contends that changes in the composition of the nonosseous components of the bone tissue and changes in body composition may have a significant impact on BMD changes measured in our study of astronauts undergoing spaceflight. With DXA, apparent changes in BMD may be introduced artifactually through severe weight loss or severe changes in body composition that might alter the regional spatial distribution of fat around the body circumference.(1)In QCT, artifactual changes might be produced, for example, if the fraction of adipose tissue in the bone marrow (2,3) increased as a function of spaceflight. This would result in an artifactual increase in the measured amount of bone loss. However, there is no evidence to support the existence of such changes in spaceflight. Subjects undergoing longduration spaceflight undergo weight loss but not of the magnitude known to introduce significant measurement errors into DXA. Whole body composition studies during 4-to 6-month flights on the MIR (N ס 14) have shown slight gains in body fat mass (0.5 kg) with losses in lean mass of 2.1 kg for a total body loss of 1.7 kg or 2.2%.(4) Furthermore, a study of crew members during a 17-day spaceflight showed no significant change in the cellular fraction of spinal marrow.(5) This has also been confirmed for longer-duration MIR missions.Even if Dr Bolotin's contention was correct, QCT and DXA BMD measurements, which have greatly different sources of error, ought to show different results in our spaceflight cohort. In fact, our measured changes in DXA and in integral QCT BMD are very similar. In the spine, average rates of loss for QCT and DXA were identical at 0.9% per month. In the total femur region of the hip, the results for integral QCT and DXA BMD were 1.4%/month and 1.5%/month, respectively. Thus, two independent measurements of BMD showed almost identical results. The results observed for DXA in spaceflight are supported by results from bed rest studies, (6,7) in which BMD loss is observed in a model of mechanical unloading that does not involve weightlessness. Furthermore, the total body BMD changes were shown to be significantly correlated with Ca balance, a technique that does not require assumptions regarding body composition.(6) Ca balance measurements were also conducted during Skylab, showing significant bone loss consistent with the single-photon absorptiometry (SPA) technique used at that time. (8) Finally, the bone changes observed with densitometry are in keeping with bone structural changes observed in rodent hindlimb unloading models. In these cases, microstructural changes observed in the unloaded hindlimb with CT and histology (9) show loss of cortical area and degradation of trabecular structure in the unloaded hindlimbs that are consistent with the patterns observed in our own findings, with a higher rate of trabecular compared with cortical bone loss. (10) In conclusion, we disagree with Dr Bolotin's contention that measurement artifacts are responsible for our results. The contention is not supported by st...