Ten children with either Sillence type III or IV osteogenesis imperfecta participated in a randomized prospective study evaluating discontinued use of long leg braces. Each patient was paired with a partner determined to have similar clinical characteristics. Within each of the five pairs, one child received braces for 16 months and the other did not. Then the assignment was switched. The children were evaluated every 4 months at NIH. At each admission, children were measured for height, weight, bone density (vertebral...radius), and range of motion, and were tested for muscle strength. Also, x rays were taken of the long bones of the lower extremities and spine, and the children underwent video-based gait analysis. In addition, we received monthly information, completed by parents, describing each child's activities for a 2- week period. A clinical severity rating was performed by two clinicians at each NIH visit. Data show that there was no significant carryover effect from being braced first (implying the braced and unbraced intervals could be compared). During the study muscle strength declined in both braced and unbraced intervals. The latter showed a more substantial decline. The quadriceps was most affected. Clinical severity scores, a functional assessment, were higher during the unbraced interval, implying that children were more active during the braced intervals (P=0.02). While in braces, children participated in more activity at the moderately active level of intensity and less minimal activity (i.e., less sedentary and more upright activity) (P=0.02). More lower extremity fractures occurred during the unbraced than braced intervals, but this finding was not statistically significant (P=0.21).
