Bacteremia complicating gram-negative urinary tract infections: A population-based study

Majdi N. Al-Hasan, Jeanette E Eckel-Passow, Larry M. Baddour

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Background: Urinary tract infection (UTI) is common and bacteremia complicating this infection is frequently seen. There has been limited data published that characterize bacteremic UTI in a population-based setting over an extended period. We therefore examined the incidence rate, microbiology, outcome, and in vitro antimicrobial resistance trends of bacteremic UTI due to gram-negative bacilli in Olmsted County, Minnesota, from 1/1/1998 to 12/31/2007. Methods: We used Kaplan-Meier method to estimate mortality rates, Cox proportional hazard regression to determine risk factors for mortality, and logistic regression to examine temporal changes in antimicrobial resistance rates. Results: We identified 542 episodes of bacteremic gram-negative UTI among Olmsted County residents during the study period. The median age of patients was 71 years and 65.1% were females. The age-adjusted incidence rate per 100,000 person-years was 55.3 (95% confidence interval [CI]: 49.5-61.2) in females and 44.6 (95% CI: 38.1-51.1) in males. Escherichia coli was the most common pathogen (74.9%). The 28-day and 1-year all-cause mortality rates were 4.9% (95% CI: 3.0-6.8) and 15.6% (95% CI: 12.4-18.8), respectively. Older age was associated with higher mortality; community-acquired infection acquisition and E. coli UTI were both independently associated with lower mortality. During the study period, resistance rates increased linearly from 10% to 24% for trimethoprim-sulfamethoxazole and from 1% to 8% for ciprofloxacin. Conclusions: To our knowledge, this is the first population-based study of bacteremic gram-negative UTI. The linear trend of increasing antimicrobial resistance among gram-negative isolates should be considered when empiric therapy is selected.

Original languageEnglish (US)
Pages (from-to)278-285
Number of pages8
JournalJournal of Infection
Volume60
Issue number4
DOIs
StatePublished - Apr 2010

Fingerprint

Bacteremia
Urinary Tract Infections
Mortality
Confidence Intervals
Population
Escherichia coli
Community-Acquired Infections
Incidence
Sulfamethoxazole Drug Combination Trimethoprim
Ciprofloxacin
Microbiology
Bacillus
Logistic Models
Infection

Keywords

  • Antimicrobial resistance
  • Bloodstream infection
  • Epidemiology
  • Fluoroquinolones
  • Gram-negative
  • Incidence
  • Mortality
  • Urinary tract infection

ASJC Scopus subject areas

  • Microbiology (medical)
  • Infectious Diseases

Cite this

Bacteremia complicating gram-negative urinary tract infections : A population-based study. / Al-Hasan, Majdi N.; Eckel-Passow, Jeanette E; Baddour, Larry M.

In: Journal of Infection, Vol. 60, No. 4, 04.2010, p. 278-285.

Research output: Contribution to journalArticle

@article{a0bd80e5bd3b47d3ae9303b181357de8,
title = "Bacteremia complicating gram-negative urinary tract infections: A population-based study",
abstract = "Background: Urinary tract infection (UTI) is common and bacteremia complicating this infection is frequently seen. There has been limited data published that characterize bacteremic UTI in a population-based setting over an extended period. We therefore examined the incidence rate, microbiology, outcome, and in vitro antimicrobial resistance trends of bacteremic UTI due to gram-negative bacilli in Olmsted County, Minnesota, from 1/1/1998 to 12/31/2007. Methods: We used Kaplan-Meier method to estimate mortality rates, Cox proportional hazard regression to determine risk factors for mortality, and logistic regression to examine temporal changes in antimicrobial resistance rates. Results: We identified 542 episodes of bacteremic gram-negative UTI among Olmsted County residents during the study period. The median age of patients was 71 years and 65.1{\%} were females. The age-adjusted incidence rate per 100,000 person-years was 55.3 (95{\%} confidence interval [CI]: 49.5-61.2) in females and 44.6 (95{\%} CI: 38.1-51.1) in males. Escherichia coli was the most common pathogen (74.9{\%}). The 28-day and 1-year all-cause mortality rates were 4.9{\%} (95{\%} CI: 3.0-6.8) and 15.6{\%} (95{\%} CI: 12.4-18.8), respectively. Older age was associated with higher mortality; community-acquired infection acquisition and E. coli UTI were both independently associated with lower mortality. During the study period, resistance rates increased linearly from 10{\%} to 24{\%} for trimethoprim-sulfamethoxazole and from 1{\%} to 8{\%} for ciprofloxacin. Conclusions: To our knowledge, this is the first population-based study of bacteremic gram-negative UTI. The linear trend of increasing antimicrobial resistance among gram-negative isolates should be considered when empiric therapy is selected.",
keywords = "Antimicrobial resistance, Bloodstream infection, Epidemiology, Fluoroquinolones, Gram-negative, Incidence, Mortality, Urinary tract infection",
author = "Al-Hasan, {Majdi N.} and Eckel-Passow, {Jeanette E} and Baddour, {Larry M.}",
year = "2010",
month = "4",
doi = "10.1016/j.jinf.2010.01.007",
language = "English (US)",
volume = "60",
pages = "278--285",
journal = "Journal of Infection",
issn = "0163-4453",
publisher = "W.B. Saunders Ltd",
number = "4",

