# Integration of gut microbiota, Stroke Dysbiosis Index, and inflammatory biomarkers in assessing prognosis of acute ischemic stroke

## TL;DR

This study explores how gut microbiota changes and inflammatory markers can predict infections and outcomes in patients with acute ischemic stroke.

## Contribution

The study introduces a novel integration of gut dysbiosis indices and inflammatory biomarkers for prognosis in acute ischemic stroke.

## Key findings

- Infected stroke patients showed reduced gut diversity and enrichment of harmful bacteria like Klebsiella and Escherichia.
- Elevated SDI/MDI and altered biomarkers (like TMAO and LPS) strongly correlated with post-stroke infections.
- The model showed strong predictive ability with AUCs over 0.80 after validation.

## Abstract

Acute ischemic stroke (AIS) is often complicated by systemic infections that worsen prognosis. Emerging evidence suggests gut microbiota dysbiosis, inflammatory biomarkers, and gut–barrier dysfunction play pivotal roles in post-stroke outcomes. This study aimed to integrate Stroke Dysbiosis Index (SDI), Microbial Dysbiosis Index (MDI), and inflammatory biomarkers to evaluate their prognostic value in AIS patients.

An observational prospective cohort was conducted at a tertiary stroke center in Jakarta (September 2023–September 2024). Eighty AIS patients admitted within 24 h of onset were enrolled. Fecal samples underwent 16S rRNA sequencing for microbiota profiling and SDI/MDI calculation. Blood biomarkers (NMDAR NR2B, butyrate, TMAO, RANKL, iFABP, LPS) and platelet-to-lymphocyte ratio (PLR) were measured. Clinical severity was assessed with NIHSS. Outcomes included infection within 7 days, and complications. Statistical analyses comprised correlation, regression, and ROC curve modeling.

Infection occurred in 46.3% of patients, predominantly older (>60 years) and female. Infected patients showed reduced microbial diversity, enrichment of pathogenic taxa (Klebsiella, Escherichia, Salmonella), elevated SDI/MDI, and depleted SCFA producers. Biomarkers revealed increased NMDAR, TMAO, iFABP, and LPS, with reduced butyrate and RANKL (all p < 0.001). These markers correlated strongly with infection status. ROC analysis demonstrated promising discriminative ability in this cohort (bias-corrected AUCs > 0.80 after internal bootstrapping validation). Internal validation using 1,000 bootstrap resamples was performed to estimate optimism in AUC values and obtain bias-corrected confidence intervals.

Gut dysbiosis, elevated dysbiosis indices, and inflammatory biomarker derangements were strongly associated with post-stroke infections and adverse prognosis. Integrating microbiota and biomarker profiles with clinical parameters may provide a robust framework for risk stratification and open avenues for microbiota-targeted therapies in AIS.