Lesion-site-dependent responses to therapy after aphasic stroke

Stroke survivors with language difficulties (aphasia) vary: some recover quickly while others suffer long-term impairments, and different patients respond differently to the same speech and language therapies.1 In recent years, we and others have shown that much of the variability in language outcomes after stroke can be explained by reference to the details of the brain damage that individual patients have suffered.2 Here, we show that this same information can be used to predict responses of patients who had stroke to treatment.Detailed methods are provided in online supplementary material: here, we summarise the key points. Our focus is on a novel treatment for Central Alexia (CA): an acquired reading disorder in the context of a general language impairment (aphasia). Patients with CA are slow to read, make frequent errors and have additional problems with spoken and written language. Our intervention is a computerised therapy embodied in an application called ‘iReadMore’, which uses multimodal cueing and massed practice to improve patients’ single-word reading skills.3### Supplementary file 1[jnnp-2017-317446-SP1.pdf]Our study included 23 participants with CA after left hemisphere stroke (see online supplementary table), recruited through both the PLORAS project2 and the outpatient speech and language therapy services at the National Hospital for Neurology and Neurosurgery, University College London Hospitals. Before the treatment began, each participant’s cognitive skills were assessed with an extensive protocol including linguistic and non-linguistic tests, yielding a total of 28 pretreatment behavioural variables per patient. We also acquired structural MRI from each patient, extracting lesion images using the Automatic Lesion Identification toolbox.4 The outputs (binary lesion images) were …