Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Longitudinal Language Recovery Following Repeated Treatment in Chronic Post-Stroke Aphasia
Poster Session C, Thursday, October 1, 10:45 am - 12:45 pm, Wangari Maathai
Sigfus Kristinsson1, Sophia Coolsen1, Saeed Ahmadi1, Roger Newman-Norlund2, Leonardo Bonilha3, Chris Rorden2, Julius Fridriksson1; 1Department of Communication Sciences and Disorders, University of South Carolina, Columbia, South Carolina, United States, 2Department of Psychology, University of South Carolina, Columbia, South Carolina, United States, 3Department of Neurology, University of South Carolina School of Medicine, Columbia, South Carolina, United States
Introduction: Although behavioral speech/language therapy improves language recovery in chronic post-stroke aphasia, it remains unclear whether treatment gains stabilize at ceiling performance or accumulate across successive rounds of treatment. In the absence of empirical evidence, this knowledge gap acts as a bottleneck for advancing the clinical management of post-stroke aphasia, including healthcare insurance and best-practice recommendations. To this end, we tested the hypothesis that repeated treatment cycles are associated with continued naming improvement, with no evidence of attenuation across successive rounds of treatment. Methods: We analyzed a retrospective dataset from the Center for the Study of Aphasia Recovery including 35 participants with post-stroke aphasia who participated in and completed at least two aphasia treatment studies. Participants completed two to four treatment studies, yielding 157 assessment observations nested within 81 treatment rounds. The primary outcome was raw Philadelphia Naming Test (PNT) score. The primary longitudinal analysis modeled visit-level PNT scores for treatment rounds 1-3 using a linear mixed-effects model with fixed effects for timepoint, treatment round, and their interaction, and random intercepts for participant and treatment round. Treatment round 4 was summarized descriptively but excluded from inferential modeling because only one participant contributed a fourth treatment study. A treatment round-level predictor model examined post-minus-pre PNT change using days post-stroke, sex, age at stroke, race, and baseline Western Aphasia Battery Aphasia Quotient (WAB-AQ). A participant-level secondary analysis computed overall PNT change as each participant’s last observed PNT minus first observed PNT and examined predictors using univariate models and AICc model selection. Results: PNT scores improved significantly from pre- to post-treatment across treatment rounds 1-3, with an estimated mean gain of 5.56 points (95% CI [3.44, 7.67], p < .001) per study. Significant gains were observed in round 1 (3.86 points, 95% CI [1.16, 6.56], p = .006); round 2 (5.07 points, 95% CI [2.33, 7.81], p < .001); and, round 3 (7.75 points, 95% CI [2.70, 12.80], p = .003). The timepoint-by-round interaction was not significant (F(2, 79.1) = 0.97, p = .384), indicating no evidence that treatment-associated naming gains diminished across repeated treatment rounds. In the treatment round-level predictor model, male sex was associated with larger PNT change (b = 4.70, 95% CI [0.46, 8.94], p = .030); days post-stroke, age at stroke, race, and baseline WAB-AQ were not significant predictors. Overall PNT change averaged 24.94 points (SD = 22.32; median = 23; range = 0-96), with 33 of 35 participants improving and none declining. In participant-level analyses, first recorded WAB-AQ was the strongest univariate predictor of overall PNT change, and AICc selection favored a model including number of treatment rounds completed, baseline WAB-AQ, and baseline PNT. Conclusion: In this selected retrospective sample, individuals with post-stroke aphasia continued to demonstrate naming gains across repeated treatment rounds. The findings argue against a simple inherent ceiling for treated naming recovery after one or two treatment episodes and support further prospective research on repeated rounds of aphasia treatment and individualized prediction of sustained treatment response.
Topic Areas: Disorders: Acquired, Speech-Language Treatment