If the objective is to minimize peri-implant strain in the crestal alveolar bone, a wide and relatively long, untapered implant appears to be the most favorable choice. Narrow, short implants with taper in the crestal region should be avoided, especially in low-density bone.

The aim of this study was to compare implant-bone interface stresses and peri-implant principal strains in anisotropic versus isotropic three-dimensional finite element models of an osseointegrated implant in the posterior mandible. We obtained anisotropic (transversely isotropic) elastic constants for mandibular bone and derived equivalent isotropic constants by averaging over all possible spatial orientations. A finite element model was constructed using ten-node tetrahedral p-elements, providing curved edges where necessary and increasing the accuracy of the results in regions of high stress gradients. Perfect bonding was assumed at the implant-bone interface. An oblique load was applied at the coronal aspect of the crown with 100 N vertical and 20 N bucco-to-lingual components. Implant-bone interface stresses exceeded reported bond strengths and principal strains reached yield strain levels in the cortical crest. Anisotropy increased what were already high levels of stress and strain in the isotropic case by 20 to 30% in the cortical crest. In cancellous bone, anisotropy increased what were relatively low levels of interface stress in the isotropic case by three- to four-fold to exceed bond strength levels. Anisotropy has subtle, yet significant effects on interface stresses and peri-implant strains and careful consideration should be given to its use in finite element studies of dental implants.

The elastic moduli have not been reported for cancellous bone from the edentulous mandible. Accurate values are needed for finite element modeling of the mandible. The aim of this study was to determine elastic modulus values in three orthogonal directions for cancellous bone taken from an edentulous jaw and to relate these values to apparent density and volume fraction. Seven samples were obtained from the edentulous mandible of a 74-year-old female. Young's modulus was determined by compression testing of cubes cut with the faces aligned with the anatomic axes. Bone volume fraction averaged 0.33 (SD 0.14) and apparent density averaged 0.55 g/cc (SD 0.29). Young's modulus was greatest in the mesio-distal direction (mean 907 MPa, SD 849 MPa), followed by the bucco-lingual (mean 511 MPa, SD 565 MPa) and infero-superior direction (mean 114 MPa, SD 78 MPa). The infero-superior direction was less than the bucco-lingual (P = 0.03) and mesio-distal (P = 0.002). The mesio-distal and bucco-lingual directions could not be shown to be different (P = 0.32). This suggests a model of transverse isotropy for cancellous bone in the jaw, where the symmetry axis is along the infero-superior (weakest) direction.

Structure, biomechanical competence, and incremental NMR line broadening (R') of water in the intertrabecular spaces of cancellous bone were examined on 22 cylindrical specimens from the lumbar vertebral bodies of 16 human subjects 24-86 years old (mean, 60 years old). A strong association (r = 0.91; P < 0.0001) was found between Young's modulus of elasticity and R' for a wide range of values corresponding to cancellous bone of very different morphologic composition. NMR line broadening is caused by the inhomogeneity of the magnetic field induced as a consequence of the coexistence of two adjacent phases of different diamagnetic susceptibility-i.e., mineralized bone and water in the marrow spaces. Structural analyses performed by means of NMR microscopy and digital image processing indicated that the variation in R' is closely related to the trabecular microstructure. Mean trabecular plate density measured along the direction of the magnetic field was found to play a major role in predicting Rj (r = 0.74; P < 0.0001). This behavior was confirmed when the plate density was varied in individual specimens, which was achieved by rotating the specimen, making use of the bone's structural anisotropy. It is concluded that the NMR transverse relaxation rate in human cancellous bone of the spine is significantly determined by trabecular structural parameters relevant to biomechanical strength. The results further underscore the important role played by the transverse trabeculae in contributing to cancellous bone strength. The work has implications on possible in vivo use of quantitative magnetic resonance for the assessment of fracture risk in osteoporotic patients.It is widely accepted that the mechanical strength of the skeleton in vertebrates is largely determined by the material density of trabecular bone, generally referred to as bone mineral density (BMD). Osteopenia, characterized by a loss in bone mass, leads to impaired bone strength, which, in turn, has been associated with atraumatic vertebral fractures suffered by patients with osteoporosis (1). The association between bone strength and BMD has been suggested by a large number of studies, with bone strength statistically showing a quadratic dependence on density (2). However, in vivo BMD measurements by either quantitative computed tomography or dual-energy x-ray absorptiometry do not very well predict fracture risk in osteoporotic patients (3).More recently, the search for a better predictor of bone strength, and thus fracture risk, has shifted toward trabecular bone morphology (4-8). Kleerekoper (5) found in a study in which postmenopausal women with vertebral fractures were compared to age-matched normal women of equal mean BMD that osteoporotics had significantly lower trabecular plate density and, as a corollary, higher trabecular plate thickness. The data were interpreted as suggesting that anThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accorda...

No abstract

Despite recent interest in the study of shoulder kinematics, there is considerable controversy in the literature regarding translations at the glenohumeral joint. The purpose of this study was to investigate the key factors that control shoulder motions, thus leading to a better understanding of joint function. Translation and rotation patterns were studied in fresh-frozen glenohumeral joints of human cadavers with a six-degrees-of-freedom magnetic tracking device. Shoulders were positioned from maximal internal to external rotation at several arm positions (various elevations and planes of motion). In order to determine the effect of muscle forces, joints were positioned both actively and passively. Additionally, articular surface geometry and ligament origin-insertion wrap lengths were measured to assess their influences on joint kinematics. When joints were positioned passively, large translations were observed at the extremes of motion. With active positioning, muscle forces tended to limit humeral head translations, principally by restricting rotational ranges of motion. However, when data from the passive model were reanalyzed by considering only the rotational ranges of motion seen actively, no significant differences in translation were found between the two models. Joint conformity was found to have a significant influence on translations during active positioning but not during passive positioning. Glenohumeral ligament wrap lengths, however, correlated with translations when joints were positioned passively but not when positioned actively. Findings from this study emphasize the importance of muscle forces in keeping the humeral head centered in the glenoid. Although large translations are possible, they can be achieved only with increases in rotational ranges of motion associated with the removal of muscle force. Additionally, joint conformity appears to play a role in controlling translations during active motions, whereas capsular constraints become more important during passive motions.

Pulse transmission ultrasound was used to determine the longitudinal wave speed along the direction of trabecular alignment in 32 water-saturated anisotropic tibial bovine cancellous bone samples and in one cortical bone sample also from the bovine tibia. These results are compared to published ultrasound wave-speed data obtained from bovine femoral specimens. Nonlinear regression was used to fit Biot's theory to the data. The correlation coefficient for regression analysis between the experimental ultrasound velocities and the velocities predicted by Biot's theory was r = 0.78.

No abstract