Keywords
archeology - forensic medicine - humerus
Introduction
Identity is considered an important aspect in anthropology, forensic science and social
demography. The stature is believed to be an integral part of identity. However, estimation
of the identity of body parts and the stature of a person is complex, indeed. In archeology,
the stature estimation of human skeletal remains is an essential step in assessing
the general body size, health and sexual dimorphism.[1]
[2] However, variation exists among the intra- and interpopulation, as well as between
the male and female individuals.[3]
[4] The ethnic, ancestral and geographical differences exist because of the hereditary,
environmental and social factors. Krishan[5] reported that the stature of an individual is the variable that can be estimated
with the greatest accuracy, even from the smallest bone available. However, long bones
are preferred for stature determination because of their better accuracy.[5]
Steele[6] described that estimation of living stature can be done by using the humeral length
in the absence of more accurate long bones, such as the femur or tibia. Salles et
al.[7] opined that the forensic analysis of the modern population cannot be based on the
formulas that were obtained from the ancient population. This is because of the rapid
diachronic secular changes of limbs, and it has been observed by Salles et al.[7] that human beings are growing taller. In this context, developing a set of morphometric
data from the modern population is considered to be essential to forensic investigations.
There are several methods that can be used to estimate the stature of an individual
by using his bones, among which the most reliable one is the regression analysis.[8]
[9] Regression analysis is more appropriate in defining the relationship between the
length of long bone and living stature of an individual, as well as the relationship
between the measurements of bone fragments and bone length.[9] The present study to collect the dimensions of the proximal segments of the humerus
in the South Indian population and to obtain the regression equations that will enable
us to predict the whole length of humerus.
Materials and Methods
The present study included 166 (82 right sided and 84 left sided) dried adult human
cadaveric humeri, which were obtained from the collections of anatomy and forensic
medicine departments of Kasturba Medical College, Mangalore, India. The humeri were
carefully observed with respect to the proximal segments, which are vital in this
present study. The humeri that presented significant deformities at the proximal end
were excluded from the present study. The gender and age determination of the humeri
were not performed in the present study. The present study was approved by the Time-Bound
Research Ethics Committee of Kasturba Medical College, Mangalore (A Constituent Unit
of Manipal Academy of Higher Education, Manipal, Karnataka, India) on July 28, 2010.
The same researcher performed all the measurements, which prevented inter observer
variation. Each measurement was performed three times and the average was calculated.
The maximum length of humerus (MLH) was measured by using the osteometric board. This
was the maximum distance between the most proximal points over the caput humeri and
the most distal point of the trochlea. The measurements of the proximal segments of
the humeri were performed by using a digital Vernier caliper (Mitutoyo Corporation
150 mm/6 inch, model number 500-196-20, Kawasaki, Japan) ([Fig. 1]).
Fig. 1 Measurements of the proximal segments of the humerus performed in the present study.
-
S1 - distance between the most proximal part of the head of the humerus and the most
distal part of the anatomical neck
-
S2 - oblique length between the most proximal and distal points in the anatomical
neck
-
S3–distance between the most medial aspects of the lesser tubercle and the most lateral
aspect of the greater tubercle
-
S4 - horizontal breadth of the humeral head, at its center
-
S5 - largest breadth of the greater tubercle
-
S6 - largest breadth of the lesser tubercle
-
S7 - the widest part of the proximal end of the humerus
The morphometric data were tabulated separately for the right and left sides of the
humeri. The data were statistically analyzed by using the SPSS software, version 15
(SPSS Inc., Chicago, IL, USA). After obtaining the mean and standard deviation (SD.)
for each of the parameters, the association between the variables and the length of
the humerus was investigated by means of the Pearson correlation coefficient (r).
The linear regression was applied for the right and left humeri separately. The simple
linear regression shows the regression coefficient (COE) and the significance (p-value) for the dimensions of the proximal segments of the right and left humeri,
separately. The simple linear regression analysis shows the relationship of the dimensions
of the individual proximal segment with the MLH. This analysis shows the coefficient
of correlation (Pearson correlation coefficient) between a dependent variable and
an independent variable. The Pearson coefficient determines the strength of the relationship
between the variables. The p-value determines the statistical significance. A p-value < 0.05 was considered statistically significant. The simple linear regression
equations were formulated from the obtained data, which would predict the mean length
of the humerus (MHL).
Results
From the 166 humeri (82 of right side and 84 of left side), the MHL on the right side
was 30.75 cm, with a SD of 2.03 cm. The MHL on the left side was 30.27 cm, with a
SD of 2.28 cm.
The descriptive statistics represented in [Table 1] shows the mean values of the proximal segments of the humeri of both the sides.
