Brief communication Bone remodeling rates in Pleistocene humans are not slower than the rates observed in modern populations A reexamination of Abbott et al. (1996)код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 141:315–318 (2010) Brief Communication: Bone Remodeling Rates in Pleistocene Humans are not Slower Than the Rates Observed in Modern Populations: A Reexamination of Abbott et al. (1996) Margaret Streeter,1* Sam Stout,2 Erik Trinkaus,3 and David Burr4,5 1 Department Department 3 Department 4 Department 5 Department 2 of of of of of Anthropology, Boise State University, Boise, ID 83725 Anthropology, Ohio State University, Columbus, OH 043210 Anthropology, Washington University, St. Louis, MO 63130 Anatomy and Cell Biology, Indiana University School of Medicine, Indianapolis, IN 46202 Orthopedic Surgery, Indiana University School of Medicine, Indianapolis, IN 46202 KEY WORDS bone histomorphometry; human; osteon ABSTRACT Bone histomorphometry has been applied to the lower limb cortical bone of Pleistocene humans to establish age at death and to determine bone remodeling rates (Abbott et al.: Am J Phys Anthropol 226 (1996) 307– 313). Both of these procedures require the determination of osteon density and mean osteon size. Previous analyses of Middle and Late Pleistocene human lower limb bones have produced bone remodeling rates that are slower than those determined in a more recent archeological sample. Recalculation of the data reported in Abbott et al.: Am J Phys Anthropol 226 (1996) 307–313) has revealed mathematical errors in the remodeling rates reported for the Pleistocene humans. The corrected remodeling rates for the Pleistocene group are similar to the values obtained in the more recent comparative sample. Am J Phys Anthropol 141:315–318, 2010. V 2009 Wiley-Liss, Inc. It is generally held that Middle and Late Pleistocene humans exhibit slower bone remodeling rates than modern populations. This interpretation is based primarily on the comparison of the histomorphology of fossil and modern human cortical bone samples (Thompson and Trinkaus, 1981, Trinkaus and Thompson, 1987; Abbott et al., 1996; Pfeiffer and Zehr, 1996). The most widely cited study is Abbott et al. (1996), who undertook a comprehensive histomorphological analysis of ten lower limb cortical bone samples representing both archaic and early modern human Middle and Late Pleistocene fossils. In their study, several parameters of cortical bone remodeling were determined histomorphometrically using mean osteon area and the density (number per unit area) of whole and fragmentary osteons. Abbott et al. (1996) reported smaller osteons and slower bone turnover rates in the Pleistocene humans compared to a more recent late prehistoric, agricultural, sample from Pecos Pueblo, New Mexico (Burr et al., 1990). They interpreted their ﬁndings to mean that the greater bone mass typical of the Pleistocene populations was the result of the lower bone turnover. We have recently reexamined the data reported by Abbott et al. (1996), and ﬁnd some discrepancies in the calculations published. Our purpose is to report the results of our reexamination of the data collected for the Abbott et al. (1996) article and to report the results of our reanalysis of the bone remodeling rates in the Pleistocene humans compared with the more recent Pecos Pueblo sample. tested and veriﬁed by Stout and Paine (1994). The algorithm employs the following histomorphometric parameters. C Bone Area (B.Ar) The entire area of bone in the sample excluding the marrow cavity (in mm2). Mean Osteon Cross-Sectional Area (On.Ar) The average area of bone (including Haversian canals) contained within the cement lines of structurally complete secondary osteons (in mm2). Mean Cross-Sectional Diameter (On.Dh) The average diameter of complete secondary osteons (in mm), determined from On.Ar using the