Frontal Aslant Tracts as Correlates of Lexical Retrieval in MS

Abstract

Introduction:

Previous studies reveal that a newly described white matter pathway, the frontal aslant tract (FAT), connecting inferior and superior frontal gyri has a role in speech and language functions. We explored the role of this tract in a phonemic and semantic fluency tasks in a cohort of multiple sclerosis patients diagnosed with cognitive impairment.

Methods:

Thirty-five right-handed English-speaking MS patients with varying degrees of cognitive impairment underwent diffusion tensor imaging (DTI) and the Controlled Associated Word Test (COWAT). Fractional anisotropy (FA) of FAT and arcuate fasciculus (AF) were obtained through a supervised, atlas-based tissue segmentation and parcellation method. Phonemic and semantic fluency scores were obtained from COWAT. We ran a multivariate regression model, and partial correlation analyses adjusted for age, education, and lesion load, and corrected for multiple comparisons. False discovery rate (FDR) was used for the correction of multiple comparisons.

Results:

Bilateral FAT FA showed significant association with phonemic verbal fluency task (Left; r=0.46, p=0.0058 and right; r=0.46, p=0.0059) but not semantic fluency task and this relation remained significant after FDR correction (p=0.02 bilaterally). Although left AF showed some significant association with phonemic fluency task (r=0.38, p=0.02), this relation was insignificant after FDR correction (p=0.07).

Conclusion:

We show that bilateral FAT are correlates of phonemic verbal fluency task but not semantic in an MS cohort with cognitive impairment. This finding suggests that FAT is more specialized in lexical retrieval function as semantic fluency test encompasses all the functions except the lexical retrieval.

Introduction

Frontal aslant tract (FAT) has been recently described as a pathway connecting posterior parts of inferior frontal gyrus to the supplementary motor area (SMA) and pre-SMA regions in the superior frontal gyrus. The FAT shows significant leftward asymmetry in right-handed subjects [1,2]. Subsequent studies reveal prominent lateralization of the functions of this newly described pathway. Functional profiles of the right FAT are more heterogeneous, whereas the left FAT is responsible for language-related functions. The right FAT has been associated with automatic-voluntary dissociation of the orofacial musculature in glioma [3], depression in sports-related concussion [4], syllable insertion errors and visual memory deficits in autism spectrum disorder [5–6], visuo-constructive deficits in Alzheimer’s Disease [7]. Interestingly, right FAT connectivity is maladaptively associated with poorer speech production performance in persistent developmental stuttering [8]. Maladaptive changes of right frontal lobe are also shown in post-stroke aphasia literature where therapeutic inhibitory stimulation to right hemispheric promotes the speech recovery [9]. Left FAT dysfunction is related to verbal fluency impairment in primary progressive aphasia (PPA) [10] and MS [11]; speech initiation, verbal and phonemic verbal fluency impairments and stuttering in frontal gliomas [12–15], speech fluency in persistent developmental stuttering [16]; lexical retrieval in anaplastic oligoastrocytoma [17]; social interaction in ASD [18]; and phonological and semantic fluency in ischemic stroke [19]. Interestingly, Corivetti and colleagues further divided left FAT into anterior and posterior parts during awake intra-operative mapping in glioma patients and demonstrated the role of the posterior FAT being more in the motor-speech domains and anterior FAT in lexico-semantic domains [20]. Another study shows that the prognosis and progression of the speech impairment in non-fluent/agrammatic variant of PPA is related to left FAT connectivity [21].

In this study, we investigated the bilateral FAT in multiple sclerosis (MS) as an immune-mediated pathology that is not traditionally known to cause primary speech and language disturbances. Recent studies show that cognitive impairment, such as verbal fluency impairment in MS, is more common than previously thought [22], and cognitive impairment relates significantly to frontal white matter pathway abnormalities [23]. We specifically tested the role of the FAT in lexical retrieval as has been attempted in other disease conditions [10–15]. We utilized phonemic, semantic verbal fluency tests to reveal the specificity of the FAT role in lexical and semantic retrieval functions. We used another speech pathway with frontal projections, AF, as a control tract to determine the functional specificity of the FAT.

Methods

Our institutional review board approved the study, and informed consent was obtained from 35 right-handed English-speaking MS (26 females) patients with different degrees of cognitive impairment (none, mild, moderate, and severe), average age 41.86 years (range 18–58) (Table 1). Per the enrollment criteria of the study, MS patients with comorbid brain disease, and recent clinical depression or other psychiatric disorders were not enrolled in the study.

Table 1.

Summary of demographics, fluency scores, and diffusion tensor imaging metrics. Cognitive impairment index ranges 0 and 1 (higher is worse) and calculated based on the Minimal Assessment of Cognitive Function in Multiple Sclerosis (MACFIMS) as described somewhere else [32].

