Utilizing Quantitative Analysis of CSF Volume from Clinical T1-Weighted MRI to Predict Thrombectomy Outcomes

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Endovascular thrombectomy (EVT) is standard for acute ischemic stroke with large vessel occlusion.
• Functional independence post-EVT varies, and brain atrophy (higher cerebrospinal fluid volume - CSFV) may influence outcomes.
• Baseline CSFV could potentially predict EVT benefit by reflecting brain health.

Purpose of the Study:

• To quantify total CSFV from T1-weighted MRI scans.
• To assess global brain atrophy using CSFV measurements.
• To determine the association between baseline CSFV and functional outcomes after successful EVT.

Main Methods:

• Retrospective analysis of 432 patients from a stroke thrombectomy registry (2015-2021).
• Included 214 patients with MRI within 14 days of EVT and 90-day mRS data.
• Used SynthSR for image enhancement and FreeSurfer for CSFV estimation, calculating CSFV percentage (CSFV%) to intracranial volume.

Main Results:

• Baseline CSFV% significantly predicted 90-day modified Rankin Scale (mRS) in ordinal regression models (p < 0.001).
• CSFV% remained an independent predictor of 90-day mRS after adjusting for clinical factors (p = 0.012).
• CSFV% did not significantly predict hemorrhagic transformation.

Conclusions:

• Increased CSFV% correlates with poorer functional outcomes following EVT.
• Total CSFV% shows potential as an imaging biomarker for prognostication in EVT patients.
• This metric may aid clinicians in determining patient prognosis before EVT.