CLINICAL PHENOTYPES OF SEPSIS-ASSOCIATED ENCEPHALOPATHY: A RETROSPECTIVE COHORT STUDY

Xin Lü, Mubing Qin, Joseph Harold Walline, Yanxia Gao, Shiyuan Yu, Zengzheng Ge, Chao Gong, Huadong Zhu, Djillali Annane, Yi LiShock 2023 · DOI

TL;DR

This study identifies four distinct clinical phenotypes of sepsis-associated encephalopathy (SAE) using large ICU databases. The "mixed" phenotype, characterized by both ischemic-hypoxic and metabolic dysfunction, was found to be associated with the highest risk of in-hospital mortality, suggesting that tailored treatment strategies based on these phenotypes could improve patient outcomes.

What Problem Does This Paper Address?

Sepsis-associated encephalopathy (SAE) is a common and serious complication of sepsis, involving central nervous system dysfunction. However, SAE presents with a wide range of clinical and pathophysiological features, making it challenging to understand its underlying mechanisms and to develop effective treatments. The heterogeneity of SAE makes it difficult to design targeted clinical trials. This paper aims to address this by identifying distinct SAE phenotypes that correlate with different clinical outcomes, thereby paving the way for more personalized management and research.

Methods

The researchers conducted a retrospective cohort study using two large, publicly available critical care databases: MIMIC-IV and eICU. They identified adult patients admitted to the ICU with sepsis, defined by Sepsis-3 criteria. SAE was identified by a Glasgow Coma Scale (GCS) score less than 15 or the presence of delirium within 24 hours of ICU admission. Based on hypothesized mechanisms and previous research, they defined four SAE phenotypes: ischemic-hypoxic, metabolic, mixed (combining both), and unclassified. Clinical outcomes, primarily in-hospital mortality, were analyzed in relation to these phenotypes. Statistical analyses included descriptive statistics, chi-squared tests, and multivariable logistic regression to adjust for confounding factors like age, sex, GCS, and modified SOFA score.

Datasets

  • MIMIC-IV (Medical Information Mart for Intensive Care IV): A freely accessible database containing de-identified health-related data associated with over 400,000 intensive care unit admissions from the Beth Israel Deaconess Medical Center between 2008 and 2019.
  • eICU Collaborative Research Database: A multi-center database containing de-identified data from over 200,000 ICU admissions across 335 ICUs in the United States between 2014 and 2015.

These are existing, large-scale critical care datasets, not newly introduced benchmarks.

Key Insights and Ideas

  • SAE is not a monolithic entity: The study demonstrates that SAE can be broadly categorized into at least four distinct clinical phenotypes, each with different underlying pathophysiological implications.
  • The "mixed" phenotype is particularly dangerous: The finding that the mixed phenotype (combining ischemic-hypoxic and metabolic insults) carries the highest mortality risk is a critical insight. This suggests a synergistic or additive detrimental effect of these two types of dysfunction on the brain during sepsis.
  • Phenotypes have differential resource utilization: While not the primary focus, the data hints that different phenotypes might require different levels of care (e.g., mechanical ventilation, RRT), which could inform resource allocation and management strategies.
  • Unclassified phenotype is common but less severe: The large proportion of patients falling into the "unclassified" category, and their relatively better outcomes, suggests that not all SAE cases involve severe, identifiable organ dysfunction beyond encephalopathy. This group might represent a milder form or a different spectrum of SAE.

Key Results

  • The study analyzed 4,120 sepsis patients, with 2,780 identified as having SAE.
  • The four identified SAE phenotypes were: ischemic-hypoxic (37%), metabolic (39%), mixed (15%), and unclassified (39%). These distributions were consistent between the MIMIC-IV and eICU cohorts.
  • In-hospital mortality rates varied significantly by phenotype:
  • Ischemic-hypoxic: 33.9%
    • Metabolic: 28.4%
    • Mixed: 41.5%
    • Unclassified: 14.2%
  • The mixed phenotype was associated with the highest risk of in-hospital mortality (adjusted OR 2.11, 95% CI 1.67-2.67, P < 0.001) even after adjusting for key clinical variables.
  • The ischemic-hypoxic phenotype was associated with increased use of mechanical ventilation, and the metabolic phenotype with increased use of renal replacement therapy.

