The influence of recombinant human insulin-like growth factor-I (rhIGF-I), its binding protein-5 (IGFBP-5) or their equimolar complexes on calvarial osteogenesis was investigated by quantitative radiography and histomorphometry after local administration to adult mice or mature rats. The systemic effects of these proteins were investigated in aged Sprague-Dawley rats with regard to their ability to prevent or restore bone mass in ovariectomy induced osteopenia as assessed by radiography, dual-energy X-ray absorptiometry (DEXA) analyses, peripheral computerized tomography (pQCT) and mineral analyses after daily s.c. administration for 3 or 8 weeks following a bone depletion period of 8 weeks. Bone mass of murine calvariae was significantly increased in a dose-dependent manner by the complex 7 days after discontinuation of local administration for 19 days in mice, whereas IGF-I alone expressed only weak effects. IGFBP-5 alone was ineffective in this respect. In the same model, only the complex had a weak osteogenetic potential in 7 week or 5 month old rats. Systemic long-term treatment with the complex of rhIGF-I/IGFBP-5 (2.0/7.6 mg/kg/day, s.c.) for 8 weeks resulted in significantly increased cortical thickness, area and mineral density in femoral midshaft or tibial metaphysis suggesting periosteal bone formation. This was obviously related to increased muscle strength since these effects were parallelled by increased body weight. No effect on trabecular bone occurred as demonstrated by site-specific analyses (vertebrae, proximal tibia) using DEXA, pQCT and radiography. This selective action of rhIGF-I/IGFBP-5 on periosteal bone formation is unique for an IGFBP. Femoral ash and calcium content, both corrected for tissue volume, increased slightly. However, when the increase in cortical thickness and bone mass was corrected for bone size, the effects are nearly abolished, suggesting an additional effect of bone growth. This potential deserves further evaluation in order to differentiate between effects on cortical bone via muscle strength and lack of efficacy on trabecular bone balance.