Host Response Stratification in Malarial and Non-malarial Sepsis: A Prospective, Multicenter Analysis From Uganda

  • Matthew J. Cummings
  • , Julius J. Lutwama
  • , Nicholas Owor
  • , Alin S. Tomoiaga
  • , Jesse E. Ross
  • , Moses Muwanga
  • , Christopher Nsereko
  • , Irene Nayiga
  • , Stephen Kyebambe
  • , Joseph Shinyale
  • , Thomas Ochar
  • , Kai Nie
  • , Hui Xie
  • , Sam Miake-Lye
  • , Bryan Villagomez
  • , Jingjing Qi
  • , Steven J. Reynolds
  • , Martina Cathy Nakibuuka
  • , Xuan Lu
  • , John Kayiwa
  • Mercy Haumba, Joweria Nakaseegu, Xiaoyu Che, Pauline Byakika-Kibwika, Misaki Wayengera, Jane Achan, Seunghee Kim-Schulze, W. Ian Lipkin, Max R. O'donnell, Barnabas Bakamutumaho

Research output: Contribution to journalArticlepeer-review

Abstract

OBJECTIVES: Globally, the burden of sepsis is highest in malaria endemic areas of sub-Saharan Africa. The influence of malaria on biological heterogeneity inherent to sepsis in this setting is poorly understood. We sought to determine shared and distinct features of the host response in malarial and non-malarial sepsis in sub-Saharan Africa. Design and Setting: Analysis of Olink proteomic data from prospective observational cohort studies of sepsis conducted at public hospitals in Uganda (discovery cohort [Entebbe, urban], n = 238; validation cohort [Tororo, rural], n = 253). PATIENTS: Adults (age ≥ 18 yr) hospitalized with sepsis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The frequency of malaria-associated (malarial) sepsis was 20% in the discovery cohort and 28% in the validation cohort. In both cohorts, a shared host response was predominant, with less than or equal to 8% of proteins differentially expressed (Benjamini-Hochberg-adjusted p ≤ 0.05) between malarial and non-malarial sepsis, after adjustment for demographic variables and HIV and tuberculosis coinfection. In both cohorts, malarial sepsis was associated with increased expression of immunosuppressive proteins (interleukin-10, leukocyte immunoglobulin-like receptor B1, killer cell immunoglobulin-like receptor 3DL1), including those associated with Tcell exhaustion and apoptosis (lymphocyte activation gene 3, T cell immunoglobulin and mucin domain containing 4). A classifier model including these immunosuppressive proteins showed reasonable discrimination (area under the receiver operating characteristic curves, 0.73 [95% CI, 0.65-0.81] and 0.72 [0.65-0.79]) and calibration (Brier scores 0.14 and 0.18) for stratification of malarial sepsis in the discovery and validation cohorts, respectively. CONCLUSIONS: Host responses are largely conserved in malarial and non-malarial sepsis but may be distinguished by a signature of relative immunosuppression in the former. Further investigations are needed to differentiate mechanisms of malarial and non-malarial sepsis, with the goal of informing pathogen-stratified and pathogen-agnostic treatment strategies.

Original languageEnglish
Pages (from-to)e984-e991
JournalCritical Care Medicine
Volume53
Issue number4
DOIs
StatePublished - 1 Apr 2025

Keywords

  • Africa
  • immunology
  • malaria
  • proteomics
  • sepsis

Fingerprint

Dive into the research topics of 'Host Response Stratification in Malarial and Non-malarial Sepsis: A Prospective, Multicenter Analysis From Uganda'. Together they form a unique fingerprint.

Cite this