In this study we examined whether the effect of continuously combined hormone replacement therapy (HRT) on bone metabolism is influenced by dydrogesterone dose, smoking and initial degree of bone turnover. In a double-blind randomized study, 123 healthy postmenopausal women (mean age 51.7 years; range 30-61 years) received 17 beta-estradiol, 2 mg orally per day, continuously combined with either 2.5, 5, 10 or 15 mg of dydrogesterone daily. At baseline and at 3 and 6 months of therapy, bone formation was assessed by determining total alkaline phosphatase (TAP), bone-derived alkaline phosphatase (BAP), and the carboxy-terminal propeptide of collagen type I (PICP) in serum; bone resorption was assessed by the calcium/creatinine (Ca/Creat) and hydroxyproline/creatinine (Hp/Creat) ratio in 2-h fasting urine, and the serum carboxy-terminal pyridinolyne cross-linked telopeptide of collagen type I (ICTP). Dydrogesterone dose did not influence the effect of HRT on any of the bone markers. Combining the data of the four treatment groups, the decrease in each marker, compared to baseline values, was significant. However, in non-smokers, compared to smokers, after 6 months of therapy the decline in BAP and TAP was significantly more pronounced and the plasma estradiol level was significantly higher. For each bonemarker at baseline, women in the highest quartile, compared to women in the lowest quartile, showed a significantly stronger decrease in this marker in response to HRT. We conclude that dydrogesterone dose does not modify the effectiveness of replacement therapy. However, smoking and a low bone turnover at baseline may diminish its beneficial effect on bone.