Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neuroimaging Predictors of Language Outcome in Moderate to Late Preterm and Early Term Infants
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster A97 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Peggy Hiu Ying Chan1,2, Patrick Chun Man Wong1, Winnie Chiu Wing Chu3, Hugh Simon Hung San Lam2, Xiujuan Geng1; 1Brain and Mind Institute, The Chinese University of Hong Kong, 2Department of Paediatrics, The Chinese University of Hong Kong, 3Department of Imaging and Interventional Radiology, the Chinese University of Hong Kong
Most preterm infants are born at moderate to late preterm (MLPT) gestations. Studies have shown that MLPT infants and even early-term (ET) infants are at risk for poorer language outcomes compared to their full-term peers. We hypothesize that a reduction in volume and thickness of the left language network encompassing language regions in the frontal and temporal lobes is most predictive of adverse language outcome in MLPT and ET infants. Alternatively, because preterm birth affects cortical structures more globally, the relationship between the brain and outcomes might be more diffuse. This study aims to identify neuroimaging features that predict MLPT and ET infants’ future language ability. Forty-nine healthy Chinese MLPT and ET infants (mean (range) gestation age at birth: 36 (32- 39) weeks) were recruited. Subjects were scanned at approximate two months of chronological age and during natural sleep without sedation on a Siemens Magnetom Prisma 3T scanner. High-resolution T1- and T2-weighted images were acquired by a 16-channel paediatric coil. Regional volumes for 90 ROIs were obtained by using AAL, and the cortical thickness and surface area on each vertex was obtained by using InfantFS. Infant’s language ability, including expressive and receptive language, and language composite scores were assessed at 12 months old using Bayley Scales of Infant and Toddler Development III. In ROI-based analysis, four regions showed negative correlations between regional volume and language composite scores: left parahippocampal (r2= 0.2, p=0.002), superior frontal (r2= 0.19, p=0.01), middle frontal (r2= 0.18, p=0.02), and fusiform (r2= 0.14, p=0.05). However, none of these regions reach significance after correction for multiple comparisons. GLM models with and without neural data were built to predict the language composite score. When we compared the non-neural model (Model 1) which controls for sex, gestational age at birth, and chronological age at scan with the neural model (Model 2) which additionally controls for parahippocampal volume, the neural model has a significantly higher model fit (Model 1: r2= 0.06; Model 2: r2=0.2; F=7.43, p=0.009). In cortical surface-based analysis, we found that the right inferior parietal and inferior temporal, right precentral, and right orbital frontal cortices showed negative correlations between thickness and language composite scores. The right lingual cortex showed positive correlations between thickness and language composite scores. The right orbital frontal cortex showed negative correlations between thickness and expressive scaled scores. The left lateral occipital cortex showed positive correlations between area, expressive scaled scores, and language composite scores. The left supramarginal cortex showed negative correlations between area and expressive scaled scores. The right middle temporal and right precentral and superior frontal regions showed negative correlations between area and receptive scaled scores. (All corrected p<0.05). We identified several distinct ROIs and cortical features in the left language network are associated with infants' future language abilities. Future research with a larger sample size would be needed to further investigate the precise neurological underpinnings of preterm infants' language problems. For now, our results suggest that MRI is a promising avenue for predicting future language outcomes beyond non-neural measures.
Topic Areas: Language Development/Acquisition,