Abstract WP61: Enhancing Stroke Trial Efficiency Using Nihss Scores To Predict Persistence Of Treatable Deficits

Introduction: A persistent challenge in stroke clinical trials is predicting which patients will respond optimally to treatment. Negative trials may result when patients’ severity of deficits contributes to a floor effect, or paucity of deficits leaves no room for the treatment to exert a benefit. A further challenge arises from the fact that impairments in distinct neurological domains are known to respond differently to treatments such as recanalization. Methods: Here we present a secondary analysis of 624 ischemic stroke patients enrolled in the NINDS TPA trial. Patients who were dead at 90 days or had missing data at any time point were excluded. Our analysis in the remaining 487 cases summarizes relevant differences in the proportions of patients with persistent deficits at 90 days when stratified according to their baseline scores. Results: For arm motor impairments (see Figure), 75% of patients with a baseline score of 2 recovered to a score of 0 at 90 days. Patients with this recovery profile could contribute to a ceiling effect if enrolled in restorative trials to treat motor deficits. Our analysis demonstrates contrasting patterns in other domains including sensory, language and perceptual deficits. Conclusions: Evaluating these patterns can help to predict whether patients being enrolled in a restorative trial will retain a treatable deficit in the domain that the study treatment targets. Commonly collected data from the NIHSS can guide the selection of patients with deficits in relevant recovery domains to promote more economical and efficient study operations for clinical stroke trials.