Maternal Licorice Consumption During Pregnancy and Pubertal, Cognitive, and Psychiatric Outcomes in Children
Katri Räikkönen Silja Martikainen Anu-Katriina Pesonen Jari Lahti Kati Heinonen Riikka Pyhälä Marius Lahti Soile Tuovinen Karoliina Wehkalampi Sara Sammallahti ... Show more
Am J Epidemiol (2017) 1-12. DOI: https://doi.org/10.1093/aje/kww172, 03 February 2017
- Note: Most of the references for Attention Deficit Disorder in VitaminDWiki concern low Magnesium
- ADHD and Vitamin D Deficiency
- Licorice During Pregnancy Tied to Problems in Children New York Times description of the study
Vitamin D is not mentioned
- "Is ADD still a diagnosis?
The term was formally changed in 1994 in the new Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV), to "ADHD predominantly inattentive" (ADHD-PI or ADHD-I) – though the term 'attention deficit disorder' is still widely used."
Download the PDF from VitaminDWiki
Vitamin D is not mentioned
Earlier puberty, especially in girls, is associated with physical and mental disorders. Prenatal glucocorticoid exposure influences the timing of puberty in animal models, but the human relevance of those findings is unknown. We studied whether voluntary consumption of licorice, which contains glycyrrhizin (a potent inhibitor of placental 11β-hydroxysteroid dehydrogenase type 2, the “barrier” to maternal glucocorticoids), by pregnant women was associated with pubertal maturation (height, weight, body mass index for age, difference between current and expected adult height, Tanner staging, score on the Pubertal Development Scale), neuroendocrine function (diurnal salivary cortisol, dexamethasone suppression), cognition (neuropsychological tests), and psychiatric problems (as measured by the Child Behavior Checklist) in their offspring. The children were born in 1998 in Helsinki, Finland, and examined during 2009–2011 (mean age = 12.5 (standard deviation (SD), 0.4) years; n = 378). Girls exposed to high maternal glycyrrhizin consumption (≥500 mg/week) were
- taller (mean difference (MD) = 0.4 SD, 95% confidence interval (CI): 0.1, 0.8), were
- heavier (MD = 0.6 SD, 95% CI: 0.2, 1.9), and had
- higher body mass index for age (MD = 0.6 SD, 95% CI: 0.2, 0.9).
They were also 0.5 standard deviations (95% CI: 0.2, 0.8) closer to adult height and reported more advanced pubertal development (P < 0.04). Girls and boys exposed to high maternal glycyrrhizin consumption
- scored 7 (95% CI: 3.1, 11.2) points lower on tests of intelligence quotient, had poorer memory (P < 0.04), and had
- 3.3-fold (95% CI: 1.4, 7.7) higher odds of attention deficit/hyperactivity disorder problems
compared with children whose mothers consumed little to no glycyrrhizin (≤249 mg/week). No differences in cortisol levels were found. Licorice consumption during pregnancy may be associated with harm for the developing offspring.
ADHD, 11β-hydroxysteroid dehydrogenase type 2, cognition, glucocorticoids, glycyrrhizin, puberty
Earlier puberty, especially in girls, is associated with physical and mental disorders, including hormone-sensitive cancers, cardiometabolic disorders, and depression (1–3). According to the “developmental origins hypothesis” of health and disease, prenatal exposure to environmental adversity “programs” long-lasting structural and functional changes in key organs and homeostatic systems, including the brain and its control of behaviors and reproductive function (4, 5).
Overexposure of the fetus to maternal glucocorticoids is probably a key mechanism underlying these “programming” events. While maternal and fetal cortisol levels are correlated (6), fetal levels are up to 10 times lower. This gradient is due to the placental “barrier” enzyme 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2) (7–9). This enzyme catalyzes metabolism of 80%–90% of active cortisol in the maternal bloodstream to inactive cortisone during passage to the fetal circulation (8, 9).
Animal models of fetal glucocorticoid overexposure include genetic lack of fetoplacental 11β-HSD2, its inhibition by glycyrrhizin (glycyrrhetinic acid and glycyrrhizic acid—natural constituents of licorice) or by the synthetic analog carbenoxolone, maternal stress, maternal high- or low-protein diet, or bypass with poor-substrate glucocorticoids such as dexamethasone or betamethasone (7–13). In these models the offspring have alterations in timing of puberty, reproductive function, and hypothalamic-pituitary-adrenal (HPA) axis function as well as having cardiometabolic risk factors, attenuated learning and memory, and increased anxiety- and depression-like behaviors (7–13).
