Influence of the Gut Microbiota on Acute Ischemic Stroke Functional Outcomes at Three Months

Abstract

Background: Functional recovery from ischemic stroke (IS), the main cause of adult disability worldwide, is influenced by many factors, and a portion of interindividual variability remains unexplained.

Methods: Observational study in a tertiary stroke centre of patients with IS analyzed using shotgun metagenomic sequencing (January 2020-March 2022). Functional outcomes were assessed according to modified Rankin Scale (mRS) scores 3-months post-IS, considering 0-2 favorable and 3-6 unfavorable. The causal relationship between several bacteria and post-IS outcomes was explored via two-sample Mendelian randomization (MR) analyses using Genome-Wide Association Analysis (GWAS) summary statistics.

Results: Comparing 128 patients with favorable and unfavorable post-IS functional outcomes, β-diversity analysis showed a separation in microbial structure, and α-diversity measures revealed greater bacterial richness in the favorable outcomes group. Taxonomic profiling of the samples showed that a greater abundance of pathogenic bacteria (e.g., Pseudomonas, Finegoldia, Porphyromonas) was associated with an unfavorable outcome. Functional profiling of the samples revealed differences in the ethylbenzene degradation pathway and in 16S rRNA (uracil1498-N3)-methyltransferase. MR confirmed increased pyruvate levels to be causally associated with post-IS favorable outcomes (β = -0.50, 95% CI: -0.91, -0.10).

Conclusions: Our study points to gut microbiota differences in patients with unfavorable versus favorable 3-month post-IS outcomes. Patients with unfavorable outcomes presented gut microbiota dysbiosis and alterations in multiple metabolic pathways.

Trial registration: This study was registered on 3 October 2021 with https://clinicaltrials.gov. Access number: NCT04795687.

Keywords: functional outcome; gut‐brain axis; ischemic stroke; microbiota; shotgun metagenomics.

FIGURE 1

Gut microbiota differences in α‐ and β‐diversities between favorable (mRS 0–2) and unfavorable (mRS 3–6) outcomes in patients with ischemic stroke. (A) Boxplots representing class‐level α‐diversity measures (left to right): observed species, Chao1, and ACE indexes. Boxes represent the first to third quartiles, horizontal lines within the boxes show the median, and whiskers indicate variability outside the upper and lower quartiles. (B) PCoA at species‐level. The two components explain 30.0% and 11.9% of the variance. ACE: abundance‐based coverage estimator; mRS: Modified Rankin Scale; PCoA: principal coordinates analysis.

FIGURE 2

Taxonomic differences as determined using LEfSe between favorable (mRS 0–2) and unfavorable (mRS 3–6) outcomes in patients with ischemic stroke. (A) Significantly different taxa. (B) Differential relative abundance of Butyrate‐producing bacterium SL6/1/1. LDA: linear discriminant analysis; LEfSe: LDA size effect; mRS: Modified Rankin Scale.

FIGURE 3

Functional differences between favorable (mRS 0–2) and unfavorable (mRS 3–6) outcomes in patients with ischemic stroke. (A) Differential relative ethylbenzene degradation KEGG pathway abundance. (B) Differential relative abundance of 16S rRNA (uracil1298‐N3)‐methyltransferase. KEGG: Kyoto Encyclopedia of Genes and Genomes; mRS: Modified Rankin Scale.