	Mean ± Standard Deviation
Gender	26 Females, 9 Males
Age	41.85 ±10.49
Education	14.68±2.37
Disease Duration	13.43±9.08
Expanded Disability Status Scale (EDSS)	3.54±2.02
Cognitive Impairment Index	0.46±0.25
Phonemic Verbal Fluency (F,A,S letters) Raw Scores	35.80±13.85
Phonemic Verbal Fluency (F,A,S letters) Percentiles	29.60±28.62
Semantic Verbal Fluency (Animals) Raw Scores	17.28±4.69
Semantic Verbal Fluency (Animals) Percentiles	22.71±20.78
T1w lesions (in mL)	4.25±3.96
T2w-FLAIR lesions (in mL)	15.93±15.20
Left Frontal Aslant Tract Fractional Anisotropy	0.36±0.02
Right Frontal Aslant Tract Fractional Anisotropy	0.36±0.02
Left Arcuate Fasciculus Fractional Anisotropy	0.37±0.03
Right Arcuate Fasciculus Fractional Anisotropy	0.37±0.03

Behavioral Assessments

Controlled oral word association test (COWAT) was used to evaluate for the phonemic (F, A, S letters) and semantic (animal category) fluencies [24]. Raw scores were obtained from the study subjects and were converted to percentiles as described elsewhere [25] to be used in the statistical analyses.

Magnetic Resonance Imaging Acquisition and Processing

Whole-brain MRI data were acquired on a Philips 3.0T Intera scanner using a SENSE receive head coil. The MRI protocol included conventional and nonconventional MRI sequences (dual-echo turbo spin echo, fluid attenuation by inversion recovery [FLAIR], and 3-dimensional T1-weighted magnetization prepared rapid acquisition with gradient echo). The T1-weighted sequence spatial resolution was 1 mm × 1 mm × 1 mm, and field-of-view is 256 mm × 256 mm. Diffusion-weighted image (DWI) data were acquired axially from the same graphically prescribed conventional MRI volumes using a single-shot multislice 2-dimensional spin-echo diffusion sensitized, and fat-suppressed echo-planar imaging (EPI) sequence, with the balanced Icosa21 tensor encoding scheme The b-factor = 1000 s/mm2, replication and echo times TR/TE = 7100/65 milliseconds, FOV = 256 mm ×256 mm, and slice thickness/gap/#slices = 3 mm/0 mm/44. The EPI phase encoding used a SENSE k-space undersampling factor of 2, with an effective k-space matrix of 128 × 128, and an image matrix after zero-filling of 256 × 256. The DWI images were converted to nifti format with dcm2niix software (https://www.nitrc.org/plugins/mwiki/index.php/dcm2nii:MainPage) and uploaded to DSIstudio (http://dsistudio.labsolver.org). DWI images were visually inspected for significant motion artifact, and eddy/motion correction was performed. The images then scaled into 1 mm isotropic, and the b-table was checked by an automatic quality control routine to ensure its accuracy [26]. Diffusion tensor imaging was then calculated.

Lesion Segmentation

The whole-brain lesion load is mapped and quantified in all patients using the coregistered multispectral dual fast spin-echo (FSE) and the FLAIR volumes. The lesion probability masks are computed in MRIcron (http://www. nitrc.org/projects/mricron/) [27]. The T1w and T2w lesion volumes are saved as binary masks to enable fusion with other multimodal volumes acquired from the same subject.

Atlas Based Tractography

DSI Studio was used to process the DWI data and obtain fractional anisotropy (FA; μ ± σ), radial, axial and mean diffusivities (AD, RD, and MD; ×10⁻³ mm² s⁻¹). Also, DSI was used to conduct atlas-based segmentation of the white matter tracts [29–30]. FAT and AF pathways are obtained from the Human Connectome Project white matter atlas (http://www.humanconnectomeproject.org/) and warped into native DTI space through the DSI studio software (Figure 1). Tracts are then edited with neuroanatomy guidance on a slice by slice basis on DTI maps and underwent thresholding with an FA value of 0.15.

Figure 1.

The frontal aslant tracts (in yellow) connecting superior frontal gyrus (SFG) to inferior frontal gyrus (IFG) are illustrated with the arcuate fasciculus (in red) on T1weighted background.

Statistical Analyses

A multivariate linear regression model was used to account for age, education, T1w, and T2w lesions. The residual DTI and clinical values obtained were obtained in to be used in the Pearson correlation analyses (partial correlation analysis). As FA is a composite value calculated from axial, radial, and mean diffusivities, only FA was used in the statistical analyses. False discovery rate (FDR) analysis of 5 % was also conducted for multiple comparison analyses, and a p-value of less than 0.05 after FDR correction was considered to indicate statistical significance. The statistical software package R (https://www.r-project.org/) was used for the data analyses.