Animal and human placentation and gestation, however, differ, as does the ontogeny of 11β-HSD2 in the placenta (14), precluding predictions applicable to humans. Using a dual-circuit immediate ex vivo perfusion in fresh, intact, term human placentas, we demonstrated that very low doses (10−8 M) of glycyrrhetinic acid, coperfused in the maternal circulation with cortisol, resulted in near complete inhibition of placental cortisone production and free passage of cortisol to the fetal circulation (9). Because glycyrrhizin also inhibits maternal 11β-HSD2, it increases cortisol access to renal mineralocorticoid receptors, sometimes causing maternal hypertension. Although other actions of glycyrrhizin include inhibition of 15-hydroxyprostaglandin dehydrogenase, this probably does not occur systemically with licorice consumption in vivo (15).
We have studied the offspring of pregnant women from a unique longitudinal cohort of Finnish mothers and children, in which nearly 50% of mothers reported licorice consumption during pregnancy (16–18). At the age of 8 years, children of women who consumed high amounts of licorice during pregnancy, when compared with children of women who consumed low amounts of licorice or none, scored lower on tests measuring intelligence and memory (17), had an increased risk of externalizing behavior problems (17), and had higher HPA-axis activity upon awakening and during psychosocial stress (18). We examined these children in adolescence to explore associations with pubertal maturation and to learn whether the cognitive, behavioral, and neuroendocrine changes persist.
Study design and participants
The participants were from an urban community-based cohort of 1,049 women and their healthy, singleton infants born in 1998 in Helsinki, Finland (16). Between 2009 and 2011, all 920 cohort members who had given permission to be contacted and whose addresses were traceable were invited to participate in a follow-up study; 692 (75.2%) were contacted by phone, and 451 (65.2% of the women who were contacted) participated.
Of the follow-up participants, 327 of the children were prenatally exposed to zero–low maternal consumption of glycyrrhizin (≤249 mg/week; mean = 47 mg/week) and 51 to high maternal consumption (≥500 mg/week; mean = 845 mg/week). They represented 48.0% (of 681) and 56.7% (of 90) of those invited, respectively (P = 0.12 for difference in participation rates). The children were assessed at follow-up visits held during 2009–2011. At the follow-up visit, children exposed to zero–low amounts of glycyrrhizin had a mean age of 12.5 (standard deviation (SD), 0.4) years, and children exposed to high amounts of glycyrrhizin had a mean age of 12.5 (SD, 0.4) years (P = 0.26 for group difference). For girls, mean difference (MD) for age in years was −0.15 (95% confidence interval (CI): −0.32, 0.01) between the exposure groups; for boys, MD = 0.04 (95% CI: −0.14, 0.21) between the exposure groups. Among those exposed to zero–low and high amounts of glycyrrhizin and who were invited (n = 771), nonparticipation (n = 393) was related to older maternal age (MD = 0.72 years, 95% CI: 0.03, 1.40), higher level of maternal stress (on a 100-mm visual analog scale, MD = 5.3 mm, 95% CI: 1.7, 9.0), and weekly coffee consumption during pregnancy (MD = 1.3 cups/week, 95% CI: 0.3, 2.4). However, it was not related to the other 25 variables that we tested, including maternal weekly licorice consumption, chocolate consumption, cacao consumption, tea consumption, salt consumption, alcohol consumption, smoking status, blood pressure during pregnancy, maternal or paternal weight, maternal height or body mass index (BMI) at delivery, or delivery mode, nor was it related to the offspring's birth order, body size at birth, length of gestation, or Apgar score at 1 minute (P > 0.08 for all).
Due to this slightly differential dropout, we computed stabilized inverse-probability-of-censoring weights (IPCWs) (19) using the aforementioned characteristics associated with nonparticipation. These IPCWs were subsequently included in all statistical models in order to reduce the possibility of selection bias affecting our results. The assumptions of positivity and misspecification of the weights were checked as recommended (20).
The study protocol was approved by the ethical committees of the City of Helsinki and the Uusimaa Hospital District. Written informed consent was obtained from the mothers after delivery and from the parent/guardian and adolescent at the follow-up assessment.
Maternal licorice consumption
While in the maternity ward, the mothers reported the brand(s) and weekly consumption of licorice during pregnancy (glycyrrhizin intake calculated as mg/week), using a list of all brands of licorice-containing confectionery available in Finland in 1998 (16). The list was prepared by the National Food Administration in 1993 and updated with information from manufacturers (16). The zero–low-exposure (0–249 mg/week) and high-exposure (≥500 mg/week) groups comprised 75 percent and 11 percent of births in the initial cohort (16).
