Cortical bone maintenance and geometry of the tibia in prehistoric children from Nubia's Batn el Hajar.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 66275-280 (1985) Cortical Bone Maintenance and Geometry of the Tibia in Prehistoric Children From Nubia’s Batn el Hajar DENNIS P. VAN GERVEN, JAMES R. HUMMERT AND DAVID B. BURR University of Colorado, Boulder 0 . P . V,J . RH.), West Virginia University, Morgantown D.B.B.) KEY WORDS Nubia, Kulubnarti, Cortical bone maintenance, Bone geometry, Growth. ABSTRACT The relationship between advancing age in adults and patterns of cortical bone maintenance has been extensively documented for archaeological populations (Dewey, et al., 1969; Van Gerven et al., 1969; Perzigian, 1973). Most recently, this research has been expanded to include a more thorough consideration of the geometric properties of bone in relationship to adult age changes (Martin and Atkinsin, 1977; Ruff and Hayes, 1983). To date, however, few studies have documented subadult patterns of cortical bone maintenance in archaeological populations and none have incorporated the relationship between patterns of cortical bone loss and gain and the changing geometric properties of growing bone. Using a sample of 172 tibias from children excavated from the Medieval Christian site of Kulubnarti, located in Nubia’s Batn el Hajar, the present research examines the relationship between percent cortical area, bone mineral content, and cross-sectional moments of inertia. Among these children, bone mineral content increases steadily from birth in spite of a reduction in percent cortical area during early and late childhood. It appears, therefore, that tissue quality of the bone is not adversely affected by the reduction. Furthermore, the reduction in percent cortical area in later childhood corresponds to a dramatic increase in bending strength measured by cross-sectional moments of inertia. Thus, whether this cortical reduction is due entirely or in part to either normal modeling or nutritional stress, the tissue and organ quality of the bone is not adversely affected. Few statements in the annals of physical anthropology have proven to be as incorrect and yet as useful as Jarcho’s (1964) suggestion that ancient human populations did not experience osteoporosis. Beginning with Dewey’s initial study of cortical thickness in adult males and females from three ancient Nubian populations, a host of research has demonstrated a broadly universal pattern of bone loss with aging in ancient times generally consistent with patterns observed for modern adults (Dewey, et al., 1969; Van Gerven et al, 1969 Perzigian, 1973). While ancient females tended to lose bone sooner and few survived to experience the consequences of clinical osteoporosis, then as now females lost significant amounts of bone with advancing age while men showed less advanced-age changes. 0 1985 ALAN R. LISS, INC. Beyond extending our knowledge of agerelated bone loss in time and space, the analysis of ancient skeletal remains has contributed important methodological and theoretical insights into skeletal biology as a whole. Unlike data derived from extant populations where direct measurement is frequently limited to biopsy or dissecting room materials, archaeological remains have provided large samples of whole bones suitable for a variety of direct and indirect methods of measurements. Van Gerven and co-workers (1969) used archaeological remains to conduct a controlled comparison of direct and x-ray measurement techniques as applied to the measurement of cortical thickness and Received June 25,19&1, accepted September 17,1984. 276 D.P. VAN GERVEN, J.R. HUMMERT, AND D.B. BURR cortical area in the human femur. By comparing estimates of cortical area calculated from antero-posterior x-rays with direct measurements made from cut cross-sections of the same bones, the research demonstrated a serious bias in the x-ray technique. By failing to account adequately for differential bone loss around the circumference of the bone as well as differences in cross-sectional geometry resulting from sex differences and age changes in femur morphology, the x-ray technique seriously overestimated cortical area in older females and underestimated it in males. Carlson et al. (1976)later sectioned the same femurs at five sites along the diaphysis in order to further assess differential rates of remodeling and bone loss within the femur. Most recently Ruff and Hayes (1983) sectioned a sample of adult femurs and tibias from Pecos Pueblo at ten diaphyseal sites (five femur and five tibia). In order to assess age changes in the geometric and mechanical properties of the bones in addition to patterns of tissue loss and gain with aging, they measured first and second moments of area in addition to cortical, total subperiosteal, and medullary areas. Their results revealed an interesting interaction between age, cortical area, and geometry of the diaphysis. While females