Frontiers in Psychology (Apr 2015)
Treatment-Induced Neuroplasticity Following Intensive Speech Therapy and a Home Practice Program in Fifteen Cases of Chronic Aphasia
Abstract
Introduction Evidence suggests that intensive aphasia therapy can take advantage of the brain’s potential to ‘overcome learned non-use’ (Pulvermuller & Berthier, 2008). Many patients have shown significant language improvements, but few studies have examined the neuroplastic changes accompanying such improvements over time. The current study examined changes in behavior and fMRI activity in 15 aphasic individuals who were tested pre-/post-treatment and six months after use of a tablet-based home practice program. Methods Twenty participants with chronic moderate-to-severe post-stroke aphasia completed a 2-week intensive treatment program. The primary outcome measure was confrontation naming on sets of treated (TR) or untreated (UNTR) drawings of common objects (Snodgrass & Vanderwart, 1980) and picturable actions (Masterson & Druks, 1998). Participant characteristics included a range of time post-onset (MPO=6-142; mean=26.8); auditory comprehension (BDAE-3 percentile: 6th-83rd; mean=47th); naming (BNT % correct=0-78; mean=44.4); sentence repetition (BDAE-3 percentile: 5th-80th; mean=39th). Following treatment, participants autonomously practiced retrieving words for half TR and half UNTR actions and objects with an iPad-based home practice (HP) program for six months. Functional MRI data were acquired on a 3T system at three time-points from 15 MR-compatible participants who completed the HP program (S1:pre-treatment; S2:post-treatment; S3:6mos post-HP). Participants overtly named blocks of TR, matched UNTR, and consistently correct (CORR) pictures of objects and actions in S1 and S2. Pictures were blocked by practice condition, i.e., practiced (PR) or unpracticed (UNPR) in S3. SPM8 was used for pre-processing (realign, co-register with 3D MPRAGE, segment, indirectly normalize to MNI template and smooth data) and statistical modeling. Lesions were masked and excluded from the normalization process. First-level t-tests examined task-related activation during CORR, TR, and UNTR picture naming (S1/ S2) and during CORR, PR, and UNPR picture naming (S3). Results Aphasic participants made significant gains in naming TR targets following the treatment phase and PR targets following the HP phase, even in the scanner. Contrasts by training condition and over time revealed partially overlapping ‘signature’ activity patterns that were unique to each aphasic individual. Within subject significant clusters had nearly identical voxels of peak activity across time and training/practice conditions. We present here two participants, one non-fluent (JBR) and one fluent (ACL), who both made behavioral progress during treatment and HP program (see Figure). JBR’s ‘signature’ pattern of activity (p<.05, FWE) included: • Generally smaller amplitude and spatial extent of activity over time, but slightly greater for TR/PR and UNTR/UNPR than CORR; ACL’s ‘signature’ pattern included: • Generally smaller amplitude and spatial extent of activity in network post-treatment, with greatest and most asymmetrically right hemisphere activity post-HP; • Activity in R posSTG decreased over time for CORR pictures while increasing over time for TR/PR pictures Discussion Short-term intensive treatment followed by a home practice program can produce enduring language improvements that provide rich opportunities for investigating treatment-induced neuroplasticity in aphasia. Given the high degree of individual variability in lesion location/extent, and the resulting variability in aphasia type/severity, it makes sense to examine treatment-induced changes in neural activity patterns within subjects where ‘signature’ patterns of activity are remarkably reliable across time.
Keywords
