The ancestral shape hypothesis: an evolutionary explanation for the
occurrence of intervertebral disc herniation in humans
Corresponding author: Mark Collard mcollard@sfu.ca
BMC Evolutionary Biology 2015, 15:68 doi:10.1186/s12862-015-0336-y
The electronic version of this article is the
complete one and can be found online at: http://www.biomedcentral.com/1471-2148/15/68
Abstract
Background - Recent studies suggest there is a relationship between
intervertebral disc herniation and vertebral shape. The nature of this
relationship is unclear, however. Humans are more commonly afflicted with
spinal disease than are non-human primates and one suggested explanation for
this is the stress placed on the spine by bipedalism. With this in mind, we
carried out a study of human, chimpanzee, and orangutan vertebrae to examine
the links between vertebral shape, locomotion, and Schmorl’s nodes, which are
bony indicators of vertical intervertebral disc herniation. We tested the
hypothesis that vertical disc herniation preferentially affects individuals
with vertebrae that are towards the ancestral end of the range of shape
variation within Homo sapiens and therefore are
less well adapted for bipedalism.
Results - The study employed geometric morphometric techniques.
Two-dimensional landmarks were used to capture the shapes of the superior
aspect of the body and posterior elements of the last thoracic and first lumbar
vertebrae of chimpanzees, orangutans, and humans with and without Schmorl’s
nodes. These data were subjected to multivariate statistical analyses.
Canonical Variates Analysis indicated that the last thoracic and first
lumbar vertebrae of healthy humans, chimpanzees, and orangutans can be
distinguished from each other (p<0 .028="" and="" but="" cannot="" chimpanzees="" humans="" of="" p="" pathological="" vertebrae="">0.4590). The Procrustes distance between
pathological humans and chimpanzees was found to be smaller than the one
between pathological and healthy humans. This was the case for both vertebrae.
Pair-wise MANOVAs of Principal Component scores for both the thoracic and
lumbar vertebrae found significant differences between all pairs of taxa
(p<0 .029="" except="" humans="" i="" pathological="">vs
Conclusions - The results support the hypothesis that intervertebral
disc herniation preferentially affects individuals with vertebrae that are
towards the ancestral end of the range of shape variation within H. sapiens and therefore are less well adapted for
bipedalism. This finding not only has clinical implications but also
illustrates the benefits of bringing the tools of evolutionary biology to bear
on problems in medicine and public health.
Keywords: Back
pain; Disc herniation; Vertebral shape; Bipedalism; Geometric morphometrics;
Schmorl’s nodes
Background
Back pain is an important health issue. It has been estimated that
22-65% of people will experience back pain at some point in their lives [1], making it one of the most common health
problems [2].
Back pain is also one of the most serious health problems. Recent work suggests
that it is the greatest contributor to disability on a global scale [3]. The prevalence of back pain and the frequency with
which it causes disability mean that it can impose a substantial economic
burden on countries [4].
For example, the annual cost of back pain in the UK has been estimated to
exceed £1.5 billion per year [5].
Given the importance of back pain, there is a need for greater understanding of
the underlying factors that cause it.
Intervertebral disc herniation is a widespread but poorly understood
cause of back pain [6].
It is defined as a prolapse of the gelatinous substance inside the disc, the
nucleus pulposus, either horizontally through the fibrous outer disc layers or
vertically into the vertebral endplate[6].
Intervertebral disc herniation is frequent among adults, with recent studies
suggesting that prevalence rates range from 20% to 78%, depending on
population [7]-[9]. Numerous potential causes of intervertebral disc
herniation have been proposed, including genetic predisposition, disc
composition, developmental issues, and physical strain or trauma [10]-[15], but the aetiology and pathogenesis of the
condition remain unclear [16].
Recently, a number of studies have suggested that vertebral shape may
affect the propensity to experience intervertebral disc herniation. Pfirrmann
and Resnick [17] found that Schmorl’s nodes were associated with a
flat vertebral endplate as opposed to the more common concave endplate in a
sample of cadavers. Schmorl’s nodes are depressions on the upper and lower
surfaces of the vertebral body that result from vertical intervertebral disc
herniation [18]. They can be identified with the use of medical
imaging technology [19],[20] or on dry bone [21]-[23]. Harrington et al. [24] obtained similar results to Pfirrmann and
Resnick [23]. They found that the size and shape of the
vertebral body was associated with lower lumbar intervertebral disc herniation
in a large sample of clinical patients. Most recently, Plomp et al. [25] found a correlation between lower thoracic
vertebral shape and the presence of Schmorl’s nodes in Medieval and
Post-Medieval skeletons. They concluded that the shape of the pedicles and
vertebral body might play a role in the development of Schmorl’s nodes [25].
Given that several studies have suggested a link between vertebral shape
and the propensity to experience intervertebral disc herniation, there is
reason to investigate possible explanations for why certain vertebral shapes
should predispose for this condition. Humans display substantially more
degenerative and traumatic spinal pathologies than non-human primates[26],[27]. This has led some researchers to hypothesize that
our unique mode of locomotion, bipedalism, may influence the development of
these conditions [28]-[30]. With this theory in mind, we carried out a
cross-species study of vertebral shape variation in humans and non-human apes
to examine the links between vertebral shape, locomotor behaviour and vertical
intervertebral disc herniation. Specifically, we tested the hypothesis that
intervertebral disc herniation preferentially affects individuals whose
vertebral shape are towards the ancestral end of the range of shape variation
within Homo sapiens and therefore are less well adapted
for bipedalism.
