Aphasia, Cerebrovascular Accident
Conditions
Keywords
aphasia, therapies, investigational, rehabilitation of speech and language disorders, magnetic resonance imaging, functional, language
Brief summary
The purpose of this study is to determine if an intentional act improves treatment response for patients with nonfluent aphasia. The treatment involves naming pictures and saying members of categories. The intentional act requires initiating picture naming or category member trials with a left-hand movement sequence. Nonfluent aphasia is a disorder of language production in which patients with damage to the brain's language system have trouble initiating and maintaining spoken communication. All patients participating in the study take part in functional MRI scans to determine how treatments affect brain systems.
Detailed description
A new treatment manipulating intention substrates for language production in nonfluent aphasia patients was developed. The intention component involves initiating word-finding trials with a complex left-hand movement. The study addresses (1) whether or not the intention manipulation (complex left-hand movement) makes a unique contribution to treatment outcome and (2) whether or not the intention manipulation helps to shift word production mechanisms from the left to the right frontal lobe. All study participants take part in functional magnetic resonance imaging (fMRI) scans of word-finding before and after treatment and at 3-month follow-up to measure changes in lateralization of frontal lobe activity during word finding. Only patients with a substantial degree of left frontal activity on the pre-treatment fMRI scan can participate. There are three specific aims: (1) to determine if repetitive initiation of word production with a complex left-hand movement leads to increased right-hemisphere lateralization of frontal activity and if these changes can be attributed to the intention component of treatment, (2) to determine whether activity in posterior perisylvian cortices that is entrained to right frontal activity shows a greater increase in right-hemisphere lateralization from pre- to post-treatment fMRI when the intention component is included in treatment, and (3) to determine whether onset of hemodynamic responses (HDRs) in right motor/premotor cortex becomes more closely associated with the temporal onset of participants' spoken responses across treatment when the intention component is included in treatment. If successful, the treatment can provide a new treatment vehicle for increasing language function in patients with nonfluent aphasia.
Interventions
BEHAVIORALWord-finding with intention component
Word-finding trials (picture-naming) with intention manipulation (initiating word-finding trials with a complex left-hand movement). 8 (or more) baseline sessions over 4 days followed by 30 treatment sessions (2 sessions/day, 5 days/week for 3 weeks).
BEHAVIORALWord-finding with no intention component
Word-finding trials with no intention manipulation. 8 (or more) baseline sessions in 4 days followed by 30 treatment sessions (2 sessions/day, 5 days/week for 3 weeks).
Sponsors
National Institute on Deafness and Other Communication Disorders (NIDCD)
University of Florida
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
SINGLE (Subject)
Eligibility
Sex/Gender
ALL
Age
21 Years to 95 Years
Healthy volunteers
No
Inclusion criteria
* Nonfluent aphasia caused by stroke * Moderate to severe word-finding problems * 6 or more months post stroke * Right handed prior to stroke * All strokes in left hemisphere * Native English speaker * Capable of following verbal directions
Exclusion criteria
* Severe impairment of word comprehension * Brain injury or disease in addition to stroke * Drug or alcohol abuse within past 6 months * Schizophrenia or other psychiatric disorder necessitating hospitalization * History of learning disability * Claustrophobia * Cardiac pace-maker * Ferrous metal implants not attached to bone, metal fragments in body * Profound hearing loss
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Lateralization of Frontal Lobe (and Posterior Perisylvian) Activity During Word Production | immediately post-treatment scan minus pre-treatment baseline scan | Functional MRI laterality indices (LIs)were calculated for lateral frontal, medial frontal, and posterior perisylvian cortex regions of interest (ROIs): L=number of active voxels in left hemisphere ROI and R=number of active voxels in right hemisphere ROI using the following formula: (L-R)/(L+R). LIs could vary from -1 (completely right lateralized) to +1 (completely left lateralized). Then, change in LIs was calculated by subtracting the pre-treatment from the post-treatment and 3-mo follow-up LI. It was expected the intention manipulation would show a rightward shift in LI. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Picture Naming Probe Scores (% Accuracy) | trend for time series of 8 baseline + 30 treatment sessions | Improvement for the time series from 8 baseline sessions through 30 treatment sessions for naming probes was computed with the C statistic using Tryon's (1982, 1983) formula for each subject. C statistics were converted to Z scores, using the formula provided by Tryon (1982). Z scores indicated treatment change for each subject, with a positive and significant Z score indicating substantive treatment gains. Z scores were then compared between groups with a t statistic. It was expected that the intention manipulation would lead to greater treatment gains than when it was not used. |
| Category Member Generation Probe Scores (% Accuracy) | trend for time series of 8 baseline + 30 treatment sessions | Improvement for the time series from 8 baseline sessions through 30 treatment sessions for naming probes was computed with the C statistic using Tryon's (1982, 1983) formula for each subject. C statistics were converted to Z scores, using the formula provided by Tryon (1982). Z scores indicated treatment change for each subject, with a positive and significant Z score indicating substantive treatment gains. Z scores were then compared between groups with a t statistic. It was expected that the intention manipulation would lead to greater treatment gains than when it was not used. |
Countries
United States
Participant flow
Recruitment details
Subjects were recruited from December 2006 through August 2009. Fourteen subjects with minimally nonfluent aphasia, anomia, and at least some degree of frontal lesion participated in the study.
Pre-assignment details
Prior to trial assignment, subjects underwent a functional MRI scan to determine degree of lateralization of frontal activity during word generation. Subjects with a high degree of right frontal lateralization of activity were excluded. Further, patients whose naming scores indicated that they were no more than minimally anomic were excluded.
