Portal de Periódicos da CAPES

Dual-energy X-ray absorptiometry for the measurement of bone and soft tissue composition

1995; Elsevier BV; Volume: 14; Issue: 5 Linguagem: Inglês

10.1016/s0261-5614(95)80062-x

1532-1983

P. Tothill,

Bone health and osteoporosis research

The scene was set for absorptiometry using ionising radiation by the discovery of X-rays by Roentgen in 1895 and radioactivity a year later by Becquerel. It was apparent from the first that X-rays would penetrate tissue, but unequally, so that there was more attenuation by bone than by soft tissue. The attenuation depends on the thickness of the tissue, on the atomic number of the constituent elements and on the energy of the radiation. Early development of radiology concentrated on improvements in resolution and speed. Densitometry to estimate amounts of bone was at first based on the use of film as the detector, but its non-linearity and non-uniformity limited accuracy. These limitations were overcome for peripheral sites by Cameron and Sorensen (1) who introduced what came to be known as single-photon absorptiometry (SPA). They used a mono-energetic gamma-ray source, a finely collimated beam which scanned the area of interest and a scintillation detector. Variations in the thickness of tissue traversed were eliminated by immersing the forearm in a water bath. Such a technique of eliminating one variable is not feasible for measurements of the axial skeleton, so dual-photon absorptiometry (DPA) was introduced (2). Again, radionuclide sources were used initially, the most successful being 153Gd, which provides two energies, of about 40 and 100 keV. Good measurements of bone mineral in spine and femur were obtained. Some progess was made in measuring totalbody bone mineral (3), but measurement times of around one hour were required. Another disadvantage of radionuclide DPA was the decay of the 153Gd, leading to a varying intensity and the need for expensive replacements. These factors led to a reversion to X-ray sources with two different effective energy beams obtained by very rapid switching of the high energy supply, or division of the spectrum by absorption-edge filtration, leading to DXA (4, 5). DXA is now the most widely practised means of assessing bone mineral status, giving excellent precision (reproducibility) and rapid measurement, with whole-body scanning taking no more than 20 min. In the latest developments of equipment, a fine pencil beam is replaced by a 'fan beam' and the single detector by an array of detectors, giving increased speed and finer resolution, allowing morphometry of the lumbar spine. The accuracy of DXA, or its ability to obtain the correct answer, is not easy to achieve or to assess. The reasons lie in the complicated mixture of tissues and the assumptions that are necessary. Measurements of the transmission of the two different energy beams and a knowledge of the attenuation coefficients provide the basis for the determination of two components, for example bone and soft tissue. Unfortunately, in X-ray attenuation terms, there are three major components bone mineral, lean soft tissue and fat soft tissue. Compared with lean tissue, fat appears as 'negative bone'. DXA provides only two equations to solve the unknowns, so a completely valid solution is not possible. The nature of the attenuation processes is such that it is not feasible to add a third energy to solve the three unknowns, It is necessary therefore to make assumptions about fat distribution in order