close

Вход

Забыли?

вход по аккаунту

?

Cortical bone remodeling rates in a sample of African American and European American descent groups from the American Midwest Comparisons of age and sex in ribs.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 130:214–226 (2006)
Cortical Bone Remodeling Rates in a Sample
of African American and European American Descent
Groups from the American Midwest: Comparisons
of Age and Sex in Ribs
Helen Cho,1* Sam D. Stout,2 and Thomas A. Bishop3
1
Department of Anthropology, Davidson College, Davidson, North Carolina 28035
Department of Anthropology, Ohio State University, Columbus, Ohio 43210-1364
3
Department of Statistics, Ohio State University, Columbus, Ohio 43210-1364
2
KEY WORDS
bone loss; histomorphometry; remodeling; African American; European American
ABSTRACT
This study employs regression analysis
to explore population and sex differences in the pattern
of age-associated bone loss, as reflected by histomorphometric variables that are measures of intracortical and
endocortical bone remodeling. A comparison of an African American sample from the Washington Park Cemetery in St. Louis, Missouri, and a European American
rib sample composed of cadavers, autopsies, and forensic
cases from Missouri reveals the existence of complex
age-associated patterns for differences in measures of
intracortical remodeling and cortical area. Females from
the two samples express similar bone dimensions and
dynamics. The African American females appear to lose
more bone than their male counterparts, but this difference is absent in the European American sample. When
age-associated patterns are considered, it is in the
younger cohorts that African Americans exhibit greater
relative cortical area than European Americans, but this
is reversed in the older ages, when the latter group man-
ifests greater bone mass. The European American males
consistently differ in the slopes and intercepts for the
variables compared to the other groups, and differences
are highly significant with African American females,
with the former group maintaining bone mass while the
latter exhibit a more rapid bone loss. Achieving larger
relative cortical area due to smaller endosteal area,
coupled with better bone quality due to lower intracortical porosity early in life, may be a mechanism by which
African Americans, especially females, maintain adequate bone mass in older ages, which buffers them from
bone loss and related fragility fractures despite higher
rates of intracortical remodeling and endosteal expansion later in life. These results suggest that both genetic
and environmental factors are responsible for the differences in bone remodeling and bone mass observed
between these samples. Am J Phys Anthropol 130:214–
226, 2006. V 2005 Wiley-Liss, Inc.
The loss of bone mass with age appears to be a universal phenomenon in humans, and low bone mass (osteopenia) has become a serious concern in recent populations
that enjoy increased longevity. It is generally held that a
significant deficiency in bone substance increases the
susceptibility to spontaneous fractures, or osteoporosis,
especially of the vertebrae, forearm, and hip (Parfitt,
1984; Bruce and Stevenson, 1990). An important mechanism of bone loss is the uncoupling or imbalance in the
bone-remodeling process (Plato et al., 1994). Bone remodeling is a metabolic process involving the spatial and
temporal coordination of bone resorbing (osteoclasts) and
forming (osteoblasts) cells, collectively referred to as a
basic multicellular unit (BMU) of bone remodeling. Each
BMU involves its activation, followed by the resorption
and formation of bone, and produces a distinct histomorphological feature. In cortical bone, remodeled features
are called osteons or Haversian systems. Heaney (2003)
proposed that bone remodeling at rates higher than optimal for maintaining bone strength is the primary cause
of osteoporotic bone fragility, and that reduced bone
mass is a factor that predisposes individuals to the
harmful effects of ‘‘excessive remodeling.’’
Osteoporosis has been associated with increasing age,
genetics (including sex and biological ancestry), hormones, diet, and level of physical activity. Numerous
researchers focused on causes or risk factors for osteoporosis. One of the risk factors is European ancestry, when
compared to those of African ancestry who have a lower
prevalence of osteoporosis (Nordin, 1966; Nelson et al.,
1993; Anderson and Pollitzer, 1994). There is an abundance of clinical literature on population variability in
age-associated bone loss. In American populations, African Americans are reported to have denser cortical bone
and greater bone mass, and therefore a lower incidence
of osteopenia (Garn et al., 1972; DeSimone et al., 1989;
Pollitzer and Anderson, 1989; Nelson et al., 1991, 1993,
2004; Anderson and Pollitzer, 1994; Baron et al., 1994;
Stini, 1995). African Americans have lower urinary calcium levels than European Americans, indicating lower
rates of bone resorption (Anderson and Pollitzer, 1994).
Compared to European Americans, African Americans
C 2005
V
WILEY-LISS, INC.
C
*Correspondence to: Helen Cho, Department of Anthropology,
Davidson College, Box 6934, Davidson, NC 28035-6934.
E-mail: hecho@davidson.edu
Received 16 September 2004; accepted 28 March 2005.
DOI 10.1002/ajpa.20312
Published online 19 December 2005 in Wiley InterScience
(www.interscience.wiley.com).
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
have a slower bone turnover rate and a greater rate of
subperiosteal apposition, leading to larger bone volume
(Solomon, 1979; Weinstein and Bell, 1988; Kleerekoper
et al., 1994; Bell et al., 1995). The different ratios of subperiosteal apposition and endosteal resorption between
the two groups result in larger bone mass through all
ages in African Americans (Garn, 1981).
Similar results for population differences in cortical
bone histomorphology were found (Ericksen, 1979; Richman et al., 1979; Thompson and Gunness-Hey, 1981;
Ubelaker, 1986; Yoshino et al., 1994; Stout and Lueck,
1995; Cho et al., 2002). Kerley (1965), however, found no
differences in osteon counts between European Americans and African Americans, but his sample for the latter was small (N ¼ 11).
If both increased bone remodeling rates and low bone
mass are important factors determining the risk of
osteoporotic fracture, then populations that differ in
this risk should also differ with respect to these two
factors. Given the possible effects of both genetic and
environmental factors on bone remodeling and bone
mass, it is hypothesized that population differences
should vary with age. In this paper, we present the
results of research designed specifically to determine
whether differences in age-associated patterns for bone
remodeling rates and bone mass exist between European Americans and African Americans that relate to
reported population differences in the incidence of
osteopenia and fracture risk. Additionally, remodeling
dynamics are compared between males and females in
each population, as older postmenopausal females are
at greatest risk for osteopenia and related fragility fractures due to estrogen deficiency (Nguyen et al., 1995;
O’Neill et al., 1997; Frost, 1999; LeBoff and Glowacki,
1999; Nevitt, 1999).
MATERIALS AND METHODS
The rib is an ideal site for histomorphometric research,
and a considerable amount of bone remodeling research
was undertaken using the midshaft of the rib (Wu et al.,
1970; Frost, 1987; Stout and Paine, 1994; Stout and
Lueck, 1995; Cho et al., 2002). It is easily accessible during autopsies, and is especially useful in bioarchaeology,
since sectioning the rib is less invasive than sectioning a
larger bone such as the femur. Still, intraskeletal variability in bone remodeling is an issue that should be considered in bone loss studies, and it is addressed in the
Discussion below.
Rib samples from the middle third of bones identified
as being the sixth rib were obtained for histomorphometric analysis. The majority of the African American rib
sample is from the Washington Park Cemetery, located
on the outskirts of St. Louis, Missouri. It is the largest
African American cemetery in the St. Louis area, and
was founded in the early 1920s. The last body was
buried around 1990. A portion of the cemetery was excavated during the construction of Metrolink, the city’s
light-rail system. The rib samples were obtained from
the State of Missouri Department of Natural Resources,
which was involved in the excavation, identification, and
relocation of the human remains. Eight African American ribs from autopsy and forensic cases from the Boone
County Medical Examiner’s Office (Columbia, MO) were
subsequently added to this sample. The total African
American sample consists of midsections of ribs from 135
individuals of known age. Age at death of the Washing-
215
ton Park sample is based on cemetery records. The age
range is 17–95 years, with a mean age of 54.3 years. The
side of the body from which the rib sample originated
could not be determined because this information was
not collected during sampling. The left and right ribs
undergo homogeneous loading from respiration, which
probably results in similar bone turnover on both sides.
