JOURNAL OF BONE AND MINERAL RESEARCH Volume 9, Number 10, zyxwvutsrq 1994 Mary Ann Liebert, Inc., Publishers zyxwvutsrqp Brittle Bones in Spontaneously Diabetic Female Rats Cannot Be Predicted by Bone Mineral Measurements: Studies in Diabetic and Ovariectomized Rats J. VERHAEGHE,'.* A.M.H. SUIKER,3 T.A. EINHORN,4 P. GEUSENS,' W.J. VISSER,3 E. VAN HERCK,' R. VAN BREE,' S. MAGITSKY,4 and R. BOUILLON' ABSTRACT Spontaneously diabetic BB rats were sham operated (SO) or ovariectomized (OVX) within days after onset and studied after 4, 8, and 12 weeks. Analyses included histomorphometry of proximal tibial metaphyses, biochemical analyses of humeri, DXA analyses, and biomechanical testing of femora. In SO diabetic rats, no osteoblasts, osteoid tissue, or osteoclasts were present on the trabecular bone surface, but trabecular bone volume (TBV) remained normal compared with control BB rats. The concentration of IGF-I per dry weight of humerus was decreased after 12 weeks of diabetes, whereas the concentrations of calcium and osteocalcin did not change. DXA analysis showed normal bone mineral density (BMD) at both diaphyseal and metaphyseal femoral areas. On biomechanical testing, angular deformation, energy absorption, and torsional strength of the femora were decreased after 8-12 weeks of diabetes, but stiffness was normal. Ovariectomy in diabetic rats caused a decrease in femoral BMD especially at the metaphysis, and there was a trend toward decreased TBV in the tibial metaphysis; TBV loss was less marked than in control OVX rats, however. The increase in BMD at the femoral diaphysis, measured after 12 weeks of OVX in control rats, was absent in diabetic rats. Multiple- regression analysis indicated that the presence of diabetes but not ovariectomy, weight, and mineral content correlated with decreased energy absorption, angular deformation, and strength of the femora. The data infer that the (near) absence of unmineralized bone matrix in severely diabetic rats alters bone microarchitecture and ultimately results in brittle bones, which is not predicted by BMC or BMD measurements. INTRODUCTION T REMAINS UNCERTAIN whether insulin-dependent diabetes I mellitus constitutes a risk factor for the development of skeletal fractures. We have concluded from a survey of pub- lished reports that the risk is probably about twofold increased in diabetic patients."' However, in most of these reports no distinction was made between insulin- and non-insulin-depen- dent diabetes. Non-insulin-dependent diabetes is frequently associated with obesity and is, in effect, a protective factor against appendicular bone 10~s.'~' Second, there are no data available on variables that may influence the incidence of fractures in insulin-dependent diabetic subjects: the duration of diabetes, long-term diabetes control, and the presence of com- plications, such as frequent hypoglycemia or neuropathy zy . In diabetic rats, there is a more consistent picture of increased bone fragility. Dixit and Ekstrom reported decreased breaking strength after 8 weeks of alloxan-induced diabetes."' Our groups have demonstrated that all classically studied biome- chanical parameters are diminished in male BB rats after 12 'Labnratorium voor Experimentele Geneeskunde en Endocrinologie, Katholieke Universiteit Leuven, Lnuvain, Belgium 'Department of Obstetrics and Gynaecology, Katholieke Universiteit Leuven, Lnuvain, Belgium. 'Afdeling Botpathologie, A.Z. Utrecht, Utrecht, The Netherlands. 4Department of Orthopaedics, Mount Sinai School of Medicine, New York, New York. 'Arthritis and Metabolic Bone Disease Research Unit, Katholieke Universiteit Leuven, Louvain, Belgium. 1657