550 - Left lateralization of event-related desynchrony during verb generation positively correlates with expressive vocabulary in extremely preterm children

Poster Number: 550
Publication Number: 550.1457

Background: Children born extremely preterm (EPT, < 28 weeks gestation) are at increased risk for language deficits versus term children (TC, > 37 weeks gestation). Neural mechanisms underlying language development in prematurity are not well understood. Beta-band event-related desynchrony (ERD) indexes neural engagement for tasks. In TC, beta ERD during verb generation becomes increasingly left-lateralized with age. Previously, we reported increased functional connectivity in EPT children versus TC during verb generation in magnetoencephalography (MEG). This increased connectivity positively associated with Expressive Vocabulary Test (EVT2) scores in the EPT group specifically. In the current experiment, we assess lateralization of beta-band ERD in EPT and TC and assess correlations with EVT2 performance for EPT.

Objective: To investigate differences in ERD between EPT and TC and the relationship between ERD lateralization and language scores.

Design/Methods: We studied 117 4- to 6-year-old children (65 TC, 37 EPT with no history of language delay, and 15 preterm with a history of language delay [EPT-HLD]) with multimodal neuroimaging and standardized language assessments. MEG was acquired during a verb generation task (275 channel CTF system at 1200 Hz) and registered to each participant’s 3D T1 MRI. Sources were modeled in Fieldtrip using a 10mm grid in MNI space. Data were filtered to assess beta-band power from 700 to 1200 ms after stimulus onset. For each participant, the number of left versus right hemisphere dipoles showing significant (p < 0.05) beta-band ERD was summarized in a laterality index (LI). Positive values indicate left lateralization, and negative values indicate right. Group differences were investigated using ANOVA. Relationships between ERD lateralization and language scores were investigated using correlations across and within groups.

Results: There were no significant differences in age, sex, race, ethnicity, family income, or parental education between groups (Table 1). There were no significant differences in the quantity (magnitude) of suprathreshold dipoles or lateralization of ERD between groups. There were group differences in language scores, including EVT2 (p < 0.01). Post-hoc analyses indicated this was due to the EPT-HLD group having lower scores than the EPT or TC groups. Within the EPT group, ERD LI positively correlated with EVT2 (r=0.428, p< 0.01, Figure 1).

Conclusion(s): There were no significant differences between TC, EPT, or EPT-HLD in magnitude or laterality of ERD. Within the EPT group, a more “mature” left-lateralized ERD correlated with better expressive language scores.