close

Вход

Забыли?

вход по аккаунту

?

Estimation of age at death in human males from quantitative histology of bone fragments.

код для вставкиСкачать
Estimation of Age at Death in Human Males from
Quantitative Histology of Bone Fragments '
I. J. SINGH2 AND D. L. GUNBERG
Department of A n a t o m y , Unizsersity of Oregon Medical School,
Portland, Oregon 97201
ABSTRACT
Studies were conducted to quantitate histological age changes in human
bone cortex. Ground (undecalcified ) and decalcified cross sections of mandible, femur
and tibia were measured for: (1) number of osteons in two ficlds, ( 2 ) average number
of lamellae per osteon, and ( 3 ) average Haversian canal diameter. These data were
obtained from 59 subjects (52 males) ranging in age from 39 to 87 years. Multiple
regression techniques were then utilized to estimate age at death from several combinations of these measurements. With age, the number of osteons per unit area of bone
and the number of lamellae per osteon increased but Haversian canal diameter decreased. The number of lamellae per osteon had the least significant correlation with
age. Multiple regression analyses indicated that in this age group age at death could
be estimated to within six years of the true value i n 95% of human males. Histological
measurements of the mandible provided estimates that were consistently more accurate
than those based on the two long bones. A monograph, prepared from mandibular
measurements, can be used to estimate age at death between 40 and 80 years of age
in the male human population. Significant sex differences were not noted and racial
differences were not analyzed. I t was concluded that, compared to existing morphologic methods, microstructure of bone cortex can be quantitated to estimate age at
death more accurately. More extensive studies o n race, sex and metabolic influences
on age estimation from different bones would have important medico-legal and anthropologic implications. The experimental approach utilized here should also be useful i n
diagnostic pathology and in ageing research.
Until 25 years of age, timing and sequence of appearance of ossification centers have provided the most useful determinants of age at death (Pyle and Hoerr,
'55; Greulich and Pyle, '59; Kraus, '61).
Thereafter, until about 50 years of age,
such estimates depend upon strictly subjective morphologic criteria, such as the
rate of cranial suture closure, presence of
lipping on bony margins, and syrnphyseal
changes in the pubic bone (reviews by
Stewart, '54; Krogman, '62).
Age changes in bony lipping were never
critically or extensively investigated. Chronology of cranial suture closure was recently demonstrated to exhibit too much
variability to he of much use in estimating
age at death (Singer, '53; Brooks, '55;
Cobb, '55). Currently, the most widely
used criteria for age estimation are based
on morphodifferentiation of the pubic symphysis (Todd, '20, '21). However, the
validity of those standards was recently
questioned on two grounds: first, the inherently subjective approach does not lend
itself to easy interpretation, and second,
AIM. J. PHYS. ANTHROP.,33: 373-382.
these standards show so much inter-examiner variability that large errors in age
estimation result (Hanihara, '52; Brooks,
'55).
McKern and Stewart ('57) investigated
several criteria of skeletal ageing such as
cranial suture closure, pubic symphyseal
differentiation and epiphyseal union in
several bones and, by numerically ranking
developmental stages, provided regression
equations for estimating age at death. However, error of estimation was not provided.
Recently several authors, notably Jowsey ('60), Frost ('64), and Enlow ('68), have
reported histomorphologic age changes in
human bone cortex. Their studies were not
conducted for estimating age at death and,
from that point of view, they do not provide pertinent information, although they
did demonstrate measurable age changes
in bone cortex. It has been suggested (Lacroix and Dhem, '67; Ahlqvist and Dam1 Supported i n part by the National Institutes of
Health Training grant GM 445.
2 Present address: Institute for Dental Research,
New York University Dental Center, New York, New
York 10010.
373
374
I. J. SINGH AND D. L. GUNBERG
sten, '69) that histological measurements in a formalin-formic acid mixture (10%
of bone might be useful determinants of neutral buffered formalin and 20% forage at death, and Kerley ('65, '69) applied mic acid, in distilled water), and paraffin
the first quantitative histological approach embedded so as to provide 10 thick cross
to this problem. In rats, it was demon- sections. Four such slides were prepared
strated that models for estimating age at from each bone specimen; each slide condeath can be succcssfully postulated from tained from one to three adjacent sections.
histological measurements of bone cortex One slide per bone specimen was stained
(Singh and Gunberg, '70a).
with carbol-thionin (modified after Schmorl,
The objective of this study is to derive '00) and examined.
a practical and useful model for the esti3 . Histologic measiirements
mation of age at death from quantitative
histology of human bone cortex.
