Background Pre-intervention perfusion imaging is increasingly becoming part of stroke triage. Small studies supporting imaging based patient selection have been published. The goal of this larger study was to determine if perfusion imaging could impact on functional outcomes in patients undergoing stroke interventions.

Methods All patients who had undergone endovascular therapy for anterior circulation strokes over a 7 year period were retrospectively analyzed. The pre-intervention perfusion imaging was assessed for size of cerebral blood volume (CBV), cerebral blood flow and mean transit time (MTT) abnormalities. A perfusion mismatch for irreversible versus reversible ischemia was based on CBV and MTT. Clinical outcome and mortality were based on the 90 day modified Rankin Scale. An analysis of the pre-intervention perfusion parameters was then performed to determine any impact on functional outcomes.

Results 110 patients underwent endovascular therapy for anterior circulation strokes. A younger age and lower National Institutes of Health Stroke Scale score were important clinical predictors of favorable outcome (modified Rankin Scale ≤2). The extent of the CBV abnormality and percentage of CBV/MTT mismatch were the strongest imaging predictors of outcome and mortality. A CBV area of 229.5 mm² (±290) was seen for favorable outcomes versus 968 mm² (±1173) for poor outcomes (p<0.0001). A CBV/MTT mismatch of 91% (±10.7) was seen for favorable outcomes versus 72.5% (±31.6) for poor outcomes (p=0.0001). The CBV area was 273 mm² (±392) in patients without mortality versus 1401.1 mm² (±1310) in patients with mortality (p<0.0001). Patients who survived had a mean CBV/MTT mismatch of 90.2% (±12.5) versus 61.1% (±35.2) for those who did not (p<0.0001). A CBV lesion approximately greater than one-third of the middle cerebral artery distribution predicted a poor outcome and mortality.

Conclusion The extent of pre-intervention CBV abnormality is a strong predictor of functional outcomes following endovascular stroke therapy. This information can aid in patient selection and improve procedure efficacy.