Rates of obesity and associated diabetes have increased for children of all age groups. As an example, when one considers the glucose metabolism among the youth in the United States of America, the prevalence of prediabetes / diabetes has recently increased from 9% to 23% 15
. Few studies have reported on the effects of prediabetes on bone metabolism. The study of Afghani et al. 11
was the first in the pediatric population that sought the influence of prediabetes on bone in overweight children. After the study of Dimitri et al. 16
that suggested a relationship between childhood obesity and skeletal fractures, several studies were conducted regarding the relationship between obesity, prediabetes, and BMD 8,12,17,18
Our study showed no results depicting a significant difference in BMD in prediabetic versus normal subjects. These results did not seem to change when the effect of sex and pubertal status were analyzed. Although these findings are in line with those of Afghani et al, they conducted their research in prepubertal children and defined the presence of prediabetes only with IGT 11. In our study fasting glucose, 2-h glucose, insulin, 2-h insulin, HOMA-IR, HBA1c, did not affect BMD and these results did not change when we adjusted for important covariates. This finding suggests that abnormal glucose regulation does not have major harmful effects on bone health in obese children. This result is consistent with some previous studies. Firstly, Lee et al. found no association between any prediabetes criteria and BMD in adult subjects 13. Another study in adolescent children with metabolic syndrome failed to show a relationship between glucose levels and BMD 18. However, several studies found a significant association between markers of glucose regulation and BMD, suggesting a detrimental effect of insulin resistance on bone health in children 8,11,12,17.
Among the factors that underlie these controversies, inhomogeneity between diagnostic criteria used to define prediabetes or glucose dysregulation may be important. Some of these studies only used the presence of IFG or IGT to indicate prediabetes and this may be considered an important limitation. However, Cheng et all suggested that fasting plasma glucose (FPG) is inadequate for viewing prediabetes in obese young people and the OGTT is the best way to detect the prediabetic state 19. Another study that investigated screening markers of IGT revealed that HbA1C was the only predictor of IGT 20. This data can explain why all markers of glucose regulation have not been consistently associated with BMD in these or other studies.
Studies in adults have shown that triglycerides and HDL were negatively correlated with BMD 21. Pollock et als 87 studies found no association between lipids and BMD in prepubertal children. Silva 18 et al found that the association between lipids and BMD changes were dependent on sex; triglycerides were inversely correlated with BMD only in female adolescents. We observed no association between lipid levels and BMD, and this did not change according to the pubertal status or sex.
We found one significant relationship in this study, there was a positive correlation between BMI and BMD. Body size has been suggested as the strongest sole determinant of BMD. This association was attributed weight bearing and to the endocrine properties of adipose tissue, which forms estrogen from androstenedione 11. The study of Lorentzon and et al. 21 suggested that cortical bone parameters at the tibia were correlated with fat mass while this was not valid for the radius. This finding supports the positive effect of weight bearing on BMD. This could also explain the positive correlation that we found between BMI and the BMD of the calcaneus.
Visceral adiposity has been shown to be deleterious for bone health 17,18,22. This relationship is attributed to the strong association between metabolic disturbance and visceral rather than subcutaneous adiposity 18. Although we did not directly measure visceral adiposity, we assessed this parameter through measurement of waist circumference, which is often used as a substitute for visceral adiposity, and found no correlation between BMD and waist circumference. As with the direct effect of insulin resistance parameters, the indirect effects of these parameters such as waist circumference were also not correlated with BMD.
Despite high BMD in patients with type 2 diabetes 23, diabetes has been associated with an increased risk of bone fracture, possibly reflecting the effect of falls due to impairment of visual sharpness or proprioception caused by diabetes 24.
Obesity has been suggested to be a risk factor for Vitamin D deficiency. It is estimated that as much as 87.5% of obese children have vitamin D insufficiency / deficiency 25. We could not analyze the influence of vitamin D status on BMD between the two groups because of our small sample size, similar levels of vitamin D in the normal and prediabetic groups. Furthermore, the results of all patients whose vitamin D levels were studied were recorded as insufficient. It may be important to consider vitamin D status when analyzing the effect of obesity and the prediabetes on BMD. Vitamin D status has not been taken into account by earlier studies that have investigated the relationship between BMD and prediabetes.
Our study has some limitations. Firstly this is a cross-sectional study. Secondly, ultrasonography was used to determine BMD in place of dual-energy X-ray absorptiometry (DEXA). However, ultrasonography is a reliable diagnostic tool for identifying osteoporosis 26 and quantitative ultrasound is commonly used in clinical practice because it is less costly and transportable then DEXA 27,28. Obesity is associated with elevated systemic inflammatory markers. A relation has been reported between rising inflammatory markers in obese patients and osteocalcin, a bone turnover marker 29.
On that basis, we might have obtained more enlightening data concerning BMD if we had investigated inflammatory markers and bone turnover markers in our study. Furthermore, data regarding dietary habits and physical exercise, the major risk factors for low BMD were not available in this study.