This “ancestral shape hypothesis” is derived from work on the evolution
of bipedalism. It is now generally accepted that humans and other hominins are
more closely related to chimpanzees (Pan troglodytes),
and bonobos (Pan paniscus) than they are to any other living
species [31]. At the moment, the locomotor behaviour of the
common ancestor of the hominin and chimpanzee/bonobo lineages is debated. A
number of different locomotor behaviours have been suggested to be antecedent
to bipedalism [32]-[34]. The most frequently cited suggestion is that the
common ancestor was a knuckle-walker like chimpanzees, bonobos, and gorillas (Gorilla gorilla) [35]. However, it has also been argued that the common
ancestor of the hominin and chimpanzee/bonobo lineages was an arboreal
quadrumanous climber like orangutans (Pongo pygmaeus) [36]. Depending on which of these hypotheses is correct,
the hominin lineage shifted from knuckle-walking to bipedalism or from
quadrumanous climbing to bipedalism. In both cases, the demands placed on the
vertebrae would have changed. Selection likely acted to improve the ability of
the vertebrae to cope with the new demands, but given that vertebral shape is
almost certainly influenced by multiple genes and that the spine is
multifunctional, we can also expect that within a hominin species, some
individuals will have vertebrae that are closer in shape to those of the common
ancestor than others. Given that the ancestral vertebral shape would not have
been adapted for bipedalism, individuals whose vertebrae are towards the
ancestral end of the range of shape variation can be expected to suffer
disproportionately from external load-related spinal pathologies.
In our study, we employed geometric morphometrics (GM) to record and
analyze vertebral shape. Being based on coordinate data as opposed to the
inter-landmark distances of standard morphometrics, GM methods allow patterns
of shape variation to be investigated within a well-understood statistical
framework that yields easily interpreted numerical and visual results[37]-[40]. To identify vertebral shapes associated with
bipedalism, we adopted the approach employed by Russo [41] and compared human vertebrae to the vertebrae of a
knuckle-walker (the chimpanzee) and the vertebrae of a quadrumanous climber
(the orangutan). Humans, chimpanzees, and orangutans vary in modal vertebral
formulae, with 12 thoracic and 5 lumbar vertebrae in humans, 12 thoracic and 4
lumbar vertebrae in orangutans, and 13 thoracic and 3 to 4 lumbar vertebrae in
chimpanzees [42]. Consequently, the last thoracic (T12/13) and the
first lumbar (L1) vertebrae were included in the study to ensure positional
homology between vertebrae of different species and to represent the
functionally distinct thoracic and lumbar spines. Another important
consideration was that human T12s and L1s are commonly afflicted by Schmorl’s
nodes [20] and previous studies have found their shapes to
correlate with the presence of these lesions [25],[43].
Following Plomp et al. [25], the presence of Schmorl’s nodes was used as an
indicator of vertical intervertebral disc herniation. We tested two predictions
of the ancestral shape hypothesis: 1) there should be differences in shape
between healthy human, chimpanzee, and orangutan vertebrae; and 2) human
vertebrae with evidence of vertical intervertebral disc herniation should be
more similar in shape to the vertebrae of chimpanzees or orangutans than are
human vertebrae without evidence for intervertebral disc herniation.
Last thoracic and first lumbar vertebrae from 71 humans, 36 chimpanzees,
and 15 orangutans were included in the sample (Table 1). Only adult individuals were included in the
analysis. Due to preservation issues and curation practices, not all
individuals had both vertebrae present. In total, the sample comprised 114
human vertebrae (59 thoracic, 55 lumbar), 56 chimpanzee vertebrae (25 thoracic,
31 lumbar), and 27 orangutan vertebrae (12 thoracic, 15 lumbar). The human
vertebrae analysed in this study are the same as those analysed by Plomp et
al. [43]. They are Medieval-period specimens from the sites
of Fishergate House, York [44], and Coach Lane, North Shields [45], and are curated at Durham University, UK (see Additional
file 1 for details). Of the 114 human vertebrae, 54
exhibited Schmorl’s nodes, and 60 did not. For the purposes of this paper, we
will refer to the former as “pathological” and the latter as “healthy”. The
chimpanzee and orangutan vertebrae are housed at the American Museum of Natural
History, New York, and the Smithsonian National Museum of Natural History,
Washington DC, and are a mixture of zoo and wild-caught animals. None of the
non-human ape vertebrae exhibit signs of pathology.
Table 1. Composition of sample of
vertebrae from the 71 humans, 36 chimpanzees, and 15 orangutans included in
study
The dataset comprised the 2D Cartesian coordinates of 17 landmarks
recorded on 197 dry-bone vertebrae (Figure 1). The landmarks were based on those used by Plomp et
al. [25]. As we explained earlier, these authors found an
association between certain vertebral shapes and the presence of Schmorl’s
nodes in humans. The landmarks capture the outline shape of the pedicles, the
neural foramen, and the superior aspect of the vertebral body [25]. Eight are Type II landmarks; the remainder are
semi-landmarks [46]. The landmarks were recorded on standardized
digital photographs with the aid of TPS Dig [47].
The first step in a geometric morphometric analysis is to reduce the
effects of confounding factors [38]. As vertebrae are symmetrical along the sagittal
midline, we followed the protocol outlined by Klingenberg et al. [48] to remove the influence of asymmetry on the
results. To begin with, we created two datasets, one comprising the original
landmark coordinates and the other the reflected and relabelled landmark
coordinates [49]. We then slid the semi-landmarks to remove shape
differences arising from the small differences that occur in the placement of
semi-landmarks [50],[51]. Next, we subjected the coordinates of the Type II
landmarks and slid semi-landmarks of both datasets to generalized Procrustes
analysis (GPA) [50],[51]. GPA is designed to remove translation, rotational,
and size effects [38]. Lastly, asymmetry was removed by calculating the
average Procrustes coordinates between the original and reflected landmarks.
