Using behaviour to predict stroke severity in conscious rats: post-stroke treatment with 3', 4'-dihydroxyflavonol improves recovery.

Prognostic models are used to predict outcome in stroke patients and to stratify treatment groups in clinical trials. No one has previously attempted to use such models in stroke recovery studies in animals. We have now shown the predictive value of assigning stroke severity ratings, based on behaviours displayed in conscious rats during infusion of endothelin-1 to constrict the middle cerebral artery, on neurological and histological outcomes. The validity of prior stratification of treatment groups according to stroke ratings was tested by assessment of the protective potential of synthetic flavonol, 3',4'-dihydroxyflavonol (DiOHF). Neurological deficits and performance on the sticky label test were evaluated before and at 24, 48 and 72 h post-stroke. Histopathology was assessed at 72 h. Positive correlations between stroke ratings and neurological deficit scores were found at 24 (r=0.58, P<0.001), 48 (r=0.53, P<0.001) and 72 (r=0.56, P<0.001) h post-stroke, with more severe strokes associated with worse deficit scores. Similar correlations were observed with the sticky label test. Higher stroke ratings also correlated with greater infarct volumes (total infarct volume: r=0.74, P<0.0001). Treatment with DiOHF (10 mg/kg i.v. given 3, 24 and 48 h post-stroke) significantly reduced infarct volume and restored neurological function in rats with modest stroke ratings (P<0.01), but not in rats with high stroke ratings. These results suggest that stroke ratings, based on behavioural assessment as the stroke develops, reliably predict histopathological and functional outcomes and allow stratification of treatment groups. DiOHF given after stroke improves outcomes in moderate strokes, and therefore has cytoprotective potential.