}

TY - JOUR

T1 - Bacteremia complicating gram-negative urinary tract infections

T2 - A population-based study

AU - Al-Hasan, Majdi N.

AU - Eckel-Passow, Jeanette E

AU - Baddour, Larry M.

PY - 2010/4

Y1 - 2010/4

N2 - Background: Urinary tract infection (UTI) is common and bacteremia complicating this infection is frequently seen. There has been limited data published that characterize bacteremic UTI in a population-based setting over an extended period. We therefore examined the incidence rate, microbiology, outcome, and in vitro antimicrobial resistance trends of bacteremic UTI due to gram-negative bacilli in Olmsted County, Minnesota, from 1/1/1998 to 12/31/2007. Methods: We used Kaplan-Meier method to estimate mortality rates, Cox proportional hazard regression to determine risk factors for mortality, and logistic regression to examine temporal changes in antimicrobial resistance rates. Results: We identified 542 episodes of bacteremic gram-negative UTI among Olmsted County residents during the study period. The median age of patients was 71 years and 65.1% were females. The age-adjusted incidence rate per 100,000 person-years was 55.3 (95% confidence interval [CI]: 49.5-61.2) in females and 44.6 (95% CI: 38.1-51.1) in males. Escherichia coli was the most common pathogen (74.9%). The 28-day and 1-year all-cause mortality rates were 4.9% (95% CI: 3.0-6.8) and 15.6% (95% CI: 12.4-18.8), respectively. Older age was associated with higher mortality; community-acquired infection acquisition and E. coli UTI were both independently associated with lower mortality. During the study period, resistance rates increased linearly from 10% to 24% for trimethoprim-sulfamethoxazole and from 1% to 8% for ciprofloxacin. Conclusions: To our knowledge, this is the first population-based study of bacteremic gram-negative UTI. The linear trend of increasing antimicrobial resistance among gram-negative isolates should be considered when empiric therapy is selected.

AB - Background: Urinary tract infection (UTI) is common and bacteremia complicating this infection is frequently seen. There has been limited data published that characterize bacteremic UTI in a population-based setting over an extended period. We therefore examined the incidence rate, microbiology, outcome, and in vitro antimicrobial resistance trends of bacteremic UTI due to gram-negative bacilli in Olmsted County, Minnesota, from 1/1/1998 to 12/31/2007. Methods: We used Kaplan-Meier method to estimate mortality rates, Cox proportional hazard regression to determine risk factors for mortality, and logistic regression to examine temporal changes in antimicrobial resistance rates. Results: We identified 542 episodes of bacteremic gram-negative UTI among Olmsted County residents during the study period. The median age of patients was 71 years and 65.1% were females. The age-adjusted incidence rate per 100,000 person-years was 55.3 (95% confidence interval [CI]: 49.5-61.2) in females and 44.6 (95% CI: 38.1-51.1) in males. Escherichia coli was the most common pathogen (74.9%). The 28-day and 1-year all-cause mortality rates were 4.9% (95% CI: 3.0-6.8) and 15.6% (95% CI: 12.4-18.8), respectively. Older age was associated with higher mortality; community-acquired infection acquisition and E. coli UTI were both independently associated with lower mortality. During the study period, resistance rates increased linearly from 10% to 24% for trimethoprim-sulfamethoxazole and from 1% to 8% for ciprofloxacin. Conclusions: To our knowledge, this is the first population-based study of bacteremic gram-negative UTI. The linear trend of increasing antimicrobial resistance among gram-negative isolates should be considered when empiric therapy is selected.

KW - Antimicrobial resistance

KW - Bloodstream infection

KW - Epidemiology

KW - Fluoroquinolones

KW - Gram-negative

KW - Incidence

KW - Mortality

KW - Urinary tract infection

UR - http://www.scopus.com/inward/record.url?scp=77952569812&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952569812&partnerID=8YFLogxK

U2 - 10.1016/j.jinf.2010.01.007

DO - 10.1016/j.jinf.2010.01.007

M3 - Article

C2 - 20114061

AN - SCOPUS:77952569812

VL - 60

SP - 278

EP - 285

JO - Journal of Infection

JF - Journal of Infection

SN - 0163-4453

IS - 4

ER -