Estimation of pubertal maturation was based on 3 measures of growth and development. 1) The difference was taken between the child's height-for-age standard-deviation score (21) (using the current measured height without shoes, measured with a Seca stadiometer) (Model 213; Seca GmbH & Co KG, Hamburg, Germany) and the standard-deviation score of the midparental target height (22); this is a measure of remaining growth potential and, consequently, the timing of the pubertal growth spurt. 2) The Tanner staging questionnaire was administered by a research nurse; using schematic drawings of 2 secondary sex characteristics (pubic hair development in girls and boys and breast development in girls or development of genitalia in boys), the examiner derives two 5-stage scores ranging from prepubertal (stage I) to postpubertal (stage V) (23). 3) The Pubertal Development Scale is a self-report on secondary sex characteristics (growth spurt, body hair that is not specifically pubic hair, and skin changes in girls and boys; menarche and breast development in girls; and facial hair and voice change in boys) and yields one 4-stage score ranging from no development (1) to full completion of development (4) (24).
We also measured weight in light clothing without shoes (Model 8; Seca GmbH & Co KG) and calculated BMI (weight (kg)/height (m)2). We transformed the values into weight-for-age and BMI-for-age standard-deviation scores (21).
The cognitive test battery comprised tests of intelligence, memory and learning, social perception, attention, and executive function. The short form of the Wechsler Intelligence Scale for Children III (25) included vocabulary, similarities, block design, and picture arrangement subtests, for which we used age-standardized scores as well as estimated age-standardized total intelligence and verbal and performance intelligence quotients (IQs) (26). For the Developmental Neuropsychological Assessment for children (27), we used age-standardized scores for word generation, design fluency, free and cued narrative memory, memory of names, memory of faces, and theory of mind subtests. Measures of attention and executive function included the Continuous Performance Test II (28), the Trail Making Test (29), and the Wisconsin Card Sorting Test (30).
Mothers completed the Child Behavior Checklist, a standardized and validated rating scale screening for psychiatric problems (31). We calculated scores for the scales oriented to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, and used the 82nd percentile as the cutoff to identify adolescents with borderline clinically significant problems (31).
Samples of saliva were collected using cotton swabs (Salivette; Sarstedt, Nümbrecht, Germany). On the first of the 2 consecutive days, samples were collected upon awakening and 15, 30, 45, and 60 minutes thereafter, at 12:00 noon, at 5:00 p.m., and at bedtime. Dexamethasone was administered after the bedtime saliva sample, and a sample was collected upon awakening the next day. We used a low dose of dexamethasone (3 µg/kg of total body weight) to attempt to detect individual variation in HPA-axis suppression (32).
Salivary cortisol concentrations were determined by solid-phase, time-resolved fluorescence immunoassay with fluorometric end-point detection (DELFIA; Wallac, Turku, Finland). The intraassay and interassay coefficients of variation varied between 4.0% and 7.7%, and the mean coefficient of variation between duplicate analyses was 5.9%.
Covariates and confounders
Covariates and confounders included the adolescent's age (years), highest educational level of either parent (secondary or less, vocational, university) reported at follow-up, gestational length (weeks) as confirmed by ultrasonography, birth weight (grams) of the adolescent as derived from birth records, maternal age (years) and BMI, calculated from weight and height derived from medical records, maternal smoking status (none, 1–10 cigarettes/day, >10 cigarettes/day), maternal weekly alcohol consumption (no, yes, no answer; g/week for those reporting yes), coffee consumption (cups/week), tea consumption (cups/week), cacao consumption (cups/week), salt consumption (no, yes, low-sodium), chocolate consumption (never, seldom, weekly, daily), and stress (measured using a 100-mm visual analog scale) during pregnancy, which was reported while the mother was on the maternity ward. They also included the adolescent's own licorice consumption (never, less than once a week, once a week, 2–4 days a week, daily, no answer), reported in the follow-up assessment.
In addition, we conducted analyses of pubertal maturation adjusting for maternal self-reported age at menarche (years) as a crude proxy of the genetic component of pubertal development and analyses of HPA-axis for time at awakening and time at dexamethasone intake.
Differences between the zero–low- and high-exposure groups in the continuous, categorical, and binary outcomes were tested with and without covariates (described in the Results section), using generalized linear models specifying a Gaussian, ordinal logistic, or binary logistic reference distribution for quantitative, ordinal, or binary responses, respectively. All models were weighted by IPCW and computed using robust estimators, which provides consistent covariance estimates even in the case of misspecification of the variance and link functions. By using this analytical strategy, we also tested whether glycyrrhizin intake as a continuous variable (natural log–transformed, to improve linear model fitting)—among those who reported consumption of at least some licorice—was associated in a dose-response manner with the continuous, categorical, and binary outcomes. All P values were 2-sided.
Analyses of pubertal development were conducted separately in girls and boys because adolescent girls are, on average, ahead of boys in pubertal development. Other outcomes were tested in girls and boys combined because sex × exposure-level group interactions were not significant in any analyses (P values > 0.05).
For more see the attached PDF