The data were compared by using, independent samples test. The analysis showed that
the comparison between the right and left sides was not statistically significant.
The 2-tailed p-values were higher than 0.05 (p > 0.05).
Table 1
Morphometric data of the proximal segments of the humerus (n = 166)
|
Segment
|
Right side (n = 82)
|
Left side (n = 84)
|
P-value
|
|
S1
|
3.28 ± 0.31
|
3.25 ± 0.32
|
0.43
|
|
S2
|
4.12 ± 0.36
|
4.08 ± 0.35
|
0.45
|
|
S3
|
3.26 ± 0.47
|
3.29 ± 0.40
|
0.72
|
|
S4
|
3.84 ± 0.31
|
3.80 ± 0.35
|
0.46
|
|
S5
|
2.92 ± 0.29
|
2.84 ± 0.27
|
0.08
|
|
S6
|
1.36 ± 0.17
|
1.36 ± 0.19
|
0.78
|
|
S7
|
4.50 ± 0.36
|
4.42 ± 0.37
|
0.16
|
(values are given in cm, mean ± SD, independent samples test)
The Pearson coefficient dictates the quantitative relation of each of the segment
with the length of humerus. The Pearson coefficient, coefficient of determination
(R2) and p-values obtained in the present study are given in [Table 2]. The present study observed that the relationship between the dimensions of the
proximal segments of the humerus and the length of the humerus were proportional.
The relationship was real and did not occur by chance (p = 0.00, which is statistically highly significant).
Table 2
Pearson coefficient and p-values of the right (n = 82) and left (n = 84) sides of the proximal humeral segments
|
Segment
|
Pearson coefficient
|
R2
|
Significance (p-value)
|
|
Right side
|
Left side
|
Right side
|
Left side
|
Right side
|
Left side
|
|
S1
|
0.54
|
0.50
|
0.30
|
0.25
|
0.00
|
0.00
|
|
S2
|
0.78
|
0.77
|
0.62
|
0.59
|
0.00
|
0.00
|
|
S3
|
0.39
|
0.41
|
0.15
|
0.17
|
0.00
|
0.00
|
|
S4
|
0.71
|
0.72
|
0.50
|
0.51
|
0.00
|
0.00
|
|
S5
|
0.63
|
0.73
|
0.40
|
0.53
|
0.00
|
0.00
|
|
S6
|
0.38
|
0.46
|
0.15
|
0.21
|
0.00
|
0.00
|
|
S7
|
0.77
|
0.77
|
0.60
|
0.59
|
0.00
|
0.00
|
[Table 3] shows the Pearson coefficient in decreasing order of values. Among all the measurements
performed, the S2 segment of both sides was the best parameter. The Pearson coefficient
was 0.78 on the right side and 0.77 on the left side. The second best parameter was
the S7 segment (The Pearson coefficient was 0.77 on both the sides). The lowest Pearson
coefficient value was for the S6 segment on the right side, which was 0.38, and on
the left side, it was for the S3 segment, which was 0.41.
Table 3
Pearson coefficient of the proximal segments in decreasing order
|
Right humerus
|
Pearson coefficient
|
P-value
|
Left humerus
|
Pearson coefficient
|
P-value
|
|
S2
|
0.78
|
0.00
|
S2
|
0.77
|
0.00
|
|
S7
|
0.77
|
0.00
|
S7
|
0.77
|
0.00
|
|
S4
|
0.71
|
0.00
|
S5
|
0.73
|
0.00
|
|
S5
|
0.63
|
0.00
|
S4
|
0.72
|
0.00
|
|
S1
|
0.54
|
0.00
|
S1
|
0.50
|
0.00
|
|
S3
|
0.39
|
0.00
|
S6
|
0.46
|
0.00
|
|
S6
|
0.38
|
0.00
|
S3
|
0.41
|
0.00
|
The simple regression was formulated, Y = (a + bX) ± SD, in which Y is the maximum humeral length (dependent variable); X is the dimension of the proximal segment of the humerus (independent variable), b is the multiplying factor and a is the constant, which was obtained by using the SPSS software (SPSS Inc.). The simple
regression formula, which has highest multiplying factor, is considered to be the
best. The simple regression formulae, which were obtained in the present study are
given in [Table 4]. The formula applied to the S4 segment was the best for predicting the length of
humerus on the right side (the multiplying factor was 4.67). On the left side, the
formula that was applied to the S5 segment was considered the best (the multiplying
factor was 6.19).