Participants by arm
| Arm | Count |
|---|---|
| Word-finding With Intention Manipulation Treats word-finding (picture naming, category member generation) with an intention manipulation (complex left-hand movement to initiate word-finding trials) | 7 |
| Word-finding With no Intention Manipulation Word-finding trials similar to intention mediated treatment, but without intention manipulation | 7 |
| Total | 14 |
Baseline characteristics
| Characteristic | Total | Word-finding With Intention Manipulation | Word-finding With no Intention Manipulation |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 7 Participants | 5 Participants | 2 Participants |
| Age, Categorical Between 18 and 65 years | 7 Participants | 2 Participants | 5 Participants |
| Age Continuous | 66.93 years STANDARD_DEVIATION 11.24 | 70.86 years STANDARD_DEVIATION 12.35 | 63.00 years STANDARD_DEVIATION 9.22 |
| Region of Enrollment United States | 14 participants | 7 participants | 7 participants |
| Sex: Female, Male Female | 5 Participants | 4 Participants | 1 Participants |
| Sex: Female, Male Male | 9 Participants | 3 Participants | 6 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 7 | 0 / 7 |
| serious Total, serious adverse events | 0 / 7 | 0 / 7 |
Outcome results
Primary
Lateralization of Frontal Lobe (and Posterior Perisylvian) Activity During Word Production
Functional MRI laterality indices (LIs)were calculated for lateral frontal, medial frontal, and posterior perisylvian cortex regions of interest (ROIs): L=number of active voxels in left hemisphere ROI and R=number of active voxels in right hemisphere ROI using the following formula: (L-R)/(L+R). LIs could vary from -1 (completely right lateralized) to +1 (completely left lateralized). Then, change in LIs was calculated by subtracting the pre-treatment from the post-treatment and 3-mo follow-up LI. It was expected the intention manipulation would show a rightward shift in LI.
Time frame: immediately post-treatment scan minus pre-treatment baseline scan
Population: Data for all subjects completing the protocol for their respective arm were analyzed.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Word-finding With Intention Manipulation | Lateralization of Frontal Lobe (and Posterior Perisylvian) Activity During Word Production | -0.2802 laterality index | Standard Deviation 0.1077 |
| Word-finding With no Intention Manipulation | Lateralization of Frontal Lobe (and Posterior Perisylvian) Activity During Word Production | -0.1832 laterality index | Standard Deviation 0.2768 |
Comparison: H1: The Intention Group will show a significant rightward shift in lateral frontal laterality from pre-treatment to post-treatment.~H0: The Intention Group will show no shift in lateral frontal laterality from pre-treatment to post-treatment.~Since this is a repeated measures t test, the data are presented as the mean and standard deviation for for the post-treatment laterality index minus the pre-treatment laterality index.p-value: <0.05t-test, 1 sided
Comparison: H1: The Control Group will show a significant rightward shift in lateral frontal lateral indices from pre-treatment to post-treatment.~H0: The Control group will show no shift in lateral frontal laterality indices from pre-treatment to post-treatment.~Since the analysis to test these hypotheses is a repeated-measures t-test, the mean and standard deviation are given for post-treatment laterality index minus pre-treatment laterality index.p-value: <0.05t-test, 1 sided
Secondary
Category Member Generation Probe Scores (% Accuracy)
Improvement for the time series from 8 baseline sessions through 30 treatment sessions for naming probes was computed with the C statistic using Tryon's (1982, 1983) formula for each subject. C statistics were converted to Z scores, using the formula provided by Tryon (1982). Z scores indicated treatment change for each subject, with a positive and significant Z score indicating substantive treatment gains. Z scores were then compared between groups with a t statistic. It was expected that the intention manipulation would lead to greater treatment gains than when it was not used.
Time frame: trend for time series of 8 baseline + 30 treatment sessions
Population: Data from all subjects completing their respective arms were analyzed, with the exception of one control subject whose baseline was not stable.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Word-finding With Intention Manipulation | Category Member Generation Probe Scores (% Accuracy) | 3.50 Z score | Standard Deviation 0.83 |
| Word-finding With no Intention Manipulation | Category Member Generation Probe Scores (% Accuracy) | 1.94 Z score | Standard Deviation 1.75 |
Comparison: H1: The Intention Group would show more improvement across treatment than the Control Group.~H0: The Intention Group and the Control Group would not show any difference in improvement across treatment.p-value: <0.05t-test, 1 sided
Secondary
Picture Naming Probe Scores (% Accuracy)
Improvement for the time series from 8 baseline sessions through 30 treatment sessions for naming probes was computed with the C statistic using Tryon's (1982, 1983) formula for each subject. C statistics were converted to Z scores, using the formula provided by Tryon (1982). Z scores indicated treatment change for each subject, with a positive and significant Z score indicating substantive treatment gains. Z scores were then compared between groups with a t statistic. It was expected that the intention manipulation would lead to greater treatment gains than when it was not used.
Time frame: trend for time series of 8 baseline + 30 treatment sessions
Population: Data from all subjects completing their respective arms were analyzed, with the exception that one control subject was eliminated because of an unstable baseline measure.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Word-finding With Intention Manipulation | Picture Naming Probe Scores (% Accuracy) | 3.68 Z score | Standard Deviation 1.61 |
| Word-finding With no Intention Manipulation | Picture Naming Probe Scores (% Accuracy) | 3.92 Z score | Standard Deviation 1.18 |
Comparison: H1: The Intention Group would show more improvement across treatment than the Control Group.~H0: The Intention Group and the Control Group would not show any difference in improvement across treatment.p-value: <0.05t-test, 1 sided