Associations of sedentary behavior, physical activity, body composition, sleep and vitamin D with bone stiffness

Abstract

In children and adolescents, bone modeling and remodeling is highly active in order to expand bone in length and width, to increase bone mass, and to maintain bone shape. Although bone mass acquisition is relatively slow throughout childhood, with the onset of puberty and the growth spurt of height in adolescence the rate of bone mineral accumulation increases, reaching a peak bone mass shortly after a peak height. The peak bone mass is an important predictive factor of osteoporosis in the later life due to the bone loss during ageing. Except for genetic factors, there is an estimated 20 to 40% of the peak bone mass variation contributed to modifiable factors e.g., mechanical loading, physical activity, sedentary behavior, sleep and nutritional factors. However, we are only beginning to identify the specific dimensions and doses of these modifiable factors needed for the short-term and long-term beneficial effects on bone health. The lack of longitudinal epidemiological studies and the conflicting results in the intervention studies among healthy pediatric populations limit our knowledge. Hence, the present thesis aims to provide a better understanding on the associations between physical activity, sedentary behavior, sleep, nutrition and bone health in children and adolescents. The present thesis is based on the data from the IDEFICS/I.Family cohort including children and adolescents aged 2 to 15 years from eight European countries. Three examination waves with repeated measurements were conducted in 2007/2008, 2009/2010 and 2013/2014. In the subgroups, bone stiffness index was measured using calcaneal quantitative ultrasound (QUS) in all examination waves, serum bone formation marker osteocalcin was analyzed using chemiluminescence assays in the first examination wave while serum bone resorption marker C-terminal telopeptides of type I collagen was analyzed in the first and third examination waves. Calcaneal QUS as a validated method to estimate bone health is becoming popular in pediatric populations since it is non-radiating, quick and cost-effective. The measured parameters of broadband ultrasound attenuation and speed of sound, as well as the derived stiffness index are related to bone mass and bone structural properties. Meanwhile, bone resorption and formation markers have been suggested to be sensitive to the changes in environmental factors, hormone levels and treatments. Therefore, stiffness index, osteocalcin and C-terminal telopeptides of type I collagen were considered as bone health outcomes. Body composition in terms of fat mass and fat free mass was derived from objectively measured skinfold thickness. Weight status was estimated from objectively measured body height and weight by calculating body mass index z-scores and cut-offs. Physical activity and sedentary behavior were measured using both self-administrated questionnaires and accelerometers. Sleep duration and quality were evaluated using questionnaires. Consumption frequency of dairy products and usual calcium intake were collected using food frequency questionnaires and 24h-dietary recalls, respectively. Serum 25-hydroxyvitamin D was analyzed using chemiluminescence assays. Linear mixed-effect models were used with adjustments for a cluster effect of country and potential confounders. The major findings were presented and discussed in four published or submitted original papers, final sample sizes that varied in each paper depended on the analysis strategies for different research questions. First, the longitudinal results indicated a positive relationship between fat free mass and stiffness index during growth. Specifically, baseline fat free mass was observed to predict two-year and six-year changes in stiffness index, and six-year changes in fat free mass was also positively associated with change in stiffness index. Meanwhile, the association between six-year changes in fat mass and stiffness index differed by sex and pubertal status, suggesting an inverse association in boys and girls before menarche, but a positive association in girls after menarche (Lan Cheng, et al., Bone. 2019). Second, objectively measured moderate-to-vigorous physical activity (MVPA) was positively associated with the increase of stiffness index over two years and six years of follow-up. These results were supported by the comparable albeit weaker positive associations between self-reported time spent at sports clubs and stiffness index. However, the inverse associations between screen time as a surrogate for sedentary behavior and stiffness index depended on weight status. Specifically, the cross-sectional association between weekly duration of watching TV and stiffness index was observed to be inverse only in thin/normal weight group. Both baseline and two years change in weekly duration of watching TV, and six years change in weekly duration of playing computer/games were inversely associated with corresponding changes in stiffness index in the overweight/obese group (Lan Cheng, et al., Int J Behav Nutr Phys Act. 2020). Third, the association between sleep and stiffness index was analyzed using data from two follow-up examination waves with the interval of approximately four years. Total sleep duration was calculated and further classified into short, adequate and long based on the recommendation from the National Sleep Foundation. Poor sleep quality was estimated by reporting either having trouble to get up in the morning, or having difficulty to fall asleep, or have no regular bedtime routine. The positively cross-sectional associations between nocturnal sleep duration, daytime napping and stiffness index were only observed in participants with adequate sleep duration. After four years of follow-up, the positive association between daytime napping and stiffness index was more pronounced in participants with short sleep duration. Moreover, long-term detrimental effect of extreme sleep duration (short or long) on stiffness index only existed in participants with poor sleep quality (Lan Cheng, et al., Osteo Int. 2020). At last, only cross-sectional analyses were conducted in the associations between vitamin D, bone turnover markers and stiffness index using merged datasets based on the first and third examination waves. Serum 25-hydroxyvitamin D, calcium intake and dairy products consumption were observed to be inversely associated with bone resorption marker but not formation marker. MVPA modified the association between 25-hydroxyvitamin D and stiffness index, suggesting that serum 25-hydroxyvitamin D no less than 20 ng/ml would be a protective factor for calcaneal stiffness index only if children met the MVPA guideline of one hour MVPA per day on average (Lan Cheng, et al., Am J Clin Nutr. submitted). In summary, the present cumulative thesis provides a comprehensive understanding on the associations between lifestyle-related factors and bone health indicators in children and adolescents. Future prevention and intervention studies with regard to improving childhood bone health should put emphasis on promoting sufficient MVPA, maintaining adequate sleep duration and calcium intake, and improving consumption frequency of dairy products. Some bone health determinants in specific groups i.e. excess fat mass in boys and pre-pubertal girls, long screen time in overweight/obese children, and insufficient vitamin D level in inactive children particularly need attention.