Additionally, bone remodeling dynamics were reported to
be similar in the same section of the sixth and eleventh
ribs (Santoro and Frost, 1967, 1968; Frost, 1969), suggesting homogeneity in the ribs.
The European American sample is composed of 35
autopsy ribs used by Stout and Paine (1992) to develop
their histological age-estimating formulae, plus an additional 29 samples from cadavers, forensic, and autopsy
cases. The autopsy cases were from the Boone County
Medical Examiner’s Office, the cadaver ribs were from
the Washington University School of Medicine in St.
Louis, and the forensic cases were from numerous counties in the state of Missouri (also obtained from the
Boone County medical examiner). The age range for the
European American sample is 17–102 years, with a
mean age of 43.8 years. In the combined African American and European American samples, there are 117
males and 74 females (8 individuals of unknown sex).
This study did not include individuals less than 17
years of age for several biological reasons. Transverse
cortical drifts that occur in the ribs of juveniles involve
continuous removal and formation of bone in the pleural
and cutaneous surfaces of the rib, respectively, resulting
in differing ‘‘tissue ages’’ within the cortical bone. Cortical drift ceases in the early teens, when the effective age
of the adult compacta is achieved (Wu et al., 1970). In
addition, subadult bone remodeling rates differ from
those of adults, and remain an area of further research.
Rib samples were prepared for histomorphometric
analysis as described by Stout and Paine (1992). Each
rib section was embedded in CastoliteTM resin (Buehler,
Ltd., Lake Bluff, IL) to ensure structural integrity during cutting and grinding. Embedded sections were placed
in a vacuum to infiltrate the sample and to eliminate air
bubbles that would otherwise obscure the visibility of
microscopic structures. After the resin had dried, parallel-sided transverse sections were removed using an Isomet low-speed gravity-feed petrographic saw (Buehler,
Ltd.). These sections were ground manually to a final
thickness of 50–100 lm, using a variable-speed grinder
and silicon carbide sandpaper. The resulting thin sections were then cleaned with acetone, cleared with xylene,
mounted on microscopic slides with PermountTM, and
coverslipped.
The following histomorphometrics were determined for
each thin section, as described by Stout and Paine
(1994), using an Olympus BX50 light microscope (Olympus Optical Co., Ltd., Tokyo, Japan) fitted with a Merz
eyepiece reticule. A microscopic ‘‘field’’ (or unit area) is
defined as the area of the Merz grid superimposed on
the rib thin section in view. Approximately 50% of each
entire rib cross-section was read to ensure representative sampling by reading every other microscope field
(the overall sampling pattern is analogous to a checkerboard). To minimize sampling error, two cross-sections
per individual were prepared. The results of the two sections per individual were averaged, as suggested by Wu
et al. (1970), to obtain representative data from the individual. For a review and illustration of this method
of histomorphometric analysis, see Robling and Stout
American Journal of Physical Anthropology—DOI 10.1002/ajpa
216
H. CHO ET AL.
(2000). The following histomorphometrics
computed.
1
were then
1. Total cortical area of bone sampled per section
(Sa.Ar): the area of cortical bone read.
2. Intact osteon density (N.On) in number/mm2: the
total number of osteons per unit area (Sa.Ar) that
have their Haversian canal perimeters intact or
unremodeled. Half or more of an osteon’s area must
fall within the Merz grid to be counted:
N.On ¼
Total Number of Intact Osteons
:
Sa.Ar
3. Fragmentary osteon density (N.On.Fg) in number/
mm2: the total number of osteons per unit area
(Sa.Ar) that lack a Haversian canal or for which the
perimeters of their Haversian canals, if present,
have been remodeled by subsequent generations of
osteons. Half or more of the fragmentary osteon
must fall within the Merz grid to be counted:
N.On.Fg ¼
activity eventually produces an asymptote in OPD
when osteonal bone occupies the entire cortex, and
each new creation removes the evidence for an earlier creation. The contribution of missing osteons to
the OPD.Cd equation increases exponentially until
the asymptote is reached. The algorithm of Frost
(1987) uses a scaling operator b, which when multiplied by an observed OPD provides an estimate of its
corresponding OPD.Cd. Beta is defined by the equation
Total Number of Fragmentary Osteons
:
Sa.Ar
4. Mean osteonal cross-sectional area (On.Ar) in mm2:
the average area of bone contained within the cement
lines of structurally complete osteons for each rib
specimen. Osteons were considered to be structurally
complete if their reversal lines were intact. Complete
osteons with Haversian canals that deviated significantly from circular were excluded. Mean area was
calculated as the average cross-sectional area of a
minimum of 25 complete osteons per cross-section.
5. Osteon population density (OPD) in number/mm2:
the sum of N.On and N.On.Fg:
OPD = N.On + N.On.Fg:
6. Mean cross-sectional diameter (Dm), in mm, of the
intact osteons of a specimen determined from mean
osteonal cross-sectional area, using the formula
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4On:Ar
Dm ¼
:
p
During bone remodeling, older osteons are continuously removed by new osteon creations. As a result,
the observed number of osteons in a cross-section
becomes a progressively smaller fraction of the total
osteon creations accumulated over the lifetime of an
individual. In order to compute the annual rate of
bone remodeling of an individual, the ‘‘missing’’ osteons must be accounted for. Frost (1987) suggested
an algorithm for estimating the missing osteons and
bone remodeling rates. The following variables are
derived using the algorithm as adapted by Stout and
Paine (1994).
7. Accumulated osteon creations (OPD.Cd ¼ N.On þ
N.On.Fg þ missing osteons) in number/mm2 that
correspond to a given OPD: continuous remodeling
1
Abbreviations for histomorphometric nomenclature are adopted
from Parfitt et al. (1987) with the exception of OPD, for which no
comparable variable exists in the new system. They therefore differ
from those used in Stout and Paine (1992, 1994).
American Journal of Physical Anthropology—DOI 10.1002/ajpa
b¼
1
1 ax
where alpha (a) is an OPD normalized to its predicted asymptote, as described by the expression
a¼
OPD
OPD asymptote
and the exponent has the value of 3.5, as suggested
by Frost (1987). The OPD asymptote for a given
specimen is estimated by the formula
OPD asymptote ¼
k
Dm2
:
It is based on the relationship between osteon size
(Dm2) and the unit of measurement (mm2). k is a
packing factor that accounts for the fact that a unit
of area of bone can actually contain more intact
osteons and their fragments than predicted by a theoretical orthogonal distribution (Frost, 1987). It is
estimated by the formula
k ¼ ðOPD asymptoteÞðDm2 Þ:
For this study, the value of k was determined to be 1.7.
Following the suggestion of Frost (1987), this value is
based on an independent sample of rib cross-sections
from older individuals in which all primary lamellar
bone was replaced by secondary osteonal bone. The
sample consisted of ribs from 21 dissecting-room cadavers with a mean age of 73.6 years, and an age range of
60–102 years. The sample’s maximum observed OPD of
36.25/mm2 was used as the asymptotic value to determine k. The value of 0.047 mm used for Dm2 is based
on a mean osteonal cross-sectional area of 0.037 mm2
for the sixth rib reported by Wu et al. (1970) for a large
clinical sample. It should be noted that the Dm2 value
used to estimate the asymptote for a particular specimen is based on that specimen’s mean osteonal crosssectional area, not the 0.037 mm2 that was used to
determine k. OPD.Cd for individual bone specimens
can then be estimated by multiplying their observed
OPD by b:
OPD:Cd ¼ b OPD:
8. Mean activation frequency (Ac.f) in number/mm2/
years: the mean number of osteons created annually
per mm2 of bone, estimated by dividing OPD.Cd by
age. Transverse cortical drifts occurring during
growth make the effective mean age of adult compacta appear younger than the chronological age.
