Recent Decline in Age at Breast Development: The Copenhagen Puberty Study

Kaspar Sørensen, Jørgen Holm Petersen, Niels E. Skakkebæk, Anders Juul

PEDIATRICS · 2009 · 627 citations · 41 references

Concepts

TL;DR

Recent studies of early breast development in American girls have sparked global debate, yet secular trend analyses are hampered by inconsistent data across studies. This study provides systematically collected European data from a single region over a 15‑year period to examine changes in pubertal timing. We evaluated 2,095 girls by breast palpation, anthropometry, and hormone assays, estimating age distributions for breast, pubic hair, and menarche stages in 1991–1993 and 2006–2008 cohorts. Pubertal onset (Tanner stage 2+) occurred significantly earlier in the 2006 cohort (9.86 y vs 10.88 y) even after BMI adjustment, while menarche ages were similar and hormone profiles differed only by lower estradiol in 8‑10‑year‑olds, indicating factors beyond hormones and BMI drive the shift.

Abstract

Recent publications showing unexpectedly early breast development in American girls created debate worldwide. However, secular trend analyses are often limited by poor data comparability among studies performed by different researchers in different time periods and populations. Here we present new European data systematically collected from the same region and by 1 research group at the beginning and end of the recent 15-year period.Girls (N = 2095) aged 5.6 to 20.0 years were studied in 1991-1993 (1991 cohort; n = 1100) and 2006-2008 (2006 cohort; n = 995). All girls were evaluated by palpation of glandular breast, measurement of height and weight, and blood sampling (for estradiol, luteinizing hormone, and follicle-stimulating hormone). Age distribution at entering pubertal breast stages 2 through 5, pubic hair stages 2 through 5, and menarche was estimated for the 2 cohorts.Onset of puberty, defined as mean estimated age at attainment of glandular breast tissue (Tanner breast stage 2+), occurred significantly earlier in the 2006 cohort (estimated mean age: 9.86 years) when compared with the 1991 cohort (estimated mean age: 10.88 years). The difference remained significant after adjustment for BMI. Estimated ages at menarche were 13.42 and 13.13 years in the 1991 and 2006 cohorts, respectively. Serum follicle-stimulating hormone and luteinizing hormone did not differ between the 2 cohorts at any age interval, whereas significantly lower estradiol levels were found in 8- to 10-year-old girls from the 2006 cohort compared with similarly aged girls from the 1991 cohort.We found significantly earlier breast development among girls born more recently. Alterations in reproductive hormones and BMI did not explain these marked changes, which suggests that other factors yet to be identified may be involved.

References

41