DIET, BONE MASS, AND OSTEOCALCIN - A CROSS-SECTIONAL STUDY

To determine the relationships among nutrient intake, bone mass, and b one turnover in women we have investigated these issues in a populatio n-based, cross-sectional, observational study in one county in central Sweden. A total of 175 women aged 28-74 at entry to the study were in cluded. Dietary assessment was made by both a semiquantitative food fr equency questionnaire and by four 1-week dietary records. Dual energy X-ray absorptiometry was performed at five sites: total body, L2-L4 re gion of the lumbar spine, and three regions of the proximal femur. Ser um concentrations of osteocalcin (an osteoblast-specific protein refle cting bone turnover) were measured by a radioimmunoassay. Linear regre ssion models, with adjustment for possible confounding factors, were u sed for statistical analyses. A weak positive association was found be tween dietary calcium intake as calculated from the semiquantitative f ood frequency questionnaire and total body bone mineral density (BMD) among premenopausal women. No association emerged between dietary calc ium intake and site-specific bone mass, i.e., lumbar spine and femoral neck, nor was an association found between dietary calcium intake and serum osteocalcin. BMD at some of the measured sites was positively a ssociated with protein and carbohydrates and negatively associated wit h dietary fat. In no previous studies of diet and bone mass have dieta ry habits been ascertained so carefully and the results adjusted for p ossible confounding factors. Neither of the two methods of dietary ass essment used in this study revealed any effect of calcium intake on BM D at fracture-relevant sites among these healthy, mostly middle-aged w omen. A weak positive association was found between calcium intake est imates based on the food frequency questionnaire and total body BMD. I n this study population the preventive effect of high dietary calcium on osteoporosis is probably very weak. The independent significance of protein, carbohydrates, and fat is uncertain.