eventually lose cortical area as endosteal resorption exceeds subperiosteal gain, bending strength of the diaphysis in both the femur and tibia is not adversely affected. In fact, the age-related redistribution of cortex from the endosteum to the subperiosteal surface results in an increase in both maximum and minimum bending strength (maximum and minimum moments of area). Beyond demonstrating through direct measurement an interaction between bending strength and geometry initially measured indirectly through x-ray (Smith and Walker, 1964), Ruff and Hayes have added an important geometric and mechanical perspective to the question of osteoporotic bone loss. While the role of archaeological materials in the study of adult cortical bone maintenance has been clearly extensive, the analysis of subadult remains has been less so. This is unfortunate in view of a substantial body of data derived largely through radiographic techniques suggesting that cortical bone loss is not restricted to adults. For example, a number of researchers (Garn et al., 1964, 1969; Garn,1970; Frisancho et al., 1970 a,b; Himes et al., 1975, 1976; Himes, 1978) have found that among living populations relative reductions in bone mass occur among subadults. Such reductions are frequently associated with nutritional stress. For example, Garn et al. (1964) radiographed Guatemalan children suffering from protein+alorie malnutrition. Relative to healthy children, Garn observed a significant reduction in compact bone. Of particular interest in light of Ruff and Hayes’ recent mechanical and geometric assessment of adult long bones, Garn found that among malnourished children there was a trend toward greater subperiosteal diameters and reduced cortical bone (Garn, 1969). Among the few studies of prehistoric children that have been done, a similar pattern emerges. Cook (19791, for example, observed a pronounced loss of cortical thickness in the femur of Late Woodland infants during what she estimated to be the weaning period. Hummert (1983) examined the relationship between longitudinal growth of the tibia1 diaphysis and percent cortical area (cortical bone aredtotal subperiosteal area) in a large series of ancient Nubian children from the site of Kulubnarti in Nubia’s Batn el Hajar. He found that among these children growth in bone length as well as cortical area measured at the midshaft was well maintained from birth through age 16 but that percent cortical area reflected excessive endosteal resorption during early and later childhood. In other words, while absolute cortical area increased steadily with bone growth, it lost ground to the medullary canal as a percentage of total area within the periosteum. Hummert concluded that this relative reduction in bone was generally consistent with cortical bone loss observed among malnourished living children and interpreted his results as additional evidence of the dietary stress previously reported for the population (Van Gerven et al., 1981). Taken together, analyses of cortical bone maintenance in living as well as archaeologically derived subadults suggest a complex interaction between growth, periosteal apposition, endosteal resorption, and cortical bone maintenance similar to that observed for adults. Research on subadults has not, however, taken the next logical step suggested by Ruff and Hayes’ analysis of the adult femur and tibia, that is, to ask how changes in 277 CORTICAL BONE MAINTENANCE IN NUBIAN CHILDREN the geometric properties of growing bone affect bone strength during periods of bone loss and gain. The purpose of the present research is to examine the relationship between percent cortical area, bone mineral content, and bending strength (proportional to antero-posterior and medio-lateral cross sectional moments of inertia) in the subadult tibia. By examining the relationship between these variables, it should be possible to further assess the impact of cortical bone changes on the organ and tissue properties of this long bone in children. MATERIALS AND METHODS The materials selected for study were excavated by the senior author from the Medieval Christian site of Kulubnarti located some 80 miles south of Wadi Halfa in Nubia's Batn el Hajar (now part of the Republic of Sudan). They represent the same subadult sample examined by Hummert (1983) for changes in percent cortical area. The total sample of 172 individuals ranged in age from birth (0 years) to age 16. Age categories and sample sizes within categories is presented in Table 1. Age at death was determined by patterns of dental formation and eruption as discussed by Van Gerven et al. (1981). While sex could be determined in a few cases as a result of preserved genitalia, the sample was predominantly unsexable. In all cases the quality of preservation was excellent and in most cases some soft tissues such as skin and hair were also preserved. Upon sectioning at midshaft, many tibias contained some dried blood and other soft tissues