Sections were examined under low power
with 1OX objective and l o x widefield
MATERIAL AND METHODS
ocular lenses. This combination of objec1. Material
tive and ocular lenses permitted examination
of a microscopic field that was 2.0 mm
Human material was obtained from 59
cadavers from Anatomy dissection rooms. in diameter. Measurements were made on
The sample was well documented as to the 120 thionin-stained and 19 unstained
sex, medical history, and age at death; and ground sectioned bone specimens. Two miin no case was there gross or microscopic croscopic fields were selected at random
pathology associated with any mandible, from the periosteal third of bone sections.
femur or tibia. Samples of all three bones Wherever possible, each field was from a
were available from 40 subjects (33 males), different section. On most, but not all, dein the remaining 19 males only mandibu- calcified sections this could be done; on
lar samples could be collected. There were ground sections it was not possible since
only seven females; therefore models for only one section pcr specimen was made.
estimation of age at death were derived However, even within the same section, seexclusively from the males. The females lection of the two fields was random and
were used to test the models and for some no attempt at a special selection of fields
preliminary observations on sex differ- was made. Measurements made on each
section were as follows:
ences.
( a ) X,, total number of osteons (com2. Histologic technique
plete Haversian systems) counted in two
Approximately 1 cm x 1 cm fragments fields (not an average). Histologically, each
of bone were removed from the midshaft osteon is clearly demarcated from the
of anterior surface of femur and tibia and neighbouring ones as well as from the infrom the posterior border of the mandibu- terstitial systems by prominent cement
lar ramus opposite the lingula. Although lines. Osteons that were partially obscured
bone fragments did not extend through by the periphery of the field, or were cut
the complete circumference of any bone obliquely or seen only as fragments, were
each bone sample was complete from the included only if the complete Haversian
periosteal to the endosteal border.
canal was seen.
In the 19 males where only mandibular
(b) X,, average number of lamellae per
samples were available, bone fragments osteon determined by a count on all oswere cleaned of soft tissues, dehydrated in teons encountered in two fields.
absolute alcohol, and embedded in Caro( c ) Xs, average Haversian canal d i m e plastic ( a methacrylate manufactured by ter. The shortest diameter of each HaverCarolina Biological Supply). Thick sec- sian canal in two fields was measured
tions, 150 to 200 p, were cut with a rotary using a calibrated ocular micrometer consaw and hand ground to approximately 30 sisting of a 1 cm scale divided into 100
to 50 p thickness. One ground section per equal parts which had been standardized
bone specimen was examined. Bone frag- against a stage micrometer. Diameter was
ments from the remaining 40 subjects obtained only when the complete Haverwere simultaneously fixed and decalcified sian canal was visible in the field. In
375
AGE ESTIMATION FROM BONE
obliquely cut osteons this measurement was
not made when the canal was three or
more times longer than it was wide. An
average value for each bone specimen was
computed.
4. Data analysis
Simple correlations for each of the independent variables (histological measurements) with the dependent variable, age
in years, were first computed. Then multiple linear regression equations of the form :
A
plication to problems of histological studies, such methods were reported by Singh
and Gunberg ('70a). From each bone, separate series of regressions on age were
postulated; their design is presented in
table 1. One nomograph for the estimation
of age at death was constructed from these
regressions.
RESULTS
Y = Pa + ,Or, XI + . . . . . . . . . . . + p p X p
were derived in which the value of p was
determined a priori. Regression relationships were tested at the 5% level of significance. Draper and Smith ('68) discuss
regression techniques extensively; in apTABLE 1
Design of problems for regression analyses
Dependent variable:
Y
Selection of independent variables:
I XI, x2, x3
I1
I11
xi, xz
XI,
x3
IV X l ,
v
x3
XI
VI X?
VII x3
List of measurements:
Y, Age in years
XI, Total number of osteons in two fields
X2, Average number of lamellae per
osteon
X3, Average diameter of Haversian canal
1
These designs were followed thrice; once each for
the mandible, femur and tibia.
