Smoking may impair the bone protective effects of nutritional calcium: a population-based approach

Postmenopausal women were randomly selected to investigate the effects of smoking on prevention of bone loss with nutritional calcium. DXA was performed twice, and smoking and calcium intake habits were inquired through the mail in 954 women. Smoking dampened the bone protective effects of nutritional calcium. This may reflect the pathophysiology underlying smoking-induced bone loss postmenopause. This study evaluated the effect of smoking on the bone protective properties of nutritional calcium. Of the random sample of 954 peri- and postmenopausal women selected from the Osteoporosis Risk Factor and Prevention (OSTPRE) study cohort (n = 13,100) in Kuopio, Finland, 182 had smoked at some time (ever smokers) and 772 had never smoked. Women were divided in tertiles according to self-reported dairy nutritional calcium intake (mg/day): < 648 (1st), 648-927 (2nd), > 927 (3rd). Bone mineral density at lumbar spine (LS) and femoral neck (FN) was measured with DXA at baseline in 1989-1991 and at the 5-year follow-up in 1994-1997. In a linear regression model, nutritional calcium intake did not predict annual bone loss in smokers. These results were similar in the subanalysis on 71 current smokers (at both baseline and 5-year measurements) and on 85 past smokers. In never smokers, a statistically significant linear trend was observed between calcium intake and annual bone loss at LS, but at FN only after adjustment for age, weight, hormone replacement therapy (HRT), and other covariates. In analysis of covariance (ANCOVA), no differences in bone loss rate were observed between calcium intake tertiles among smokers. In nonsmokers, the annual bone loss rate was lower in the second (-0.41%) and the third (-0.35%) tertile compared with the first tertile (-0.61%) at LS (p < 0.05) and lower in the third tertile (-0.55%) than in first tertile (-0.72%) at FN after adjustment for age, weight, HRT, and other covariates (p < 0.05). When smokers were added to the nonsmoker group, the differences in bone loss rate between calcium intake tertiles disappeared. In addition, in ANCOVA, the term of interaction between smoking and calcium intake was statistically significant at LS only. In conclusion, smoking seems to impair the bone protective effects of nutritional calcium in postmenopausal women, more clearly in LS than FN.