Results

The patients’ age was 41.85 ±10.49 (as mean ± standard deviation), disease duration 13.43±9.08, average expanded disability status scale (EDSS) 3.54±2.02, average education 14.68±2.37, average T1weighted lesions 4.25±3.96 mL, T2weighted lesions lesion load 15.93±15.20 mL as summarized in Table 1 along with other demographics.

We have not observed any lateralization in the FA values of FAT and AF (Table 1). Phonemic verbal fluency test scores showed significant correlations with the left (r=0.46, p=0.0058) (Figure 2A) and right FAT FA (r=0.46, p=0.0058) (Figure 2B) and remained significant after FDR correction (p=0.02 bilaterally). Neither left FAT FA (r=0.15, p=0.38) (Figure 3A) or right FAT (r=0.25,p=0.14) (Figure 3B) showed significant association with semantic fluency task. Although the left AF FA was correlated significant with phonemic fluency task (r=0.38, p=0.026) (Figure 2C), this correlation did not survive FDR correction (p=0.07). No significant correlation was observed between phonemic fluency and right AF FA (r=0.26, p=0.13) (Figure 2D) or between semantic fluency scores and bilateral AF (left: r=0.063,p=0.72 and right: r=−0.062, p=0.72) (Figure 3C–D).

Figure 2.

Scatter plots showing the relation between phonemic fluency task and the fractional anisotropy (FA) of the bilateral frontal aslant tracts and arcuate fasciculus. Of note, the represented values are the standardized residuals out of the linear regression model adjusted for age, education, and lesion load were. Rho (R) and p-values are generated from a partial correlation.

Figure 3.

Scatter plots showing the relation between semantic fluency task and the fractional anisotropy (FA) of the bilateral frontal aslant tracts and arcuate fasciculus. Again, the represented values are the standardized residuals out of the linear regression model adjusted for age, education, and lesion load were. Rho (R) and p-values are generated from a partial correlation.

Discussion

Phonemic verbal fluency is a complex task that encompasses attention, working memory, and lexical retrieval functions and generally relies on frontal lobe functions [30]. Semantic fluency is a similar task that includes the same functions, except that it has semantic retrieval rather than lexical retrieval [25]. In this study, we showed that bilateral FAT is associated with phonemic fluency but not semantic fluency task, and this finding supports the hypothesis that FAT is more specialized in lexical retrieval function in MS as semantic fluency test encompasses all the functions except the lexical retrieval. Previous studies investigated this fiber tract in disease pathologies such as neurodegenerative [7,10] traumatic [4], developmental [5,6], infiltrative [12,15], and inflammatory [11] pathologies. Our findings align well with previous studies [10; 12–14] showing the importance of FAT in phonemic fluency tasks. In our cohort, the AF, which projects into inferior frontal gyrus, did not show a similar degree of correlation with phonemic fluency, and this finding potentially suggests that FAT is more specialized in the lexical retrieval than AF as well.

In the previous studies, right FAT was reported to be associated with mostly non-language impairments such as visual memory deficits [6], visuo-constructive deficits [7], or depressed mood [4]. Additionally, FAT showed leftward lateralization in healthy controls [1,2]. Neef and colleagues reported right FAT connectivity is associated with poorer speech production [8]. In contrary to these studies, we have not observed any leftward lateralization of the FAT, and the right FAT was found to be correlated with better phonemic fluency performance in our cohort. This finding is not surprising as some of the post-stroke aphasia studies reported the positive role of right hemispheric homologous language pathways in fluency-based tasks [31,32]. Also, another study investigating the role of FAT in verbal fluency performance in MS reported a correlation of bilateral FA of FAT and better phonemic fluency task performance [11].

We previously published a supervised atlas-based segmentation of thalamic nuclei in an MS cohort [23] and language pathways in post-stroke aphasia [33] and utilized a similar approach for the segmentation of FAT and AF in this study. Similar to deterministic fiber tractography [34], our supervised atlas-based segmentation methodology is neuroanatomy-oriented in handling white matter pathways parcellations.

The anatomy of language specific pathways versus those pathways connecting attention/executive function to the language area are unclear. Both areas, however, must contribute to verbal fluency. With further exploration of these pathways in patients with different lesions, we hope to better define pathways involved in language, pathways involved in attention/executive function, and pathways that connect executive function to language areas.

Our study is limited by small sample size, the lack of longitudinal data, and other speech and language-related tests such as naming and comprehension as well as a control group. Future studies, with larger sample sizes and clinical scores of different speech and language tests, can examine other functions of the FAT in MS.

In conclusion, the left FAT showed a high degree of specialization in performing lexical retrieval, which is essential for phonemic verbal fluency tasks across different pathologies, including MS. We can speculate that FAT can be targeted by therapeutic modalities, such as brain stimulation, to improve verbal fluency in different disease pathologies. As previously noted [10], the FAT may be one of the main speech and language pathways based on accumulating evidence.