formula On:Dh ¼ ð4On:Ar=pÞ1=2 Intact Osteon Density (N.On) The number of complete (unremodeled) Haversian systems (osteons/mm2) *Correspondence to: Margaret Streeter, Department of Anthropology, Boise State University, 1910 University Drive, Boise, ID 83725. E-mail: email@example.com Received 26 October 2008; accepted 26 August 2009 MATERIALS AND METHODS Bone remodeling rates are calculated using the algorithm proposed by Wu et al. (1970) and Frost (1987a) and later C 2009 V WILEY-LISS, INC. DOI 10.1002/ajpa.21192 Published online 19 November 2009 in Wiley InterScience (www.interscience.wiley.com). 316 M. STREETER ET AL. TABLE 1. Histomorphometric parameters calculated using OPD as reported in Tables 1, 4, and 5 of Abott et al. (1996) Name Archaics Shanidar 2 Shanidar 3 Shanidar 4 Shanidar 5 Shanidar 6 Tabun 1 Broken Hill Early modern Skhul 3 Skhul 6 Skhul 7 Agea (yr) On.Arb (mm2) OPD (mm2) AOC (mm2) On.Bc (mm2) Ac.Fd (#/mm2/yr) BFR (%/yr) 25 45 37.5 42.5 27.5 22.5 – 0.038 0.018 0.030 0.024 0.023 0.030 0.036 11.46 24.29 10.34 13.5 10.05 9.29 13.9 11.92 25.26 10.44 13.65 10.05 9.38 15.01 38.4 34.6 25.4 27.3 18.5 22.9 43.5 0.93 0.78 0.42 0.46 0.67 0.63 – 3.57 1.38 1.25 1.12 1.54 1.89 – – 30 35 0.016 0.039 0.023 8.02 5.7 13.6 8.02 5.70 13.75 11.3 20.5 25.1 – .33 .61 – 1.30 1.39 See text for explanation of acronyms. a Mean of age range. b Converted from lm2 for this study. c Formula used was On.B 5 (On.Ar 3 On.N). d Calculated by subtracting 12.5 yr from median age. Fragmentary Osteon Density (N.On.Fg) 2 The number of fragmentary osteons/mm . Fragmentary osteons are those osteon segments with a cement line but no Haversian canal. Osteon Population Density (OPD) 2 The total number of intact and fragmentary osteons/mm OPD ¼ N:On þ N:On:Fg=mm2 Accumulated Osteon Creations (AOC) The total number of osteon creations (both visible and removed osteons and fragments) corresponding to a given OPD. AOC is expressed in #/mm2 AOC ¼ bðOPDÞ Continuous remodeling eventually reaches an asymptote of OPD at which each newly created osteon removes evidence of previous osteon creations. The number of missing osteons, those that have been remodeled out, increases exponentially as the asymptote is reached. The algorithm proposed by Frost (1987a) uses a scaling operator, b, which when multiplied by the OPD, gives an estimation of AOC. b is deﬁned by the equation b ¼ ð1 ax Þ1 where a is an OPD normalized to its predicted asymptote a 5 (OPD)(OPD asymptote1). The exponent x is equal to 3.5 as suggested by Frost (1987b). The formula for the OPD asymptote of a given specimen is OPD asymptote ¼ kððOn:DhÞ2 Þ1 To calculate the OPD asymptote, the value of the osteon fragment packing factor, k, is needed. The value of k accounts for the fact that a unit (1 mm2) of bone can contain more intact osteons and osteon fragments than theoretically predicted, as a result of overlap. The value American Journal of Physical Anthropology of k is speciﬁc for each bone and is determined by the formula k ¼ ðOPD asymptoteÞ ðOn:DhÞ2 The k value for the tibia has not been established, but because of gross morphological similarities between the femur and tibia, the femoral value of 1.38 was used for both bones in Abbott et al. (1996). Mean Activation Frequency (Ac.f) The mean number of osteons created annually expressed in #/mm2/yr. The annual osteon bone formation rate is equal to the mean osteon area multiplied by all of the osteons previously created (AOC), divided by the number of years over which these osteon creations accumulated. Cortical drift (movement of the bone cortex through space as a result of change in shape and size of bone during growth) causes a signiﬁcant proportion of the original bone that was present at birth to be removed before skeletal maturity. Because of this, the actual age of the bone is younger than its chronological age. It is necessary, therefore, to