We agree with Dr. Kamel that reversal of renal calcium leak may be one mechanism whereby hydrochlorothiazide preserves bone density in older adults, a possibility noted in our discussion. We also noted that evidence supports other mechanisms as well.

In our trial, calcium deficiency did not explain the observed beneficial effects of thiazide. Although we did not provide calcium supplements, we did inform participants of their baseline calcium intake. Subsequently, at every visit, we assessed calcium intake and actively encouraged participants to maintain intakes of at least 1000 mg/d. As shown in the Figure, the mean total daily calcium intake during the study was 1600 mg/d among women in the placebo group compared with 1400 mg/d among women in the two hydrochlorothiazide groups. In men, the mean calcium intake was about 1200 mg/d in all three study groups during the 3-year follow-up. Therefore, the significant treatment effect seen with hydrochlorothiazide is over and above any effect of adequate calcium intake. On the basis of data from Dawson-Hughes and colleagues (1), the relatively stable levels of bone density in our placebo group might be attributed to adequate calcium intake. We are currently investigating the renal physiologic effects of thiazide in these participants. This analysis is complex, involving such factors as calcium, sodium, and protein intakes; urine sodium, phosphate, and creatinine levels; and serum levels of calcium, creatinine, and parathyroid hormone.

View larger version (33K): [in this window] [in a new window]   Figure. Total calcium intake from diet plus supplements during 3 years of follow-up among 205 women (top) and 115 men (bottom), according to study group.

We agree that evidence from a fracture outcome trial is lacking as a basis for recommending hydrochlorothiazide for fracture prevention in healthy older adults. On the basis of available experimental and epidemiologic evidence, we continue to believe that low-dose thiazide could play a role in prevention of bone loss.