Article Figures & Data
Figures
- Figure 1.
A, Depiction of the structural lesions in native space using a T1-w axial image. The lesions can be seen as hypo-intense areas and are highlighted with a red circle. Neurologic convention is used. B, Individual DMNs for each patient, obtained using group spatial ICA on the resting state data (see Materials and Methods). Individual data are in z-scores (p < 0.05, uncorrected) and presented over each patient’s T1 registered to a common space (four-year-old MRI brain template obtained from the Neurodevelopmental MRI Database; see Materials and Methods). This was done as a demonstration of data quality, given that propofol can reduce connectivity in cortical areas of significance for the current study (Boveroux et al., 2010). On the right, the average DMN for all patients is depicted, using t values (p < 0.01, uncorrected) over a canonical four-year-old MRI brain template. C, Probability templates for the left arcuate (anterior, posterior, and long segment) were extracted from the Tractotron atlas (Thiebaut de Schotten et al., 2011; Rojkova et al., 2016), thresholded at a 80% (only voxels having a 80% probability of being part of the AF according to the atlas are shown) and registered to a common space (the four-year-old MRI brain template obtained from the Neurodevelopmental MRI Database). For each patient, here we depict the arcuate probabilistic atlas (in red), the lesion (in blue), and the overlap between them (in pink) over a sagittal slice of the left hemisphere. Neurologic convention is used. L, left hemisphere.
- Figure 2.
Cognitive outcomes. A, Results of the PPVT-III test for each patient. The bars show the mean and SD. B, Means and SDs barplots of the composite scores from the NEPSY-II. The gray slots indicate normative scores range (8–12).
- Figure 3.
Means and SDs boxplots in the different subtests of the NEPSY-II (CI, comprehension of instruction; BPNI, body-part naming and identification; WG, word generation; PP, phonological processing; SN, speeded naming; MD, memory for designs; NM, narrative memory; SR, sentence repetition; BC, block construction; DC, design copying; IHP, imitating hand positions; VMP, visuomotor precision) grouped in domains (green, language; orange, memory and learning; purple, visuospatial processing; blue, sensorimotor processing). The gray slot indicates normative scores range (8–12).
- Figure 4.
Expressive language measures. Means and SDs boxplots for children with PAIS and controls, for proximity and complexity scores. The graphs represent individual scores for the proximity score (A) for the pMLU scores (B) and from the referent group (N = 50), the group mean value (in black) and the score from each patient (in red). Maximum score for the pMLU score (complexity of the segmental structure of the words (consonants) is 8.31; maximum score for proximity score is 1. C, Utterance complexity in spontaneous speech with the MLU in words and the MLU of the five longest utterances (max-L) for each patient. The gray slots indicate the scores range from an age-matched control group of N = 50 Spanish-speaking children at age 4 (Vázquez and Alonso, 2007).
- Figure 5.
fMRI activation in the passive listening task for the language versus control contrast in patients. Two significant clusters with enhanced fMRI activity are found over the right inferior frontal and middle temporal gyri. Color-coded results are shown in yellow and at a p < 0.001, in red at p < 0.005, in green at p < 0.01, and in blue at p < 0.05 uncorrected threshold with five voxels of cluster extent, over the Neurodevelopmental MRI Database four-year-old template from Richards et al. (2016). Neurologic convention is used.
- Figure 6.
Deterministic tractography. DTI in vivo dissections of the four tracts of interest AF/ILF/UF/IFOF shown in native space on the T1-w images in each patient (A) and four controls (B). Neurologic convention is used. The two-ROI approach revealed that for all the children with a left PAIS (A), the three segments of the AF were well preserved in the right hemisphere. Meanwhile, we were not able to dissect the left AF in most of them (L4 and L6 presented the posterior segment only, whereas L1 presented the posterior and the long segment but not the anterior one). We did not observe any substantial damage for the ventral WM pathways (ILF, IFOF, and UF) in either hemisphere. By contrast, the three segments of the left AF were always present in control children (B). However, while the anterior and posterior segments of the right AF were always present in these children, we were not able to reconstruct the long segment of the AF in three controls.
