Abstract
The primate scapula has been studied widely since its shape has been shown to correlate with how the forelimb is used in daily activities. In this study, we expand on the existing literature and use an image‐based methodology that was originally developed for orthopaedic practice to quantify and compare the three‐dimensional (3D) morphology of the scapula across humans and great apes. We expect that this image‐based approach will allow us to identify differences between great apes and humans that can be related to differences in mobility and loading regime of the shoulder. We hypothesize that gorillas and chimpanzees will have a similar scapular morphology, geared towards stability and weight‐bearing in knuckle‐walking, whilst the scapular morphology of orangutans is expected to be more similar to that of humans given their high glenohumeral mobility associated with their suspensory lifestyle. We made 3D reconstructions of computed tomography scans of 69 scapulae from four hominid genera (Pongo, Gorilla, Pan and Homo). On these 3D bone meshes, the inferior glenoid plane was determined, and subsequently, a set of bony landmarks on the scapular body, coracoid, and acromion were defined. These landmarks allowed us to measure a set of functionally relevant angles which represent acromial overhang, subacromial space and coracoacromial space. The angles that were measured are: the delto‐fulcral triangle (DFT), comprising the alpha, beta, and delta angle, the acromion‐glenoid angle (AGA), the coracoid‐glenoid centre‐posterior acromial angle (CGA), the anterior tilt (TA CGA) and the posterior tilt of the CGA (PT CGA). Three observers placed the landmarks on the 3D bone meshes, allowing us to calculate the inter‐observer error. The main differences in the DFT were found between humans and the great apes, with small differences between the great apes. The DFT of humans was significantly lower compared to that of the great apes, with the smallest alpha (32.7°), smallest delta (45.7°) and highest beta angle (101.6°) of all genera. The DFT of chimpanzees was significantly higher compared to that of humans (p < 0.01), with a larger alpha (37.6°) and delta angle (54.5°) and smaller beta angle (87.9°). The mean AGA of humans (59.1°) was significantly smaller (p < 0.001) than that of gorillas (68.8°). The mean CGA of humans (110.1°) was significantly higher (p < 0.001) than in orangutans (92.9°). Humans and gorillas showed mainly a posterior tilt of their coracoacromial complex whilst chimpanzees showed mainly an anterior tilt. The coracoacromial complex of the orangutans was not tilted anteriorly or posteriorly. With our image‐based method, we were able to identify morphological features of the scapula that differed significantly between hominid genera. However, we did not find an overall dichotomy in scapular morphology geared towards high stability (Pan/Gorilla) or high mobility (Homo/Pongo). Further research is needed to investigate the functional implications of these differences in scapular morphology.
Keywords: functional morphology, hominid, primates, scapula, shoulder
Overview of the mean acromion‐glenoid angle of each genus on a frontal view of the left scapula (ventral view). Red (a) Pongo, green (b) Gorilla, blue (C) Pan and purple (d) Homo. Gorilla, a knucklewalker, shows the greatest value, compared to both Homo and Pan. One representative three‐dimensional mesh of each genus is used.
1. INTRODUCTION
The wing‐shaped scapula is an intriguing bone with specific morphological attributes that have been studied intensively across mammals (e.g. Astúa, 2009; Matsuo et al., 2019; Monteiro & Abe, 1999; Oxnard, 1968; Zhang et al., 2012). Also in primates, the scapula has received much attention as its shape correlates strongly with how the forelimb is used in daily activities (e.g. Preuschoft et al., 2010; Schmidt, 2008; Selby & Lovejoy, 2017; Young, 2006, 2008). In great apes and humans, the scapula is positioned dorsally on the thorax, providing the forelimb with a large range of motion during arboreal locomotion, whilst also providing adequate stability for quadrupedal locomotion. It has been shown by several authors that the scapular morphology of great apes correlates with locomotor behaviour, in particular arboreal versus terrestrial locomotion (Green, 2013; Taylor, 1997; Young, 2008). For example, scapular shape changes during ontogeny in chimpanzees and gorillas, along with a shift from arboreal to terrestrial locomotion. Such ontogenetic changes in scapular shape are, however, not found in orangutans and humans which do not show marked changes in locomotor behaviour during ontogeny (Green, 2013). These findings indicate that there is a correlation between locomotor behaviour and scapular morphology. Moreover, this leads us to expect that morphological traits of the scapula are similar in Gorilla and Pan and distinct from that of Pongo and Homo.
