Bone histomorphometric reproducibility in normal patients

To study bone histomorphometry reproducibility in normal subjects, we performed during orthopedic surgery bone biopsies in 16 postmenopausal women. Each woman had four bone biopsies, two at the usual site in the iliac crest, one on the left and one on the right side, and two other biopsies just behind the usual site, one at each side. We performed measurements of trabecular bone volume, relative osteoid volume, osteoid surfaces, osteoclastic resorption surfaces and calcification front. The average values of the 16 patients were compared, on the one hand, two by two, by a student test, and on the other hand, by a variance analysis. By these two methods the results showed no significant difference between the average values of the 16 patients at each location for any of the histomorphometric parameters studied. However, there was a location variation which was estimated by the intra-individual variation for a given patient. On the other hand, we calculated from the variance analysis the location variance for a group of 10 to 100 patients. In any case all the parameters had a location variation which was high for osteoclastic resorption surfaces and relative osteoid volume when expressed in % of the absolute value of these parameters. The variation of the trabecular bone volume was 0--46. 15% (95% confident limit interval) in a single patient and the hypothetical value of the location variation was 41.6% for a group of 10 patients and 13.0% for a group of 100 patients.