1. Age changes in histological
measurements
Characteristics of the various histological measurements of the mandible, femur
and tibia are presented in table 2. Sample
size for the mandible is considerably larger
than that for the other two bones and is
more skewed toward the later ages; consequently, mean age is slightly higher than
that for femur and tibia.
Although all three parameters show age
associated changes, considerable variability is present. As determined by the coefficient of variation, variability is least for
number of osteons (XI) and higher for the
other two. Mean diameter of Haversian
canal is significantly higher for the mandible ( a = 0.05); other measurements do
not vary significantly among bones. Age
changes are illustrated only by a scattergram of the number of lamellae per osteon
(X,) in the femur (fig. 1); figures of age
changes in XI and X3 are not included. As
a function of age, an increase in XI and X2
and a concomitant decrease in X3 occurs.
In all three measurements some non-line-
TABLE 2
Characteristics of human data
Bone
Variable
Osteons
SID.
N
Lamellae/Osteon
S.D.
N
Hav. Can. Dia.
S.D.
(Pm>
N
T
r
Mandible
Femur
Tibia
64.25
12.14
52
62.333
10.80
33
62.333
10.80
33
55.02
7.00
52
56.85
7.39
33
58.48
8.65
33
11.78
2.30
52
11.68
1.86
33
11.74
2.01
33
63.44
18.72
52
43.24
16.42
33
45.54
15.87
33
376
I. J. SINGH AND D. L. GUNBERG
Number
[,
r
.
Ago
I"
..
.
yeors
Fig. 1 Age changes in the average number of
lamellae per osteon ( X r ) in the human femur.
arity, especially at later ages, seems to
exist. For further analysis, however, a linear model was assumed and any nonlinearity ignored for reasons which will be
discussed later.
2. Regression equations for the
estimation of age
Simple correlation matrix for XI, Xz and
X, with age shows that the number of osteons and the Haversian canal diameter
are better correlated with age than the
average number of lamellae per osteon
(table 3). Also, slightly higher correlations
with age are obtained from mandibular
measurements than from femoral or tibia1
data. From each of the three bones, multiple regression equations were computed
with age as the dependent variable and
either three, two, or one independent variable(s) : the predicting equations are provided in table 4. In some equations a negative intercept is seen which is to be
expected because age distribution of the
sample is skewed toward the later ages and
every subject is above 38 years of age. SigTABLE 3
Simple correlations of human histological
measurements with age in years
Variable
Osteons
Lamellae/Osteon
Hav. Can. Dia.
Bone
Mandible
Femur
Tibia
0.969
0.950
-.966
0.945
0.890
-.937
0.919
0.908
--.935
A l l correlations are significantly
zero, P < .05.
different
from
nificant regressions result from every combination of independent variables investigated and the residuals resulting thereform
do not show any systematic bias. Mandibular measurements provide the best regressions with age and with the least standard
errors of estimate for every combination of
independent variable. In every bone, the
error in estimation is least when all three
predicting variables are regressed on age
and is maximum when determination of
age is based only on the average number
of lamellae per osteon. When these regression equations derived from males are used
to estimate age in the seven females in the
sample the resulting errors are within experimental and measurement errors and
are not significant.
3. Nomograph for the estimation
of age
Because of sample limitations, a nomograph was constructed only from mandibular measurements; figure 2 presents one
based on the mandibular measurements
XI and Xs which were chosen because they
are better correlated to age than X,. A sample size of 33 for the femur and tibia was
not considered adequate for construction
of such nomographs. This regression relationship provides a multiple correlation
( R ) of 0.978 with a standard error of estimate of 2.58 years. Thus between the ages
of 40 to 80 years, age may be estimated
to the order of accuracy of -1- 2.58 years in
6 7 % , and within
5.16 years in 95% of
the subjects, of the true value, from this
nomograph.