Wu et al. (1970) suggested that the effective birth of
217
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
adult compacta in the middle third of the sixth rib
occurs at approximately 12.5 years of age:
Ac:f ¼
OPD:Cd
:
chronological age 12.5 years
9. Average osteonal bone formation rate (BFR), in mm2/
mm2/year, can be estimated by multiplying the mean
activation frequency by the mean osteonal cross-sectional area for a specimen, as described by the expression
BFR = Ac.f On:Ar:
BFR is an estimate of the bone formed per unit area
of the adult compacta per year. Additional variables
relating to cortical area were measured from one
thin section per individual.
10. Total subperiosteal area (Tt.Ar): total cross-sectional
area or area under the subperiosteum, including the
marrow cavity, or endosteal area.
11. Cortical area (Ct.Ar): the amount of cortical bone in
a cross-section of bone, excluding the endosteal area.
12. Endosteal area (Es.Ar): the area of the marrow cavity obtained by subtracting cortical area from total
area:
Es.Ar = Tt.Ar Ct:Ar:
13. Relative cortical area (Ct.Ar/Tt.Ar): the relative
amount of cortical bone in cross-sectional area of
bone, or the ratio of cortical bone area (Ct.Ar) to
total area (Tt.Ar) of a rib cross-section:
Relative Cortical Area ¼
Cortical Area
:
Total Subperiosteal Area
Total subperiosteal area and cortical area were
directly computed by scanning the rib thin section
on a flatbed scanner and using NIH Image analysis
software. Endosteal area (Es.Ar) and relative cortical area (Ct.Ar/Tt.Ar) were derived from cortical
area and total area measurements.
Statistical methods
The software packages STATISTICA (StatSoft, Inc.,
Tulsa, OK) and MINITAB (Minitab, Inc., State College,
PA) were used to compute the variables and to conduct
the statistical analysis required to compare the key characteristics across samples. The following eight response
variables were analyzed in this study: OPD, On.Ar, 1/Ac.f,
1/BFR, Tt.Ar, Ct.Ar, Es.Ar, and Ct.Ar/Tt.Ar. The primary
interests are 1) the assessment of the relationship of
these variables to the age of individuals; and 2) the determination of whether this relationship is dependent on the
biological ancestry and/or sex of individuals. The behavior
of these response variables as a function of age was analyzed for four distinct groups, identified by biological
ancestry and sex as African American females (AAF),
African American males (AAM), European American
females (EAF), and European American males (EAM).
Since the primary questions of interest relate to the relationship of the response variables to age, and whether
Fig. 1. Age distribution within ancestry/sex groups. AAF,
African American female; AAM, African American male; EAF,
European American female; EAM, European American male
(here and in subsequent figures).
this relationship is dependent on ancestry and/or sex, it is
important that the distributions or ranges of age within
the four ancestry by sex groups are comparable. Figure 1
demonstrates that the distributions of age within groups
are comparable.
Based on an empirical examination of the relationship
of the responses to age, a simple linear regression model
was used to approximate the relationship of each
response variable to age within each of the four groups.
The linear regression model took the standard form:
Yij ¼ b0i þ b1i xAgeij þ eij i ¼ 1; 2; 3; 4 and j ¼ 1; 2; . . . ; Ni
where Yij denotes the response, Ageij denotes the age of
the jth rib sample in ancestry by sex group i, and Ni
denotes the number of bone samples in ancestry by sex
group i. Figure 2 contains the empirical plots of each of
the eight responses vs. age for the ancestry/sex groups.
There is clear empirical evidence that a linear relationship exists between these variables and age. Due to the
nonlinear relationship of both Ac.f and BFR with age, a
transformation was sought to linearize the model. It was
determined that both the reciprocals 1/Ac.f and 1/BFR
are linearly related to age, so they were used as the
response variables in the analysis. In both cases, however, it was also determined that the nonconstancy of
the error variances required a weighted regression analysis to be performed, using weights equal to 1/Age.
Comparison of regression lines across the four
ancestry by sex groups. The least-squares point esti^ 0i and b
^ 1i were obtained and reported for the
mates b
eight response variables for each of the four ancestry by
sex groups. For each of the eight response variables, a
statistical test of the equality of the four regression lines
was also carried out. P-values associated with the test of
the hypothesis that all four regression lines are the
same
H0 : b01 ¼ b02 ¼ b03 ¼ b04
and
b11 ¼ b12 ¼ b13 ¼ b14
vs. the alternative hypothesis
HA : Not all regression lines are the same
were calculated and used to test the null hypothesis
against the alternative hypothesis at the 0.05 level of
significance. If the test is not statistically significant,
American Journal of Physical Anthropology—DOI 10.1002/ajpa
218
H. CHO ET AL.
Fig. 2. a: Empirical plots of responses vs. age for African American females. b: Empirical plots of responses vs. age for African
American males. c: Empirical plots of responses vs. age for European American females. d: Empirical plots of responses vs. age for
European American males.
TABLE 1. Results of test of equality of regression lines
Response variable
OPD
On.Ar
1/Ac.f
1/BFR
Tt.Ar
Ct.Ar
Es.Ar
Ct.Ar/Tt.Ar
*
*
F statistic
P value
2.796
1.162
2.984
1.055
20.959
7.351
20.035
15.827
0.013
0.328
0.008
0.391
0.000
0.000
0.000
0.000
Significant at P < 0.05.
then there is no evidence in the data that the four
regression lines can be declared different. If the test is
significant, then the data provide evidence that a detectable difference in regression lines exists across groups.
In that case, Bonferroni confidence intervals are constructed to determine where the differences lie.
All tests rejected the hypothesis of equal regression
lines, except for the response variables On.Ar and 1/BFR
(Table 1). Comparisons of the fitted regression lines for
the test of the equality of regression lines are presented
in Figures 3–10.
The slope and intercept estimates obtained for each
response variable were compared across each of the
ancestry by sex groups by creating confidence intervals
for the differences in point estimates for both the slope
and intercept. Since there were four groups, there were
six comparisons of slopes and six comparisons of intercepts, defined by the following combinations of ancestry
and sex group categories: AAF-EAF, AAF-EAM, AAFAAM, EAF-EAM, EAF-AAM, and AAM-EAM. Therefore,
12 comparisons were made for each response variable
except for On.Ar and 1/BFR, that have similar regression
lines between groups. The Bonferroni method of multiple
comparisons was used to approximately control the overall confidence coefficient to 0.95. The form of confidence
intervals for this procedure was as follows. For comparing the intercepts and slopes across group i and j,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^ 0i b
^ 0j Þ t:025=12;14 x r
^ 2b^
^ 2b^ þ r
ðb
0i
0j
and
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^ 1j Þ t:025=12;14 x r
^ 1i b
^ 2b^ þ r
^ 2b^ ;
ðb
1i
1j
^ b^ 0;i and r
^ b^ 1i are the standard errors of the point
where r
estimates obtained from the regression models. In order
to produce conservative estimates, 14 degrees of freedom
for the t-values were used because this was the smallest
number of degrees of freedom available for estimating
the error variance for any of the regression models.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
219
Fig. 3. Fitted regression lines of osteon population density
(OPD in number/mm2) and age in years for African Americans
and European Americans by sex.
