Does the sex of one’s co-twin affect height and BMI in adulthood? A study of dizygotic adult twins from 31 cohorts

Leonie H. Bogl, Aline Jelenkovic, Eero Vuoksimaa, Linda Juel Ahrenfeldt, Kirsi H. Pietiläinen, Maria A. Stazi, Corrado Fagnani, Cristina D’Ippolito, Yoon‐Mi Hur, Hoe‐Uk Jeong, Judy L. Silberg, Lindon J. Eaves, Hermine H. Maes, Gombojav Bayasgalan, Narandalai Danshiitsoodol, Tessa L. Cutler, Christian Kandler, Kerry L. Jang, Kaare Christensen, Axel Skytthe, Kirsten Ohm Kyvik, Wendy Cozen, Amie E. Hwang, Thomas M. Mack, Cathérine Derom, Robert Vlietinck, Tracy L. Nelson, Keith E. Whitfield, Robin P. Corley, Brooke M. Huibregtse, Tom A. McAdams, Thalia C. Eley, Alice M. Gregory, Robert F. Krueger, Matt McGue, Shandell Pahlen, Gonneke Willemsen, Meike Bartels, Maties Torrent, Zengchang Pang, Qihua Tan, Dongfeng Zhang, Nicholas G. Martin, Sarah E. Medland, Grant W. Montgomery, Jacob Hjelmborg, Esther Rebato, Gary E. Swan, Ruth E. Krasnow, Andreas Busjahn, Paul Lichtenstein, Sevgi Yurt Öncel, Fazil Alıev, Laura A. Baker, Catherine Tuvblad, Sisira Siribaddana, Matthew Hotopf, Athula Sumathipala, Frühling Rijsdijk, Patrik K. E. Magnusson, Nancy L. Pedersen, Anna K. Dahl Aslan, Juan R. Ordoñana, Juan F. Sánchez-Romera, Lucía Colodro‐Conde, Glen E. Duncan, Dedra Buchwald, Ádám Domonkos Tárnoki, Dávid László Tárnoki, Yoshie Yokoyama, John L. Hopper, Ruth J. F. Loos, Dorret I. Boomsma, Thorkild I. A. Sørensen, Karri Silventoinen, Jaakko KaprioBiology 2017 · DOI

TL;DR

This large-scale international study investigated whether growing up with a twin of the opposite sex (OS) influences adult height and BMI, testing the "twin testosterone transfer" hypothesis. The findings revealed only very small, statistically significant differences in height between OS and same-sex (SS) twins, with no meaningful differences in BMI or overweight/obesity prevalence, suggesting limited impact of prenatal hormonal exposure or postnatal socialization on these traits.

What Problem Does This Paper Address?

The "twin testosterone transfer" (TTT) hypothesis posits that male hormones can transfer between fetuses in utero, potentially influencing the development of their co-twins. Specifically, it's hypothesized that females with a male co-twin might be exposed to higher levels of prenatal testosterone, leading to more "masculinized" traits. While animal studies provide some support for such prenatal hormonal influences on metabolic set points and body composition, evidence in humans has been inconsistent, often limited by small sample sizes, lack of replication, and potential biases. This study aimed to rigorously test the TTT hypothesis and the role of postnatal socialization on two key anthropometric traits: height and Body Mass Index (BMI) in adulthood, using a large, diverse dataset of dizygotic twins.

Methods

The researchers utilized data from the Collaborative project of Development of Anthropometrical measures in Twins (CODATwins) database, a large international collaboration pooling data from 31 twin cohorts across 19 countries. They analyzed data from 122,302 dizygotic (DZ) twin individuals (68,494 from same-sex pairs and 53,808 from opposite-sex pairs) aged 20 years and older.

Key statistical methods included:

  • Multiple Regression: Used to compare mean height and BMI between twins from OS and SS pairs, adjusting for cohort, age, and birth year. Regression coefficients and their 95% confidence intervals (CIs) were reported.
  • Logistic Regression: Employed to compare the prevalence of being overweight or obese between OS and SS twin pairs, also adjusted for cohort, age, and birth year. Adjusted odds ratios (ORs) and 95% CIs were calculated.
  • Random-Effects Meta-Analysis: Used to assess heterogeneity across cohorts and to pool effect sizes for height and BMI, providing an overall estimate of the differences between OS and SS twins.
  • Zygosity Determination: Zygosity was determined by questionnaires or DNA genotyping, depending on the original study.

Datasets

The study utilized data from the CODATwins (Collaborative project of Development of Anthropometrical measures in Twins) database. This database is a compilation of data from 31 twin cohorts across 19 countries. The final sample for this analysis comprised 122,302 dizygotic twin individuals. No new datasets were introduced.

