Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Predicting Treatment Response in Chronic Post-Stroke Aphasia: Pre-Treatment Variability in Naming Indicates Neuroplastic Potential of the Middle Temporal Gyrus
Poster Session B, Friday, October 25, 10:00 - 11:30 am, Great Hall 3 and 4
Sigfus Kristinsson1, Janina Wilmskoetter2, Roger Newman-Norlund1, Chris Rorden1, Leonardo Bonilha1, Julius Fridriksson1; 1University of South Carolina, SC, USA, 2Medical University of South Carolina, SC, USA
Introduction Recent findings indicate that variability in lexical-semantic processing may be associated with language recovery in aphasia1; however, the mechanism driving this effect remains unclear. Several established models of speech production offer that otherwise unexplained variability in performance (e.g., object naming) may indicate flexibility within the semantic network.2 This notion is neurobiologically grounded in the middle temporal gyrus (MTG); the MTG is a critical hub for lexical-semantic processing3, and MTG lesion is associated with particularly poor language recovery in post-stroke aphasia.4 In this study, we tested the hypothesis that treatment-induced language recovery in aphasia relies on upregulation of functional activity in the MTG and, critically, that pre-treatment variability in lexical-semantic processing is a reliable indicator MTG-mediated language recovery. Method A total of 107 participants with chronic post-stroke aphasia following a left hemisphere lesion underwent extensive neuropsychological assessment and neuroimaging prior to receiving six weeks of restorative anomia treatment. Recovery was assessed as raw change in performance on the Philadelphia Naming Test (PNT) from pre- to post-treatment. The PNT was administered twice at baseline and lexical-semantic processing was measured as the difference in performance across the two administrations (henceforth, ‘lexical-semantic variability’). Multiple regression was used to determine the effect of lexical-semantic variability, pre- to post-treatment upregulation of activity within the MTG, and their interaction on treatment-induced recovery, while adjusting for lesion volume, proportional lesion within the MTG, and baseline PNT score. Results Participants improved their performance by 7.7 items on average and average lexical-semantic variability was 6.8 items. The primary statistical model (F(6, 26)=2.94, p=.025, R2=.42) revealed a significant interaction between MTG activation change and lexical-semantic variability (B=-2.06, p=.024), in addition to a significant main effect of lexical-semantic variability (B=1.18, p<.01), whereas MTG activation change was not independently associated with recovery (B=4.59, p=.30). Additionally, we tested the potential complementary role of the contralateral MTG in a post hoc analysis; briefly, a significant interaction was observed between functional recruitment and lesion size (p<.002), suggesting a complementary role in case of smaller lesions, but a maladaptive role in larger lesions. With the addition of this term, the effect of left hemisphere MTG activation change was also significant (B=9.28, p<.05) in the final model (F(8, 24)=4.40, p<.01, R2=.60). Conclusion Consistent with our hypotheses, these findings suggest that lexical-semantic variability, measured as the inconsistency in accurate confrontation naming prior to treatment, predicts treatment-induced language recovery in chronic aphasia and, specifically, that the effect of lexical-semantic variability is mediated through upregulation of activity in the MTG bilaterally. These findings hold promise to inform prognostication efforts in the clinical management of aphasia, and support the development of neurobiologically-informed aphasia treatments.
Topic Areas: Disorders: Acquired, Language Production