Table 4
Simple regression formulae to determine the mean humeral length (MHL) from the data
of the proximal segments
|
Right humerus
|
Left humerus
|
|
MHL = 19.06 + 3.55 (S1) ± 1.71
|
MHL = 18.60 + 3.60 (S1) ± 1.98
|
|
MHL = 12.35 + 4.46 (S2) ± 1.26
|
MHL = 9.89 + 5.00 (S2) ± 1.47
|
|
MHL = 25.28 + 1.67 (S3) ± 1.88
|
MHL = 22.62 + 2.33 (S3) ± 2.09
|
|
MHL = 12.84 + 4.67 (S4) ± 1.44
|
MHL = 12.32 + 4.73 (S4) ± 1.60
|
|
MHL = 18.06 + 4.35 (S5) ± 1.58
|
MHL = 12.68 + 6.19 (S5) ± 1.56
|
|
MHL = 24.57 + 4.55 (S6) ± 1.89
|
MHL = 22.78 + 5.55 (S6) ± 2.03
|
|
MHL = 11.13 + 4.36 (S7) ± 1.30
|
MHL = 9.27 + 4.75 (S7) ± 1.47
|
Discussion
In the absence of the cranium and the pelvis, the fragments of long bones can be used
during the anthropology and forensic science investigations.[10]
[11]
[12] The simple regression formulae are considered important during the determination
of the stature from the available anthropometric dimensions.[13]
[14] Singhal and Rao[15] reported that the length of the humerus can be used to estimate the stature of an
individual with an error margin of less than 2 cm. They also reported that their regression
formula, which was derived from the longer segments of the humerus, can be used with
other samples of the Indian population. However, if there are shorter segments, new
equations are required. Somesh et al[16] studied the distance between the most proximal point of the humeral head and the
greater tuberosity. They also determined the distance between the head of the humerus
and the surgical neck of the humerus in the South Indian population. The present study
did not measure these segments of the humerus. The best parameters in the present
study were, the oblique length between the most proximal and the most distal points
on the anatomical neck (S2), the horizontal breadth of the humeral head at its center
(S4), and the widest part of the proximal end of the humerus (S7). These horizontal
dimensions are different from the ones reported by Somesh et al,[16] who measured the vertical segments, which had lower coefficient values. This suggests
that the horizontal segments of the present study gave better results. Salles et al[7] also reported that the oblique length between the most proximal and the most distal
points on the anatomical neck, and the horizontal breadth of the humeral head at its
center had good correlation to the MLH.
The forensic, anthropologic and archaeological studies suggest that the MHL offers
important data to study the characteristics of a population.[17] In the present South Indian study, the MHL was 30.75 ± 2.03 cm on the right side
and 30.27 ± 2.28 cm on the left side, respectively. These data are almost similar
to the data from the Turkish population.[18] However, the MHL of the present study was lower in comparison to that of other European
population. This is due to the ancestral variation, as the Europeans are tall and
robust.[13]
[19] In a Brazilian study, the oblique length between the most proximal and the most
distal points on the anatomical neck were 4.9 ± 0.5 cm and 4.8 ± 0.4 cm for the right
and left sides, respectively.[7] These dimensions are much higher than the ones obtained in the present study, which
were 4.12 ± 0.36 cm and 4.08 ± 0.35 cm, respectively. However, the horizontal breadth
of the humeral head at its center was 3.84 ± 0.31 cm for the right side and 3.8 ± 0.35
cm for the left side, in the present study. This is similar to the data observed in
the Brazilian study by Salles et al, which was 4.0 ± 0.4 cm and 3.9 ± 0.3 cm, respectively.[7]
By using the derived regression formulae, one can fairly estimate the full length
of humerus. This is useful when only a few segments of a long bone are available.
By using the MLH, it is possible to determine the stature of an individual. In the
present study, the correlation between the measurements of the proximal segments of
the humerus and the stature of an individual was not possible, due to lack of information
about the dried bones. The stature of an individual is extremely variable and can
be affected by ethnic differences. The regression formula of one population cannot
be applied to another.[16] In this context, the data and formulae of the present study are important as they
provide data about the South Indian population. The morphometric data of the humeral
segments have implications in the identification of missing persons during the medico-legal
investigations.[17]
[20] The morphometric data of the humerus segments are enlightening to orthopedic surgeons
during the treatment of humeral fractures and reconstruction of the humerus.[16] The data are also enlightening during procedures like prosthetic designing, sizing
and positioning.[21]
[22]
Conclusions
We believe that the data in the present study will contribute to estimation of the
humeral length and the length of its proximal segments in a subset of the South Indian
population. The derived formulae of the present study may be useful in forensic investigations
in which the stature of an individual has to be determined and there are only few
segments of bone are available. The data can be of help in archaeological and anthropological
studies in which excavations often yield only a few incomplete skeletal remains. The
data in the present study are also essential to orthopedicians, who can utilize them
during the planning of reconstructive surgeries involving the proximal end of the
humerus.
