Correlation analysis of cognitive dysfunction and fluctuations in serum fibroblast growth factor 21 (FGF21) levels in the overweight group.

Background: To explore the changing characteristics of cognitive function in overweight/obese (OWO) adolescents and analyse its relationship with the level of serum fibrob-last growth factor 21 (FGF21). Methods: A total of 175 adolescents were selected and divided into a normal body mass index (BMI) group (n = 50), an overweight BMI group (n = 50), and an obese BMI group (n=75). All participants underwent assessment of anthro-pometric indicators (height, weight, waist circumference, BMI Z score). Fasting venous blood was collected to meas-ure the level of serum FGF21 (by enzyme-linked immuno-sorbent assay (ELISA)), as were metabolic parameters such as fasting plasma glucose (FPG), fasting insulin (FINS), gly-cated haemoglobin (HbA1c), and lipid profiles (TC, TG, LDL-C, HDL-C). Overall cognitive function was evaluated via the Chinese version of the Montreal Cognitive Assess-ment Foundation Scale (MoCA-B), executive function was assessed via the Wisconsin Card Sorting Test (WCST) (with a focus on analysing the number of persistent errors/PE and the number of completed classifications/CC), and working memory was evaluated via the number span test (DST). Independent sample t tests or Mann Whitney U tests were used to compare the differences between groups. Pearson or Spearman correlation analysis was used to explore the relationships between serum FGF21 levels and cognitive indicators and metabolic parameters. Multiple linear regres-sion was used to analyse the independent association between serum FGF21 and cognitive function scores (after adjusting for potential confounding factors such as age, sex, BMI Z score, and HOMA-IR). Results: Compared with those of normal individuals, the sys-to-lic blood pressure, diastolic blood pressure, fasting blood glucose, glycated haemoglobin and triglyceride levels of ado-lescents in the obese group were greater (all P<0.05). Under the consistent or inconsistent stimulation conditions of the Flanker task, there was no statistically significant differ-ence in the ACC between any two groups of adolescents. Compared with those in the normal body type group and the overweight group, the reaction time of adolescents in the obese group was prolonged (all P<0.05). In the n-back task, there was no statistically significant difference in the ACC between any two groups of adolescents. However, the response time of adolescents in the obese group in the 1-back and 2-back tasks was longer than that in the normal body type group and the overweight group (all P<0.05). Compared with those in the normal body type group, the serum FGF21 levels of adolescents in the obese group were greater (P=0.001). The results of the partial correlation analysis revealed that the reaction time of adolescents in the Flanker and n-back tasks was correlated with their BMI, body fat mass, waist circumference, waist hip ratio, FGF21 level, etc. (all P<0.05). Multiple linear regression analysis further confirmed that BMI was associated with prolonged response time in cognitively related behavioural tasks in adolescents (all P<0.05), and the level of FGF21 was correlated with the ACC in the 2-back task (P=0.001) and the response time to inconsistent stimuli (P=0.048). Conclusions: Overweight adolescents have significant cogni-tive impairment, with significantly elevated serum FGF21 lev-els, and elevated FGF21 levels are independently associated with poorer overall cognitive and executive functions.