Furthermore, a previous study in ALSPAC found an inverse relationship of parental social position with offspring BMC and BA at age 9.9 years, also acting via the pathway of offspring weight [26]. It therefore seems most plausible that our associations are not explained by intrauterine AZD2281 effects, but rather that unmeasured aspects of the shared

family environment which are associated with parental smoking, such as diet or level of physical activity, influence increased weight gain and greater bone mass in the children. Studies have shown that overweight children and adolescents have higher whole body and spinal bone mass [27–29] and that BMC is positively related to both lean and fat mass in childhood [30, 31]. CHIR-99021 cell line Fat mass has been demonstrated to stimulate bone growth in prepubertal children previously in the ALSPAC [32, 33]. There has been a greater association reported between fat mass and bone mineral accrual in girls than in boys during puberty [34, 35], which may in part explain why we found no associations in boys, although one study suggests that this sex difference is not present in prepubertal children [35]. In our cohort, there was also a weaker univariate relationship between maternal smoking and offspring weight in sons than in daughters, so it is also possible that the social characteristics in families where parents

smoke have a lesser influence on adiposity in boys than girls. In analysis adjusted for pubertal stage (both genders) and age at menarche (in girls), the associations between maternal smoking and bone outcomes in girls were attenuated, whereas the paternal associations remained similar. This suggests that these positive maternal associations may partly be explained by the association between maternal smoking in pregnancy and earlier age at menarche, which has been shown previously in ALSPAC [36]. Adjustment for pubertal stage in boys did not affect the associations between parental smoking and bone outcomes, and parental smoking was not related to pubertal stage at age 10 years in boys. Our findings conflict

with the study by Jones et al. [7] which indicated negative relationships between maternal smoking in pregnancy and bone mass in 8-year-olds for the total Methane monooxygenase body, femoral neck and lumbar spine, with relationships at the femoral neck and lumbar spine remaining after adjustment for the child’s height and weight. However, they studied a Tasmanian Cell Cycle inhibitor cohort identified at birth as at increased risk of sudden infant death syndrome which contained 65% male offspring and a higher prevalence of maternal smoking during pregnancy (49%) compared with ours (21%). Children of mothers who smoked were lighter at age 8 years in Jones’ study, whereas we found a strong positive relationship between maternal smoking and offspring weight. Jones et al. do not make comparison with paternal smoking or give sex-specific findings.