Stroke Dysbiosis Index (SDI) in Gut Microbiome Are Associated With Brain Injury and Prognosis of Stroke

Background: Significant dysbiosis occurs in the gut microbiome of stroke patients. Condensing these broad, complex changes into one index would greatly facilitate the clinical usage of gut microbiome data. Here, we formulated a gut microbiota index in patients with acute ischemic stroke based on their gut microbiota dysbiosis patterns and tested whether the index was correlated with brain injury and early outcome.

Methods: A total of 104 patients with acute ischemic stroke and 90 healthy individuals were recruited, and their gut microbiotas were compared and to model a Stroke Dysbiosis Index (SDI), which representing stroke-associated dysbiosis patterns overall. Another 83 patients and 70 controls were recruited for validation. The association of SDI with stroke severity (National Institutes of Health Stroke Scale [NIHSS] score) and outcome (modified Rankin scale [mRS] score: favorable, 0–2; unfavorable, >2) at discharge was also assessed. A middle cerebral artery occlusion (MCAO) model was used in human flora-associated (HFA) animals to explore the causal relationship between gut dysbiosis and stroke outcome.

Results: Eighteen genera were significantly different between stroke patients and healthy individuals. The SDI formula was devised based on these microbiome differences; SDI was significantly higher in stroke patients than in healthy controls. SDI alone discriminated stroke patients from controls with AUCs of 74.9% in the training cohort and 84.3% in the validation cohort. SDI was significantly and positively correlated with NIHSS score on admission and mRS score at discharge. Logistic regression analysis showed that SDI was an independent predictor of severe stroke (NIHSS ≥8) and early unfavorable outcome (mRS >2). Mice receiving fecal transplants from high-SDI patients developed severe brain injury with elevated IL-17 ⁺ γδ T cells in gut compared to mice receiving transplants from low-SDI patients (all P < 0.05).

Conclusions: We developed an index to measure gut microbiota dysbiosis in stroke patients; this index was significantly correlated with patients' outcome and was causally related to outcome in a mouse model of stroke. Our model facilitates the potential clinical application of gut microbiota data in stroke and adds quantitative evidence linking the gut microbiota to stroke.