These coordinates were used in all further analyses. The reflection, sliding
procedure, and GPA were applied separately to the T12/T13 and L1 vertebrae. The
semi-landmarks were slid, and the GPA performed, with the aid of TPSRelW [47].
Intra observer error was assessed as per Neubauer et al. [52]. A T12 vertebra and an L1 vertebra were each
digitized ten times, and the greatest Procrustes distance between the repeated
measurements for a given specimen was then compared to the smallest Procrustes
distance among all specimens of the same type. In both analyses, the
between-specimen distances were close to three times greater than the
within-specimen distances. According to Neubauer et al. [52], this level of difference indicates that
intra-observer error is unlikely to be a confounding factor. This analysis was
carried out with Morphologika [53].
The impact of allometry was assessed by regressing the Procrustes
coordinates on log centroid size. The statistical significance of male–female
shape differences was determined using MANOVAs on all principal component (PC)
scores obtained through principal components analyses (PCA). These analyses
were performed in SPSS 16.0 [54] and MorphoJ [55], and carried out separately for the last thoracic
and first lumbar vertebrae. Allometry was found to be a factor in vertebral
shape (T12/T13: r2 = 0.092, p < 0.001; L1: r2 = 0.072,
p < 0.001), but sexual dimorphism was not (p > 0.10). The frequency of
Schmorl’s nodes between the two human populations was not statistically
different (χ2 p > 0.339) and there was no statistical
difference in vertebral shape between human populations (p > 0.108). In
light of these results, we opted to employ allometry-free regression residuals
derived from pooled-sex samples in the remainder of the analyses [56], with humans analyzed as a homogeneous population.
Following Klingenberg and Monteiro [57], we applied canonical variates analysis (CVA) to
the pooled-sex regression residuals to determine the maximum Procrustes
distances among taxa. The significance of differences was assessed using
permutations of pair-wise Procrustes distances among all possible pairs of
taxa. We carried this out initially for the last thoracic vertebrae and
repeated it for the first lumbar vertebrae. The analyses were conducted in
MorphoJ [55].
We used PCA to explore the pattern of inter-taxon shape variation [38]. Only PCs representing at least 5% of the total
variance were considered in order to minimize noise from higher
components [58]. The statistical significance of inter-taxon PCA
score differences was assessed using MANOVAs. As in the previous analysis, the
last thoracic and first lumbar vertebrae were analyzed separately. This
analysis was performed in TPSRelW [47] and SPSS 16.0 [54].
The CVA of the Procrustes coordinates for the last thoracic vertebrae
returned three CVs (canonical vectors). The first accounts for 67.1% of the
variance, the second 24.1%, and the third 8.8%. There is little separation
among taxa when CV3 is plotted against CV1 (Additional file 2: Figure S1). When CV1 is plotted against CV2
(Figure 2a), it is apparent that the shape of the last thoracic
vertebrae of orangutans is different from the shape of the last thoracic
vertebrae of not only healthy and pathological humans but also of chimpanzees.
It is also apparent when CV1 is plotted against CV2, that pathological human
vertebrae have more in common with chimpanzee vertebrae than do healthy humans.
All the inter-taxon Procrustes distances are significant except for the one between
pathological humans and chimpanzees (Table 2). Pair-wise analyses using permutations of
Mahalanobis distances produce the same pattern (Additional file 3: Table S1).
The PCA yielded six PCs that met the ≥5% of variance criterion. PC1
accounts for 30.3% of the variance, PC2 25.6%, PC3 18.1%, PC4 9.6%, PC5 5.1%,
and PC6 4.8%. There is considerable overlap among the taxa on PC1, PC4, PC5,
and PC6 (Additional file 2: Figure S2-S4). However, the taxa are distinguishable
when PC2 and PC3 are plotted against each other. Healthy human vertebrae tend
to score more positively on PC2 and negatively on PC3, while orangutan
vertebrae tend to score more negatively on PC2 and more positively on PC3
(Figure 2b). Pathological humans and chimpanzees plot between
healthy humans and orangutans on both PCs. The deformation grids in
Figure 2b illustrate the shape differences between the
negative and positive extremes of PC2 and PC3. Moving from the positive extreme
of PC2 to the negative one, there is a transition from heart-shaped vertebral
bodies with flared pedicles to rounder vertebral bodies without flared
pedicles. There is also a decrease in neural foramen size relative to the
vertebral body, and a translation of the posterior margin of the body into the
neural canal. The shape differences that occur as we move from negative to
positive scores on PC3 are a relative decrease in neural foramen size and a
relative increase in the width of the pedicles. Thus, compared to healthy
humans, pathological humans and chimpanzees have relatively smaller neural
foramina, shorter, wider pedicles, and rounder vertebral bodies, whereas
compared to orangutans, they have relatively larger neural foramina, longer,
narrower pedicles, and more heart-shaped vertebral bodies. The MANOVA on the
PCs that met the criterion for inclusion is significant (p < 0.0001).
Pair-wise MANOVAs are significant for all inter-taxon comparisons, except those
between pathological humans and chimpanzees (Table 3).
