PEDIATRICS Vol. 121 No. 5 May 2008, pp. e1229-e1239 (doi:10.1542/peds.2007-0871)
ARTICLE |
Trends in Bacteremia in the Pre- and Post-Highly Active Antiretroviral Therapy Era Among HIV-Infected Children in the US Perinatal AIDS Collaborative Transmission Study (1986–2004)
Division of Infectious Diseases, Departments of a Pediatrics
b Medicine, Emory University School of Medicine
c Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, Georgia
d Department of Pediatrics, Harlem Hospital Center, New York, New York
e Department of Pediatrics, University of Maryland, Baltimore, Maryland
f Department of Pediatrics, University of Medicine and Dentistry, Newark, New Jersey
g Division of HIV/AIDS Prevention, National Center for HIV, STD, and TB Prevention, Centers for Disease Control and Prevention, Atlanta, Georgia
OBJECTIVE. HIV-infected children are at high risk for bacteremia. Highly active antiretroviral therapy has reduced rates of opportunistic infections; less is known about its effect on pediatric bacteremia rates. Thus, we sought to determine its impact on bacteremia incidence in HIV-infected children.
METHODS. Children born during 1986–1998 were followed until 2004 in the Perinatal AIDS Collaborative Transmission Study. We determined the pre–and post–highly active antiretroviral therapy (before and after January 1, 1997) incidence of bacteremia among HIV-infected children and characterized the CD4% temporal declines and mortality among patients with and those without incident bacteremias.
RESULTS. Among 364 children, 68 had 118 documented bacteremias, 97 before and 21 after January 1, 1997. Streptococcus pneumoniae constituted 56 (58%) pre–and 13 (62%) post–highly active antiretroviral therapy cases. The incidence rate ratio of bacteremias comparing post–versus pre–highly active antiretroviral therapy was 0.3 overall and 0.2, 0.2, and 0.4 among children aged 0 to 24, 25 to 48, and 49 to 72 months, respectively. Kaplan-Meier analysis for time to first bacteremia in children born during the pre–highly active antiretroviral therapy compared with the post–highly active antiretroviral therapy era revealed that 69% and 94%, respectively, remained bacteremia free at a median follow-up of 6 years. The Cox proportional hazards model also showed a significant reduction of bacteremias in the post–highly active antiretroviral therapy era, even after controlling for gender and race. Among children <6 years of age, those who experienced bacteremia had faster temporal CD4% decline than those who never had bacteremia. Survival analysis revealed that HIV-infected children with bacteremia experienced higher overall mortality when controlling for gender, race, and clinic site.
CONCLUSIONS. A significant decrease in bacteremia incidence and a prolongation in the time to first bacteremia incident were seen in the post–highly active antiretroviral therapy era. Children with a steeper decline of CD4 T cells were more likely to develop bacteremia. Children who experienced bacteremia had an associated higher mortality than their bacteremia-free counterparts.
Key Words: pediatric HIV/AIDS bacteremia incidence HAART
Abbreviations: OI—opportunistic infection IVIG—intravenous immunoglobulin PACTS—Perinatal AIDS Collaborative Transmission Study CDC—Centers for Disease Control and Prevention HOPE—HIV Follow-up of Perinatally Exposed Children TMP-SMX—trimethoprim-sulfamethoxazole CI—confidence interval
Accepted Oct 11, 2007.
![]()
CiteULike
Connotea
Del.icio.us
Digg
Facebook
Reddit
Technorati
Twitter What's this?