Fig. 6. Fitted regression lines of reciprocal of bone formation
rate (1/BFR in 1/mm2/mm2/years) and age in years for African
Americans and European Americans by sex.
Fig. 4. Fitted regression lines of osteon area (On.Ar in mm2)
and age in years for African Americans and European Americans by sex.
Fig. 7. Fitted regression lines of total subperiosteal area
(Tt.Ar in mm2) and age in years for African Americans and
European Americans by sex.
Fig. 5. Fitted regression lines of reciprocal of activation frequency (1/Ac.f in 1/number/mm2/years) and age in years for
African Americans and European Americans by sex.
Fig. 8. Fitted regression lines of cortical area (Ct.Ar in
mm2) and age in years for African Americans and European
Americans by sex.
The Bonferroni method. Let {Y1, Y2, . . . , YK} be normally distributed random variables with means {l1,
l2, . . . , lK} and variances {r12, r22, . . . , rK2}, respectively. The Yis may or may not be independent. Let {s12,
s22, . . . , sK2} be v2 estimators of {r12, r22, . . . , rK2},
based on {m1, m2, . . . , mK} degrees of freedom, respectively.
The {s12, s22, . . . , sK2} may or may not be independent,
but the {Yi} are independent of the {si2}, so that
follows a t-distribution with ui degrees of freedom. Then
the probability is greater than or equal to 1 a that,
simultaneously, the following events hold:
Ti ¼
Yi li
;
si
i ¼ 1; 2; . . . ; K
li 2 Yi tva=2K
si
i
for i ¼ 1; 2; . . . :; K:
In this application, we take li to be the difference in true
values of slopes (intercepts) across any two sex/ancestry
groups, Yi to be the difference in slope (intercept) point
estimates across two sex/ancestry groups, and si to be
the standard error of the difference obtained from the
American Journal of Physical Anthropology—DOI 10.1002/ajpa
220
H. CHO ET AL.
Fig. 9. Fitted regression lines of endosteal area (Es.Ar in
mm2) and age in years for African Americans and European
Americans by sex.
regression analysis. In order to be conservative, the
degrees of freedom ui were all set to 14, the smallest
degree of freedom for all the regression analyses.
RESULTS
Results of the regression analysis and pairwise comparisons of the intercepts and slopes are presented for
each response variable in Tables 2 and 3.
All slopes for OPD are statistically different from zero
and positive, indicating that OPD increases with age for
the African American and European American males and
females. The rate of increase in OPD with age is greatest
for African American females (slope ¼ 0.412), and slowest
for European American males (slope ¼ 0.143). Based on yintercept results, European American males begin adult
life with the highest OPD, and African American females
the lowest observed values for OPD at 14.1/mm2 and 1.1/
mm2, respectively. Pairwise comparisons of intercepts and
slopes find significant differences between African American females and European American males (Fig. 3). The
intercepts differ (P ¼ 0.004), with the latter group having
larger OPD values in younger ages, and the African American female slope is steeper (P ¼ 0.002).
The overall F test for equality of regression lines for
mean osteon size (On.Ar) failed to reject the null hypothesis, indicating that regression lines are comparable for the
African American and European American males and
females. The slopes for On.Ar were all statistically significant and negative, indicating mean osteon size decreases
with age in both sexes and biological ancestries (Fig. 4).
The relationship between Ac.f and age was found to be
nonlinear, particularly for European and African American males. For regression analysis, therefore, a reciprocal transformation (1/Ac.f) was used to achieve linear
relationships between the response variable with age.
Activation frequency appears to decrease with age for all
groups, but most significantly for males. Examining individual slopes reveals that the European American males
exhibit the steepest rate of decrease in Ac.f with age
(slope ¼ 0.033), but have the highest initial Ac.f rates
(intercept ¼ 0.211/mm2/year).2 African American females exhibit the slowest rate of decline in Ac.f with age
(slope ¼ 0.017), and have the lowest initial Ac.f rates
2
Since the values of intercepts for Ac.f and BFR are based on the
reciprocal transformation, a larger value translates into a smaller
nontransformed value.
Fig. 10. Fitted regression lines of relative cortical area
(Ct.Ar/Tt.Ar) and age in years for African Americans and European Americans by sex.
(intercept ¼ 0.762/mm2/year) (Fig. 5). Significant differences in intercepts between groups are observed between
the African American females and European American
males (P ¼ 0.004).
The relationship between BFR and age is nonlinear,
and a reciprocal transformation was applied to these
data. The overall F test for equality of regression lines
for 1/BFR failed to reject the null hypothesis, indicating
that regression lines are comparable for the African
American and European American males and females.
In general, BFR decreases with age, and appears to level
off around 50 years of age for all four groups (Fig. 6).
In cross-sectional area measures, as expected, males
have larger ribs than females, regardless of biological
ancestry. None of the slopes for the regressions of Tt.Ar
on age by group differ from zero, indicating that for the
adult age range included in this study, overall rib size
does not change with age (Fig. 7).
For absolute cortical area, all groups have negative
and statistically significant slopes with age, except for
European American males, for whom the slope does not
differ significantly from zero. The decrease in Ct.Ar with
age occurs at the highest rate for African American
females (slope ¼ 0.262), and the slowest rate for European American males (slope ¼ 0.054). The African
American males exhibit the greatest initial Ct.Ar (intercept ¼ 34.4 mm2), and European American males the
smallest (intercept ¼ 25.51 mm2). Comparison of slopes
finds a significant difference between African American
females and European American males (Fig. 8).
Only the Es.Ar slope for the African American females
is statistically significant, so there are no changes with
age in other groups. The primary difference across
groups is the significant difference in average values
(intercepts). Males are larger with respect to Es.Ar than
females (Fig. 9).
Lastly, the relative amount of cortical area in rib
cross-sections decreases with age for all groups except
European American males, for whom the slope does not
differ significantly from zero. The highest rate of decrease in relative cortical area is found for African American females (slope ¼ 0.006) compared to European
American males, who essentially show no significant
decrease in relative cortical area with age. Significant
slope differences occur between African American
females and African American males, and African American females and European American males. The African
American females initially exhibit the largest relative
cortical areas (intercept ¼ 0.693), and European Ameri-
American Journal of Physical Anthropology—DOI 10.1002/ajpa
221
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
1
TABLE 2. Regression analysis results for biological ancestry/sex groups
Measurement
OPD
On.Ar
1/Ac.f
1/BFR
Tt.Ar
Ct.Ar
Es.Ar
Ct.Ar/Tt.Ar
Ancestry/sex
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
AAF
EAF
AAM
EAM
Intercept value2
1.096
8.340
6.925
14.061
0.052
0.051
0.048
0.046
0.762
0.449
0.221
0.211
0.664
7.253
9.985
12.486
44.240
48.247
80.106
86.327
32.006
26.588
34.400
25.510
12.235
21.659
45.706
60.817
0.693
0.556
0.439
0.302
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
Intercept P value*
6.906
7.840
4.924
3.040
0.006
0.010
0.006
0.004
0.510
0.954
0.336
0.238
15.184
42.028
13.768
7.282
11.786
19.632
11.434
11.014
4.416
5.540
4.098
3.860
10.014
19.812
10.656
9.830
0.074
0.172
0.058
0.050
0.752
0.052
0.006
0.000
0.000
0.000
0.000
0.000
0.004
0.362
0.194
0.084
0.931
0.735
0.151
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.018
0.048
0.000
0.000
0.000
0.000
0.000
0.000
Slope value2
Slope P value*
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
0.000
0.001
0.000
0.000
0.000
0.006
0.000
0.000
0.002
0.069
0.000
0.000
0.000
0.029
0.000
0.000
0.231
0.812
0.161
0.188
0.000
0.007
0.000
0.205
0.000
0.281
0.378
0.326
0.000
0.048
0.000
0.905
0.412
0.280
0.276
0.143
0.0004
0.0003
0.0003
0.0002
0.017
0.019
0.028
0.033
1.042
1.029
1.240
1.137
0.127
0.045
0.142
0.161
0.262
0.167
0.225
0.054
0.389
0.212
0.083
0.107
0.006
0.004
0.002
0.000
0.122
0.142
0.086
0.066
0.0001
0.0002
0.0001
0.00008
0.010
0.020
0.006
0.006
0.302
0.846
0.274
0.200
0.210
0.374
0.200
0.240
0.078
0.106
0.072
0.084
0.178
0.376
0.186
0.214
0.002
0.004
0.002
0.002
1
AA, African American; EA, European American; M, male; F, female.
