White matter developmental trajectories associated with persistence and recovery of childhood stuttering

Abstract

Stuttering affects the fundamental human ability of fluent speech production, and can have a significant negative impact on an individual's psychosocial development. While the disorder affects about 5% of all preschool children, approximately 80% of them recover naturally within a few years of stuttering onset. The pathophysiology and neuroanatomical development trajectories associated with persistence and recovery of stuttering are still largely unknown. Here, the first mixed longitudinal diffusion tensor imaging (DTI) study of childhood stuttering has been reported. A total of 195 high quality DTI scans from 35 children who stutter (CWS) and 43 controls between 3 and 12 years of age were acquired, with an average of three scans per child, each collected approximately a year apart. Fractional anisotropy (FA), a measure reflecting white matter structural coherence, was analyzed voxel-wise to examine group and age-related differences using a linear mixed-effects (LME) model. Results showed that CWS exhibited decreased FA relative to controls in the left arcuate fasciculus, underlying the inferior parietal and posterior temporal areas, and the mid body of corpus callosum. Further, white matter developmental trajectories reflecting growth rate of these tract regions differentiated children with persistent stuttering from those who recovered from stuttering. Specifically, a reduction in FA growth rate (i.e., slower FA growth with age) in persistent children relative to fluent controls in the left arcuate fasciculus and corpus callosum was found, which was not evident in recovered children. These findings provide first glimpses into the possible neural mechanisms of onset, persistence, and recovery of childhood stuttering. Hum Brain Mapp 38:3345-3359, 2017. © 2017 Wiley Periodicals, Inc.

Keywords: Stuttering; brain development; language; longitudinal study; speech disorder.

Figure 1

FA reductions in children with persistent stuttering and children who recovered from stuttering relative to controls (group effect). (A, B) The left column shows the locations where significant FA reductions were found in the persistent group (blue) and the recovered group (green). Significant FA reductions in both persistent and recovered groups are indicated in red. The right column illustrates the white matter fibers passing through the regions showing FA reductions based on DTI tractography of a 9‐year‐old female control participant. (C) To illustrate the growth trajectories of each group, individual FA values in the clusters of FA reductions located in the left arcuate fasciculus and the mid body of the corpus callosum were plotted against age. White circles, blue squares and green triangles indicate individual FA values in the control, persistent and recovered groups, respectively. Data points acquired from the same participant are connected by thin solid lines. The fixed effects of no interest including sex, IQ, and socioeconomic status were removed from the FA values. Linear trend lines were added to illustrate the developmental trajectories of FA in each group (controls: gray line, persistent: blue dashed line, and recovered: green dotted line). Abbreviations: arc‐fp, arcuate fasciculus in the frontoparietal areas; arc‐t, arcuate fasciculus in the temporal lobe; cc, corpus callosum; cg, cingulum; FA, fractional anisotropy; ilf, inferior longitudinal fasciculus; L.H., left hemisphere.

Figure 2

Reductions of FA growth rate in children with persistent stuttering and children who recovered from stuttering relative to controls (group by age interactions). (A–D) The left column shows the locations where significant lower FA growth rate were found in the persistent group (blue) and the recovered group (green). Significant reductions of FA growth rate in both persistent and recovered groups are indicated in red. The right column illustrates the white matter fibers passing through the regions showing reductions of FA growth rate based on DTI tractography of a 9‐year‐old female control participant. (E) To illustrate the growth trajectories of each group, individual FA values in the speech‐motor regions showing reductions of FA growth rate were plotted against age. White circles, blue squares and green triangles indicate individual FA values in the control, persistent and recovered groups respectively. Data points acquired form the same participant are connected by thin solid lines. The fixed effects of no interest including sex, IQ and socioeconomic status were removed from the FA values. Linear trend lines were added to illustrate the developmental trajectories of FA in each group (controls: gray line, persistent: blue dashed line, and recovered: green dotted line). Abbreviations: arc‐fp, arcuate fasciculus in the frontoparietal areas; arc‐t, arcuate fasciculus in the temporal lobe; atr, anterior thalamic radiation; cb, cerebral peduncle; cc, corpus callosum; cg, cingulum; FA, fractional anisotropy; fs, frontal lobe short fibers; ilf, inferior longitudinal fasciculus; L.H., left hemisphere; str, superior thalamic radiation.