in vascular spaces within the bone. After sectioning, each individual was examined under a x 10 magnifying lens while positioned beneath a transparent millimeter grid. Calculations were made for total subperiosteal area, cortical area, and medullary area by counting grid intersections as discussed by Martin and Armelagos (1979) and Hummert (1983). In order to reduce error, each count was repeated three times and averaged. Percent cortical area was then calculated by dividing cortical area by total subperiosteal area. Measurements of bone mineral content and cross-sectionalmoments of inertia were made using a Norland Cameron Bone Mineral Analyzer with a standard 118-inch detector TABLE 1 Sample size Age 0-1.5 2-4 5-7 8-10 11-13 14-16 Total 46 43 40 21 16 6 172 collimator, interfaced to a MITS microcomputer with an A/D input. Each tibia was wrapped in tissue equivalence bag and scanned at midshaft in the AP and ML planes. Two scans were made in each plane and the values averaged. The photon absorption curve was recorded for each scan, and the bone mineral content (BMC) and crosssectional moments of inertia about the ML and AP axes were calculated for this curve. The antero-posterior cross-sectional moment of inertia (AP CSMI) is therefore proportional to the antero-posterior bending strength, and the medio-lateral cross-sectional moment of inertia (ML CSMn is proportional to medio-lateral bending strength. Details of these techniques and the method of calculating the variables are described elsewhere (Martin and Burr, 1984). RESULTS Figure 1presents tibia length and percent cortical areas by age group for the Nubian sample. While growth in tibia1 length appears well maintained from birth onward, agechanges in percent cortical area reveal a more complex pattern. For example, there is a considerable decline in percent cortical area after birth followed by a steady increase to the 12-year age group. While such a decline followed by steady increase is normal for modern children, Hummert (1983) has argued that the overall Nubian values are low in comparison to well-nourished modern children. Also, while modern children show a decline after birth, the pattern is typically reversed by 9 months of age. The continuing decline among the Nubian children to age 3 appears to reflect the consequences of weaning stress (Hummert, 1983). Following age 12, the midchildhood trend toward an increase in percent cortical area with age is reversed. Percent cortical area in the 15-yearage group is on average 4% below that of the 12-year group (72% compared to 278 D.P. VAN GERVEN. J.R. HUMMERT, AND D.B. BURR from birth through age 16. Standardizing by cortical area produced a Merent result. Mineral density (g/cm3) undergoes a steady increase to age 6 and then levels off through the remaining age groups. The difference between the two techniques can be explained I60 in terms of the geometry of the tibia at midshaft. At birth and through the early years of life, the tibia is basically tubular in shape. , , , 00 Consequently, bone width measurements and cortical area produce comparable estimates 550 2 4 0 LO 12 14 16 of the bone actually present in the cross secw tion. In the present data this is reflected in Fig. 1. The relationship between age, tibial length, the basically parallel values from birth and percent cortical area (PCA). through age 6. However, as the geometry of the bone changes in later childhood, bone width measurements increasingly underes68%)and 2% below the 9-year age group (70% timate the amount of bone actually present. compared to 68%).This change corresponds Thus, while BMC/BW continues to rise after to a period of rapid longitudinal growth in age 6, BMC/CA levels off. In both cases, howlength off the tibial diaphysis (Hummert, ever, bone mineral content, as a measure of 1983). Huss-Ashmore et al. (1982) observed a tissue quality, does not undergo the negative similar relationship between accelerated changes observed for percent cortical area. In growth and cortical reduction in a sample of fact, bone mineral values by age 16 correancient Nubians from Wadi Halfa. According spond closely to values reported for modern to Hummert, this early teenage reduction at populations. It appears, therefore, that reKulubnarti is the result of rapid periosteal duction in percent cortical area do not result apposition associated with overall bone in a corresponding reduction in tissue qualgrowth offset by an even more pronounced ity as measured by bone mineral content. Age changes in cross-sectional moments of period of endosteal resorption and increased inertia are presented in Figure 3. As indimedullary area. Figure 2 presents bone mineral content cated, both AP and ML CSMI undergo pro(BMC) for the Kulubnarti sample. Inasmuch gressive increase from birth through age 12 as the bone mineral analyser measures bone with a dramatic increase from 12 to 16 years. mineral additively across the antero-poste- In AP CSMI there is