62 SE.
*Significant at P < 0.05.
2
can males the least (intercept ¼ 0.302). Significant differences between intercepts also occur between African
American females and African American males, African
American females and European American males, and
African American males and European American males
(Fig. 10).
DISCUSSION
It is well-established that African Americans have
slower bone remodeling rates, denser cortical bone, and
greater bone mass, both histologically and macroscopically, than European Americans, and therefore are not
as susceptible to osteoporosis (Garn et al., 1972; Cohn et
al., 1977; DeSimone et al., 1989; Luckey et al., 1989; Pollitzer and Anderson, 1989; Nelson et al., 1991, 1994;
Anderson and Pollitzer, 1994; Baron et al., 1994). Bone
density is a measure of the extent of mineralization of
the bony matrix by inorganic substances, mainly calcium. Bone mass refers to the amount of lamellar bone,
and both cortical thickness and bone density are measures of bone mass. Some of the results from this histological study were unexpected and indicate that the bone
dynamics between the two ancestries are more complex
than simply African Americans exhibiting slower remodeling rates and greater bone mass. Each ancestry and
sex group has a relatively unique pattern for the ageassociated variables, but the European American male
sample alone exhibits the greatest differences in slopes
and intercepts for the variables when compared to the
other groups.
In our study, the osteon population density increases
with age, as expected in both male and female African
American and European American samples. This
increase in OPD with age results in a net loss of bone
from intracortical porosity created by Haversian canals,
and is responsible for some of the normal age-associated
decrease in bone density. The most significant difference
across the four ancestry by sex groups is the difference
between African American females and European American males. The European American males also have the
largest OPD value at the youngest cohort (y-intercept)
that is significantly higher than African American
females, who have the smallest value. A small OPD
value for the latter group reflects lower intracortical
porosity, which contributes to better bone quality, a factor that is essential for maintaining skeletal health in
older ages. A high OPD value is an indicator of higher
activation of remodeling foci and bone turnover in
younger European American males.
The results for BFR do not support the notion that
European American males have larger OPD values at all
ages due to faster remodeling, but Ac.f results indicate a
higher frequency of osteon creations for this group at
younger ages, as indicated by the lowest transformed
value at the y-intercept. The African American females
have a higher reciprocal value of activation frequency,
which translates into a small nontransformed value.
However, there are no significant differences in 1/BFR,
and all groups exhibit similar remodeling rates and patterns. Therefore, bone dynamics and, in particular, bone
turnover are more active in younger European American
American Journal of Physical Anthropology—DOI 10.1002/ajpa
222
H. CHO ET AL.
TABLE 3. Results of pairwise comparisons of intercepts and slopes1
Measurement
OPD
On.Ar
1/Ac.f
1/BFR
Tt.Ar
Ct.Ar
Es.Ar
Ct.Ar/Tt.Ar
Comparison
groups
Difference in
intercepts
Intercept
difference
significant?*
Difference
in slopes
Slope
difference
significant?*
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
AAF-EAF
AAF-AAM
AAF-EAM
EAF-AAM
EAF-EAM
AAM-EAM
7.244
5.829
12.965
1.414
5.721
7.136
0.001
0.004
0.006
0.003
0.005
0.002
0.312
0.541
0.973
0.228
0.661
0.432
6.590
9.321
11.822
2.731
5.232
2.501
4.006
35.865
42.087
31.859
38.080
6.222
5.417
2.394
6.496
7.811
1.078
8.890
9.424
33.471
48.583
24.047
39.159
15.112
0.137
0.254
0.391
0.116
0.254
0.138
No
No
Yes
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
No
No
No
No
No
No
Yes
Yes
No
No
No
No
No
No
No
No
No
No
Yes
Yes
No
Yes
No
No
Yes
Yes
No
No
Yes
0.131
0.136
0.268
0.004
0.137
0.133
0.0001
0.0001
0.0002
0.0000
0.0001
0.0001
0.002
0.011
0.016
0.009
0.014
0.006
0.014
0.198
0.094
0.212
0.108
0.104
0.082
0.269
0.288
0.187
0.207
0.020
0.095
0.037
0.208
0.058
0.112
0.170
0.177
0.306
0.496
0.129
0.319
0.190
0.003
0.004
0.006
0.001
0.004
0.002
No
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
No
Yes
No
No
No
No
Yes
Yes
No
No
No
1
AA, African American; EA, European American; M, male; F, female.
*Significant at P < 0.05.
males than African American females. Overall, relatively
slower remodeling in older ages is beneficial for maintaining bone mass, as remodeling activity introduces
porosity with the creation of new osteons, and continuously forms transient bone resorption cavities. Nevertheless, slower remodeling rates can also allow a greater
and more rapid accumulation of microcracks that could
compromise the mechanical strength and quality of
bone. Given the biomechanical environment of the rib
compared to that for long bones such as the femur, it
seems unlikely that the lower Ac.f observed for some
groups is a result of less targeted remodeling due to
lower biomechanical strain levels for these groups, but
rather factors affecting more ‘‘global,’’ systemically based
remodeling.
All groups exhibited similar-sized osteons that decrease
in size at a comparable rate with increasing age. Age,
therefore, has similar effects on osteon size in all individuals, regardless of biological ancestry and sex. This
may mean that older African American and European
American males and females remodeled smaller packets
of bone than younger age cohorts. Smaller osteons may
be disadvantageous, in that this would allow a greater
number of osteons to exist per unit area of bone and
provide a greater opportunity for osteon debonding.
But a greater number of osteons per unit area can also
be advantageous because there are more osteons for
energy absorption, and a greater number of cement
lines to arrest microcrack propagation (Martin et al.,
1998).
American Journal of Physical Anthropology—DOI 10.1002/ajpa
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
As bone mass is also a function of the amount of cortical bone, results of cross-sectional area measurements
provide additional information on how these factors
account for greater bone mass in the African American
skeleton. Interestingly, our results demonstrate that the
European American male group maintains bone mass,
whereas others lose bone with age.
Due to sexual dimorphism, young males have larger
total subperiosteal areas than females in both ancestries.
None of the groups exhibit significant age-related
changes in rib size, but the positive slope in females may
indicate the normal subperiosteal apposition that occurs
with increasing age (Garn, 1970), perhaps as a compensatory mechanism for endosteal bone loss. The negative
slopes for Tt.Ar in both African American and European
American male samples are difficult to explain. They
may reflect a secular change in rib size, but only for
males, or that males with larger ribs are more likely to
die at a younger age. It is unlikely that larger ribs are a
cause for reduced longevity, but rather that the observed
differences between males and females reflect some
unknown environmental factors that do, and that also
correlate with rib size. This may be similar to observations for correlations between tooth size, health, and lifespan discussed by Larsen (1997).
