Motor callosal disconnection in early relapsing-remitting multiple sclerosis

Abstract

In relapsing-remitting multiple sclerosis (RRMS) the corpus callosum (CC) is often and early affected by macroscopic lesions when investigated by conventional MRI. We sought to determine to which extent microstructural and effective disconnection of the CC are already present in RRMS patients at the earliest stages of the disease prior to evidence of macroscopic CC lesion. We compared 16 very early RRMS patients (median expanded disability status scale (EDSS), 1.5; range, 0-2.0) to an age-matched group of healthy controls and focused analysis to the motor CC, i.e. that part of the CC relaying interhemispheric motor information. A combined functional magnetic resonance imaging/diffusion tensor imaging fiber-tracking procedure was applied to identify the callosal motor fibers (CMFs) connecting the hand areas of the primary motor cortices of the two hemispheres. Fractional anisotropy (FA) within the motor CC (FA-CC) assessed the CMF microstructural integrity. Bifocal paired transcranial magnetic stimulation (TMS) tested short-interval interhemispheric inhibition (S-IHI), an established measure of CMF effective connectivity. FA-CC and S-IHI were significantly reduced in early RRMS compared to healthy controls. Furthermore, a significant linear correlation between microstructure (FA-CC) and function (S-IHI) in the controls was broken down in the patients. These abnormalities were obtained in the absence of macroscopic CMF lesion in conventional MRI, and whilst motor hand/arm function in the nine-hole-peg test and corticospinal conduction time were normal. Findings suggest that reductions in FA and S-IHI may serve as surrogate markers of motor callosal disconnection at the earliest stages of RRMS prior to development of macroscopic lesion.

Figure 1

FA of the hand CMFs within the CC (FA‐CC) is significantly reduced in RRMS patients (black bar) compared to healthy subjects (white bar, *P = 0.005). Data are means + 1 S.E.M.

Figure 2

(A) S‐IHI (y‐axis) is given as (1 − conditioned MEP/unconditioned MEP) × 100% and plotted against intensity (in % of the resting motor threshold, RMT) of the conditioning stimulus (x‐axis). All data (white circles: healthy controls; black circles: RRMS patients) are group means ± 1 S.E.M. Note that RRMS patients have a significantly reduced mean S‐IHI intensity curve compared to healthy controls (P = 0.047). (B) In contrast, the mean L‐IHI intensity curve is not different between RRMS patients and healthy controls.

Figure 3

Macrostructural (MRI, DTI), microstructural (FA‐CC) and effective connectivity (S‐IHI) of callosal motor fibers in one representative healthy control (upper row of diagrams) and one representative RRMS patient (lower row of diagrams). Left panels: MRI shows individual hand CMFs as determined by combined fMRI/DTI fiber‐tractography in the coronal plane. Middle panels: the corresponding FA‐CC of the hand CMFs (red lines) are shown in relation to the mean ± 95% CI (black lines and grey areas) of the healthy control group. Right panels: corresponding S‐IHI (1 − conditioned MEP/unconditioned MEP) × 100%, y‐axis) is plotted against S1 intensity (in %RMT, x‐axis) in relation to the mean S‐IHI intensity curve (white circles) and 95% CI (grey area) of the healthy control group. Note normal macroscopic appearance of CMF with DTI tractography, but reduced FA‐CC and S‐IHI in the RRMS patient.

Figure 4

Regression analysis of FA‐CC (x‐axis) with S‐IHI measured at a conditioning stimulus intensity of 140% of resting motor threshold (y‐axis) revealed a significant direct linear correlation for the healthy subjects (white circles, regression line plotted, r = 0.65, P = 0.021). This correlation was absent in the RRMS patients (black circles). Coordinates of group means ± 1 S.E.M. are indicated by colored circles (blue: healthy subjects; red: RRMS patients). Note that the group means of the RRMS patients are shifted towards less FA‐CC and less S‐IHI.