Radiologically diagnosed postmenopausal osteoporotic patients with at least one nontraumatic vertebral flattening were treated for one year with either oral pamidronate (APD), 300 mg/day plus calcium 1 g/day (n=39) or with calcium alone (n=21). Bone mineral density (BMD) was assessed in lumbar spine, femoral neck, trochanter and Ward's triangle by dual X-ray absorptiometry in order to determine the number of responders at each site. As no densitometric inclusion criteria were stipulated, wide inter- and intra-individual variations in both regional basal BMD and response to therapy were found. However, the APD-treated group showed significant mean BMD increases in spine (+3.1%; p < 0.001) and femoral neck (+3.2%; p < 0.002) versus basal level, whereas the calcium only group failed to exhibit significant differences. The entire 60-strong population was then split into two groups, according to whether individual BMD content was greater or less than the mean basal value for each skeletal site evaluated. For either treatment, subpopulations with lower basal BMD tended to achieve greater bone gain, though statistically significant differences were only disclosed at trochanter (p < 0.004) with APD and at femoral neck (p < 0.002) in the calcium only group. Globally speaking, increases in BMD were observed in 60-80% of patients receiving either treatment - who were thus defined as responders - at each particular skeletal area assessed. However, when only skeletal areas with low basal BMD were considered, the number of responders reached 60-100%. Responsive sites varied among patients: out of 56 cases, 9 (24%) on APD and 6 (32%) on calcium alone responded in all 4 areas evaluated, while a single case on the latter treatment failed to show BMD response at any site. Overall, the mean number of responsive sites was 2.7. Odds ratios were calculated considering treatment modality and high or low basal BMD as parameters, but no significant differences were found in the number of responders. It may be concluded that APD induces moderate lumbar and femoral neck bone mass gain in severe postmenopausal osteoporosis, whereas calcium alone leads to non significant variations, both findings being in agreement with reported data. Therefore, evaluated APD doses enhance mineralization in responsive sites alone, but fail to increase the total number of responders. Interestingly, responsive sites seem to be those relitively spared by the course of the disease.