Also reflecting body size differences, marrow cavity
area is larger in males of both ancestries. While the African American females have the smallest endosteal areas
in the youngest age cohort, the areas of their marrow
cavity are not significantly different from those of European American females, as was expected. Only the African American females exhibit a significant endosteal
expansion with age, while other groups do not exhibit
significant relationships between Es.Ar and age; these
findings are inconsistent with the previous literature.
Although the female regression lines appear to have
steeper slopes, perhaps reflecting increased rates of
endosteal expansion after menopause, the slopes do not
differ significantly with those of males from their respective ancestries. In her study of femoral medullary cavities from the Terry Collection, Ericksen (1979) similarly
reported that African American females have smaller
marrow cavities at younger age groups but greater endosteal expansion after the fifth decade compared to
European American females. Males maintain the trend
of smaller medullary cavities in African Americans than
in European Americans. Plausible explanations that
Ericksen (1979) suggested for this inconsistency are
methodological differences with other studies, intraskeletal variability in age-related bone turnover, or peculiarities of the African American females from the Terry Collection. To our knowledge, none of the individuals in the
current study sample died from metabolic diseases that
could have affected bone remodeling dynamics. These
unexpected findings merit further research and explanation in the future.
Relative cortical area (Ct.Ar/Tt.Ar) is a more relevant
variable in bone loss studies than absolute cortical area
because it is independent of bone size. The relative
amount of cortex in the younger African American males
is significantly greater than observed in the European
American males. But similar to findings for absolute
cortical area, the reverse is seen from the seventh decade onward. That is, for older ages, the European Americans have more bone mass. Though the African American females lose bone at a faster rate than the African
American males, as reflected in the largest slope value,
223
females begin adult life with significantly more bone.
The African American females also exhibit the highest
Ct.Ar/Tt.Ar values from the second to seventh decades,
whereas the European American males exhibit the lowest. That is to say, the African American females have
greater relative bone mass at younger ages than all
other individuals of the same age cohort, but the pattern
also reverses after the seventh decade, which is postmenopausal. Interestingly, the European American males
demonstrate relatively no change in bone mass with age,
while other groups lose bone mass with increasing age.
Achieving larger relative cortical area early in life
owing to small endosteal area, coupled with better bone
quality due to low intracortical porosity, may be a mechanism by which African Americans, especially females,
maintain adequate bone mass in the face of relatively
high rates of endosteal expansion later in life. At older
postmenopausal ages, African American females in the
present study seem to be ‘‘worse off ’’ or less advantaged
in bone mass than all other groups. These findings are
similar to those of Ericksen (1979) in her study of agerelated medullary expansion in the proximal femur of
African American females. In a recent study, also of the
proximal femur, Nelson et al. (2004) found more typical
results: African American and South African postmenopausal black females have more bone strength owing to
smaller endosteal areas and thicker cortices when compared to their European counterparts.
The lack of significant differences in the association of
the response variables with age between African American and European American females may be due to our
small European American sample. Larger European
American samples may yield different results for
females, and similar studies should be conducted on a
larger skeletal sample.
Another caveat in the present study is that the rib
remodeling dynamics may not be representative of bone
turnover in other skeletal sites. A large body of clinical
literature exists on age-associated bone loss and intraskeletal variability between various skeletal sites (Melsen and Mosekilde, 1981; Frisch and Eventov, 1986; Cornell et al., 1988; Pødenphant et al., 1988; Nordin et al.,
1996), as well as concordance between skeletal sites
(Pødenphant et al., 1986; Hernandez et al., 1991). A
recent study concluded that there is significant difference between most of the histomorphometric variables
between the rib and femur midshaft cross-sections of an
Imperial Roman population, possibly due to differences
in the loading environment (Cho, 2002). Previous studies
indicate that the loading environment of a nonweightbearing skeletal site (e.g., nonweight-bearing rib) has no
effect on other skeletal sites undergoing different loading
environments (e.g., weight-bearing femur) (Tommerup
et al., 1993; Robling, 1998). The skeleton is regulated by
both systemic and local biochemical factors. Since bone
responds to the local loading environment, activity level
does not have a systemic effect on the skeleton. Mechanical loading in the femur, therefore, should have no effect
on the rib. Robling (1998) and Tommerup et al. (1993)
analyzed the rib and femur and found significant differences between the two elements. Robling (1998) found a
strong correlation in bone remodeling dynamics between
two elements, but a 31% unexplained variance. Similarly, Tommerup et al. (1993) concluded that activity
level does not have a systemic effect on the skeleton, and
that direct loading on the femur has no effect on the rib.
Since the rib is nonweight-bearing, the magnitude of the
American Journal of Physical Anthropology—DOI 10.1002/ajpa
224
H. CHO ET AL.
load it experiences is considerably less than the femur.
The rib, however, undergoes more continuous loading
cycles compared to the femur, which has more intermittent loading cycles. In addition, because of their involvement in respiration, ribs undergo loading ‘‘throughout
life and should have reduced susceptibility to disuse
osteoporosis,’’ and rib remodeling dynamics reflect general systemic metabolism rather than local effects from
weight-bearing or physical activity (Stein and Granik,
1976). Thus, our results may differ from those of other
studies because we employed rib samples. Further
research utilizing numerous skeletal sites should be conducted to understand intraskeletal variability in bone
remodeling and bone loss.
Additionally, when comparing the skeletal biology of
different biological ancestries, numerous lifestyle and
genetic factors must be considered. A Finnish study concluded that the nonweight-bearing radius benefited from
a high calcium intake early in life, with no effects from
physical activity. The weight-bearing tibia did not benefit
from high calcium intake, but benefited from physical
activity, although the effects were not apparent until
older ages (Uusi-Rasi et al., 2002). Therefore, lifestyle
factors such as diet and activity level may have site-specific effects. Genetic factors are estimated to account for
70–80% of bone mass during the early years of life
(Anderson and Pollitzer, 1994). It was estimated that
African Americans are an admixture of African (variety
of tribal populations primarily from West Africa), European, and Native American ancestries (Wienker, 1987;
Parra et al., 1998), and populations of African ancestry
appear to be inherently advantaged in skeletal health
compared to others. For example, African American children have more rapid skeletal development (including
height) than European American children, even though
they are generally smaller at birth (Garn et al., 1972;
Owen and Lubin, 1973). This more rapid development
occurs despite the common lower socioeconomic status of
African American children and the limitations that accompany low status, such as deficient diets.
Several studies of African populations may shed some
light on the issue of the effects of genetics and lifestyle
on bone. Walker et al. (1971) compared metacarpal bone
dimensions in South African Bantu children and adults
of both sexes to European counterparts, and found unexpected results. Adult Bantu females have a much lower
intake of calcium, more pregnancies, and longer lactation periods compared to Europeans (Walker, 1972), and
thus have ‘‘a very high drain of calcium’’(Walker et al.,
1971). Nonetheless, the Bantu have similar dimensions
to their European counterparts and, more importantly, a
notably lower incidence of osteoporosis. Bantu children,
however, have smaller dimensions than their European
counterparts, as was expected. The authors attributed
the unexpected findings in Bantu adults to more efficient
utilization of calcium and higher levels of physical activity, even in the elderly. Schnitzler (1993) reported lower
fragility fracture rates in both American and African
blacks than in American and African whites, but stated
that the two groups achieve this differently. Although
African Americans generally exhibit lower bone turnover
rates than European Americans, African blacks have a
higher bone turnover than African whites. It was suggested that a higher bone turnover in African blacks, in
conjunction with greater weight-bearing activity, leads to
better bone quality due to frequent removal of fatigue
damage, and thus lower fracture rates (Schnitzler, 1993;
Schnitzler and Mesquita, 1998). Schnitzler (1993) concluded that both American and African blacks have ‘‘a
genetic tendency to greater bone density than whites,’’
but because of nutritional factors (i.e., a calcium-deficient diet), African blacks do not achieve this potential
for denser bone than African whites. Our findings for
African Americans are consistent with their conclusions.
