580 - Brain DTI-ALPS index Association with Cognitive Outcome in Early School–Age Children Born With Extremely Low Birth Weight Who Exposed Systemic Inflammation in Critical Period

Publication Number: 580.4150

SaeYun Kim, College of Medicine, The Catholic University of Korea, Seoul St Mary's Hospital, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Huijin Song, Seoul National Uniersity Hospital, Jongno-gu, Seoul-t'ukpyolsi, Republic of Korea; Ee-Kyung Kim, Seoul National University Children's Hospital, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Jung-Eun Cheon, Seoul National University Hospital, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Bung Nyun Kim, Seoul National University, College of medicine, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Seung Han Shin, Seoul National University Hospital, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Han-Suk Kim, Seoul National University, College of medicine, Seoul, Seoul-t'ukpyolsi, Republic of Korea; Ji Won Koh, St. Mary/s hospital, Cheonnan, Ch'ungch'ong-namdo, Republic of Korea

Background: The glymphatic system, flushing the waste products out of the brain into the systemic circulation, plays a pivotal role in neuroinflammation as well as brain homeostasis.

Objective: This study aims to evaluate the glymphatic system using diffusion tensor imaging along the perivascular space (DTI-APLS) and cognitive outcomes of early school-aged preterm-born children with or without systemic inflammation (SI) in critical period compared to full-term (FT) control.

Design/Methods: Preterm born children at gestational age < 32 weeks and FT-born children as controls were included in this prospective cohort study. Brain MRI with DTI and neuropsychological tests were performed aged 6 to 8 years.

Results: Twenty-five children in SI, 15 children in noSI, and 20 FT children were recruited. The mean full scale Full-Scale Intelligence Quotient (FSIQ) of SI group was 85.3(15.3), that of noSI group was 93.5(19.1), and that of FT group was 106.1(14.2). The mean executife function (EF) composite score of SI group was 40.7 (8.4), that of noSI group was 46.5 (6.5), and that of FT group was 51.2 (7.0). For both scores, significant mean difference were only observed between SI and FT group (both P< 0.001). The DTI-ALPS index were 1.45 (0.106) for SI group, 1.59 (0.179) for noSI group, and 1.57 (0.177) for FT group. There were significant mean differences between SI and other groups (both P<.05). Multiple cortical and subcortical brain volumes differed significantly in three groups: left pars orbitalis, right pars orbitalis, right isthmus of cingulate gyrus, and left/right caudate nucleus. Mediation analysis within the combined group suggested that the DTI-ALPS index fully mediated the relationship between the volume of left caudate nucleus and executive function (c’=0.223, P=0.096), explaining 32.71% of the effect. And for right caudate nucleus, the mediating effect size was 36.09% (c’=0.185, P=0.162). Additionally, mediation analysis among the SI group showed that the DTI-ALPS index partially mediated the association predicting EF through the volume of left pars orbitalis, explaining 41.16% of the effect (c’=0.358, P=0.033).

Conclusion(s): DTI-ALPS index demonstrated a more severe reduction of glymphatic activity in SI group. Reserved caudate nuclei predict the better EF mediated by DTI-ALPS index. However, in children who were exposed to SI in critical period, the volume of left pars orbitalis is associated the EF via mediating DTI-ALPS index. Our findings suggest that the glymphatic system rather than the brain regional volume could serve as a EF predictor specifically in preterm children.

Table 1. Neurodevelopmental Outcomes and DTI-ALPS indices of the Three Groups
*P <.05, **P <.01, ***P <.001
aP-values were calculated using general linear model (1-way analysis of variance) with adjustment SES, pairwise comparison was conducted by post hoc Bonferroni test for multiple comparison. bThere were 22 in the SI group, 13 children in the noSI group, and 19 children in the FT group who completed EF tests.
DTI-ALPS index, diffusivity along the perivascular space; EF, executive function; FSIQ, Full-Scale Intelligence Quotient; FT, control children who were born at full-term; noSI, children exposed no systemic inflammation in neonatal period, SES, socioeconomic status, PRI, perceptual reasoning index; PSI, processing speed index; VCI, verbal comprehension index; WMI, working memory index; SI, children exposed systemic inflammation in neonatal period

Table 2. Mediation analysis of the brain volume as predictor of executive function contrasted with DTI-APLS index as mediating variables
*P <0.05, ** P<0.01, ***P <0.001

DTI-ALPS, Diffusion tensor image analysis along the perivascular space; EF, executive function; ROI, region of interest;


Figure 1. Mediation analysis of the brain volume as predictor of EF contrasted with DTI-APLS index as mediating variables
Mediation analyses suggested that DTI-ALPS index mediated the relationship between left caudate nucleus volume and the EF composite score (total effect β =0.331, 95% CI 1.076-8.169, P=0.011; direct effect β =0.223, 95% CI -0.550-6.766, P =0.096; indirect effect β =0.108, 95% CI 0.061-6.199, P=0.082) with mediating effect size of 32.7% (A), Right caudate nucleus volume predict the EF composite score, mediating via DTI-ALPS index (total effect β =0.289, 95% CI 0.453-7.927, P=0.028; direct effect β =0.185, 95% CI -1.081-6.440, P =0.162; indirect effect β =0.104, 95% CI -0.186-3.207, P=0.081) with explaining 36.1% of the effect (B). Furthemore, mediation analysis of SI group suggested that DTI-ALPS index mediated the relationship between left pars orbitalis and the EF composite score (total effect β =0.608, 95% CI 5.589-19.696, P<0.001; direct effect β =0.358, 95% CI 0.5839-14.3, P =0.033; indirect effect β =0.250, 95% CI 0.1477-10.253, P=0.044) with mediating effect size of 41.16% (C).