There have been conflicting reports about the effect of diabetes on bo
ne density. In 1978, we studied 109 patients, 46 with type I and 63 wi
th type II diabetes; similar to 12 years later we restudied 35 of the
66 surviving patients. In the original study, radial bone density did
not differ significantly between patients with either type of diabetes
but was significantly lower than in nondiabetic control subjects. In
eight osteopenic patients, bone formation rate and other histological
indexes of osteoblast recruitment and function were markedly depressed
compared with those in nondiabetic control subjects. In ,patients rem
easured similar to 2.5 years (41 patients) and similar to 12.5 years (
35 patients) after baseline, bone loss had continued at the expected r
ate in patients with type I diabetes, with maintenance of the same def
icit, but was slower than expected in patients with type II diabetes,
such that the initial deficit had been completely corrected. In six of
the eight patients who had undergone bone biopsy, one with type I and
five with type II diabetes, the mean bone mineral density z-score of
the spine and femoral neck similar to 12 years later was >0 and in one
subject was significantly higher than normal at both sites. Based on
these data and on previous studies, we propose that in patients with d
iabetes, low bone formation retards bone accumulation during growth, m
etabolic effects of poor glycemic control lead to increased bone resor
ption and bone loss in young adults, and low bone turnover retards age
-related bone loss. These effects account for low bone density in youn
g patients with type I diabetes and normal or increased bone density i
n older patients with type II diabetes.