ESPE Abstracts

Background and aim: Menarche is a milestone of female pubertal development as well as an important sociocultural and psychological event during adolescence. Age of menarche has changed with time, and there is a broad individual variation in timing. Many factors are associated with menarcheal age, where further research is needed. The QEPS-growth model makes it possible to conduct detailed analyses of growth^1,2. The aim of the study was to investigate anthropometric factors explaining variations in menarcheal timing.

Material and Methods: The timing of menarche was analysed in a longitudinally followed population, the GrowUp1990Gothenburg cohort (community-based setting)^3,4. The analysed study group included 755 healthy females born at gestational week (GA) 36+2–41+6, with information of menarcheal age. Height at 7 years and specific pubertal growth was calculated by the QEPS-model². The individual timing of menarche was related to birthweight and birth-length (kg, cm), GA, maximal BMI between 3.5–7 years (SDS)4, parental heights, height at pubertal onset (P5%), duration of pubertal growth (P5%–95% of specific pubertal growth spurt) and adult height. Variables with univariate significant relation, was allowed into linear multivariable regression models for explaining the variation of the outcome menarcheal age. Data given as mean (± SDS) or Beta-coefficient (B).

Results: Mean menarcheal age was 12.94 (± 1.33) years. Birthweight correlated with age at menarche (B-coefficient 0.24), whereas birth-length and GA did not. A negative linear correlation was seen between BMI and menarcheal age (B -0.23), and of height at 7 years (B -0.36) – higher BMI and taller childhood height was associated with earlier menarche. In analysis of parental heights, taller maternal height correlated with later menarche (B 0.029), whereas no association was seen with paternal height. In multivariable linear regression models, explanatory significant factors were height at 7 years, height at pubertal onset (P5%) and duration of pubertal growth (time P5%-95%); together explaining 44% of the variation of age at menarche (R-square 0.44). As a control the nonparametric robust regression was performed, with R-square 0.41.

Conclusion: In a cohort of healthy Swedish girls with longitudinal growth data born in the 1990s, timing of menarche was correlated with birth-weight, childhood-BMI, height at 7yrs and at pubertal onset, duration of pubertal growth and maternal height.

1. Nierop et al. Journal of Theoretical Biology 2016. Oct 7;406:143-65. 2. Holmgren et al. BMC Pediatrics 2017. Apr 9;17(1):107. 3. Sjöberg et al. Acta Paediatrica 2012 Sep;101(9):964-72. 4. Holmgren et al. Pediatric Research 2017:81,448–454