We have been taught that Legg-Calvé-Perthes disease usually results in femoral head deformity because collapse within the area of bone necrosis occurs before new bone formation gives adequate support. During our training, we were exposed to a number of strategies to avoid loss of femoral head sphericity. Mechanical treatments included non-weight-bearing, bed rest, a variety of braces, and traction. Most surgical approaches were reserved for older children. We were taught that with proper femoral head coverage and force protection, creeping substitution would be more likely to gradually replace dead bone with new living bone before severe deformity occurred1.
The article by Kim et al. describes a piglet model for investigating weight-bearing (WB) compared with non-weight-bearing (NWB) management. A hind-limb amputation above the knee was necessary to make the animal non-weight-bearing. With the animal forced into non-weight-bearing, compliance was not an issue. The piglets were the age equivalent of a four to five-year-old child. Humans usually develop Legg-Calvé-Perthes disease after that age, and younger children do better than those who are older. Eight weeks after placing a tight ligature around the femoral neck and transecting the ligamentum teres, measurements of bone resorption, new bone formation, and vascularity were made in the devascularized femoral heads in the NWB and WB groups. The contralateral hips in the WB group served as the controls. The femoral epiphyses in the NWB group had less bone but greater vascularity compared with the WB group at eight weeks. Compared with the controls, new bone formation was reduced in the devascularized hips and there was no difference between the WB and NWB groups. Since the study was ended at eight weeks, it is unknown whether non-weight-bearing would have had any long-term effect on femoral head collapse.
In an earlier article this year, Kim reviewed currently used force-protective and surgical treatments for Legg-Calvé-Perthes disease and investigations on pharmacologic approaches2. The most interesting and encouraging topic was a review of antiresorptive therapies to prevent bone loss. Osteoprotegerin blocks RANKL (receptor activator of nuclear factor kappa-B ligand). In animal models, binding of osteoprotegerin to RANKL prevents the RANK-RANKL interaction and effectively inhibits osteoclast formation, activation, and survival. Aminobisphosphonates accelerate osteoclast resorptive activity and accelerate apoptosis. To avoid systemic effects of bisphosphonates, local injections have been tested in an animal model. Reduced osteoclastic activity occurs following both treatments, but the vascularity required to deliver the drug affects the response. Anabolic strategies were also discussed since the antiresorptive interventions fail to be accompanied by vigorous new bone formation.
The coupling of bone resorption with new bone formation maintains bone structure. There are many determinants to such homeostasis3. Osteoporosis is a classic example of a gradual negative shift of this balance. Bone resorption does not always result in replacement with new bone. Osteoarthritis is a focal example of that phenomenon. The variation in radiographic and pathologic patterns in osteoarthritic femoral heads occurs because bone may be replaced with bone, cartilage, fibrocartilage, myxomatous tissue, or cysts4. The risk factors and cause of vascular interference in Legg-Calvé-Perthes disease are unknown, whereas the factors and mechanisms of vascular interference in adult osteonecrosis are well known. The size and location of femoral head lesions are the predominant predictors of deformity and degenerative disease5. Similar to the situation in osteoarthritis, the bone replacement response in osteonecrosis varies widely and is under diverse influences that could affect interventions6. Because an osteoclastic drive is the initial process in these disorders, antiresorptive agents are being investigated for Legg-Calvé-Perthes disease and osteonecrosis. The uncoupling of new bone formation from bone resorption implies that some biochemical features are not characteristic of the normal bone turnover state seen in healthy individuals and during fracture-healing. The road to more appropriate interventions will be paved by a better understanding of why this uncoupling occurs and how the uncoupling in each type of disorder varies among individuals.
With good reason, Kim et al. feel that antiresorptive and anabolic strategies for Legg-Calvé-Perthes disease will require rigorous investigation. Little is known regarding the long-term effect of antiresorptive or anabolic agents on children. Understanding the biologic factors involved in Legg-Calvé-Perthes disease will likely affect how any individual therapy will affect the long-term outcome for a specific individual. Given the broad spectrum of conditions associated with bone resorption and failure of new bone formation, many orthopaedic subspecialties will be involved in future studies.