Surface-Based Morphometry of Cortical Thickness and Surface Area Associated with Heschl's Gyri Duplications in 430 Healthy Volunteers

Abstract

We applied Surface-Based Morphometry to assess the variations in cortical thickness (CT) and cortical surface area (CSA) in relation to the occurrence of Heschl's gyrus (HG) duplications in each hemisphere. 430 healthy brains that had previously been classified as having a single HG, Common Stem Duplication (CSD) or Complete Posterior Duplication (CPD) in each hemisphere were analyzed. To optimally align the HG area across the different groups of gyrification, we computed a specific surface-based template composed of 40 individuals with a symmetrical HG gyrification pattern (20 single HG, 10 CPD, 10 CSD). After normalizing the 430 participants' T1 images to this specific template, we separately compared the groups constituted of participants with a single HG, CPD, and CSD in each hemisphere. The occurrence of a duplication in either hemisphere was associated with an increase in CT posterior to the primary auditory cortex. This may be the neural support of expertise or great abilities in either speech or music processing domains that were related with duplications by previous studies. A decrease in CSA in the planum temporale was detected in cases with duplication in the left hemisphere. In the right hemisphere, a medial decrease in CSA and a lateral increase in CSA were present in HG when a CPD occurred together with an increase in CSA in the depth of the superior temporal sulcus (STS) in CSD compared to a single HG. These variations associated with duplication might be related to the functions that they process jointly within each hemisphere: temporal and speech processing in the left and spectral and music processing in the right.

Keywords: FreeSurfer; Heschl's gyrus; MRI; anatomy; cortical surface area; cortical thickness; gyrification; hemispheric specialization.

Figure 1

An example of manual corrections applied in case of wrong estimation of pial and/or white cortical surfaces. In this individual, the FreeSurfer software did not correctly estimate both white and pial surfaces in the upper part of the temporal pole. Manually adding checkpoints to the white matter surface results in better surface estimation after a second extraction process (yellow line, white surface; red line, pial surface; green dots, checkpoints).

Figure 2

Variability maps of the curvature (CURV) parameter in the left (A) and right (B) hemispheres. The color scale represents the variance of the CURV parameters computed for each HG gyrification pattern in each hemisphere. The zoomed part is centered in the temporal transverse region where the HG duplication is located.

Figure 3

Significant variations of Cortical Thickness (CT) and Cortical Surface Area (CSA) in the “CPD–single HG”, “CSD–single HG” and “CPD–CSD” contrasts in the left and right hemispheres showing the spatial overlap of CT and CSA variations. (A). Significant variations in cortical thickness. (B). Significant variations in cortical surface area, the lateral view of the right temporal lobe is provided to illustrate the increase in CSA located in the mid-part of the STS in the “CSD–single HG” comparison. Significant variations are superimposed on the mean patch of each corresponding group (“CPD–single” and “CPD–CSD” contrasts: mean patch of 48 left CPD, mean patch of 96 right CPD; “CSD–single HG” contrast: mean patch of 98 left CSD and 95 right CSD; CSA: cortical surface area; CT, cortical thickness; CPD, common posterior duplication; CSD, common stem duplication; HG, Heschl's gyrus; p < 0.05 FDR corrected for multiple comparison; cluster extension threshold, 20 mm²; hot color indicates a positive variation; cold color, indicates a negative variation; light gray, gyri; dark gray, sulci).

Figure 4

Morphology and relationships between variations in CT and CSA. Projection of the SBM results on the volumetric means of each group of individuals defined according to their gyrification pattern. Results of “CPD-single” and “CPD-CSD” contrasts superposed on the mean images of 48 left CPD or 96 CPD for the right side. Results of “CSD-single HG” contrasts superimposed on the mean image of 98 CSD for the left hemisphere or 95 CSD for the right hemisphere. The yellow scale corresponds to increased values, and the blue scale to decreased values. From left to right a sagittal, a coronal, and an axial slice are presented passing through the significant clusters.