Prediction of white matter integrity pattern in overweight and obese adults
Author
dc.contributor.author
Reyes García, Sussanne
Author
dc.contributor.author
Peirano Campos, Patricio
Author
dc.contributor.author
Lozoff, Betsy
Author
dc.contributor.author
Algarin Crespo, Cecilia
Admission date
dc.date.accessioned
2020-10-12T21:19:37Z
Available date
dc.date.available
2020-10-12T21:19:37Z
Publication date
dc.date.issued
2020
Cita de ítem
dc.identifier.citation
Proceedings of the Nutrition Society (2020), 79 (OCE2), E189
es_ES
Identifier
dc.identifier.other
10.1017/S0029665120001378
Identifier
dc.identifier.uri
https://repositorio.uchile.cl/handle/2250/177082
Abstract
dc.description.abstract
Introduction
Obesity has been associated with lower white matter integrity (WMI) in limbic brain regions, including the fornix. Both early decrease
of WMI in the fornix (WMIf) and midlife obesity have been related to dementia incidence with advancing age. No studies have
explored early cognitive predictors of WMIf in overweight-obese (OO) adults. Aim of this study was to compare OO and normalweight
(NW) participants with respect to (a) WMIf in adulthood and (b) the relationship between cognitive performance at school-age
and in adolescence with WMIf in adulthood.
Methods
Participants were part of a cohort followed since infancy who underwent magnetic resonance imaging studies in adulthood (22.3 ± 1.3
years). Diffusion tensor imaging was performed and Tract Based Spatial Statistics (TBSS) was used to obtain fractional anisotropy
(FA) skeleton; increased FA relates to greater WMI. A mask for the fornix was created (JHU-ICBM DTI-81 Atlas) and then used to
extract the average FA for each individual. Participants also performed neurocognitive tasks: (a) school-age (10.3 ± 1.0 years): the trail
making test comprises two conditions and time difference between conditions reflects cognitive flexibility; (b) adolescence (15.6 ± 0.5
years): incentive task that test the effect of incentives (reward, loss avoidance or neutral) on inhibitory control performance (correct
responses latency). In adulthood, BMI was categorized as NW (≥ 18.5 to < 25.0 kg/m2) and OO (≥ 25.0 kg/m2) groups. A t-test and
univariate GLM were conducted. Analysis were adjusted by sex and age-specific BMI z-scores.
Results
Participants were 27 NW (41% female) and 41 OO (49% female). Compared to NW, OO participants showed decreased FA in the
fornix (0.585 vs. 0.618, p < 0.05), i.e. lower WMIf. Differences were apparent in the relationship between cognitive flexibility at
school-age (F = 2.9, p = 0.06) and loss avoidance latency in adolescence (F = 3.5, p < 0.05) with FA in the fornix in adulthood.
Increased cognitive flexibility at school-age (β = 0.335, p < 0.05) and decreased loss avoidance latency in adolescence (β = -0.581,
p < 0.001) were related to higher FA in the fornix in OO adults. No relationship resulted significant in NW adults.
Discussion
Performance in neurocognitive tasks at earlier developmental stages were related with WMIf only in OO adults, group characterized
by decreased WMIf. Our results provide evidence regarding specific neurocognitive tasks with predictive value for WMIf alterations.
Further, they could contribute to the understanding of neural mechanisms underlying obesity and also provide insight relative to neurodegenerative
risk with advancing age.