Speed of sound (SOS) and broad-band ultrasound attenuation (BUA) were measured in cubes of human trabecular bone from lumbar vertebrae, in the three major anatomical axes. There were significant differences in sos and in BUA when measured in the different axes, indicating a structural component to the ultrasonic measurement. Qualitatively different behaviour was observed in the cranio-caudal (CC) axis compared to the transverse directions: SOS was approximately 500 m s(-1) greater than in either the lateral (LT) or antero-posterior (AP) axes, and BUA was approximately 23 dB MHz(-1) cm(-1) greater. Small, but significant, differences existed between the AP and LT axes for both SOS and BUA. In the AP and LT directions, strong linear correlations existed between sos and apparent density (r = 0.90), and between BUA and apparent density (r = 0.96). In the cc axis, correlations with density were poorer. The anomalous behaviour in the cc axis was due to a transient travelling ahead of the main wavefront, and it is suggested that this represents propagation of ultrasound directly through the trabecular framework as a bar wave. This can only occur in the cc axis where the majority of trabeculae are orientated parallel to the direction of propagation. Measurements on cubes in air, as opposed to water, supported this hypothesis. Modifications to the experimental technique necessary to consistently detect this phenomenon are described.

Degeneration of the intervertebral disc, seen radiologically as loss of disc height, is often associated with apparent remodelling in the adjacent vertebral body. In contrast, maintenance or apparent increase in disc height is a common finding in osteoporosis, suggesting the properties of the intervertebral disc may be dependent on those of the vertebral body or vice versa. We have investigated this relationship by measuring the radiological thickness of the subchondral bone and comparing it to the chemical composition of the adjacent disc. Sagittal slabs were sampled from lumbar spines obtained at autopsy and X-rayed microfocally. The thickness of the subchondral bone was measured and correlated with the composition of the adjacent intervertebral disc. Eighty-three cadaveric endplates were studied from individuals aged 17-85 years. There was regional variation in thickness of the subchondral bone, being greater adjacent to the annulus than the nucleus, and the endplates cranial to the disc were thicker than those caudal. There was a positive correlation between the thickness of the subchondral bone and the proteoglycan content of the adjacent disc, particularly in the region of the nucleus. A weaker correlation was seen here between water content and thickness, whilst there was no significant correlation at the annulus or between the bone thickness and collagen content. The positive relationship between the radiographic thickness of vertebral subchondral bone and the proteoglycan content of the adjacent disc seen in human cadaveric material could be due to the bone responding to a greater hydrostatic pressure being exerted by discs with higher proteoglycan content than by those with less proteoglycan present. It is suggested that while this is true in "normal" specimens, the relationship becomes altered in disease states, possibly because of changes to the nutritional pathway of the disc, with resultant endplate-bone remodelling affecting the flow of solutes to and from the intervertebral disc.

Bone mineral density (BMD) was measured in the lumbar spine using dual-energy X-ray absorptiometry in 222 unscreened women (aged 50-82 years), and information on back pain and historic loss of standing height was obtained at interview. Vertebral morphometry was performed on lateral spinal radiographs. The shape of the vertebral body was quantified using appropriate vertebral shape indices (VSIs), and vertebral deformities were identified using thresholds defined in terms of the means (M) and standard deviations (SD) of these VSIs for the whole group. Severity of deformity was defined as either grade 1 (M+2SD < VSI < M+3SD), grade 2 (M+3SD < VSI < M+4SD or grade 3 (VSI > M+4SD). Subjects with grade 1 vertebral deformities were older than subjects without such deformities, but did not have a reduced age-related Z-score of BMD. Grade 2 wedge and concave deformities were associated with a reduced age-related Z-score of BMD, suggesting that the aetiology of such deformities is closest to conventional concepts of 'osteoporotic fracture'. Grade 3 deformities were associated with neither increased age nor decreased BMD. Stature decreased in these subjects with age. Subjects reporting historic height loss had a higher mean number of wedge deformities. Subjects with back pain did not have a higher incidence of vertebral deformity than subjects without, confirming that many deformities were asymptomatic. Neither back pain nor historic loss of height were found to be associated with low spinal BMD.

No abstract

No abstract