Table 3. Results of pairwise MANOVAs for
T12/T13 vertebrae on PCs 1 through 6, which collectively represent 93.5% of the
total shape variance
Thus, the results of the analyses of the last thoracic vertebrae are
consistent with the test predictions. The finding of differences among healthy
human, chimpanzee, and orangutan vertebrae is in line with the prediction that
the vertebral shape of these taxa should be distinguishable due to their
locomotion. The analyses also indicate that healthy human vertebrae are
statistically distinguishable from chimpanzee vertebrae, whereas pathological
human vertebrae are not. This finding is consistent with the prediction that
human vertebrae with evidence of vertical intervertebral disc herniation should
be more similar in shape to the vertebrae of chimpanzees than are human
vertebrae without evidence of intervertebral disc herniation.
First lumbar vertebrae
The CVA of the Procrustes coordinates for the first lumbar vertebrae
returned three CVs. The first CV accounts for 68.6% of the variance, the second
20.1%, and the third 11.3%. There is little distinction among the taxa when CV3
is plotted against CV1 (Additional file 2: Figure S5). In contrast, when CV2 is plotted against
CV1, it is apparent that pathological human vertebrae are more similar in shape
to chimpanzee vertebrae than are healthy human vertebrae (Figure 3a). The Procrustes distances support these
observations. All inter-taxon Procrustes distances are significant except the
one between pathological human and chimpanzee vertebrae (Table 4). The same pattern is produced by pair-wise analyses
using permutations of Mahalanobis distances (Additional file 3: Table S2).
The PCA for the first lumbar vertebrae yielded five PCs that met the ≥5%
of variance criterion. PC1 accounts for 39.6% of the variance, PC2 23.8%, PC3
16.0%, PC4 7.4%, and PC5 5.1%. There is considerable overlap among taxa on PCs
3 through 5 (Additional file 2: Figure S6-S7). However, taxa are distinguishable on
PC1 and PC2 (Figure 3b). Healthy humans score more negatively than
orangutans on PC1 and PC2, with pathological humans and chimpanzees between
them on both PCs. Deformation grids show that the shape differences between
samples are similar to those seen in the T12/13 analysis (Figure 3b). Again, the most obvious shape differences relate
to the pedicles and vertebral body. Moving from the negative end of PC1 to the
positive end, there is a decrease in neural foramen size relative to the
vertebral body and the pedicles become shorter and wider. In addition, there is
a backward translation of the posterior margin of the vertebral body that
results in it becoming less heart-shaped and more shovel-shaped. The shape
differences captured by PC2 are a difference in pedicle orientation, with the
pedicles becoming more laterally angled from the body as we move from the
positive end of PC2 to the negative one. To reiterate, healthy humans and
orangutans score at the extremes of the shape variation on both PCs, with
pathological humans and chimpanzees between them. Thus, when compared to
healthy humans, pathological humans and chimpanzees tend to have smaller neural
foramina, wider, shorter pedicles, and more shovel-shaped bodies. When compared
to orangutans, pathological humans and chimpanzees have larger neural foramina,
narrow pedicles, and less shovel-shaped vertebral bodies. The MANOVA on the PCs
that met the ≥5% of variance criterion is statistically significant
(p = 0.001). Pair-wise MANOVAs are significant for all inter-taxon comparisons,
except between pathological humans and chimpanzees (Table 5).
Table 5. Results of pairwise MANOVAs for
first lumbar vertebra on PCs 1 through 5, which collectively represent 92.0% of
the total shape variance
Thus, the results of the analyses of the first lumbar vertebrae are also
consistent with the test predictions. The finding of differences in shape
between the healthy human, chimpanzee, and orangutan specimens is consistent
with the first test prediction, while the finding that pathological human
vertebrae are closer in shape to chimpanzees than are healthy human vertebrae
is consistent with the second test prediction.
Discussion
This study explicitly tested the ancestral shape hypothesis, which holds
that intervertebral disc herniation preferentially affects individuals with
vertebrae that are towards the ancestral end of the range of shape variation
within H. sapiens and therefore are less well adapted for
bipedalism. We tested two predictions of this hypothesis with shape data
recorded on the last thoracic and first lumbar vertebrae of orangutans,
chimpanzees, healthy humans, and humans with Schmorl’s nodes, which are bony
indicators of intervertebral disc herniation. The first prediction was that
there should be differences in shape between healthy human vertebrae,
chimpanzee vertebrae, and orangutan vertebrae, due to the different modes of
locomotion of the taxa. The second prediction was that pathological human
vertebrae should share more similarities in shape with chimpanzee or orangutan
vertebrae than do healthy human vertebrae. The results of the analyses were
consistent with both predictions. We found that the last thoracic and first lumbar
vertebrae of healthy humans, orangutans, and chimpanzees differ significantly
in shape, which is in line with the first prediction. We also found that human
vertebrae with Schmorl’s nodes share more similarities in shape with chimpanzee
vertebrae than do healthy human vertebrae, which is consistent with the second
prediction. Thus, the study supports the ancestral shape hypothesis.
A potential alternative explanation for our findings needs to be
considered. The vertebral shapes associated with Schmorl’s nodes may be a
consequence of intervertebral disc herniation rather than its cause. It is
certainly the case that vertebrae can remodel. For example, the shape of the
vertebral body is known to change with increasing age. Body height tends to
decrease and there is often an increase in surface concavity as the endplate
collapses [59]. However, we do not consider intervertebral disc
herniation causing changes in vertebral shape to be a good explanation for our
results. One of the main shape differences identified between healthy human
vertebrae and those with Schmorl’s nodes relates to the neural foramen [25]. Previous work indicates that the shape of the
neural foramen does not change after the neural arch fuses to the vertebral
body [60],[61] at around six years of age in humans [62]. Therefore, any factor that influences the shape of
the neural foramen must act during spinal development. Bone remodelling during
development could influence the shape of the vertebrae, including the neural
foramen. Although this could explain why there is a difference in shape between
pathological and healthy human vertebrae, it does not explain the relationship
identified between pathological human and chimpanzee vertebrae. This
explanation would require that bone remodelling result in vertebral shape
changes that systematically approach a shape functionally related to
quadrupedal locomotion. This, we submit, is less parsimonious than the
ancestral vertebral hypothesis.
