The human body is made up of hard and soft tissues. The role of the hard tissue, the skeleton, besides being the main reservoir of minerals serving homeostatic mechanisms, is mechanical. Bones offer protection, support, and, most importantly for the purpose of this discussion, they provide anchorage points for striated muscles, which permit posture and movement. We take for granted that bones are hard and will resist the forces normally applied to them. We also accept that, with age, bone will be affected by senescence and will lose mass and resistance, resulting in fractures, decreased mobility, and chronic pain. But nobody expects such a clinical picture in a neonate, which is why many parents who take children with such symptoms to hospital are often suspected of abuse before severe osteogenesis imperfecta is diagnosed. Osteogenesis imperfecta, or brittle bone disease, is a heterogeneous syndrome. Some children have frequent fractures, which compromise survival, whereas others will have only a few fractures during childhood. There is no cure. Supportive care based on well defined occupational therapy or physiotherapy, timely corrective surgery of deformities, and control of chronic pain have helped, but they are unable to alter the natural course of the disease. Linkage studies have established an association between the disease and two genes (COL1A1 and COL1A2) that encode polypeptide chains that make up the type I collagen triple helix, the major component of bone matrix. Over 250 mutations have been characterised in both of these genes. However, trying to correlate mutations and severity of phenotype has been difficult, frustrating, and unsuccessful. Patients with identical mutations can have different degrees of disease severity. Additionally, data from laboratories that analyse collagen structure indicate that no collagen mutations can be found in about 25% of severe cases. Other factors are probably, therefore, involved in development of disease. Histological studies, for example, have allowed the characterisation of new forms of osteogenesis imperfecta with specific alterations (disorganised matrix, mineralisation defect, &c), and no detectable collagen mutations. Indeed, the pathophysiology of bone lesions in osteogenesis imperfecta remains unknown. Osteogenesis imperfecta is a disease of the osteoblast. The osteoblast produces an abnormal matrix that does not respond to mechanical loads. In compensation, the osteoblast population increases and osteoclast activity is raised, leading to a high bone turnover rate. Attempts to improve bone mass and structure with calcitonin, cortisone, growth hormone, parathyroid hormone, thyroxin, vitamins A, C, and D, and minerals (aluminium, calcium, fluoride, magnesium phosphate, and strontium) have not been encouraging. Furthermore, growth hormone increases bone mass in some patients, but the effects seem secondary to a promotion of growth rather than a direct effect on mass. Increased rate of bone turnover is compounded by secondary bone loss induced by immobilisation caused by lower limb fractures. In adults, bisphosphonates decrease this bone loss and slow bone turnover. In a group of 30 severely affected patients, bone mineral density and physical activity greatly increased, fracture rate decreased, and chronic pain disappeared, resulting in improved quality of life. Furthermore, several children confined to wheelchairs were able to walk (figure). No adverse effects on growth or bone modelling and repair have been seen. Such treatment can also be used in infants, in whom results are even better. Growth rate seems to be stimulated under therapy, suggesting that bisphosphonates might have a positive effect on bone formation. At the bone level, the main effect of the drug seems to be a thickening of cortical bone, with less of an effect on cancellous bone. There is also a decrease in bone turnover that might have immediate benefit, but whose long-term effects remain to be assessed. Bisphosphonates will probably be the treatment of choice for some time. However, because they are a symptomatic treatment and not a cure, bisphosphonates will eventually be superseded by treatment of underlying defects. Many thought that bone marrow transplantation, aimed at introducing normal stem cells, could result in cure. This approach, however, proved hazardous. Other forms of somatic gene therapy are under investigation. However, because of the complexity and variability of the basic abnormalities, this approach will take time to come to fruition. Elsa at 26 months old She received pamidronate for 2 years and had surgery to correct her leg deformities.