*
DISCUSSION
Age estimation from skeletal material is
hardly a new idea, but its usage and popularity bears an inverse relationship to its
validity and reliability. The literature is as
voluminous as it is disappointing. This investigation was conceived because often
only bone fragments are available for individual identification and an important
aspect of such identification is the estimation of age at death. Traditional methods
of age estimation based upon subjectively
determined gross morphologic criteria have
been useful but, although currently utilized, they are limited in application. Also,
a prerequisite of such methods is the avail-
377
AGE ESTIMATION FROM BONE
TABLE 4
Regression equations
foT
estimating age in humans from histological measuTements
Dependent variable: age in years
Selection
+
+
+
+
+
+
+
+
+
+
+
+
Multiple
R1
20.82
0.85 XI 0.87 Xz - 0.22 x3
- 18.99 1.13 Xi 1.76 Xz
32.23
0.92 Xi - 0.30 X3
74.73
1.52 Xz - 0.45 X3
- 28.24 1.68 X i
5.31 5.00 Xz
103.99 - 0.63 X3
I
I1
I11
27.65
0.65 XI 0.78 Xz - 0.26 X3
- 14.69 1.13 Xi t 1.11XP
29.59
0.79 Xi - 0.28 X3
61.25 -C 1.74 X7 - 0.44 Xn
16.10 1.38 Xi
2.00
5.16 X2
89.01 - 0.62 X3
V
VI
VII
I
I1
I11
IV
V
VI
VII
+
+
0.979
0.976
0.978
0.969
0.969
0.950
0.966
2.55
2.69
2.58
3.04
3.02
3.83
3.16
0.958
0.948
0.957
0.949
0.945
0.889
0.937
3.24
3.55
3.25
3.52
3.60
5.01
3.82
0.964
0.936
0.957
0.960
0.919
0.908
0.935
3.02
3.93
3.22
3.12
4.33
4.59
3.88
Femur
+
+
+
+
+
+
Std. error
of estimate
Mandible
I
I1
I11
IV
V
VI
VII
IV
1
Regression
equations
Tibia
43.52
0.291 X i
1.47 Xz - 0.34 XS
- 3.40 0.67 Xi -t2.27 Xz
48.61 0.53 Xi - 0.38 X3
54.79 2.19 Xz - 0.4 X3
- 4.76 1.15 Xi
5.10+ 4.88 X z
91.32 - 0.64 X3
Significant as assessed by F-test for regression at 5% level of significance.
ability, not of skeletal fragments, but of
substantial portions of a skeleton. Given
even the best skeletal material, however,
such methods are low in reliability and
high in inter-examiner variability. Consequently, age estimation from skeletal material has not, heretofore, been dependable.
This is especially true for individuals over
25 years of age.
Quantitative histology, on the other
hand, permits objectivity and the resulting
data are amenable to statistical analyses.
The value of histological methods is further enhanced because of their limited demands on amount of material required for
examination. In this study, data were limited to a small sample with a highly skewed
age distribution; also, sample size was considerably larger for the mandible than for
the femur and tibia. As fairly reliable methods are available for the estimation of age
below 25 years (review by Krogman, '62),
this study concentrated on estimating age
at death in an older human population.
The findings of this investigation are in
opposition to those of Currey ('64) and
Barer and Jowsey ('67) who concluded
that there is either a slight age-associated
increase or no change in size of the Haversian canal. This contradiction may be due
to the fact that whereas they measured
perimeter, in this investigation diameter
was measured. The findings, however, are
in agreement with those of others (Landeros and Frost, '64; Epker et d.,'64) that
closure of Haversian canal continues into
the eighth decade.
The estimating equations for age are
based on mandibular measurements demonstrated that good regression relationships
could be proposed and from them age estimates could be made within three years of
the true value in two-thirds of all subjects.
Depending upon the parameters available,
different regressions may be used to estimate age at death. In 95% of the subjects,
such estimates would be accurate to within six years of the true age. Sample size
for the femur and tibia was even more
limited; results, therefrom, are only preliminary, and should be interpreted even
more conservatively. However, they do jus-
I. J. SINGH AND D. L. GUNBERG
75 --
55
62 --
57
60 --
58
--
--
56 -54
70
T
T
i
65 --
60 --
50 --
1-
59
61
--
5 2 --
€
55
--
63
--
65
-v
67 -69
--
71
--
73
---
75
50
--
45
--
40
-
:9' $
90
92
Fig. 2 Nomograph for the estimation of age
death in human males from histologic measurements of the mandible.
tify more extensive investigations on several different bones. The same limitations
to interpretation and extrapolation apply
here as were discussed for rat bone tissues
(Singh and Gunberg, '70a).
For the limited material available, significant sex differences were not noted and
this is in accord with the finding of Kerley
('65). Data on race distribution in the sample were not available, but further studies
along these lines should take into account
the possible variations due to sex or race.
