Interhemispheric connectivity in amyotrophic lateral sclerosis: A near-infrared spectroscopy and diffusion tensor imaging study

Abstract

Purpose: Aim of the present study was to investigate potential impairment of non-motor areas in amyotrophic lateral sclerosis (ALS) using near-infrared spectroscopy (NIRS) and diffusion tensor imaging (DTI). In particular, we evaluated whether homotopic resting-state functional connectivity (rs-FC) of non-motor associated cortical areas correlates with clinical parameters and disease-specific degeneration of the corpus callosum (CC) in ALS.

Material and methods: Interhemispheric homotopic rs-FC was assessed in 31 patients and 30 healthy controls (HCs) for 8 cortical sites, from prefrontal to occipital cortex, using NIRS. DTI was performed in a subgroup of 21 patients. All patients were evaluated for cognitive dysfunction in the executive, memory, and visuospatial domains.

Results: ALS patients displayed an altered spatial pattern of correlation between homotopic rs-FC values when compared to HCs (p = 0.000013). In patients without executive dysfunction a strong correlation existed between the rate of motor decline and homotopic rs-FC of the anterior temporal lobes (ATLs) (ρ = - 0.85, p = 0.0004). Furthermore, antero-temporal homotopic rs-FC correlated with fractional anisotropy in the central corpus callosum (CC), corticospinal tracts (CSTs), and forceps minor as determined by DTI (p < 0.05).

Conclusions: The present study further supports involvement of non-motor areas in ALS. Our results render homotopic rs-FC as assessed by NIRS a potential clinical marker for disease progression rate in ALS patients without executive dysfunction and a potential anatomical marker for ALS-specific degeneration of the CC and CSTs.

Keywords: AC, anterior commissure; ALS, amyotrophic lateral sclerosis; ALS-EX, ALS with executive impairment; ALS-NECI, ALS with non-executive cognitive impairment; ALSFRS-R, revised ALS functional rating scale; ATL, anterior temporal lobe; Amyotrophic lateral sclerosis; CC, corpus callosum; CST, corticospinal tract; Corpus callosum; DD, disease duration; DPR, disease progression rate; DTI, diffusion tensor imaging; Diffusion tensor imaging; FA, fractional anisotropy; FTD, frontotemporal dementia; HC, healthy control; Hb, hemoglobin; Interhemispheric connectivity; NIRS, near-infrared spectroscopy; Near-infrared spectroscopy; TBSS, tract based spatial statistics; WM, white matter; fMRI, functional magnetic resonance imaging; pALS, pure ALS no cognitive impairment; rs-FC, resting-state functional connectivity; rs-fNIRS, resting-state functional NIRS.

Fig. 1

Homotopic resting-state functional connectivity (rs-FC) in ALS: box whisker plot of homotopic rs-FC for measurement sites H1 to H8 (median, 25th and 75th percentile, extent of distribution, and outliers) and measurement sites (spheres) on the left hemisphere relative to the MNI152T1 brain template as obtained by probabilistic mapping in all 31 patients. Distributions are given in alternating colors orange and yellow to facilitate differentiation between adjacent measurement sites. Mean MNI coordinates were (− 30.3, 65.1, 4.3), (− 46.7, 46.5, − 0.3), (− 53.1, 17.8, − 2.9), (− 63.6, − 10.9, − 8.6), (− 65.1, − 38.5, − 7.7), (− 55.6, − 67.0, − 9.3), (− 41.2, − 88.5, − 10.5), and (− 24.7, − 101.9, − 9.5) for sites H1 to H8, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Inter-individual correlation matrices of homotopic connectivities at different measurement sites. Statistically significant correlations are marked by a red box and given in a circular connectivity plot, schematically reflecting anatomical relationships. Whereas in healthy controls a significant correlation only existed between homotopic resting-state functional connectivities (rs-FCs) of the occipital and posterior temporal lobes, ALS patients displayed a significant correlation between frontal and temporo-occipital homotopic rs-FCs. Furthermore, homotopic rs-FC of the anterior temporal lobes (ATLs) significantly correlated with orbitofrontal and posterior-temporal homotopic rs-FC in ALS patients. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Color-coded representation of rank correlation ρ between homotopic resting-state functional connectivity (rs-FC) and clinical parameters for sites H1 to H8 (from top to bottom). Homotopic connections are given in grayscale or color to highlight statistical significance of the corresponding ρ values. No significant correlation existed between rs-FC and patients' functional (motor) status (left). For H4, rs-FC significantly correlated with patients' rate of motor decline (right).

Fig. 4

Mean homotopic resting-state functional connectivity (rs-FC) in healthy controls (HC, left) and rank correlation ρ between rs-FC and disease progression rate (DPR) in pure ALS (middle, red line denotes statistical significance) for sites H1 to H8. Complementary spatial patterns of rs-FC and ρ (r = − 0.78, p = 0.02), especially for sites H5 to H8, here may suggest involvement of a larger scale network in interhemispheric rs-FC. As for the ALS group as a whole, a statistically significant correlation existed only for H4. For this site a linear model (red line and 95% confidence interval in the scatter plot on the right) was fitted to the data obtained in the pALS group (red dots). Whereas data for the ALS-NECI group (blue dots) were compatible with this model, 4 out of 11 data points fell outside the 95% prediction interval (grey) in the ALS-EX group (green dots) suggesting a different disease trajectory in patients with executive dysfunction. This assumption was affirmed by Quade's rank analysis of covariance (see Section 3.3.). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Color-coded representation of voxel-wise correlation between fractional anisotropy (FA) and homotopic resting-state functional connectivity (rs-FC) of the anterior temporal lobes (ATLs, H4) embedded into the MNI152T1 brain template (top row). Only those segments of the fiber tract skeleton are shown for which a statistically significant correlation existed. These were the corticospinal tracts (CSTs) at the level of mesencephalon/diencephalon and centrum semiovale, the central corpus callosum (CC), white matter tracts extending from the central CC to the primary and pre-motor cortex, and the forceps minor. Extent of the lower CST involvement is also given in coronal (top middle) and transverse (top right) sectioning. Mean FA of this area versus homotopic rs-FC of the ATLs is given as scatter plot along with a regression line (bottom left). For comparison statistically significant correlations between rs-FC and FA are also given for the orbitofrontal cortices (H3, bottom middle and bottom right, see Section 3.4.).