Quantifying brain health in acute ischemic stroke through effective reserve

Area of Science:

• Neurology
• Radiology
• Biostatistics

Background:

• Brain health assessment often overlooks pre-existing pathology.
• Quantifying brain reserve is crucial for predicting neurological outcomes.
• Effective reserve (eR) is proposed as a novel metric integrating pathology.

Purpose of the Study:

• To develop and validate a measure of brain health, termed effective reserve (eR).
• To incorporate pre-existing brain pathology into the assessment of brain reserve.
• To evaluate the association between eR and stroke outcomes.

Main Methods:

• Analysis of two ischemic stroke cohorts (GASROS and SALVO) with neuroimaging and modified Rankin Scores (mRS).
• Automated extraction of white matter hyperintensity (WMHv), brain, and intracranial volumes to calculate brain parenchymal fraction (BPF).
• Modeling of eR using age, WMHv, and brain volume or BPF, with model comparison via Bayes Information Criterion (BIC).

Main Results:

• Brain volume outperformed BPF in modeling eR, showing a stronger inverse association with mRS (path coefficients: -0.67 vs -0.48, p < 0.001).
• Higher eR quartiles were consistently associated with better stroke outcomes (lower mRS) in both cohorts.
• The low eR group exhibited significantly worse outcomes compared to the high eR group.

Conclusions:

• Effective reserve (eR) is a promising metric for quantifying brain health, accounting for pre-existing pathology.
• The strong correlation between higher eR and improved outcomes highlights its clinical utility.
• eR has significant translational potential as a protective brain health indicator.