Such variability should be expected in view
of the voluminous literature on sexual and
racial variation in developmental age and
size (review by Tanner, '62).
Most somatic measurements are known
to show a decreasing growth rate and a
non-linear relationship to age, especially
during the older ages. This may also be
true for the histological measurements of
this study; however, any non-linear component in the human material was ignored,
because other studies had indicated that accurate estimation of age was possible from
strictly first order models (Singh and Gunberg, '70a).
Kerley ('65) estimated age from the
number of osteons, Haversian fragments,
non-Haversian systems, and the amount of
lamellar bone. In the present study, nonHaversian systems were not counted because almost all individuals in the sample
were above 40 years of age, and Kerley's
work had indicated that after this age,
such systems are rarely observed. Haversian fragments were not included for two
reasons: first, most interstitial bone in
reality consists of fragments which would
be difficult to count, and second, Haversian systems are not always oriented parallel to the long axis of a bone (Cohen and
Harris, ' 5 8 ) ; therefore, some of them are
bound to be cut obliquely. In cross sections these would appear as fragments,
and errors in counting would result. Instead, the number of lamellae per osteon
and the Haversian canal diameter were included.
At any given age and in any bone section, one can find both young and old osteons. A young osteon is defined as one
that has a large Haversian canal and few
lamellae. In fact, any bone section will generally show many different sizes and ages
of osteons and this has in the past contributed to the opinion that histologic observations of bone are not useful for age
estimation (Deslypere and Baert, '58; Krogman, '62). By relying upon traditionally
subjective histologic approach, it would be
impossible to discuss ageing in bone cortex except in terms of porosity and a general impression of more osteons in older
specimens. Even the above conclusions can
be safely made only between extreme ages,
and a finer discrimination of age or anything else would be difficult.
Theoretically speaking, if it were possible to measure all of the age dependent
histological parameters in bone, then estimation of age from them should be precise. The set of all age dependent variables,
however, is never available and, therefore,
is not measureable. On the other hand, any
one given parameter would probably manifest considerable variation and alone would
be a very poor estimator of age. The fallacy of relying upon a single criterion for
estimating age at death was dramatically
AGE ESTIMATION FROM BONE
illustrated in the case of the "Tepexpan
Man." Based on suture closure alone, the
skull was assigned an age of 55-56 years
(Romero, '49), but on revaluation with
dental radiographs, the age estimate was
revised to 25-30 years (Genoves, '60; Moss,
'60). It follows then that a judiciously
chosen subset of age dependent variables
should provide practically useful estimates
of chronological age within a certain range
of permissible error. In other words, a multivariate approach might succeed where an
univariate model is obviously inadequate.
Ageing of skeletal tissue is a complex
phenomena and, in the present state of
knowledge, it is possible that the independent estimates of the selected predicting (control or independent) variables have
a very limited functional relationship to
ageing. Thus, the ability to select control
variables with high cause and effect relationship to ageing may be limited. Nevertheless, for this problem, a predictive multivariate model can be postulated. It must
be emphasized, however, that such a model
does not assume the existence of a functional relationship between the response
variable (age in this study) and the predicting variables (histological measurements). Such an assumption is neither
implied nor required for the statistical
model proposed here (Draper and Smith,
'68). In one sense then the model is unrealistic, but as long as it is capable of
reproducing the main features of the response variable (age), it is useful. The superiority of a multivariate approach over
the univariate one has long been recognized. In fact, it is practiced, perhaps in
an intangible fashion, by almost everyone
in a diagnostic or decision making situation. A subjective multivariate approach is
inherent in Todd's ('20, '21) standards of
morphological changes in the pubic bone,
in evaluating ossification in various bones
of the hand and wrist (Greulich and Pyle,
'59) or in profile analysis for age determination from various bones as advocated by
Kerley ('65).
An important test of any measurement
system is whether or not another investigator, using the same technique and criteria, could arrive at comparable results.
Analysis of rat skeletal material indicated
that measurement errors were negligible
379
and free of any systematic bias (Singh and
Gunberg, '70a).
In this study, measurements were made
in the periosteal third of the anterior qundrant of the femur and tibia and posterior
border of mandibular ramus. Although conclusive evidence is lacking, Jowsey ('60)
suggested that the microscopic structure
of the periosteal third is different from the
middle and endosteal third, but, at least in
the human femur, the structure of different quadrants in cross section is similar.
