Skip to content
PaperFren

Antimicrobial resistance

Does antibiotic resistance kill children with blood infections?

Blomberg B, Manji KP, Urassa WK, et al. · BMC infectious diseases · 2007

Open access · cc by · source: Europe PMC

In a Tanzanian children's hospital, bloodstream infections that the standard antibiotics could not treat because of resistance independently predicted death.

Study at a glance

Design
Cohort — Prospective one-year cohort of admitted children with suspected systemic infection; blood culture, susceptibility testing, malaria smear and HIV test at admission; logistic regression for in-hospital death
N
N=1787 · 1787 children (1828 admissions) at Muhimbili National Hospital, Dar es Salaam; regression models used 1527 suspected-infection cases and 216 culture-confirmed bloodstream infections with complete data
Population
Children aged 0-7 years admitted with fever, hypothermia or other signs of systemic infection, August 2001 to August 2002
Outcome
In-hospital death; secondary: length of stay, pathogens and antimicrobial resistance patterns

Structured fields used in claim comparison tables when every cited study has a complete layer.

Key findings

Culture-confirmed bloodstream infection occurred in 13.9% of admissions and tripled the risk of dying; mortality with Gram-negative bloodstream infection (45.6%) was more than twice that with malaria (20.2%). Gram-negative bacteria were often resistant to the usual drugs (only 20% of community-acquired Enterobacteriaceae were ampicillin-sensitive, and 18% of Enterobacteriaceae produced ESBLs), and receiving antibiotics that did not cover the organism independently predicted death, as did HIV infection and malnutrition. Among survivors, inappropriate treatment was linked to a longer hospital stay (median 8 versus 6 days).

Methodology

Over one year, doctors at Tanzania's national hospital enrolled every child up to age 7 admitted with signs of serious infection, took a blood culture, a malaria smear and an HIV test, and recorded treatment and outcome. Bacteria and fungi grown from blood were tested for antibiotic susceptibility, so the team could tell whether the drugs a child actually received matched the organism. Logistic regression then identified which factors (pathogen type, inappropriate treatment, HIV, malnutrition, malaria and others) independently predicted in-hospital death.

Limitations

Most children (67.2%) had already received antibiotics before blood was cultured, and only one small-volume culture was taken without anaerobic culture, so infections with easily killed organisms such as pneumococci were probably missed and the pathogen mix is skewed toward resistant bacteria. Outcome data were missing for about a tenth of admissions and HIV status for about half of children. As an observational single-hospital study, 'inappropriate treatment' may partly reflect sicker children or harder-to-treat organisms rather than resistance alone, and the hospital-acquired label depended on timing that could misclassify cases.

How this study connects

Role on claims

Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.

  • When the empiric drug misses the bug, children die more often.

    In a prospective cohort of 1,787 febrile Tanzanian children, receiving an antibiotic that did not cover the bloodstream organism independently predicted death, alongside HIV and malnutrition; Gram-negative bloodstream infection carried 45.6% mortality, and only 20% of community-acquired Enterobacteriaceae were ampicillin-sensitive.

    Evidence for the claim as stated.

  • Whether resistance itself raises mortality is design-dependent. The Tanzanian cohort found inadequate empiric cover independently predicted death, but the Brazilian case-control study (50% vs 27.5% mortality, p = 0.085, only 20 cases) and the oncology cohort (no factor significantly predicted death) could not show an independent effect, likely because of small samples and severely ill comparison groups.

    Evidence for the claim as stated.

Open questions

Tensions this paper is part of

From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.

History

When this study was placed

Dated entries from the concept change log — when this paper was added or removed as support, challenge, or qualifier on a claim.

  1. 2026-09-24

    Placed as supporting evidence on Antimicrobial Resistance

    In a prospective cohort of 1,787 febrile Tanzanian children, receiving an antibiotic that did not cover the bloodstream organism independently predicted death, alongside HIV and malnutrition; Gram-negative bloodstream infection carried 45.6% mortality, and only 20% of community-acquired Enterobacteriaceae were ampicillin-sensitive.

  2. 2026-09-24

    Placed as supporting evidence on Antimicrobial Resistance

    Whether resistance itself raises mortality is design-dependent. The Tanzanian cohort found inadequate empiric cover independently predicted death, but the Brazilian case-control study (50% vs 27.5% mortality, p = 0.085, only 20 cases) and the oncology cohort (no factor significantly predicted death) could not show an independent effect, likely because of small samples and severely ill comparison groups.

Related papers in this topic

Same topic cluster — not a recommendation engine.