A genetic component to population differences in cortical
bone remodeling is only one of several possible contributing factors. Bone remodeling rates and cortical thickness
are also sensitive to nongenetic factors that include diet,
physical activity, and other aspects of lifestyle (Ericksen,
1991; Anderson and Pollitzer, 1994). Without knowledge
of these factors of the individuals in our study, the influences of specific environmental and hereditary factors cannot be determined. Present rib histomorphometric results
may be due to the effects of diet without the effects of
mechanical loading. Perhaps a calcium-deficient diet in
African Americans, exacerbated by pregnancy and lactation in females, accounts for the greater bone loss in the
nonweight-bearing rib of that group. The reverse may
have been in effect for European American males who had
the initial disadvantage in younger ages but were able to
conserve bone mass with age due to a better diet and its
systemic effect on the skeleton.
CONCLUSIONS
This histological analysis comparing African American
and European American rib samples had some unexpected results, but does not contradict the common
assertion that African Americans are less susceptible to
age-related bone loss and related fragility fractures than
European Americans. However, the results of the
present study suggest that bone dynamics between
ancestries are complex, and the higher bone mass and
slower turnover rates reported for African Americans in
the literature may not exist for all age ranges. European
American males consistently differed in the slopes and
intercepts for the variables compared to the other
groups, and differences are highly significant with African American females. When age-associated patterns are
considered, it was observed that African American females exhibited greater relative cortical area and lower
bone remodeling rates than European American males
for younger age cohorts, but this was reversed in older
ages, when the latter group maintained bone mass while
the former exhibited a more rapid bone loss. The results
of this study underscore the importance of considering
both genetic and environmental factors when attempting
to explain population and sex differences in bone remodeling and bone mass. More broadly, it also illustrates the
importance of looking at age-associated patterns when
making comparisons of bone remodeling rates and bone
mass between populations. In order to better understand
the interplay between genetic and nongenetic factors for
population differences in skeletal health, similar histological studies need to be undertaken on ethnically and
genetically diverse samples, including non-African American blacks. Lifestyle factors that include diet and lifestyle should be explored in conjunction with estimating
bone loss and bone mass.
ACKNOWLEDGMENTS
The rib sample from Washington Park Cemetery (St.
Louis, MO) was made available by the State of Missouri
American Journal of Physical Anthropology—DOI 10.1002/ajpa
AFRICAN AND EUROPEAN AMERICAN BONE REMODELING
Department of Natural Resources as part of an analysis
associated with the relocation of part of the cemetery.
We acknowledge Robert R. Paine, Margaret Streeter,
and Kristina Hayen for contributing some of the crosssectional area and histomorphometric data. We thank
the anonymous reviewers for their valuable comments
and critiques of the manuscript.
LITERATURE CITED
Anderson JJ, Pollitzer WS. 1994. Ethnic and genetic differences
in susceptibility osteoporotic fractures. Adv Nutr Res 9:129–
149.
Baron JA, Barrett J, Malenka D, Fisher E, Kniffin W, Bubolz T,
Tosteson T. 1994. Racial differences in fracture risk. Epidemiology 5:42–47.
Bell NH, Gordon L, Stevens J, Shary JR. 1995. Demonstration
that bone mineral density of the lumbar spine, trochanter,
and femoral neck is higher in black than in white young men.
Calcif Tissue Int 56:11–13.
Bruce R, Stevenson JC. 1990. Current understanding of osteoporosis. Compr Ther 16:9–16.
Cho H. 2002. Age-associated bone loss in an Imperial Roman
population: an histological analysis of inter-skeletal and intraskeletal variability. Ph.D. dissertation, University of MissouriColumbia.
Cho H, Stout SD, Madsen RW, Streeter MA. 2002. Populationspecific histological age-estimating method: a model for known
African American and European American skeletal remains.
J Forensic Sci 47:12–18.
Cohn SH, Abesamis C, Yasumura S, Alaia JF, Zanzi I, Ellis KJ.
1977. Comparative skeletal mass and radial bone mass in
black and white women. Metabolism 26:171–178.
Cornell CN, Schwartz S, Bansal M, Lane JM, Bullough P. 1988.
Quantification of osteopenia in hip fracture patients. J Orthop
Trauma 2:212–217.
DeSimone DP, Stevens J, Edwards J, Shary J, Gordon L, Bell
NH. 1989. Influence of body habitus and race on bone mass
density of midradius, hip and spine in aging women. J Bone
Miner Res 4:827–830.
Ericksen MF. 1979. Aging changes in the medullary cavity
of the proximal femur in American blacks and whites. Am
J Phys Anthropol 51:563–570.
Ericksen MF. 1991. Histologic estimation of age at death using
the anterior cortex of the femur. Am J Phys Anthropol 84:
171–179.
Frisch B, Eventov I. 1986. Hematopoiesis in osteoporosis—preliminary report comparing biopsies of the femoral neck and
iliac crest. Isr Med Assoc J 22:380–384.
Frost HM. 1969. Tetracycline-based histological analysis of bone
remodeling. Calcif Tissue Res 3:211–237.
Frost HM. 1987. Secondary osteon population densities: an algorithm for estimating the missing osteons. Yrbk Phys Anthropol 30:239–254.
Frost HM. 1999. On the estrogen-bone relationship and postmenopausal bone loss: a new model. J Bone Miner Res 14:1473–
1476.
Garn SM. 1970. The earlier gain and later loss of cortical bone:
in nutritional perspective. Springfield, IL: Charles C. Thomas.
Garn SM. 1981. The phenomenon of bone formation and bone
loss. In: DeLuca HF, Frost HM, Jee WSS, Johnston CC Jr, Parfitt AM, editors. Osteoporosis: recent advances in pathogenesis
and treatment. Baltimore: University Park Press. p 3–18.
Garn SM, Sandusky ST, Nagy JM, McCann MB. 1972.
Advanced skeletal development in low income Negro children.
J Pediatr 80:965.
Heaney RP. 2003. Is the paradigm shifting? Bone 33:457–465.
Hernandez ER, Gomez-Castresana F, Villa LF, Rico H. 1991.
The measurement of osteoporosis in clinical practice. Clin
Rheumatol 10:308–310.
Kerley ER. 1965. The microscopic determination of age in
human bone. Am J Phys Anthropol 23:149–164.
225
Kleerekoper M, Nelson DA, Peterson EL, Flynn MJ, Pawluszka
AS, Jacobsen G, Wilson P. 1994. Reference data for bone
mass, calciotropic hormones, and biochemical markers of bone
remodeling in older (55–75) postmenopausal white and black
women. J Bone Miner Res 9:1267–1276.
Larsen CS. 1997. Bioarchaeology: interpreting behavior from
the human skeleton. Cambridge: Cambridge University Press.
LeBoff MS, Glowacki J. 1999. Sex steroids, bone, and aging. In:
Rosen CJ, Glowacki J, Bilezikian JP, editors. The aging skeleton. San Diego: Academic Press. p 159–174.
Luckey MM, Meier DE, Mandeli JP, Dacosta MC, Hubbard ML,
Goldsmith SJ. 1989. Radial and vertebral bone density in
white and black women. J Clin Endocrinol Metab 69:762–770.
Martin RB, Burr DB, Sharkey NA. 1998. Skeletal tissue
mechanics. New York: Springer Verlag.
Melsen F, Mosekilde L. 1981. The role of bone biopsy in the
diagnosis of metabolic bone disease. Orthop Clin North Am
12:571–602.
