Behavioral Evidence for Continued Language Recovery in Chronic Post-stroke Aphasia
Elizabeth Chang1, Alycia Laks1, Andrew DeMarco1, Peter Turkeltaub1
1Center for Brain Plasticity and Recovery, Georgetown University Medical Center
Objective:
To determine whether individuals with aphasia due to left-hemisphere (LH) stroke continue to show measurable gains in language abilities in the chronic phase of recovery (>6 months post-stroke), and to identify which behavioral domains demonstrate the most robust late recovery. 
Background:
Aphasia recovery following LH stroke is traditionally assumed to plateau after six months post-stroke, but growing evidence suggests that improvement may continue well into the chronic phase. Few studies have examined within-subject behavioral change in long-term survivors or tested predictors of continued recovery. The few longitudinal behavioral studies in the chronic phase have examined global recovery using the Western Aphasia Battery Aphasia Quotient (WAB-AQ), leaving domain-specific changes in reading and language largely unexplored.  
Design/Methods:
Twenty-four adults with chronic LH stroke (mean time-since-stroke at baseline= 56.4 months, mean age at stroke = 53.8 years) completed behavioral assessments at two timepoints (mean interval = 30 months), including the WAB, Written Lexical Decision, Oral Word Reading, Oral Pseudoword Reading, Real-Word Repetition, and Pseudoword Repetition. Paired t-tests examined change over time for each behavioral measure. Pearson correlations examined whether change scores (Δ=T2-T1) were associated with age at stroke, years of education, time since stroke at baseline (months), lesion volume, and interval between timepoints.
Results:
Participants showed significant improvement in overall language ability (WAB-AQ (t(23)=-3.14, p=.005), oral real-word reading accuracy (t(23)=-2.68, p=.013), oral pseudoword reading accuracy (t(22)=-2.86,p=.009), real word repetition (t(21)=-3.50,p=.002) and pseudoword repetition (t(20)=-2.70,p=.014). Change in WAB-AQ scores (r=-0.442, p=0.035) and pseudoword repetition accuracy (r=-0.551, p=.01) were negatively correlated with time-since-stroke at baseline (i.e., more improvement with earlier first timepoint after stroke). 
Conclusions:
These findings provide preliminary evidence that aphasia can continue to improve during the chronic stage of stroke recovery, challenging the traditional plateau model. Gains across phonological and lexical-semantic tasks suggest that recovery extends across multiple domains of language processing, even in the chronic phase. 
10.1212/WNL.0000000000217299
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.