Limitations and Future Work

  • Definition Subjectivity: The definition of SAE based on GCS score or delirium assessment can be subjective and confounded by factors like sedatives or mechanical ventilation.
  • Phenotype Timing: The study focused on SAE phenotypes at ICU admission, not their dynamic evolution over time.
  • Lack of Long-Term Data: The databases did not allow for assessment of long-term outcomes, which are crucial for understanding the full impact of SAE.
  • Unclassified Phenotype: The large size of the unclassified group suggests that the current definitions might not fully capture its characteristics, and specific neuroinflammation markers were not available for verification.
  • Future Work: The authors suggest that understanding these differential phenotypes in sepsis could inform the design of future clinical trials and lead to more targeted SAE management strategies. Future research could explore specific biomarkers for these phenotypes and investigate their dynamic changes over the course of illness.

Summary updated

Abstract

Background: Sepsis-associated encephalopathy (SAE) is a dysfunction of the central nervous system experienced during sepsis with variable clinical and pathophysiologic features. We sought to identify distinct SAE phenotypes in relation to clinical outcomes. Methods: The Medical Information Mart for Intensive Care IV (MIMIC-IV) database and the eICU database were used to conduct a retrospective cohort study. Adult sepsis patients were included and SAE was defined as having a Glasgow Coma Scale (GCS) score ˂15 or delirium. The following our clinical phenotypes were defined as: ischemic-hypoxic, metabolic, mixed (ischemic-hypoxic and metabolic), and unclassified. The primary outcome was in-hospital mortality. Results: The study enrolled 4,120 sepsis patients, 2,239 from MIMIC-IV (including 1,489 patients with SAE, 67%), and 1,881 from eICU (1,291, 69%). For the SAE cohort, 2,780 patients in total were enrolled (median age, 67 years; interquartile range, 56-76.8; 1,589 (57%) were male; median GCS score was 12 [8-14]; median Sequential Organ Failure Assessment score was 6 [4-9]). The SAE phenotype distributions between the MIMIC-IV and eICU cohorts were as follows (39% vs. 35% ischemic-hypoxic, P = 0.043; 38% vs. 40% metabolic, P = 0.239; 15% vs. 15% mixed, P = 0.972; 38% vs. 40% unclassified, P = 0.471). For the overall cohort, the in-hospital mortality for patients with ischemic-hypoxic, metabolic, mixed, or unclassified phenotypes was 33.9% (95% confidence interval, 0.3-0.37), 28.4% (0.26-0.31), 41.5% (0.37-0.46), and 14.2% (0.12-0.16), respectively. In the multivariable logistic analysis, the mixed phenotype was associated with the highest risk of in-hospital mortality after adjusting for age, sex, GCS, and modified Sequential Organ Failure Assessment score (adjusted odds ratio, 2.11; 95% confidence interval, 1.67-2.67; P < 0.001). Conclusions: Four SAE phenotypes had different clinical outcomes. The mixed phenotype had the worst outcomes. Further understanding of these phenotypes in sepsis may improve trial design and targeted SAE management.
BibTeX
@inproceedings{lu2023clinical,
  author = {Xin Lü and Mubing Qin and Joseph Harold Walline and Yanxia Gao and Shiyuan Yu and Zengzheng Ge and Chao Gong and Huadong Zhu and Djillali Annane and Yi Li},
  title = {{CLINICAL PHENOTYPES OF SEPSIS-ASSOCIATED ENCEPHALOPATHY: A RETROSPECTIVE COHORT STUDY}},
  booktitle = {Shock},
  year = {2023},
  doi = {10.1097/shk.0000000000002092},
}
Figure