Mandibular specimens were selected in relation to an easily defined landmark, the
lingula; but femoral and tibial specimens
were taken from the general area of midshaft. Hence, considerably more variability in the site of sectioning could exist for
the long bones than for the mandible. Although Kerley ('65) stated that a block
almost three inches long approximately in
the middle third of adult human long bones
is histologically the same throughout its
length, evidence for this assumption was
not presented by him. In this study such a
hypothesis was not tested; therefore, an
a priori assumption was implied that the
one or two sections per bone used are representative of the histology of an easily
recognized part of a bone.
In order that age dependent parameters
of bone may be used to provide estimates
of age at death, i t is also necessary to evaluate the variability and errors introduced
in such estimation by environmental factors which are known to effect skeletal tissues. It has been suggested that with age
the closure of Haversian canals is slower
in diabetics (Landeros and Frost, '64), and
also in subjects treated with adrenal COFtical steroids (Kelin el al., '65). In osteoporotic humans the marrow cavity is probably larger (Santoro and Frost, '68), and the
resorption rate may be slightly increased
but with a normal formation rate (Jowsey
et al., '65). Other investigators have reported a reduced rate of Haversian bone
formation in such subjects (Villanueva et
al., '66; Jett et al.,'67). In recent years
several authors have developed and emphasized the view that histological parameters of bone cortex constitute a sensitive
measure of the metabolic state of the organism. The influence of various abnormal
metabolic conditions on the practical prob-
380
I. J. SINGH AND D. L. GUNBERG
lem of age estimation should be further
evaluated. The totalvariability of the Measurement system derived in this manner
would be more useful in practice where
unknown skeletal remains are to be identified and upper and lower limits on age
estimation are hazarded. Such an approach
was discussed in evaluating the effects of
essogen On rat bone tissues (Singh and
Gunberg, '70b).
Jett, S., K. WU and H. M. Frost 1967 Tetracycline-based histological measurement of cortical-endosteal bone formation in normal and
osteoporotic rib. Henry Ford Hosp. Med. Bull.,
15: 325-344.
Jowsey, J. 1960 Age changes in human bone.
CIin. Orthopaedics, 17: 210-218.
Jowsey, J., P. J. Kelly, B. L. Riggs, A. J. Bianco,
z:z
~
~
a
~
~
i
t
~
~
of
~
l
~
ma1 and osteoporotic bone. J. Bone and Jt.
Surg., 47-A: 785-806, 872.
Keiin, M., A. R. Villanueva and H. M. Frost 1965
ACKNOWLEDGMENT
A quantitative histological study of rib from
18 patients treated with adrenal cortical steA part of the sample was obtained
roids. Acta Orthop. Scandinav., 35: 171-184.
through the courtesy of Dr. E. B. Jump,
Kerley, E. R. 1964 The microscopic determinaUniversity of Oregon Dental School.
tion of age i n human bone. Am. J. Phys. An.throp., 23: 149-163.
LITERATURE CITED
1969 Age determination of bone fragAhlqvist, J., and 0. Damsten 1969 A modificaments. J. Forensic Sc., 14: 59-67.
tion of Kerley's method for the microscopic
Kraus, B. S . 1961 Sequence of appearance of
determination of age i n human bone. J. Forenprimary centers of ossification in the human
sic Sc., 14: 205-212.
foot. Am. J. Anat., 109: 103-115.
Barer, M., and J. Jowsey 1967 Bone formation
and resorption in normal human rib. A study Krogman, W. M. 1962 The human skeleton in
forensic medicine. Thomas, Springfield, pp. 18of persons from 11 to 88 years of age. Clin.
111.
Orthopaedics, 52: 241-247.
Brooks, S. T. 1955 Skeletal age at death: the Lacroix, P., and A. Dhem 1967 Le vieillissement des 0s. Etude Microradiographique de
reliability of cranial and pubic age indicators.
l'os compact. Acta Orthopaedica Belgica, 33:
Am. J. Phys. Anthrop., 13: 567-597.
745-760.
Cobb, M. W. 1955 The age incidence of suture
closure. Anat. Rec., 121: 277, Abstr.
