should maintain vigilance in evaluating and empirically starting antibiotic therapy in both younger and older children with SCD who seek care for fever or suspected infection.

Contaminated blood cultures accounted for at least one-third of our positive blood cultures. Contaminated blood cultures are costly to patients and institutions as they can contribute to administration of unnecessary antibiotics and prolong duration of hospitalization.21 When organisms that are commonly associated with blood culture contamination are recovered from blood culture specimens, distinguishing true BSI from contamination can be very challenging. Ideally, obtaining a minimum of 2 blood culture “draws” before antibiotic initiation is advisable as patients whose blood cultures grow contaminants often have only a single positive culture when multiple are obtained.22

There were several limitations to this study. Firstly, positive blood cultures with less virulent pathogens (eg, coagulase-negative Staphylococcus) may have been misclassified as “indeterminate” as opposed to a true pathogen when effective antibiotics were administered before a subsequent negative blood culture. Secondly, it is possible that a small number of cases may have been missed in the rare event that a patient from our program received care for a BSI at another institution or was transferred to Children’s Hospital of Philadelphia having had blood cultures obtained at another institution before transfer. We feel that such events would have been captured through our chart review process. Thirdly, we reported a relatively small number of BSIs (n = 48) despite a comprehensive review of over 800 patients with SCD. We feel that this reflects the high efficacy of both the pneumococcal and conjugate H. influenzae vaccine strategies in place. Finally, because the clinic census data available were limited to total number of unique SCD patients seen per calendar year (as opposed to census data stratified by age, SCD type or presence or absence of a CVAD), we were limited in our ability to analyze BSI incidence by subpopulation.

Progress has been made in reducing the incidence of BSIs caused by virulent organisms in pediatric patients with SCD in the post-PCV7 era. However, we expect that BSIs will continue to occur because of factors such as serotype replacement, suboptimal adherence and/or effectiveness of penicillin prophylaxis and the more recent use of CVAD in the SCD population. Continued use of empiric, broad-spectrum antibiotics in SCD patients with fever or suspected infection is warranted.


1. Barrett-Connor E. Bacterial infection and sickle cell anemia. An analysis of 250 infections in 166 patients and a review of the literature. Medicine (Baltimore). 1971;50:97–112
2. Overturf GD, Powars D, Baraff LJ. Bacterial meningitis and septicemia in sickle cell disease. Am J Dis Child. 1977;131:784–787
3. Powars D, Overturf G, Weiss J, et al. Pneumococcal septicemia in children with sickle cell anemia. Changing trend of survival. JAMA. 1981;245:1839–1842
4. Zarkowsky HS, Gallagher D, Gill FM, et al. Bacteremia in sickle hemoglobinopathies. J Pediatr. 1986;109:579–585
5. Gaston MH, Verter JI, Woods G, et al. Prophylaxis with oral penicillin in children with sickle cell anemia. A randomized trial. N Engl J Med. 1986;314:1593–1599
6. Adamkiewicz TV, Sarnaik S, Buchanan GR, et al. Invasive pneumococcal infections in children with sickle cell disease in the era of penicillin prophylaxis, antibiotic resistance, and 23-valent pneumococcal polysaccharide vaccination. J Pediatr. 2003;143:438–444
7. Adamkiewicz TV, Silk BJ, Howgate J, et al. Effectiveness of the 7-valent pneumococcal conjugate vaccine in children with sickle cell disease in the first decade of life. Pediatrics. 2008;121:562–569
8. Halasa NB, Shankar SM, Talbot TR, et al. Incidence of invasive pneumococcal disease among individuals with sickle cell disease before and after the introduction of the pneumococcal conjugate vaccine. Clin Infect Dis. 2007;44:1428–1433
9. American Academy of Pediatrics. Committee on Infectious Diseases.. Policy statement: recommendations for the prevention of pneumococcal infections, including the use of pneumococcal conjugate vaccine (Prevnar), pneumococcal polysaccharide vaccine and antibiotic prophylaxis. Pediatrics. 2000;106:362–366
10. Ellison AM, Ota KV, McGowan KL, et al. Pneumococcal bacteremia in a vaccinated pediatric sickle cell disease population. Pediatr Infect Dis J. 2012;31:534–536
11. McCavit TL, Quinn CT, Techasaensiri C, et al. Increase in invasive Streptococcus pneumoniae infections in children with sickle cell disease since pneumococcal conjugate vaccine licensure. J Pediatr. 2011;158:505–507
12. Critchley IA, Jacobs MR, Brown SD, et al. Prevalence of serotype 19A Streptococcus pneumoniae among isolates from U.S. children in 2005-2006 and activity of faropenem. Antimicrob Agents Chemother. 2008;52:2639–2643
13. McCready CE, Doughty HA, Pearson TC. Experience with the Port-A-Cath in sickle cell disease. Clin Lab Haematol. 1996;18:79–82
14. Abdul-Rauf A, Gauderer M, Chiarucci K, et al. Long-term central venous access in patients with sickle cell disease. Incidence of thrombotic and infectious complications. J Pediatr Hematol Oncol. 1995;17:342–345
15. Jeng MR, Feusner J, Skibola C, et al. Central venous catheter complications in sickle cell disease. Am J Hematol. 2002;69:103–108
16. Chulamokha L, Scholand SJ, Riggio JM, et al. Bloodstream infections in hospitalized adults with sickle cell disease: a retrospective analysis. Am J Hematol. 2006;81:723–728
17. Bartram JL, O’Driscoll S, Kulasekararaj AG, et al. Portacaths are safe for long-term regular blood transfusion in children with sickle cell anaemia. Arch Dis Child. 2011;96:1082–1084
18. The Hospital Infection Control Practices Advisory Committee, Center for Disease Control and Prevention, U.S.. Guidelines for the prevention of intravascular catheter-related infections. Pediatrics. 2002;110:e51
19. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1–45
20. McCavit TL, Xuan L, Zhang S, et al. Hospitalization for invasive pneumococcal disease in a national sample of children with sickle cell disease before and after PCV7 licensure. Pediatr Blood Cancer. 2012;58:945–949
21. Bates DW, Goldman L, Lee TH. Contaminant blood cultures and resource utilization. The true consequences of false-positive results. JAMA. 1991;265:365–369
22. Baron EJ, Winstein MP, Dunne M Cumitech 1C, Blood Cultures IV. 2005 Washington, DC ASM Press

bloodstream infections; sickle cell disease; pneumococcal heptavalent conjugate vaccine; central venous access device

© 2013 Lippincott Williams & Wilkins, Inc.