Incidence and characteristics of bacteremia among children in rural Ghana

PLoS One. 2012;7(9):e44063. doi: 10.1371/journal.pone.0044063. Epub 2012 Sep 10.

Abstract

The objective of the study was to describe systemic bacterial infections occurring in acutely ill and hospitalized children in a rural region in Ghana, regarding frequency, incidence, antimicrobial susceptibility patterns and associations with anthropometrical data.Blood cultures were performed in all children below the age of five years, who were admitted to Agogo Presbyterian Hospital (APH), Asante Region, Ghana, between September 2007 and July 2009. Medical history and anthropometrical data were assessed using a standardized questionnaire at admission. Incidences were calculated after considering the coverage population adjusted for village-dependent health-seeking behavior.Among 1,196 hospitalized children, 19.9% (n = 238) were blood culture positive. The four most frequent isolated pathogens were nontyphoidal salmonellae (NTS) (53.3%; n = 129), Staphylococcus aureus (13.2%; n = 32), Streptococcus pneumoniae (9.1%; n = 22) and Salmonella ser. Typhi (7.0%; n = 17). Yearly cumulative incidence of bacteremia was 46.6 cases/1,000 (CI 40.9-52.2). Yearly cumulative incidences per 1,000 of the four most frequent isolates were 25.2 (CI 21.1-29.4) for NTS, 6.3 (CI 4.1-8.4) for S. aureus, 4.3 (CI 2.5-6.1) for S. pneumoniae and 3.3 (CI 1.8-4.9) for Salmonella ser. Typhi. Wasting was positively associated with bacteremia and systemic NTS bloodstream infection. Children older than three months had more often NTS bacteremia than younger children. Ninety-eight percent of NTS and 100% of Salmonella ser. Typhi isolates were susceptible to ciprofloxacin, whereas both tested 100% susceptible to ceftriaxone. Seventy-seven percent of NTS and 65% of Salmonella ser. Typhi isolates were multi-drug resistant (MDR). Systemic bacterial infections in nearly 20% of hospitalized children underline the need for microbiological diagnostics, to guide targeted antimicrobial treatment and prevention of bacteremia. If microbiological diagnostics are lacking, calculated antimicrobial treatment of severely ill children in malaria-endemic areas should be considered.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Anthropometry
  • Anti-Bacterial Agents / pharmacology
  • Anti-Bacterial Agents / therapeutic use
  • Anti-Infective Agents / pharmacology
  • Anti-Infective Agents / therapeutic use
  • Bacteremia / drug therapy
  • Bacteremia / epidemiology*
  • Bacteremia / microbiology
  • Bacteria / drug effects
  • Bacteria / isolation & purification
  • Catchment Area, Health
  • Child, Preschool
  • Female
  • Ghana / epidemiology
  • Hospitals / statistics & numerical data
  • Humans
  • Incidence
  • Infant
  • Male
  • Microbial Sensitivity Tests
  • Patient Discharge / statistics & numerical data
  • Rural Population / statistics & numerical data*
  • Treatment Outcome

Substances

  • Anti-Bacterial Agents
  • Anti-Infective Agents

Grant support

This work was supported by the UBS Optimus Foundation. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.