Electronic Letters to:

Article: E M Clark, A R Ness, J H Tobias and the ALSPAC Study Team ADIPOSE TISSUE STIMULATES BONE GROWTH IN PREPUBERTAL CHILDREN J Clin Endocrinol Metab 2006; 0: jc.2006-0332v1 [Abstract]

eLetters: Submit a response to this article

Electronic letters published:

Reply

Emma M Clark, Jon H Tobias   (5 March 2007)

Maturation-Specific or Subpopulation Heterogeneity?

Qingju Wang   (6 February 2007)

Reply 5 March 2007
Emma M Clark, SpR Rheumatology and Honorary Lecturer University of Bristol, UK, Jon H Tobias Send letter to journal: Re: Reply emma.clark{at}bristol.ac.uk Emma M Clark, et al.	We thank Qingju Wang for his interest in our paper. We would like to start by reiterating that no observational study on its own can prove causality, and this was not the aim of our study. We wished to examine whether our previously reported association between fat mass measured at aged 9.9 years and total body less head (TBLH) bone area or bone mineral content (BMC) at aged 9.9 years (1) was independent of height and lean mass. Furthermore we wished to examine whether fat mass measured at aged 9.9 years could be used to predict subsequent gain in TBLH bone area or BMC, and whether or not this association was altered by puberty. Our clearly stated a-priori hypotheses were based on extensive literature review. We disagree that our cross-sectional analyses contradict analyses from the longitudinal data. In the cross-sectional analyses based on data collected at aged 9.9 years, virtually all of bone growth will have taken place prior to the onset of puberty. In contrast, in the longitudinal analyses examining the association between fat mass at age 9.9 years and change in TBLH bone area or BMC between ages 9.9 and 11.8 years, all of bone growth will have taken place after the onset of puberty in the 599 girls in Tanner stage 2 and 3. However, we readily acknowledge that there are many ways of examining the data statistically, and we have presented only one. Also we agree that studying puberty using the Avon Longitudinal Study of Parents and Children (ALSPAC) is problematic as not all children have entered puberty. Data is continuing to be collected from this cohort and repeating our analyses in the future when we can accurately define pubertal stages in all children (boys and girls) will add further to the literature in this controversial field. Reference 1. Clark EM, Ness AR, Tobias JH. 2005. Social position affects bone mass in childhood through opposing actions on height and weight. J Bone Mineral Research 20:2082-2089
Maturation-Specific or Subpopulation Heterogeneity? 6 February 2007
Qingju Wang Endocrine Center, Austin Health, University of Melbourne, Australia Send letter to journal: Re: Maturation-Specific or Subpopulation Heterogeneity? qingju{at}sport.jyu.fi Qingju Wang	Clark et al. (1) reported that total body fat mass (TBFM) positively correlated with total-body-less-head (TBLH) bone area in cross-sectional analyses at baseline and over the following two years in boys and Tanner stage I girls, but not in Tanner stage II girls, and that there was a negative correlation in Tanner stage III girls. The authors concluded in their title that “adipose tissue stimulates bone growth in prepubertal children”, but did these results truly support their statement? A cross-sectional correlation between two traits does not mean there is a causal relationship. Longitudinal analyses may help, but the large prospective dataset in this study was not used prospectively. Fat mass could be related to bone area by both its mechanical loading on weight-bearing sites and its endocrine function over the whole body. If mechanical loading plays a role, the increase in bone area per unit gain in TBFM should be proportionally more in weight-bearing sites, i.e., the lower limbs, than in non-weight-bearing sites, i.e., the upper limbs. The authors did not undertake such an analysis, but instead related only baseline TBFM to the change of bone area in the upper and lower limbs. This approach can provide information about the extent to which a measurement of TBFM can predict the gain of bone area over subsequent years, but it does not confirm a relationship. The authors also stated that “the relationship of fat mass and bone size was attenuated by puberty”. However, in the cross-sectional analyses, the relationship of baseline TBFM and TBLH bone area in Tanner stage III girls were similar to or even greater than in Tanner stage I girls. This contradicts the finding that the relationships between baseline TBFM and bone area gain is positive, absent and negative in girls at Tanner stage I, II, and III, respectively. This contradiction may well have been due to heterogeneity in the studied population. Specifically, Tanner stage III girls were different from Tanner stage I girls in terms of development, as evidenced by their higher body fat percentage (36% vs. 25%). To address whether the relationship between adipose tissue and bone area is dependent of maturational stage, subjects with wider range of age and maturational stages should be used instead of these with the same age and different maturation stage. We need more hypothesis-driven studies to further our understanding of the relationship between the adipose tissue and bone. Reference 1. Clark EM, Ness AR, Tobias, JH and the Avon Longitudinal Study of Parents and Children Study Team. 2006. J Clin Endocrinol Metab 91:2534-2541