A possible functional explanation for the association between vertical
disc herniation and vertebral shape is provided by Harrington et al. [24]. These authors suggest that the diameter of the
vertebral disc influences its ability to withstand tension during compression.
Their argument rests on LaPlace’s law [62], which states that the ability of a fluid-filled
tube to withstand tension decreases with increasing radius. According to
Harrington et al. [24], the rounder bodies of pathological vertebrae would
have a larger diameter than the more heart-shaped bodies seen in healthy
vertebrae, making the intervertebral disc less able to withstand stress [24],[62]. We also found that pathological vertebrae have
shorter pedicles compared to healthy vertebrae. The pedicles act as structural
buttresses for the vertebral body and play an important role in load bearing
during axial compression [63]-[68]. It has been hypothesized that the shorter pedicles
identified in vertebrae with Schmorl’s nodes may be less able to withstand
physical strain placed on the spine [25],[45]. Since bipedalism causes a large amount of axial
loading on the lower vertebrae [30], we hypothesize that the combination of round
vertebral bodies with short pedicles may provide less support for the spine
during bipedal posture and locomotion.
Our results have implications for medical science beyond shedding light
on the causes of intervertebral disc herniation. One is that vertebral shape
may be a factor that could help predict an individual’s susceptibility to
vertical intervertebral disc herniation. The shape analysis techniques used in
this study can also be used on medical images, such as CT scans. It may be
possible for clinicians to investigate an individual’s vertebral shape and
identify those who may be at risk of developing the condition. This ability
would have significant diagnostic and preventative value, especially for
high-risk individuals, such as athletes [69]. In addition, a better understanding of the role
that locomotion and posture plays in the health of the spine could aid in the
treatment of individuals afflicted with symptomatic vertical intervertebral
disc herniation. Locomotion is recognized as an important factor in
rehabilitation for sufferers of back pain [70], and understanding the role that vertebral
variation can play in spinal health could aid physiotherapists to refine
activity and exercise regimes. Thus, the findings of this study may not only
help medical practitioners to understand why some individuals are more commonly
afflicted with back problems than others, but may also lead to advances in the
identification, prevention, and treatment of people suffering from intervertebral
disc herniation.
In addition to offering these potential clinical benefits, our results
provide further support for the claim that an evolutionary perspective can shed
important light on human health problems[71]-[74]. Evolutionary medicine has identified the value of
considering evolutionary adaptations to enable better understanding of human
developmental issues, chronic diseases, and nutritional needs [74], but the influence of skeletal morphology on human
health has received little attention. Our study highlights the potential of using
osteological analyses of skeletal variation, including comparative analyses
between humans and non-human primate species, in evolutionary medical studies.
Bipedalism has been suggested to impact human spinal and joint health [28]-[30],[75],[76], but few studies have been carried out to evaluate
this proposition [30]. The identification of an ancestral vertebral shape
that influences the occurrence of a common spinal pathology supports the idea
that the relatively rapid evolution of bipedalism in the hominins may continue
to impact modern human health.
The main goal of our study was to shed light on a major contemporary
health problem with the conceptual and analytical tools of evolutionary
biology, but our results also contribute to the understanding of human
evolution. Specifically, they shed additional light on the evolution of
bipedalism, and in particular, the functional anatomy associated with it.
Previous studies have identified morphological characteristics purported to
relate to bipedalism [77]-[80]. The present findings add features to this list—a
larger neural foramen relative to body size, taller, narrower pedicles, and a
more heart-shaped vertebral body. There are two persistent debates in
palaeoanthropology regarding the evolution of bipedalism and a better understanding
of the functional anatomy of bipedal vertebrae may contribute to their
resolution. The first debate regards the timing of the emergence of bipedalism
in the evolutionary record. The understanding of how human vertebrae are unique
among hominoids enables the identification of fossil vertebrae adapted for
bipedal locomotion; this will help researchers infer which species were
bipedal, provide additional insight into how bipedalism evolved, and suggest
whether it followed a gradual or punctuated pattern of evolution. The second
debate surrounding the evolution of bipedalism is whether early bipeds walked
with their knees and hips in a flexed position, like chimpanzees, or if their
mode of bipedalism resembled our own[81]-[84]. The ability to identify vertebral shape
characteristics unique to humans and compare these with features unique to
modern chimpanzees may provide additional insight into the functional anatomy
required for habitual bipedalism and help understand the evolutionary trends
that led to the modern human gait.
With regard to future research, several possibilities suggest
themselves. Firstly, if the ancestral shape hypothesis is accepted, it prompts
the question of how this shape influences the occurrence of vertical
intervertebral disc herniation. This could be investigated with biomechanical
studies of the interaction between locomotion, vertebral morphology, and the
soft tissues of the spine. Secondly, this area of research would benefit from
the use of 3D shape analyses of human and non-human ape vertebrae to
investigate how 3D vertebral morphology relates to locomotion and human spinal
health. Lastly, the clinical value of this research would be substantially
increased with the inclusion of in-vivo medical images of individuals with and
without back problems.
