Noninvasive estimation of bone mass in ancient vertebrae

• Published in: American journal of physical anthropology 2004-10, Vol.125 (2), p.121-131
• Main Authors: Gonzalez-Reimers, E., Mas-Pascual, M.A., Arnay-de-la-Rosa, M., Velasco-Vázquez, J., Santolaria-Fernández, F., Machado-Calvo, M.
• Format: Article
• Language: English
• Subjects: Absorptiometry, Photon; Anthropometry; Atlantic Islands; bone density; Bone Density - physiology; Bone Diseases, Metabolic - diagnosis; bone histomorphometry; DEXA; Forensic Anthropology - methods; Fossils; Histological Techniques; Humans; Lumbar Vertebrae - anatomy & histology; Lumbar Vertebrae - diagnostic imaging; osteopenia; osteoporosis; prehistoric Canary Islands; Sensitivity and Specificity
• ISSN: 0002-9483; 1096-8644
• DOI: 10.1002/ajpa.10374

Histomorphometry is useful in the assessment of trabecular bone mass (TBM), and thus, in the estimation of the prevalence and intensity of osteopenia in ancient population groups. However, it is a destructive method. It is therefore necessary to explore the accuracy of nondestructive approaches, such as radiography, bone mineral density (BMD) assessed by double‐energy X‐ray absorptiometry (DEXA), bone density (BD), or optical density (OD) in the diagnosis of osteopenia. We selected 51 vertebrae out of a total sample composed of 333 T12, L1, and L2 vertebrae belonging to adult pre‐Hispanic inhabitants from El Hierro. These vertebrae underwent histomorphometrical analysis, a fine‐grained film radiography with assessment of trabecular pattern following standard methods, OD, DEXA‐assessed BMD, and BD. The presence of biconcave vertebrae and wedge‐shaped vertebrae was also assessed by measuring anterior height (a), posterior height (p), and height at the middle point of the vertebral body (m), and further calculating the indices 2m/(a + p) (“spine score”) and a/p. Significant correlations were observed between TBM and BMD (r = 0.43), TBM and BD (r = 0.49), TBM and OD (r = 0.52), BMD and OD (r = 0.51), and BMD and BD (r = 0.36), but not between TBM and the indices 2m/(a + p) and a/p. In the stepwise multiple correlation analysis between TBM and BMD, BD, and OD, OD entered into first place and BD into second place, whereas BMD became displaced; the multiple correlation coefficient was 0.63, with a standard error of 3.78. A BMD greater than 0.60 g/cm2, or a bone density greater than 0.60 g/cm3, excluded osteopenia (TBM 1.6 excluded a normal bone mass (TBM >20%) with a specificity greater than 90%. Based on radiographic criteria on the total sample, we also conclude that the overall prevalence of vertebral fractures in the adult pre‐Hispanic population of El Hierro of any age is 7.5%. Am J Phys Anthropol, 2004. © 2004 Wiley‐Liss, Inc.