a 145%increase in bendrior and medio-lateral corticies, the BMC ing strength in the 4 years following age 12. value must be standardized for bone size. There is a corresponding 130% increase in One standard procedure has been to divide ML strength. While changes with age in the BMC by the corresponding AP and ML bone AP and ML planes are basically similar, difwidths (BW). This method is limited, however, by the assumption that the bone is uniformly round in cross section. A second method for standardizing BMC is to divide by actual cortical area. Unlike the use of bone width, cortical area makes no assumptions about the cross-sectional geometry of the bone. For this reason, Ruff and Hayes (1984) have argued that cortical area is a superior measure for standardizing BMC values even though it cannot be directly obtained from living subjects using radio. 0 graphic techniques. The Kulubnarti mate0 2 4 6 0 W U 1 4 l b rials provide an opportunity to compare the AGC two methods directly. As indicated in Figure 2, standardizing Fig. 2. The relationship between age and bone minBMC to bone width creates a pattern of con- eral content (BMC) standardized to cortical area (CA) tinuous gain in mineral mass (gramdcm2) and bone width (BW. ,I=o 3 I;, , , , , , , , , , , , , 1 . . , , ' CORTICAL BONE MAINTENANCE IN NUBIAN CHILDREN 301 A6C Fig. 3. The relationship between age and antero-postenor (AP)and mediolateral (ML) cross-sectional moments of inertia (CSMI). ferences with age are again understandable in terms of changing geometric patterns. As previously observed for bone width and cortical area, the tibia begins as a tubular bone with essentially equal AP and ML strength. However, following age 6, the geometry of the bone undergoes progressive change. As the bone becomes increasingly elliptical in the AF' plane, AP bending strength increases relative to the ML plane. While differences by sex cannot be assessed with this sample, the overall pattern is similar to that reported by Ruff and Hayes (1983). DISCUSSION As has been demonstrated for adults, age changes in cortical bone in subadults must be assessed in the context of geometric and mechanical factors related to bone growth and development. In the case of the Kulubnarti sample, periods of relative cortical bone loss, expressed in percent cortical area, correspond to periods of rapid diaphyseal growth. This is true for the period from birth through age three as well as for the post-12 age period. Two hypotheses may be generated to explain this relationship. First, periodic reductions in percent cortical area may be a normal part of the modeling process associated with bone growth and development. The reduction in percent cortical area observed in modern infants between birth and 9 months of age clearly reflects this kind of change. Second. reductions in Dercent cortical area during 'periods of rapid bowth may be stress related. Garn and co-workers' comparison of 279 well nourished and proteincalorie malnourished Guatamalan children (Garnet al., 1969) has demonstrated that reductions in percent cortical area do occur as a result of nutritional stress while other aspects of bone development remain relatively unaffected. From the standpoint of the second hypothesis, evidence for dietary stress on the Kulubnarti children is abundant. Van Gerven et al. (1981)have described a close correspondence between high mortality and cribra orbitalia in the population and more recently Sandford et al. (1983)have demonstrated a direct relationship between reduced iron levels measured in hair samples from the subadults and the presence of this skeletal lesion. Furthermore, Thorp (1983)has demonstrated a delay in skeletal maturation relative to dental development consistent with protein malnutrition. Percent cortical area values for the Kulubnarti children aged birth through 3 years suggest that the modeling and dietary stress hypotheses may not be mutually exclusive. While an initial loss of bone is to be expected given the normal process of skeletal modeling, the extension of loss beyond the first year through age 3 is not typical of modern well-nourished children (Garn, 1970)and appears related to nutritional stress. An interpretation of the reduction in percent cortical area following age 12 in terms of the modeling and dietary stress hypotheses is currently hampered by a lack of comparative data. However, regardless of whether this loss is due entirely or in part to either normal modeling or nutritional stress, tissue quality measured in terms of bone mineral is not adversely affected. And, most importantly, the organ quality of the tibia measured by bending strength continues on a rapid upward trajectory consistent with the increased mechanical demands of advancing age and physical activity. It appears particularly significant that the most rapid increase in cross-sectional moments of inertia occurs during the period of pronounced percent cortical bone reduction following