Prenatal exposure to maternal stressful life events and earlier age at menarche: the Raine Study.

STUDY QUESTION

Is there an association between prenatal exposure to stressful life events and age at menarche, and does childhood BMI mediate this association?

SUMMARY ANSWER

Girls exposed to prenatal stress had a slightly earlier age at menarche, but this association did not show a dose-response effect and was not mediated by childhood offspring BMI.

WHAT IS ALREADY KNOWN

Prenatal stress may impact on reproductive function in females including age at menarche, but human data are very limited. High childhood BMI is known to be associated with earlier age at menarche. Only one small study has measured the association between maternal stress and age at menarche and reported that childhood BMI mediated the association between maternal stress and earlier age at menarche. However, neither maternal stress nor age at menarche was prospectively recorded and the study was limited to 31 mother-daughter pairs.

STUDY DESIGN, SIZE, DURATION: The Raine Study is a large prospective population-based pregnancy cohort study (n = 1414 mother-daughter pairs) continuously followed from prenatal life through to adolescence. In the present study, we examined the association between exposure to maternal stressful life events during early, late and total gestation and age at menarche in offspring using 753 mother-daughter pairs with complete case information.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Mothers prospectively reported stressful life events during pregnancy at 18 and 34 weeks using a standardized 10-point questionnaire. Exact date of menarche was assessed using a purpose-designed questionnaire at 8, 10, 14 and 17 years of age. Complete information on exposure, outcome and confounding variables was obtained from 753 mothers-daughter pairs. Multivariate linear regression complete case analysis was used to examine associations between maternal stressful life event exposure and age at menarche. Potential selection bias was evaluated using multiple imputations (50 datasets). The mediating effects of offspring childhood BMI (ages 5, 8, or 10 years) on these associations were measured in separate sub-analyses.

MAIN RESULTS AND ROLE OF CHANCE

Most (580/753, 77%) daughters were exposed to at least one prenatal stressful life event. Exposure to maternal stressful life events during the entire pregnancy was associated with a non-linear earlier age at menarche. Exposure to one event and two or more psychological stressful events was associated with a 3.5 and 1.7-month earlier onset of puberty, respectively when compared to the reference group with no exposure maternal stressful life events. The estimates from multiple imputation with 50 datasets were comparable with complete case analysis confirming the existence of an underlying effect. No separate significant effects were observed for exposure during early or late gestation. The association between prenatal stressful events and age at menarche was not mediated by childhood BMI in the offspring.

LIMITATIONS, REASONS FOR CAUTION: Stressful life events may have affected pregnant women in different ways and self-perceived maternal stress severity may have provided a more precise estimate of gestational psychological stress. The observed non-linear U-shape of the association between maternal psychological stress and age at menarche did not reflect a dose-response. This suggests that the first exposure to prenatal stress exerts a greater effect on fetal reproductive development. A potential mechanism is via dramatic initial activation of the hypothalamic-pituitary-adrenal (HPA) axis following the first stressful life event which is greater than that observed following subsequent exposure to two or more maternal stressful life events. Whilst we adjusted for a priori chosen confounders, we cannot exclude residual confounding or confounding by factors we did not include. Maternal age at menarche was not available so the effects of familial history/genetics could not be assessed. There was a large loss due to the number of girls with no information on date of menarche and missing confounder information implying risk of selection bias and multiple imputation analyses did not fully exclude this risk (similar direction but slightly weaker estimate magnitude).

WIDER IMPLICATIONS OF THE FINDINGS

Menarche is a sentinel reproductive event and earlier age at menarche carries implications for psychological, social and reproductive health and for long-term risk of common non-communicable diseases. Understanding the factors regulating age at menarche has extensive health implications. This is the first population-based cohort study in humans to demonstrate that prenatal psychological stress might directly modify age at menarche.

STUDY FUNDING/COMPETING INTEREST(S): Dr. Bräuner and Trine Koch's salaries were supported by Doctor Sofus Carl Emil Friis and spouse Olga Doris Friis foundation, The Danish Cancer Society (Kræftens Bekæmpelse, RP15468, R204-A12636, Denmark) and The Danish Health Foundation (Helsefonden, F-22181-23, Denmark). Martha Hickey was funded by NHMRC Practitioner Fellowships. The funding bodies played no role in the design, collection, analysis, or interpretation of data; in the writing of the manuscript; or in the decision to submit the manuscript for publication. Dr. Hart has received personal fees in his function as the Medical Director of Fertility Specialists of Western Australia and received educational sponsorship grants from MSD, Merck-Serono and from Ferring Pharmaceuticals. Dr Hart has also received personal fees from Shareholders in Western IVF outside the submitted work.

TRIAL REGISTRATION NUMBER

NA.