[Development of a neurological score for the clinical evaluation of sylvian infarctions].

Among the main methodological problems raised by clinical trials in cerebral vascular accidents, one of the most serious is the lack of well-established and widely accepted criteria to evaluate the course of the disease. On the basis of published scores and of several trials carried on by us in three different centres during the last few years, we have developed a neurological score devised to reflect, as accurately as possible, deficits due to infarctions in the Sylvian territory, yet rapid and simple enough to provide an objective and reproducible rating. We report here the results of a prospective study concerning the inter-rater variations of the score from one centre to the other (inter-centre study) and within each centre (intra-centre study). These results have led us to withdraw three items with insufficient response rate and poor agreement, viz, "mental confusion", "visual field" and "sensory disorders". With the remaining 10 items the mean agreement between individual items varied from 85% to 90% and the global score agreement exceeded 90%. A study comparing our score with the Barthel index showed close correlation up to the autonomy threshold (Barthel index = 60) and divergent results beyond that point. The good agreement observed between inter-rater variations suggests that the score is reliable but does not inform on its sensitivity. In view of their relative independence, the neurological and the functional score should be regarded as complementary and should be used together as criteria of clinical assessment.