Conclusions
Our study supports the hypothesis that intervertebral disc herniation
preferentially affects individuals with vertebrae that are towards the
ancestral end of the range of shape variation within Homo sapiens and therefore are less well adapted
for bipedalism. As predicted by the hypothesis, we identified a relationship
between the shape of the last thoracic and first lumbar vertebrae and
locomotion in humans, chimpanzees, and orangutans, and we found that human
vertebrae with signs of vertical intervertebral disc herniation are
indistinguishable from those of chimpanzees. When compared to healthy humans,
pathological human and chimpanzee vertebrae tend to have smaller neural
foramina, shorter, wider pedicles, and more shovel-shaped vertebral bodies. Our
study’s support for the ancestral shape hypothesis not only has clinical
implications, but also provides another illustration of the benefits of
bringing the conceptual and analytical tools of evolutionary biology to bear on
problems in medicine and public health.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors have contributed to the preparation of this manuscript. KAP
collected and analyzed the data, and led the preparation of the manuscript. MC
supervised the project and provided substantial contributions to the analysis
and interpretation of the data, and the preparation of the manuscript. USV,
DAW, and KD contributed to the project by aiding in the interpretation of the
data and the preparation of the manuscript. All authors approved the final
manuscript.
Additional files
Additional
file 1:. Archaeological site information for Fishergate House, York, and
Coach Lane, North Shields, Tyne and Wear.
We thank York Osteoarchaeology, Pre-Construct Archaeology, Durham
University, the Natural History Museum, and the American Museum of Natural
History for access to the specimens used in the study. We also thank Helgi
Pétur Gunnarsson for his assistance with the analyses. The study was funded by
the Social Sciences and Humanities Research Council, Canada Research Chairs
Program, Canada Foundation for Innovation, British Columbia Knowledge
Development Fund, MITACS, and Simon Fraser University. We thank the editor and
two anonymous reviewers for their insightful comments and suggestions on this
paper.
References
Walker BF: The prevalence of low back
pain: a systematic review of the literature from 1966 to 1998. J Spinal Disord 2000, 13:205-17.
Balague F, Mannion AF, Pellise F, Cedraschi
C: Non-specific low back pain. Lancet 2012, 379:482-491. 
Buchbiner R, Blyth FM, March LM, Brooks P,
Woolf AD, Hoy DG: Placing the global burden of low back pain in
context. Best Pract Res Cl Rh 2013, 27:575-589.
Webb R, Brammah T, Lunt M, Urwin M, Allison
T, Symmons D: Prevalence and predictors of intense, chronic,
and disabling neck and back pain in the UK general population. Spine 2003, 28:1195-1202.
Maniadakis N, Gray A: The economic burden of back pain in the UK. Pain 2000, 84:95-103.
Cowperthwaite M, Van den Hout W, Webb
KM: The impact of early recovery on long-term outcomes in a cohort of
patients undergoing prolonged nonoperative treatment for lumbar disc
herniation. J Neurosurg Spine 2013, 19:301-306.
Mann E, Peterson C, Hodler J: Degenerative marrow (modic) changes on cervical spine magnetic resonance
imaging scans. Spine 2011, 36:1081-1085.
Sharma A, Parsons M, Pilgram T: Temporal interactions of degenerative changes in individual
components of the lumbar intervertebral discs. Spine 2011, 36:1794-1800.
Takatalo J, Karppinen J, Ninimaki J, Taimela
S, Nayha S, Jarvelin MR, et al. Prevalence of degenerative imaging findings in
lumbar magnetic resonance imaging among young adults. Spine. 2009;34:1716–21.
Adams MA, Dolan P: Intervertebral disc degeneration: evidence for two distinct
phenotypes. J Anat 2012, 221:497-506.
Burke KL: Schmorl’s nodes in an American
military populations: frequency, formation, and etiology. J Forensic Sci 2012, 57:571-577.
Dar G, Masharawi Y, Peleg S, Steinberg N,
May H, Medlej B, et al.: Schmorl’s nodes distribution in the human spine and its possible
etiology. Eur Spine J 2010, 19:670-675.
Mok F, Samartzis D, Karppinen J, Luk K, Fong
D, Cheung K: Prevalence, determinants, and association of
Schmorl’snodes of the lumbar spine with disc degeneration: a population-based
study of 2449 individuals. Spine 2010, 35:1944-1952.
Williams FMK, Manek NJ, Sambrook P, Spector
TD, MacGregor AJ: Schmorl’s nodes: common, highly heritable, and
related to lumbar disc disease. Arthritis Rheum 2007, 57:855-860.
Gordon SJ, Yang KH, Mayer PJ, Mace AH, Kish
VL, Radin EL: Mechanism of disc rupture: a preliminary
report. Spine 1991, 16:450-456.
Wilkenstein B, Allen K, Setton L: Intervertebral disc herniation: Pathophysiology and emerging therapies. In The intervertebral disc:
Molecular and structural studies of the disc in health and disease.
Edited by Shapiro I, Risbud M. Springer, New York; 2014:305-320.
Pfirrmann C, Resnick D: Schmorl’s nodes of the thoracic and lumbar spine:
Radiographic-pathologic study of prevalence, characterization, and correlation
with degenerative changes of 1,650 spinal levels in 100 cadavers. Radiology 2001, 219:368-374.
Schmorl G, Junghans H: The human spine in health and disease. Grune and
Stratton, New York; 1971.
Peng B, Wu W, Hou S, Shang W, Wang X, Yang
Y: The pathogenesis of Schmorl’s nodes. J Bone Joint Surg 2003, 85:879-882.