average the formation rate over the age of the bone (mean tissue age) rather than the chronological age. The age of adult compacta has not been established for human bones other than the sixth rib (Wu et al., 1970). Failure to account for the difference in the age of observed compacta would result in the calculation of erroneous values for activation frequency. Wu et al. (1970) have suggested that the effective birth of adult compacta for most cortical bone is within ca. 3 years of the 12.5 years reported for the rib. Ac:f ¼ AOC=ðage 12:5 yearsÞ Bone Formation Rate (BFR) The annual rate of bone formation in %/yr BFR ¼ Ac:f ðOn:ArÞ3100 Net Remodeling Rate (BRh) The total amount of remodeling that occurred over the lifetime of the individual expressed in mm2/mm2. If a 317 REMODELING RATES IN PLEISTOCENE HUMANS TABLE 2. Individual values and group means recalculated in this analysis (using N.On derived from the Abbott et al. formula for On.B) for comparison with group means as reported in Table 6 of Abbott et al. (1996) N.On (mm2) Name Archaics Shanidar 2 Shanidar 3 Shanidar 4 Shanidar 5 Shanidar 6 Tabun 1 Broken Hill Group mean Early modern Skhul 3 Skhul 6 Skhul 7 Group mean Ac.F (#/mm2/yr) Abbott Present 3.57 10.1 19.2 8.5 11.4 8.0 7.6 12.1 11.0 3.37 7.1 5.3 10.9 7.8 BFR (%/yr) Abbott Present 0.10 0.42 0.44 0.23 0.27 0.29 0.34 – 0.33 0.09 – 0.18 0.32 0.25 Abbott Present 0.24 1.06 0.85 0.69 0.65 0.70 1.02 – 0.92 0.27 – 0.68 0.71 0.70 See text for explanation of acronyms. TABLE 3. Comparison of group mean values for the Pleistocene humans and the Pecos Pueblo sample as reported in Table 6 by Abbott et al. (1996) with the recalculated values from N.On (derived from their formula for On.B) N.On (#/mm2) Ac.F (#/mm2/yr) BFR (%/yr) Abbott Present Abbott Present Abbott Present 3.57 3.37 7.26 6.21 11.0 7.8 0.10 0.09 0.22 0.20 0.33 0.25 0.24 0.27 0.78 0.82 0.92 0.70 Archaic group mean Early modern group mean Pecos males group mean Pecos females group mean See text for explanation of acronyms. reasonably accurate age at death is not available, the net osteonal remodeling can be estimated as follows BRh ¼ AOC ðOn:ArÞ RESULTS Table 1 provides the histomorphometric values reported for the Pleistocene humans in Tables 1, 4, and 5 of Abbott et al. (1996). The bone remodeling rates listed in these tables were calculated using OPD (whole and fragmentary osteon density). However, for comparison of bone remodeling parameters with the more recent Pecos Pueblo sample only the total number of whole secondary osteon could be used because fragmentary osteon counts were not included in the original analysis of the Pecos Pueblo sample (Burr et al., 1990). Histological remodeling parameters calculated using OPD are not available for a recent femoral sample of known age. Furthermore, in the comparison of bone parameters between the Pleistocene humans and Pecos Pueblo, only group mean values were reported (Table 6, Abbott et al., 1996). The histomorphological values for the Pecos Pueblo sample were taken from the Abbott et al. article and were not recalculated for this article. The osteon counts reported in Table 6 of Abbott et al. are labeled as OPD but a footnote indicated that the values listed are intact osteon counts only. Whole osteon numbers (N.On) for the Middle (3.57/mm2) and Late (3.37/mm2) Pleistocene groups given in Table 6 of Abbott et al. (1996) are lower than would be expected based on the group means derived from OPD (13.26/mm2 Middle Pleistocene humans) and (9.11/mm2 Late Pleistocene humans) given in Table 4 of Abbott et al. (1996). Individual values for N.On for each of the Pleistocene humans can be derived indirectly from the formula for percent osteonal bone [On.B 5 (On.Ar 3 N.On)/(100)] given in Table 2 of Abbott et al. (1996). For example, N.On for the Shanidar 3 femur can be determined using the values given for percent osteonal bone (On.B, 34.63%), and mean osteon size (On.Ar, 0.018 mm2) in Table 3 of Abbott et al. (1996) as follows. On:B ¼ ðOn:ArÞðN:onÞð100Þ 34:63% ¼ ð0:018 mm2 ÞðN:onÞ=100 ð100Þ34:63% ¼ ð0:018 mm2 ÞðN:onÞ=100ð100Þ 0:3463 ¼ ð0:018 mm2 ÞðN:OnÞ 0:3463=0:018 ¼ ð0:018 mm2 =0:018ÞðN:OnÞ 0:3463=0:018 ¼ N:On 19:2=mm2 ¼ N:On The mean N.On for the two fossil groups (Middle and the Late Pleistocene humans) determined by this method are 11.0/mm2 and 7.8/mm2, respectively (Table 2). Recalculation of bone remodeling rates using the revised N.On values for the Pleistocene fossils also produces valAmerican Journal of Physical Anthropology 318 M. STREETER ET AL. ues that are greater than the values originally reported in Table 6 of Abbott et al. (1996). The revised values for bone remodeling in the Pleistocene humans are equal to or greater than the remodeling rates determined for the comparison sample from Pecos Pueblo (Table 3). CONCLUSIONS A recalculation of the cortical bone histomorphometry of Pleistocene humans reported in the Abbott et al.’s article (1996) ﬁnds that, contrary to the original report, the fossil human bones exhibit bone remodeling rates that are similar to those determined for the modern comparative sample from Pecos Pueblo. However, the Pecos Pueblo samples, although recent relative to the Pleistocene samples, date to the 15th and 16th Centuries and may not be representative of bone remodeling activity typical of modern human populations. The Pecos people were primarily dependent on maize agriculture, which has been shown to lead to nutritional stress (Goodman et al., 1984) and linked with changes in remodeling (Martin and Armelagos, 1979). Comparison of histological data with the Pecos Pueblo sample is also complicated because only intact osteons numbers were counted for Pecos (Burr et al., 1990). A true picture of remodeling activity requires the inclusion of fragmentary osteons (Wu et al., 1970). A more accurate comparison of bone remodeling rates between Pleistocene and modern humans will require a comparison between lower limb bones from a more recent skeletal sample that includes osteon fragments. American Journal of Physical Anthropology LITERATURE CITED Abbott S, Trinkaus E, Burr DB. 1996. Dynamic bone remodeling in later Pleistocene fossil hominids. Am J Phys Anthropol 99:585–601. Burr DB, Ruff CB, Thompson DD. 1990. Patterns of skeletal histological change through time: comparison of an archaic Native American population with modern population. Anat Rec 226:307–313. Frost HM. 1987a. Secondary osteon population densities: an algorithm for estimating the missing osteons. Yrbk Phys Anthropol 30:239–254. Frost HM. 1987b. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat Rec 219:1–9. Goodman AH, Lallo J, Armelagos G, Rose JC. 1984. Health changes at Dickson Mounds, Illinois (AD 950–1300). In: Cohen MN, Armelagos GJ, editors. Paleopathology at the origins of agriculture. Orlando: Academic Press. p 271–301. Martin DL, Armelagos GJ. 1985. Skeletal remodeling and mineralization as indicators of health: an example from prehistoric Sudanese Nubia. J Hum Evol 14:527–537. Pfeiffer S, Zehr MK. 1996. A morphological and histological study of the human humerus from Border Cave. J Hum Evol 31:49–59. Stout SD, Paine RR. 1994. Bone remodeling rates: a test of an algorithm for estimating missing osteons (Brief communication). Am J Phys Anthropol 93:123–129. Thompson DD, Trinkaus E. 1981. Age determination for the Shanidar 3 Neandertal. Science 212:575–577. Trinkaus E, Thompson DD. 1987. Femoral diaphyseal histomorphometric age determinations for the Shanidar 3, 4, 5, and 6 Neandertals and Neandertal longevity. Am J Phys Anthropol 72:123–129. Wu K, Schubeck KE, Frost HM, Villanueva AR. 1970. Haversian bone formation rates determined by a new method in a mastodon and in human diabetes mellitus and osteoporosis. Calcif Tissue Int 22:204–219.