Most research on scapular morphology in great apes has used traditional two‐dimensional measurements (Corruccini & Ciochon, 1976; Craik et al., 2014; Doyle et al., 1980; Shea, 1986) or more advanced three‐dimensional (3D) morphometric methods using a set of anatomical landmarks on the scapula (Bello‐Hellegouarch et al., 2013; Green et al., 2015; Püschel & Sellers, 2016; Young, 2008). In this study, we use a medical image‐based methodology to quantify and compare the 3D scapular morphology across great apes and humans. This methodology was originally developed for the quantification of scapular shape in humans and has allowed the identification of morphological correlates of shoulder dysfunction (Naidoo et al., 2017; Van Parys et al., 2021). One of the important concepts that were introduced by Naidoo et al. was the delto‐fulcral triangle (DFT). The DFT represents the subacromial space and is defined by three important bony landmarks; the most anterior point of the coracoid and the most lateral and posterior point of the acromion (Naidoo et al., 2017). In humans, a low subacromial space is associated with degenerative rotator cuff tears, and quantifying the DFT has both clinical and functional relevance. We believe that quantification of the subacromial space is also relevant in nonhuman primates and expect that it will allow us to identify differences between great apes and humans that can be related to differences in mobility and the loading regime of the shoulder. We hypothesize that gorillas and chimpanzees will have a similar scapular morphology, geared towards stability and weight‐bearing in knuckle‐walking, whilst the scapular morphology of orangutans is expected to be more similar to that of humans given their high glenohumeral mobility associated with their suspensory lifestyle.
2. MATERIALS AND METHODS
2.1. Specimens details
Our dataset includes 69 bone meshes of the scapula of the four hominid genera: Pongo (n = 13), Gorilla (n = 16), Pan (n = 20), and Homo (n = 20) (Table 1, and the Supplementary Table S1). Seven scapulae were obtained via computed tomography (CT) scanning of intact great ape cadavers at the faculty of Veterinary Medicine, Ghent University using a Toshiba Aquilion ONE TSX‐301C scanner (Toshiba Medical Systems Corporation). These cadaver specimens were obtained via collaborating zoos, no animals were sacrificed for research. The remaining great ape data were obtained via online image repositories (KUPRI and morphosource.org; n = 40) and via collaboration with European zoos and museums (n = 2). The human dataset was obtained from 20 volunteers, 14 from the pseudonymized database of the University Hospital Ghent (Institutional review board approval was received from the Medical Ethics Committee of University Hospital Ghent), and another six from the morphosource.org online database. All great ape and human scapulae were obtained from adult subjects without radiological signs of musculoskeletal pathology. Only one shoulder per subject was used for further analysis, the left or right shoulder was chosen randomly.
TABLE 1.
Overview of the sample composition.
| Genus | Species | Sex | L/R side | Origin |
|---|---|---|---|---|
|
Gorilla n = 16 |
G. gorilla n = 7 G. beringei n = 8 Unknown n = 1 |
Male n = 10 Female n = 5 Unknown n = 1 |
Left n = 11 Right n = 5 |
Wild n = 13 Captive n= 3 |
|
Homo n = 20 |
n/a |
Male n = 9 Female n = 10 Unknown n = 1 |
Left n = 12 Right n = 8 |
n/a |
|
Pan n = 20 |
P. paniscus n = 3 P. troglodytes n= 17 |
Male n = 10 Female n = 9 Unknown n = 1 |
Left n = 15 Right n = 5 |
Wild n = 8 Captive n = 12 |
|
Pongo n = 13 |
P. abelii n = 4 P. pygmaeus n = 8 Unknown n = 1 |
Male n = 5 Female n = 7 Unknown n = 1 |
Left n = 12 Right n = 1 |
Wild n = 9 Captive n = 4 |
2.2. Image segmentation and 3D measurements
The CT scan images derived from the intact great ape cadavers (n = 9) and online repository (KUPRI, n = 10) were segmented manually using dedicated image processing software (Mimics Innovation Suite). We segmented the scapular bone (using thresholding, region growing, and mask editing) and made a 3D mesh reconstruction of the scapula. The human dataset (n = 20) and the remaining great ape data from other online repositories (n = 30) were already available as 3D surface meshes and no further image processing was required.
