Abstract
Introduction
Many individuals with aphasia following a left-hemisphere stroke retain the ability to sing despite limited spontaneous verbal output. Researchers leveraged these capabilities to develop Melodic Intonation Therapy (MIT), a systematic prosodic and rhythmic method to foster recovery of verbal expression in individuals with nonfluent aphasia. Several reviews have summarized the evidence for the effects of MIT for language recovery, reporting positive outcomes for repetition, naming, and verbal fluency. Likewise, researchers have explored the neural correlates of MIT, anticipating right-hemisphere mediation of language recovery. Recent reviews have summarized neuroimaging outcomes following aphasia treatment, with few including MIT. The purpose of the current project was to systematically review the functional magnetic resonance imaging (fMRI) literature to explore neuroplastic changes associated with MIT. By restricting to one neuroimaging approach, we endeavored to constrain variables and better represent this complex dynamic recovering neural system.
Main Body
We searched four databases and reference lists for studies that implemented fMRI following MIT in adults with aphasia following left-hemisphere stroke. We identified seven that met inclusion criteria: five case studies and two group studies. We implemented a 20-item fMRI appraisal checklist to consider the quality of fMRI methods and statistical analyses. Reviewers extracted data across studies pertaining to aphasia participants, treatment outcomes, fMRI methods, and post-treatment activation patterns. We found that over time, fMRI methodology has improved, increasing confidence in neuroimaging findings across studies. The seven fMRI studies reported on 30 adults with aphasia (22 male, 8 female; 6 subacute, 24 chronic), all reporting improvements on language measures. The two group studies indicated considerable right-hemisphere activation post-MIT, with a laterality index in one study favoring the right in five of eight patients, more so for those with subacute aphasia, and therapy language outcomes positively correlated with right inferior frontal gyrus (IFG) increases in another. Following MIT, all five case studies reported increased right-hemisphere activation, especially in IFG and superior temporal gyrus (STG) areas. However, three of five case studies also noted increased post-MIT left activation, particularly in prefrontal and inferior parietal regions. Two case studies showed decreased left activation post-MIT, one IFG and one “perisylvian,” and one with decreased right cerebellar activation. Our right-hemisphere prediction was largely supported, as two group studies and five case studies reported increased activation in right perisylvian regions following MIT, particularly the IFG and STG. These findings are counterbalanced by findings of increased left-hemisphere activation following MIT in several studies. However, in examining left-hemisphere activation in the cases more closely, we noted discrepancies that would explain left engagement. For example, one study used a passive lexical decision task in the scanner in a patient with relatively preserved left frontal cortex, and another study delivered MIT simultaneously with ‘conventional speech therapy,’ which likely would engage left hemisphere capacities.
Conclusion
Our review suggests that important observations are evident as findings are coalesced across case studies. Group studies are particularly informative for correlating language improvements with neural changes. Overall, the preponderance of the fMRI evidence of this review supports primarily right-hemisphere mediation of successful response to MIT.