Long-term heparin treatment causes osteoporosis through, an as yet, undefined mechanism. To investigate this phenomenon and to determine the relative benefits of low-molecular-weight heparin (LMWH) use, we treated rats with once daily subcutaneous injections of either unfractionated heparin (1.0 U/g or 0.5 U/g), the LMWH, Tinzaparin (1.0 U/g or 0.5 U/g), or placebo (saline) for a period of 32 days. The effects on bone were then compared both histomorphometrically and biochemically by measuring urinary type I collagen cross-linked pyridinoline (PYD) and serum alkaline phosphatase, markers of bone resorption and formation, respectively. Histomorphometric analysis of the distal third of the right femur, in the region proximal to the epiphyseal growth plate, demonstrated that both heparin and LMWH decrease cancellous bone volume in a dose-dependent fashion, but that heparin causes significantly more cancellous bone loss than does LMWH. Although both heparin and LMWH decrease osteoblast and osteoid surface to a similar extent, only heparin increases osteoclast surface. In support of these histomorphometric findings, biochemical markers of bone turnover demonstrated that both heparin and LMWH treatment produce a dose-dependent decrease in serum alkaline phosphatase, consistent with reduced bone formation, whereas only heparin causes a transient increase in urinary PYD, consistent with an increase in bone resorption. Based on these observations, we conclude that heparin decreases cancellous bone volume both by decreasing the rate of bone formation and increasing the rate of bone resorption. In contrast, LMWH, causes less osteopenia than heparin because it only decreases the rate of bone formation.