All measurements were done on the 3D surface meshes of the scapula using Mimics Innovation Suite® and 3‐Matic® software (Materialise). The protocol consisted of defining a set of planes, subsequently used to define bony landmarks (Figure 1). These landmarks allowed us to measure seven 3D angles (Figure 2) that characterize the scapular geometry.
FIGURE 1.
Three‐dimensional mesh of a left human scapula with an indication of the bony landmarks: Point zero (PZ), inferior point glenoid (IG), most anterior point acromion (AA), most anterior point coracoid (AC), most lateral point acromion (LA), most lateral point coracoid (LC), most posterior point acromion (PA), most inferior point scapula (IS), most medial point scapula (MS). The red area indicates the inferior glenoid plane, the green area indicates the scapular plane, the purple area indicates the lateral coracoid plane and the orange area indicates the posterior acromial plane. (a) Lateral view, (b) Posterior view.
FIGURE 2.
Measurements of the three‐dimensional angles (left human scapula is shown). (a) The acromion‐glenoid angle (AGA), representing the acromial overhang; (b) the delto‐fulcral triangle (DFT), with its angles alpha, beta and delta, which characterizes the coracoacromial complex; (c) the coracoid‐glenoid centre‐posterior acromial angle (CGA) with its anterior tilt (AT CGA), representing the anterior tilt of the coracoacromial complex, and its posterior tilt (PT CGA), representing the posterior rotation of the coracoacromial complex.
The protocol consisted in defining a best‐fit circle at the inferior aspect of the glenoid following the validated method of Jacxsens et al. (2016) and the centre of the circle was indicated as point zero (PZ). The inferior rim was chosen as a representation of the glenoid because in humans the inferior glenoid plane has the least retroversion variability (De Wilde et al., 2010; Verstraeten et al., 2013). This inferior glenoid plane was defined using three points: glenoid point 1 (GP1), glenoid point 2 (GP2) and glenoid point 3 (GP3) positioned at the intersecting of the bony surface of the 3D bone mesh and the best‐fit circle. After defining the inferior glenoid plane, a set of anatomical landmarks and planes were defined on the 3D mesh of the scapula which is summarized in Figure 1 and Table 2.
TABLE 2.
Overview of the landmark points and planes used in the image‐based protocol.
| Landmark | Description |
|---|---|
| GP1 | Glenoid point 1 |
| GP2 | Glenoid point 2 |
| GP3 | Glenoid point 3 |
| PZ | Point zero, centre of the best‐fit circle of the inferior glenoid rim |
| MS | Most medial point of trigonum scapulae |
| IS | Most inferior point of the scapula, tangent to inferior glenoid plane |
| IG | Located on the best‐fit circle, defined by the intersection of the glenoid plane and scapular plane |
| AA | Most anterior point of the acromion, tangent to scapular plan |
| LA | Most lateral point of the acromion, tangent to inferior glenoid plane |
| PA | Most posterior point of the acromion, tangent to scapular plane (In a few cases the PA was not defined by a tangent plane due to anatomical variations of the scapula in great apes. Here the PA was repositioned after a visual correction towards its more anatomical location of the most posterior point of the acromion.) |
| AC | Most anterior point of the coracoid, tangent to scapular plane |
| LC | Most lateral point of the coracoid, tangent to inferior glenoid plane |
| Planes | Description |
|---|---|
| Inferior glenoid plane | Defined by GP1, GP2, GP3 |
| Scapular plane | Defined by PZ, MS, IS |
| Lateral coracoid plane | Defined by LC, PZ, MS |
| Posterior acromial plane | Defined by PA, PZ, MS |
These bony landmarks and planes were subsequently used to measure a set of angles that characterize the 3D geometry of the scapula. According to a protocol designed to quantify scapular geometry in healthy volunteers and patients, yet without including pathological scapulae directly in the study, we measured the acromion‐glenoid angle (AGA), the DFT, the coracoid‐glenoid centre‐posterior acromial angle (CGA), the anterior tilt of the CGA (AT CGA), and the posterior tilt of the CGA (PT CGA) (Naidoo et al., 2017; Van Parys et al., 2021) (Figure 2).
