Branched chain amino acid (BCAA) metabolites in maternal plasma are a reported risk factor for child autism spectrum disorder (Panjwani et al, 2019). No studies have simultaneously analysed maternal and cord plasma BCAA levels in relation to the risk of ADHD. The aims of this study were to examine the relationship between maternal and cord BCAA plasma levels and ADHD risk in childhood. Also, whether associations between BCAAs and ADHD can be modified by known or suspected life factors, such as child sex, race/ethnicity, preterm birth and maternal metabolic conditions.

Data for this study were taken from the mother–child dyads in the Boston Birth Cohort.* The dyads were recruited on a rolling basis, 24–72 hours postpartum from 1998 to 2015 and the children were followed up prospectively up to 2015.^†  At the time of study enrolment maternal blood samples were taken during a non-fasted state and umbilical cord blood was collected at birth. Child developmental outcomes were defined as ADHD, other developmental disabilities or neurotypical, and were based on physician diagnoses per International Classification of Diseases (ICD) Ninth or Tenth Revision (ICD-9 or ICD-10) in the child’s electronic medical records, including primary and subspecialty care. A range of maternal and child covariates were included in this analysis.^‡§ The data were stratified using factor analysis to create composite scores of maternal and cord BCAA, which were divided into tertiles (three parts), the first tertile being the reference group; the second and third tertiles were degrees of child ADHD risk in relation to levels of maternal or cord BCAA.

A total of 636 mother–child dyads with cord metabolites were included in the study; 297 children had a diagnosis of ADHD and 329 children were considered neurotypical. There was a significantly higher body mass index in the mothers of children with ADHD compared with mothers of neurotypical children (26.98 vs 25.97, p = 0.048). There was a significantly higher prevalence of diabetes (13.38% vs 7.95%, p = 0.027) and continuous smoking throughout pregnancy (12.71% vs 5.81%, p < 0.001) in mothers of children with ADHD compared with mothers with neurotypical children. Compared with children who were considered neurotypical, a significant proportion of the children who were diagnosed with ADHD were male (38.23% vs 76.92%, p < 0.001), born preterm (5.81% vs 16.72%, p < 0.001) and had low birth weight (12.54% vs 18.73%, p = 0.033).

There was a positive association between cord leucine, isoleucine and valine levels, as well as overall cord BCAA scores, and the risk of ADHD; more children with ADHD, vs neurotypical children, had cord blood amino acid levels (leucine [56.86% vs 46.79%, p = 0.012], isoleucine [56.52% vs 47.09%, p = 0.018], valine [55.85% vs 45.87%, p = 0.036]) and overall BCAA score (55.18% vs 46.79%, p = 0.036) above the median. This pattern can be seen in the stratified data by child sex and maternal obesity and diabetes. Only 481 maternal–infant pairs had maternal and cord blood data. Out of this subset there was no significant difference between the proportion of children with ADHD and neurotypical children with maternal blood leucine (50.52% vs 45.3%, p = 0.261), isoleucine (49.48% vs 44.95%, p = 0.328) and valine (49.48% vs 44.6%, p = 0.292) levels above median values and overall maternal BCAA score (49.48% vs 45.3%, p = 0.367). Logistical regression with pertinent co-variables found that there were a significantly higher odds ratio (OR) of ADHD in the third tertiles of cord leucine (OR 1.89, p = 0.006), isoleucine (OR 1.89, p = 0.005) and valine (OR 1.74, p = 0.015) compared with the first set of tertiles. There was a significantly higher risk of childhood ADHD with a cord BCAA score in the second (OR 1.63, p = 0.032) and third tertile (OR 2.01, p = 0.002). When the third tertile was adjusted for maternal BCAA score the pattern of ADHD risk was significant (OR 1.89, p = 0.02). There was no significant relationship between maternal BCAA score tertiles and child risk of ADHD (second tertile [OR 0.93, p = 0.738] and third tertile [OR 1.18, p = 0.422]), either before or after adjusting for cord BCAA score. There was also no significant relationship between maternal and cord BCAA scores on child risk of ADHD (second tertile [OR 0.82, p = 0.445] and third tertile [OR 1.04, p = 0.876]).

There were several limitations of this study. Firstly, even though the study sample was relatively large, this was only a small proportion of the total dataset from the Boston Birth Cohort due to only this portion having maternal and cord metabolite data. Secondly, the main analysis of the study focused on tertiles of BCAA scores and the authors were unable to perform more refined dose-response analyses due to the sample size constraints. Thirdly, instead of measuring absolute BCAA levels, the authors measured relative intensities. Therefore, further analysis using absolute concentrations of BCAA is required to understand the optimal levels for healthy neurodevelopment. Finally, the maternal blood samples were collected between 24–72 hours post-delivery and so may reflect stress-related changes or medication effects at the time of delivery, which could have affected protein homeostasis (Panjwani et al, 2019).

In summary, the authors noted an association between higher BCAA levels in cord blood, but not maternal blood, and greater risk of ADHD diagnosis in children. There was no significant relationship between cord and maternal BCAA and the risk of childhood ADHD. Overall, the results suggest that higher cord BCAA levels may signal an underlying mechanism that could be contributing to ADHD development. The authors suggested that further investigations are required to confirm the findings of this study and explain the mechanisms behind prenatal BCAA levels and ADHD pathophysiology.

Read more about the relationship between BCAA levels and childhood ADHD risk here

*The Boston Birth Cohort is an ongoing study at the Boston Medical Centre that monitors births within a predominantly urban, low-income, US minority population
^†Exclusion criteria for the study included: mothers with multiple gestation pregnancies, pregnancies due to in vitro fertilisation, and babies with chromosomal abnormalities or major birth defects
^‡Maternal covariates included: age at delivery, race/ethnicity, parity, smoking during pregnancy and education
^§Child covariates included: child’s sex, preterm and birthweight