Landeros, O., and H. M. Frost 1964 Radial rate
Cohen, J., and W. H. Harris 1958 The threeof osteon closure measured by means of tetradimensional anatomy of Haversian systems. J.
cycline labelling. Henry Ford Hosp. Med. Bull.,
Bone and Jt. Surg., 40-A: 419434.
12: 499-505.
Currey, J. D. 1964 Some effects of ageing in McKern, T. W., ahd T. D. Stewart 1957 Skelehuman Haversian systems. J. Anat., Lond., 98:
tal age changes in young American males ana69-75.
lyzed from the standpoint of age identification.
Deslypere, P., and H. Baert 1958 Assessment
Technical Report EP-45. 'Quartemaster Reof age by the measurement of the Haversian
search and Development Center, Environmencanals of human bones. A critical study of the
tal Protection Research Division, Natick, MassaBalthazaid and Lebrun method. J. Forensic
chusetts.
Med., 5: 195-199.
Draper, N. R., and H. Smith 1968 Applied re- Moss, M. L. 1960 A reevaluation of the dental
status and chronological age of the Tepexpan
gression analysis. Wiley, New York, pp. 104remains. Am. J. Phys. Anthrop., 18: 71-72,
196, 217-242.
Enlow, D. H. 1968 The human face: a n amount Pyle, S. I., and N. L. Hoer 1955 Radiographic
atlas of skeletal development of the knee: a
of the postnatal growth and development of
the craniofacial skeleton. Harper & Row, New
standard of reference. Thomas, Springfield.
York.
Romero, J. 1949 The physical aspects of TepexEpker, B. N., R. Hattner and H. M. Frost 1964
pan man. In: H. Deterra, J. Romero and T. D.
Radial rate of osteon closure: its application
Stewart. Tepexpan Man. Viking Fund Publiin the study of bone formation in metabolic
cations in Anthropology, New York, no. 11,
bone disease. J. Lab. and Clin. Med., 64: 643pp. 89-117.
653.
Santoro,
F., and H. M. Frost 1968 Correlation
Frost, H. M. 1964 Mathematical elements of
of the transverse sizes of 6th and 11th ribs in
lamellar bone remodelling. Thomas, Springfield.
normal and osteoporotic individuals. Henry
Genoves, S. T. 1960 Revaluation of age, stature
Ford Hosp. Med. Bull., 16: 87-96.
and sex of Tepexpan remains, Mexico. Am. J.
Schmorl, G. 1900 Verhandl. d. natur. forscher.
Phys. Anthrop., 18: 205-217.
geb. muchen., 2: 64. Cited by E. B. Brain: The
Greulich, W. W., and S. I. Pyle 1959 Radiopreparation of decalcified tissues. Thomas,
graphic atlas of skeletal development of the
Springfield, pp. 233-234.
hand and wrist. Stanford University, Stanford.
Hanihara, K. 1952 On the age changes in the
Singer, R. 1953 Estimation of age from cranial
male Japanese pubic bone. J. Anthropol. SOC.
suture closure. A report on its unreliability. J.
Nippon, 62: 245-263.
Forensic Med., 1: 52-59.
i
,
AGE ESTIMATION FROM BONE
Singh, I. J., and D. L. Gunberg 1970a Quantitatit-e histology of changes with age in rat bone
cortex. In preparation.
1970b Effects of estrogen treatment on
bone cortex of the rat. In preparation.
Stewart, T. D. 1954 Evaluation of evidence
from the skeleton. In: Legal Medicine. R. B. H.
Gradwohl, ed. St. Louis, pp. 407-450.
Tanner, J. M. 1962 Growth at adolescence.
Davis, Philadelphia.
381
Todd, T. W. 1920 Age changes in the pubic
bone. I. The male white pubis. Am. J. Phys.
Anthrop., 3: 285-334.
1921 Age changes in the pubic bone.
11-VI. Am. J. Phys. Anthrop., 4: 1-70, 333-424.
Villanueva, A. R., L. Ilnicki, H. Duncan and H.
M. Frost 1966 Bone and cell dynamics in
the osteoporoses: a review of measurements by
tetracycline bone labeling. Clin. Orthopaedics,
49: 135-150.
~
Документ
Категория
Без категории
Просмотров
1
Размер файла
698 Кб
Теги
histology, death, malen, age, fragmenty, estimating, human, bones, quantitative
1/--страниц
Пожаловаться на содержимое документа