Key Insights and Ideas

  • Limited Evidence for TTT Hypothesis: The study found very little support for the TTT hypothesis regarding adult height and BMI. While a statistically significant difference in height was observed, its magnitude was small and unlikely to be of public health significance.
  • Height Differences: A Nuance: Females with a male co-twin were, on average, slightly taller (0.31 cm) than females with a female co-twin. Intriguingly, males with a female co-twin were also slightly taller (0.14 cm) than males with a male co-twin, a finding that runs counter to the TTT hypothesis's prediction of masculinization in males.
  • BMI and Obesity Unaffected: Crucially, mean BMI and the prevalence of overweight or obesity did not differ between individuals from OS and SS dizygotic twin pairs, for both males and females. This suggests that any potential prenatal hormonal influences or postnatal socialization effects related to these traits are minimal.
  • Postnatal Socialization Less Likely to be a Major Factor: The study implicitly addresses postnatal socialization by comparing OS and SS twins. The lack of significant differences in BMI and obesity suggests that growing up with a sibling of the opposite sex, compared to the same sex, does not substantially alter these anthropometric outcomes in adulthood.

Key Results

  • Height:
    • OS females were, on average, 0.31 cm taller than SS females (p < 0.0001).
  • OS males were, on average, 0.14 cm taller than SS males (p = 0.025).
  • BMI: No significant differences in mean BMI were found between OS and SS females or between OS and SS males.
  • Overweight/Obesity: The odds ratios for being overweight or obese did not differ significantly between OS and SS females, nor between OS and SS males.

Limitations and Future Work

The authors acknowledge several limitations:

  • Self-Reported Data: Almost all height and weight data were self-reported, which can introduce inaccuracies, especially concerning body weight perception, which can be influenced by sibling characteristics.
  • Zygosity Determination: Zygosity was primarily determined by questionnaires, not DNA testing in most cases, which carries a small risk of misclassification.
  • Lack of Detailed Adiposity Measures: The study focused on BMI, which is a general measure of adiposity. More detailed measures of body composition and fat distribution were not available.
  • Absence of Early-Life Nutritional Data: The impact of breastfeeding status and other early-life nutritional factors on the observed differences was not assessed, though a recent study suggests this could be a factor.
  • Pubertal Timing: Data on pubertal timing, a critical developmental period, were not included, and differences in these traits might emerge during or after puberty.

Future work could explore these anthropometric traits at different developmental stages (prenatal, infancy, puberty) and incorporate more detailed measures of body composition and early-life environmental factors to further elucidate the complex interplay of genetics, prenatal environment, and postnatal influences on human development.

Summary updated

Abstract

We found no evidence to support the hypothesis that prenatal hormonal exposure or postnatal socialization (i.e., having grown up with a twin of the opposite sex) has a major impact on height and BMI in adulthood.
BibTeX
@inproceedings{bogl2017does,
  author = {Leonie H. Bogl and Aline Jelenkovic and Eero Vuoksimaa and Linda Juel Ahrenfeldt and Kirsi H. Pietiläinen and Maria A. Stazi and Corrado Fagnani and Cristina D’Ippolito and Yoon‐Mi Hur and Hoe‐Uk Jeong and Judy L. Silberg and Lindon J. Eaves and Hermine H. Maes and Gombojav Bayasgalan and Narandalai Danshiitsoodol and Tessa L. Cutler and Christian Kandler and Kerry L. Jang and Kaare Christensen and Axel Skytthe and Kirsten Ohm Kyvik and Wendy Cozen and Amie E. Hwang and Thomas M. Mack and Cathérine Derom and Robert Vlietinck and Tracy L. Nelson and Keith E. Whitfield and Robin P. Corley and Brooke M. Huibregtse and Tom A. McAdams and Thalia C. Eley and Alice M. Gregory and Robert F. Krueger and Matt McGue and Shandell Pahlen and Gonneke Willemsen and Meike Bartels and Maties Torrent and Zengchang Pang and Qihua Tan and Dongfeng Zhang and Nicholas G. Martin and Sarah E. Medland and Grant W. Montgomery and Jacob Hjelmborg and Esther Rebato and Gary E. Swan and Ruth E. Krasnow and Andreas Busjahn and Paul Lichtenstein and Sevgi Yurt Öncel and Fazil Alıev and Laura A. Baker and Catherine Tuvblad and Sisira Siribaddana and Matthew Hotopf and Athula Sumathipala and Frühling Rijsdijk and Patrik K. E. Magnusson and Nancy L. Pedersen and Anna K. Dahl Aslan and Juan R. Ordoñana and Juan F. Sánchez-Romera and Lucía Colodro‐Conde and Glen E. Duncan and Dedra Buchwald and Ádám Domonkos Tárnoki and Dávid László Tárnoki and Yoshie Yokoyama and John L. Hopper and Ruth J. F. Loos and Dorret I. Boomsma and Thorkild I. A. Sørensen and Karri Silventoinen and Jaakko Kaprio},
  title = {{Does the sex of one’s co-twin affect height and BMI in adulthood? A study of dizygotic adult twins from 31 cohorts}},
  booktitle = {Biology},
  year = {2017},
  doi = {10.1186/s13293-017-0134-x},
}
Figure