- Figure 7.
A, Correlation between DTI properties extracted from the manual reconstruction of the arcuate fasciculus bilaterally and productive aspects of language. The positive correlation between the lateralization index for the volume of the complete right AF and the maximum length of utterance (L-max) is shown. Lateralization index: values closer to –1 mean lateralization to the left, values around 0 represent a symmetrical distribution, values closer to 1 mean lateralization to the right. This result evidences that the more lateralized to the right is the integrity of the AF, the better the productive aspects of language. B, Correlation between rs-fMRI laterality indexes and productive aspects of language. The scatterplot shows the positive Spearman’s correlation between the lateralization index for the STG-IFG laterality index and the MLU. This result evidences that the more lateralized to the right the intrinsic connectivity between the regions connected by the AF, the better the productive aspects of language. A 3D render of the four-year-old template is displayed with the ROIs used in this analysis. Correlation index and p values are displayed.
Tables
Patientcode Age at scan(years) Gender Gestational age at birth(weeks) Birth weight(g) Clinical debut(hours of age) Age of MRI diagnostic (d) Vascularterritory Stroke volume at birth (ml) Motor impairment(hemiplegia) Epilepsya L1 4 M 41 3160 Clonic seizures at 12 h 4 M2 L 18890 No No L2 3.5 F 41 2960 Clonic seizures at 26 h 10 M4 L 47428 Yes No L3 4 M 40 3560 Clonic seizures at 48 h 20 M1
Post-bifurcation L17588 No No L4 4 F 41 2600 Clonic seizures at 18 h 4 M2 sup L 27882 No No L5 4 M 39 3340 Clonic seizures at 24 h 5 M1
Post-bifurcation L36512 No No L6 3.5 M 40 3025 Clonic seizures at 41 h 5 M1
Post-bifurcation L23509 Yes No ↵a Epilepsy = at least two recurrent and unprovoked seizures.
- Table 2.
Phonological complexity and proximity scores for the children with PAIS (single values) and for the control group (means and SDs)
Phonological complexity Phonological proximity L1 5.81* 0.70* L2 7.56 0.90 L3 8.03 0.97 L4 7.59 0.91 L5 7.75 0.93 L6 7.91 0.95 Controls 7.44 (0.81) 0.89 (0.09) An asterisk indicates scores significantly different from the control group (p < 0.05) resulted from the modified t test.
- Table 3.
Spontaneous utterance performance (MLU and L-max scores) for the children with PAIS and for the control group (means and SDs)
MLU L-max L1 1.74* 4* L2 4.2 9.4 L3 4.1 9.8 L4 4.1 9 L5 4.2 9.2 L6 3.6 7.8 Controls 3.66 (0.96) 8.2 (2.16) An asterisk indicates scores significantly different from the control group (p < 0.05) resulted from the modified t test.
- Table 4.
Precise lesion location for each child with PAIS with the corresponding percentage over the total of the individual lesion size
% of GM loss % of WM loss Patient code Lesion location Left Right Left Right L1 L precentral gyrus (39%)
L IFG (19%)
L postcentral gyrus (16%)
L MFG (16%)0.37 . 2.5 . L2 L SMG (56%)
L AG (33%)
L STG (10%)4.4 . 7.7 . L3 L AG (40%)
L MOG (25%)
L SMG (12%)7.8 . 10.3 . L4 L postcentral gyrus (53%)
L SMG (23%)
L precentral gyrus (20%)4.3 . 0.8 . L5 L AG (28%)
L STG (26%)
L SMG (19%)
L MTG (8%)12.1 . 7.4 . L6 L IFG (39%)
L MFG (33%)
L precentral gyrus (24%)0.6 . 1.7 . The percentage of WM and GM loss is also presented for each patient. MFG, middle frontal gyrus; SMG, supramarginal gyrus; AG, angular gyrus; MTG, middle temporal gyrus; MOG, middle occipital gyrus.
In this issue