Kyere KA, Than KD, Wang AC, Rahman SU, Valdivia-Valdivia
JM, Marca FL, et al.:Schmorl’s nodes. Eur Spine J 2012, 21:2115-2121.
Burke KL: Schmorl’s nodes in an American
military population: Frequency, Formation, and Etiology. J Forensic Sci 2012, 57(3):571-7.
Üstündağ H: Schmorl’s nodes in a
Post-Medieval skeletal sample from Klostermarienberg, Austria. Int J Osteoarchaeol 2009, 19:695-710.
Robb J, Bigazzi R, Lazzarini L, Scarsini C,
Sonego F: Social “status” and biological “status”: A comparison of grave
goods and skeletal indicators from Pontecagnano. Am J Phys Anthropol 2001, 115:213-22.
Harrington JF, Sungarian A, Rogg J, Makker
VJ, Epstein MH: The relation between vertebral endplates shape
and lumbar disc herniations. Spine 2001, 26:2133-2138.
Plomp KA, Roberts CA, Strand Vidarsdottir
U: Vertebral morphology influences the development of Schmorl’s nodes
in the lower thoracic vertebra. Am J Phys Anthropol 2012, 149:172-182.
Lovell N: Patterns of injury and illness
in the great apes: A skeletal analysis. Washington, DC, Smithsonian
Institution; 1990.
Jurmain R: Trauma, degenerative disease,
and other pathologies among the Gombe Chimpanzees. Am J Phys Anthropol 1989, 80:229-237.
Jurmain RD: Degenerative joint disease in
African great apes: an evolutionary perspective. J Hum Evol 2000, 39:185-203.
Latimer B: The perils of being bipedal. Ann Biomed Eng 2005, 33:3-6.
Filler AG: Emergence and optimization of
upright posture among hominiform hominoids and the evolutionary pathophysiology
of back pain. Neurosurg Focus 2007, 23:E4.
Ruvolo M: Genetic diversity in Hominoid
primates. Ann Rev Anthropol 1997, 26:515-540.
Lovejoy C, Suwa G, Simpson SW, Matternes JH,
White TD: The great divides:Ardipithecus ramidus reveals the
postcrania of our last common ancestors with African apes. Science 2009, 326:100-106.
Gebo DL: Climbing, brachiation, and
terrestrial quadrupedalism: historical precursors of hominid bipedalism. Am J Phys Anthropol 1996, 101:55-92.
Tuttle RH: Evolution of hominid bipedalism
and prehensile capabilities. Philos Trans Roy Soc London B 1981, 292:89-94.
Richmond BG, Strait DS: Evidence that humans evolved from a knuckle-walking ancestor. Nature 2000, 404:382-385.
Thorpe S, Crompton RH: Orang-utan positional behaviour and the nature of arboreal
locomotion in Hominoidae. Am J Phys Anthropol 2007, 131:384-401.
Adams DC, Rohlf FJ, Slice DE: Geometric morphometrics: ten years of progress following the
‘revolution’. Ital J Zool 2004, 71:5-16.
Slice DE: Geometric morphometrics. Ann Rev Anthropol 2007, 36:261-281.
Webster M, Sheets DH: A practical introduction to landmark-based geometric
morphometrics.
Quant Meth Paleobiol 2010, 16:163-188.
Zelditch ML, Swiderski DL, Sheets HD:
Fink: Geometric morphometrics for biologists: a primer.
Elsevier Academic Press, San Diego; 2004.
Russo GA: Prezygapophyseal articular
facet shape in the Catarrhine thoracolumbar vertebral column. Am J Phys Anthropol 2010, 142:600-612.
McCollum MA, Rosenman BA, Suwa G, Meindl RS,
Lovejoy CO: The vertebral formula of the last common ancestor of African apes
and humans. J Exp Zool 2010, 214B:123-134.
Holst M: Fishergate House artefacts and
environmental evidence: the human bone. UK: Arch Planning Consultancy, 2005.
Available at: http://www.archaeologicalplanningconsultancy.co.uk/mono/001/rep_bone_hum1a.html.
Accessed online, July3, 2010
Langthorne JY. Human Skeletal Remains in
Goode A, Taylor-Wilson R. Archaeological exhumation of the former Quaker burial
ground on Coach Lane, North Shields, North Tynside, Tyne and Wear: Assessment
Report, 2012: 44–65.
Plomp KA, Roberts CA, Strand Vidarsdottir U:
Technical Note: Does the correlation between Schmorl’s nodes and vertebral
morphology extend into the lumbar spine? Am J Phys Anthropol, early view.
O’Higgins P: The study of morphological
variation in the hominid fossil record: biology, landmarks and geometry. J Anat 2000, 197:103-120.
Rohlf F: TPSDig Version 1.40.
Department of Ecology and Evolution, The State University of New York at Stony
Brook, Stony Brook, NY; 2004.
Klingenberg CP, Barluenga M, Meyer A: Shape analysis of symmetric structures: Quantifying variation
among individuals and asymmetry. Evolution 2002, 56(10):1909-1920.
Meindl K, Windhager S, Wallner B, Schaefer
K: Second-to-fourth digit ratio and facial shape in boys: the lower
the digit ratio, the more robust the face. Proc R Soc B 2012, 279:2457-2463.
Bookstein F: Landmark methods for forms
without landmarks: morphometrics of group differences in outline shape. Med Image Anal 1997, 1:225-243.
Bookstein FL: Creases as local features of
deformation grids. Med Image Anal 2000, 4:93-110.