age 12.These adolescent children are behaving at the organ level much like the older postmenopausal women described by RUE and Hayes (1983).That is, while their cortex becomes relatively thinner, the greater diameter of their tibia actually increases bending strength. The adaptive implications for the skeleton Seem clear. As long as mechanisms are avail- 280 D.P. VAN GERVEN, J.R. HIJMMERT, AND D.B. BURR able to maintain adequate growth in bone size (including length and periosteal diameter) periodic reductions in cortical bone, whether due to modeling or dietary stress, need not result in reduced organ or tissue quality. CONCLUSIONS The results of the present analysis reflect the continuing importance of archaeological materials to the study of skeletal biology. By providing whole bones directly measurable by a variety of techniques, such materials provide an excellent opportunity to assess properties of internal and external morphology of vital importance to human growth and development. ACKNOWLEDGMENTS This research was supported by National Science Foundation Grant No. BNS-7800255. Special thanks to Dr.Christopher RufF for his helpful comments. LITERATURE CITED Carlson, DS, Armelagos, GJ, and Van Gerven, DP (1976) Patterns of age-related cortical bone loss (osteoporosis) within the femoral diaphysis. Hum. Biol. 48:295-314. Cook, DC (1979) Subsistence base and health in prehistoric Illinois valley: Evidence from the human skeleton. Med. Anthropol. 4:109-124. Dewey, JR, Armelagos, GJ, and Bartley, MH (1969) Femoral cortical involution in three archeological populations. Hum. Biol. 41:13-28. Frisancho, AR, Garn, SM, and Ascoli, W (1970a)Subperiosteal and endosteal bone apposition during adolescence. Hum. Biol. 42:639-664. Frisancho, AR, Garn, SM, and Ascoli, W (1970b) Unequal influence of low dietary intakes on skeletal maturation during childhood and adolescence. Am. J. Clin. Nutr. 23:1220-1227. Garn, SM, h h m a n n , CG, Behar, M, Viteri, F, and Guzman, MA (1964) Compact bone deficiency in proteincalorie malnutrition. Science 1451444-1445. Garn, SM, Guzman, MA, and Wagner, B (1969)Subperiosteal gain and endosteal loss in proteincalorie malnutrition. Am. J. Phys. Anthropol. 30:153-156. Garn, SM (1970) The Earlier Gain and Later Loss of Cortical Bone in Nutritional Perspective. Springfield, Illinois: Thomas. Himes, JH, Martorell, R, Habicht, JP, Yarbrough, C, Malina, RM, and Klein, RE (1975) Patterns of cortical bone growth in moderately malnourished preschool children. Hum. Biol. 47:337-350. Himes, JH,Martorell, R, Habicht, JP, Yarbrough, C, Maiina, RM, and Klein, RE (1976) Sexual dimorphism in bone growth as a function of body size in moderately malnourished Guatemalan preschool age children. Am. J. Phys. Anthropol. 45331-336. Himes, JH (1978) Bone growth and development in prctein calorie malnutrition. World Rev. Nutr. Diet. 28:143-187. Hummert, JR (1983) Bone growth and dietary stress among subadults from Nubia’s Batn el Hajar. Am. J. Phys. Anthropol. 62:167-176. Huss-Ashmore, R, Goodman, AH, and Armelagos, GJ (1982) Nutritional inferences from paleopathology. In MB Schiffer (ed): Advances in Archaeological Method and Theory, vol. V., N.Y.: Academic Press, pp. 395474. Jarcho, S (1964)Some observations on Disease in Prehisb r i c North America. Bull. Hist. Med. 38:l-19. Martin, DL, and Armelagos, G J (1979)Morphometrics of compact bone: An example from Sudanese Nubia. Am. J. Phys. Anthrop. 51:571-577. Martin, RB, and Atkinsin, PJ (1977)Age and sex related changes in the structure and strength of the human femoral shaft. J. Biomech. 10:223-231. Martin, RB, and Burr, DB (1984) Non-invasive measurements of long bone cross-sectional moment of inertia by photon absorptiometry. J. Biomech. 17:195-201. Penigian, A J (1973) Osteoporotic bone loss in two prehistoric Indian populations. Am. J, Phys. Anthropol. 39:87-96. Ruff, CB, and Hayes, WC (1983) Cross-sectional geometry of Pecos Pueblo femora and tibiae - a biomechanical investigation: II.Sex, age, and side differences. Am. J. Phys. Anthropl.60:383400. Ruff, CB, and Hayes, WC (1984) Bone mineral content in the lower limb Relationship to cross-sectional geometry. J. Bone Joint Surg. [Br] 66A:1024-1031. Sandford, MK, Van Gerven, DP, and Meglen, RR (1983) Elemental hair analysis: New evidence on the etiology of cribra orbitalia in Sudanese Nubia. Hum. Biol. 55:831-844. Smith, RW., Jr., and Walker, RR (1964) Femoral expansion in aging women: Implications for osteoporosis and fractures. Science 145:156-157. Thorp, S. (1983)Protein malnutrition in a Medieval population from Nubia’s Batn el Hajar. Honors Thesis, Univ. of Colorado, Boulder. Van Gerven, DP, Armelagos, GJ, and Bartley, MH (1969) Roentgenographic and direct measurement of cortical involution in a prehistoric Mississippian population. Am. J. Phys. Anthropol. 39:87-96. Van Gerven, DP, Sandford, MK, and Hummert, JR (1981) Mortality and culture change in Nubia’s Batn el Hajar. J. Hum. Evol. 10:395-408.