Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Bacteremic sepsis leads to higher mortality when adjusting for confounders with propensity score matching

Mellhammar, Lisa LU ; Kahn, Fredrik LU ; Whitlow, Caroline ; Kander, Thomas LU orcid ; Christensson, Bertil LU and Linder, Adam LU (2021) In Scientific Reports 11.
Abstract

One can falsely assume that it is well known that bacteremia is associated with higher mortality in sepsis. Only a handful of studies specifically focus on the comparison of culture-negative and culture-positive sepsis with different conclusions depending on study design. The aim of this study was to describe outcome for critically ill patients with either culture-positive or -negative sepsis in a clinical review. We also aimed to identify subphenotypes of sepsis with culture status included as candidate clinical variables. Out of 784 patients treated in intensive care with a sepsis diagnosis, blood cultures were missing in 140 excluded patients and 95 excluded patients did not fulfill a sepsis diagnosis. Of 549 included patients, 295... (More)

One can falsely assume that it is well known that bacteremia is associated with higher mortality in sepsis. Only a handful of studies specifically focus on the comparison of culture-negative and culture-positive sepsis with different conclusions depending on study design. The aim of this study was to describe outcome for critically ill patients with either culture-positive or -negative sepsis in a clinical review. We also aimed to identify subphenotypes of sepsis with culture status included as candidate clinical variables. Out of 784 patients treated in intensive care with a sepsis diagnosis, blood cultures were missing in 140 excluded patients and 95 excluded patients did not fulfill a sepsis diagnosis. Of 549 included patients, 295 (54%) had bacteremia, 90 (16%) were non-bacteremic but with relevant pathogens detected and in 164 (30%) no relevant pathogen was detected. After adjusting for confounders, 90-day mortality was higher in bacteremic patients, 47%, than in non-bacteremic patients, 36%, p = 0.04. We identified 8 subphenotypes, with different mortality rates, where pathogen detection in microbial samples were important for subphenotype distinction and outcome. In conclusion, bacteremic patients had higher mortality than their non-bacteremic counter-parts and bacteremia is more common in sepsis when studied in a clinical review. For reducing population heterogeneity and improve the outcome of trials and treatment for sepsis, distinction of subphenotypes might be useful and pathogen detection an important factor.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Scientific Reports
volume
11
article number
6972
publisher
Nature Publishing Group
external identifiers
  • pmid:33772090
  • scopus:85103600798
ISSN
2045-2322
DOI
10.1038/s41598-021-86346-4
language
English
LU publication?
yes
id
f70177cf-c95b-4312-a5de-236f9e899bfa
date added to LUP
2021-04-01 16:16:16
date last changed
2024-12-15 05:05:13
@article{f70177cf-c95b-4312-a5de-236f9e899bfa,
  abstract     = {{<p>One can falsely assume that it is well known that bacteremia is associated with higher mortality in sepsis. Only a handful of studies specifically focus on the comparison of culture-negative and culture-positive sepsis with different conclusions depending on study design. The aim of this study was to describe outcome for critically ill patients with either culture-positive or -negative sepsis in a clinical review. We also aimed to identify subphenotypes of sepsis with culture status included as candidate clinical variables. Out of 784 patients treated in intensive care with a sepsis diagnosis, blood cultures were missing in 140 excluded patients and 95 excluded patients did not fulfill a sepsis diagnosis. Of 549 included patients, 295 (54%) had bacteremia, 90 (16%) were non-bacteremic but with relevant pathogens detected and in 164 (30%) no relevant pathogen was detected. After adjusting for confounders, 90-day mortality was higher in bacteremic patients, 47%, than in non-bacteremic patients, 36%, p = 0.04. We identified 8 subphenotypes, with different mortality rates, where pathogen detection in microbial samples were important for subphenotype distinction and outcome. In conclusion, bacteremic patients had higher mortality than their non-bacteremic counter-parts and bacteremia is more common in sepsis when studied in a clinical review. For reducing population heterogeneity and improve the outcome of trials and treatment for sepsis, distinction of subphenotypes might be useful and pathogen detection an important factor.</p>}},
  author       = {{Mellhammar, Lisa and Kahn, Fredrik and Whitlow, Caroline and Kander, Thomas and Christensson, Bertil and Linder, Adam}},
  issn         = {{2045-2322}},
  language     = {{eng}},
  month        = {{03}},
  publisher    = {{Nature Publishing Group}},
  series       = {{Scientific Reports}},
  title        = {{Bacteremic sepsis leads to higher mortality when adjusting for confounders with propensity score matching}},
  url          = {{http://dx.doi.org/10.1038/s41598-021-86346-4}},
  doi          = {{10.1038/s41598-021-86346-4}},
  volume       = {{11}},
  year         = {{2021}},
}