Neubauer S, Gunz P, Hublin JJ: Endocranial shape changes during growth in chimpanzees and humans:
a morphometric analysis of unique and shared aspects. J Hum Evol 2010, 59:555-566.
O’Higgins P, Jones N: Tools for statistical shape analysis. Hull, York
Medical School; 2006.
Inc SPSS: SPSS Base 8.0 for Windows
User’s Guide. SPSS Inc, Chicago; 2007.
Klingenberg CP: MorphoJ:
an integrated software package for geometric morphometrics. Mol Ecol Res 2011, 11:353-357.
Monteiro LR: Multivariate regression models
and geometric morphometrics: the search for casual factors in the analysis of
shape. Syst Biol 1999, 48:192-199.
Klingenberg CP, Monteiro LR: Distances and directions in multidimensional shape spaces:
Implications for morphometric applications. Syst Biol 2005, 54:678-688.
Baylac M, Frieb M: Fourier descriptors, Procrustes superimposition, and data
dimensionality: An example of cranial shape analysis in modern human
populations. In Modern Morphometrics in
Physical Anthropology, Part 1 Theory and Methods (ed.). Edited by
Slice D. Kluwer, New York; 2005.
Twomey L, Taylor J: Age changes in lumbar intervertebral discs. Acta Orthop 1985, 56:496-499.
Johnson DR, O’Higgins P, McAndrew TJ: The relationship between age, size and shape in the upper thoracic
vertebrae of the mouse. J Anat 1998, 161:73-82.
Masharawi Y, Salame K: Shape variation of the neural arch in the thoracic and lumbar
spine: characterization and relationship with the vertebral body shape. Clin Anat 2011, 24:858-867.
Scheuer L, Black S: Developmental juvenile osteology. Academic, London;
2000.
Letić M: Feeling wall tension in an
interactive demonstration of Laplace’s law. Adv Physiol Educ 2012, 36(2):176.
Shapiro L: Evaluation of the “unique”
aspects of human vertebral bodies and pedicles with consideration. of Australopithecus
africanus. J Hum Evol 1993, 25:433-470.
Shapiro L: Functional morphology of the
vertebral column in primates. InPostcranial adaptation in
non-human primates. Edited by Gebo DL. Nothern Illinois University
Press, Dekalb, Illinois; 1993.
Whyne CM, Hu SS, Klisch S, Lotz J: Effect of the pedicle and posterior arch on vertebral body
strength predictions in finite element modeling. Spine 1998, 23(8):899-907.
Peloquin JM, Yoder JH, Jacobs NT, Moon SM,
Wright AC, Elliot DM: Human L3L4 intervertebral disc
mean 3D shape, modes of variation, and their relationship to degeneration. J Biomech 2014, 47(10):2452-2459.
Hongo M, Abe E, Shimada Y, Murai H, Ishikawa
N, Sato K: Surface strain distribution on thoracic and lumbar vertebrae under
axial compression: the role in burst fractures. Spine 1999, 24:1197-1202.
Hsu WK, McCarthy KJ, Savage JW, Roberts DW,
Roc GC, Micey AJ, et al.: The professional athlete spine initiative: outcomes after lumbar
disc herniation in 342 elite professional athletes. Spine J 2011, 11:180-186.
Vogt L, Pfeifer K, Portscher M, Banzer
W: Influences of nonspecific low back pain on three-dimensional lumbar
spine kinematics in locomotion. Spine 2001, 26:1910-1919.
Stearns SC: Evolutionary medicine: its
scope, interest and potential. P R Soc B-Biol Sci 2012, 279:4305-4321.
Nesse RM, Bergstrom CT, Ellison PT, Flier
JS, Gluckman P, et al.: Making evolutionary biology a basic science for medicine. Proc Natl Acad Sci U S A 2009, 107:1800-1807.
Nesse RM, Stearns SC: The great opportunity: evolutionary applications to medicine and
public health. Evol Appl 2008, 1:28-48.
Trevathan WR: Evolutionary Medicine. Ann Rev Anthropol 2007, 36:139-154.
Hutton CW: Generalized osteoarthritis: an
evolutionary problem? Lancet 1987, 329:1463-1465.
Hutton CW: Osteoarthritis: the cause not
result of joint failure? Ann Rheum Dis 1989, 48:958-961.
Davis PR: Human lower lumbar vertebrae. J Anat 1961, 95:337-344.
Boszczyk BM, Boszczyk AA, Putz R: Comparative and functional anatomy of the mammalian lumbar spine. Anat Rec 2001, 264:157-168.
Manfreda E, Mitteroecker P, Bookstein F,
Schaefer K: Functional morphology of the first cervical vertebra in humans and
nonhuman primates. Anat Rec: New Anat 2006, 289B:184-194.
Hernandez CJ, Loomis DA, Cotter MM, Schifle
AL, Anderson LC, Elsmore L, et al.:Biomechanical allometry in hominoid thoracic vertebrae. J Hum Evol 2009, 56:462-470.
Crompton RH, Weijie LY, Gunther M, Savage
M: The mechanical effectivenss of erect and “bent knee, bent hip”
bipedal walking in Australopithecus afarensis. J Hum Evol 1998, 35:55-74.
Lovejoy CO, McCollum MA: Spinopelvic pathways to bipedality: why no hominids ever relied on
a bent-hip-bent-knee gait. R P Biol Sci 2010, 65:3289-3299.
Stern JT, Susman RL: The locomotor anatomy of Australipithecus afarensis. Am J Phys Anthropol 1983, 60:279-317.
Hunt KD: The evolution of human
bipedality: ecology and functional morphology. J Hum Evol 1994, 26:183-202.