Nelson DA, Feingold M, Bolin F, Parfitt AM. 1991. Principal
components analysis of regional bone density in black and
white women: relationship to body size and composition. Am
J Phys Anthropol 86:507–514.
Nelson DA, Kleerekoper M, Peterson E, Parfitt AM. 1993. Skin
color and body size as risk factors for osteoporosis. Osteoporos
Int 3:18–23.
Nelson DA, Pettifor JM, Barondess DA, Cody DD, Uusi-Rasi K,
Beck TJ. 2004. Comparison of cross-sectional geometry of the
proximal femur in white and black women from Detroit and
Johannesburg. J Bone Miner Res 19:560–565.
Nevitt MC. 1999. Osteoporosis and fragility fractures in the elderly. In: Rosen CJ, Glowacki J, Bilezikian JP, editors. The
aging skeleton. San Diego: Academic Press. p 349–357.
Nguyen TW, Jones G, Sambrook PN, White CP, Kelly PJ, Eisman JA. 1995. Effects of estrogen exposure and reproductive
factors on bone mineral density and osteoporotic fractures.
J Clin Endocrinol Metab 80:2709–2714.
Nordin BEC. 1966. International patterns of osteoporosis. Clin
Orthop 45:17–30.
Nordin BEC, Chatterton BE, Schultz CG, Need AG, Horowitz
M. 1996. Regional bone mineral density interrelationships in
normal and osteoporotic postmenopausal women. J Bone Miner
Res 11:849–856.
O’Neill TW, Silman AJ, Diaz N, Cooper C, Kanis J, Felsenberg
D, European Vertebral Osteoporosis Study Group. 1997. Influence of hormonal and reproductive factors on the risk of
vertebral deformity in European women. Osteoporos Int 7:
72–78.
Owen GM, Lubin AH. 1973. Anthropometric differences between black and white preschool children. Am J Dis Child
126:168–169.
Parfitt AM. 1984. Age-related structural changes in trabecular
and cortical bone: cellular mechanisms and biomechanical
consequences. Calcif Tissue Int 36:123–128.
Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H,
Meunier PJ, Ott SM, Recker RR. 1987. Bone histomorphometry: standardization of nomenclature, symbols, and units.
J Bone Miner Res 2:595–610.
Parra EJ, Marcini A, Akey J, Martinson J, Batzer MA, Cooper R,
Forrester T, Allison DB, Deka R, Ferrell RE, Shriver MD. 1998.
Estimating African American admixture proportions by use of
population-specific alleles. Am J Hum Genet 63:1839–1851.
Plato CC, Fox KM, Tobin JD. 1994. Skeletal changes in human
aging. In: Crews DE, Garruto RM, editors. Biological anthropology and aging: perspectives on human variation over the
life span. New York: Oxford University Press. p 272–300.
Pødenphant J, Godfredsen A, Nilas L, Nørgård H, Brændstrup
O, Christiansen C. 1986. Iliac crest biopsy: an investigation
on certain aspects of precision and accuracy. Bone Miner 1:
279–287.
Pødenphant J, Riis BJ, Johansen JS, Leth A, Christiansen C.
1988. Iliac crest biopsy in longitudinal therapeutic trials of
osteoporosis. Bone Miner 5:77–87.
Pollitzer ME, Anderson JJB. 1989. Ethnic and genetic differences in bone mass. Am J Clin Nutr 50:1244–1259.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
226
H. CHO ET AL.
Richman BA, Ortner DJ, Schulter-Ellis FP. 1979. Differences in
intracortical bone remodeling in three aboriginal American
populations: possible dietary factors. Calcif Tissue Int 28:209–
214.
Robling AG. 1998. Histomorphometric assessment of mechanical
loading history from human skeletal remains: the relation
between micromorphology and macromorphology at the femoral midshaft. Ph.D. dissertation, University of MissouriColumbia.
Robling AG, Stout SD. 2000. Histomorphometry of human cortical bone: applications to age estimation. In: Katzenberg M,
Saunders SR, editors. Biological anthropology of the human
skeleton. New York: Wiley-Liss, Inc. p 187–213.
Santoro F, Frost HM. 1967. Osteoid seams and resorption
spaces in standard samples of human 6th and 11th ribs.
Henry Ford Hosp Med J 15:241–246.
Santoro F, Frost HM. 1968. Correlation of the transverse sizes
of 6th and 11th ribs in normal and osteoporotic individuals.
Henry Ford Hosp Med J 16:87–96.
Schnitzler CM. 1993. Bone quality: a determinant for certain risk
factors for bone fragility. Calcif Tissue Int [Suppl] 53:27–31.
Schnitzler CM, Mesquita J. 1998. Marrow composition and bone
microarchitecture and turnover in blacks and whites. J Bone
Miner Res 13:1300–1307.
Solomon L. 1979. Bone density in ageing Caucasian and African
populations. Lancet 2:1326–1329.
Stein ID, Granik G. 1976. Rib structure and bending strength:
an autopsy study. Calcif Tissue Res 20:61–73.
Stini W. 1995. Osteoporosis in biocultural perspective. Annu
Rev Anthropol 24:397–421.
Stout SD, Lueck R. 1995. Bone remodeling rates and skeletal
maturation in three archaeological skeletal populations. Am J
Phys Anthropol 98:161–171.
Stout SD, Paine RR. 1992. Histological age estimation using rib
and clavicle. Am J Phys Anthropol 87:111–115.
Stout SD, Paine RR. 1994. Bone remodeling rates: a test of an
algorithm for estimating missing osteons. Am J Phys Anthropol 93:123–129.
Thompson DD, Gunness-Hey M. 1981. Bone mineral-osteon
analysis of Yupik-Inupiaq skeletons. Am J Phys Anthropol 55:
1–7.
Tommerup LJ, Raab DM, Crenshaw TD, Smith EL. 1993. Does
weight-bearing exercise affect non-weight-bearing bone? J Bone
Miner Res 8:1053–1058.
Ubelaker DH. 1986. Estimation of age at death from histology
of human bone. In: Zimmerman MR, Angel JL, editors. Dating and age determination of biological materials. London:
Crom Helm. p 240–247.
Uusi-Rasi K, Sievänen H, Pasanen M, Oja P, Vuori I. 2002.
Associations of calcium intake and physical activity with bone
density and size in premenopausal and postmenopausal women:
a peripheral quantitative computed tomography study. J Bone
Miner Res 17:544–552.
Walker ARP. 1972. The human requirement of calcium: should
low intakes be supplemented? Am J Clin Nutr 25:518–530.
Walker ARP, Walker BF, Richardson BD. 1971. Metacarpal bone
dimensions in young and aged South African Bantu consuming a diet low in calcium. Postgrad Med J 47:320–325.
Weinstein RS, Bell NH. 1988. Diminished rates of bone formation in normal black adults. N Engl J Med 319:1698–1701.
Wienker CW. 1987. Admixture in a biologically African caste of
black Americans. Am J Phys Anthropol 74:265–273.
Wu K, Schubeck KE, Frost HM, Villanueva A. 1970. Haversian
bone formation rates determined by a new method in a mastodon, and in human diabetes mellitus and osteoporosis. Calcif
Tissue Res 6:204–219.
Yoshino M, Imaizumi K, Miyasaka S, Seta S. 1994. Histological
estimation of age at death using microradiographs of humeral
compact bone. Forensic Sci Int 64:191–198.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Документ
Категория
Без категории
Просмотров
1
Размер файла
432 Кб
Теги
group, rate, descent, comparison, bones, sex, midwest, africa, cortical, ribs, remodeling, samples, american, age, european
1/--страниц
Пожаловаться на содержимое документа