The AGA (Figure 2a) was measured between two lines (line LA‐PZ and IG‐PZ) and represented the acromial overhang. The DFT characterizes the subacromial space and is formed by three angles (Figure 2b): alpha, beta, and delta. The CGA (Figure 2c) is the angle between the lateral coracoid plane and the posterior acromial plane. The orientation of the coracoacromial complex was defined using two additional angles: the AT CGA and the PT CGA. The AT CGA (Figure 2d) is the angle between the lateral coracoid plane and the scapular plane, whereas the PT CGA (Figure 2e) is formed by the scapular plane and the posterior acromial plane.
2.3. Statistical analysis
All statistical analyses were done in R software (version 4.0.2). A Shapiro–Wilk test was used to test the normality of the data, and a Kruskal–Wallis test was used for variables that did not follow a normal distribution. A one‐way analysis of variance (ANOVA) was used to determine the statistical significance of the genus for each angle. A Levene's test was used to evaluate the homogeneity of variance and a non‐parametric Whelch ANOVA was used for variables with unequal variances. When the ANOVA showed a significant result, Tukey's post hoc analysis was done to assess which genera differed significantly from another. A p‐value <0.05 was considered statistically significant. The measurements on the great ape scapulae were performed independently by three observers following the same protocol. The interclass correlation coefficient (ICC) for three observers was calculated to quantify the inter‐observer repeatability of the image‐based protocol.
3. RESULTS
The ICC of the great ape measurements was 0.967 showing excellent agreement between the three independent observers. Therefore, only the measurements taken by one observer (V.V.) on the entire dataset were used in the comparative analysis. The results of all angles from each scapula are provided in the Supplementary Material (Table S3).
The values of the AGA, the DFT, the CGA, the AT CGA, and the PT CGA of each genus are shown as boxplots in Figure 3, and are listed in Supplementary Table S2. We also tested the differences between species within a genus, but these were only significant for a couple of angles in the genus Pan and Gorilla. The results of these interspecific analyses are provided in the Supplementary Material (Figures S1 and S2).
FIGURE 3.
Boxplots of each angle visualizing the differences between the genera: Pongo (red), gorilla (green), pan (blue), and homo (purple). Level of significance is indicated using asterisks ( **p < 0.01; ***p < 0.001). Indicated are the mean (circle), median (horizontal line) and standard deviation (whiskers).
For the AGA (Figure 4), the greatest mean value can be found in Gorilla (68.8° ± 5.9°) whilst Homo shows the lowest mean value (59.1° ± 6.4°), Pan and Pongo show similar means (61.5° ± 7.7° and 64.1° ± 6.0°, respectively). The post hoc Tukey test shows a significant difference between Gorilla and Homo (p < 0.001) and between Gorilla and Pan (p = 0.008).
FIGURE 4.
Overview of the mean AGA of each genus on a frontal view of the left scapula (ventral view). Red (a) Pongo, green (b) Gorilla, blue (c) Pan, and purple (d) Homo. Gorilla, a knucklewalker, shows the greatest value, compared to both Homo and Pan. One representative three‐dimensional mesh of each genus is used.
The DFT is defined by angles alpha, beta, and gamma. For alpha, Pan shows the greatest mean value (37.6° ± 6.2°) whilst Pongo, Gorilla, and Homo show similar means (29.6° ± 4.5°, 33.6° ± 2.2° and 32.7° ± 4.5°, respectively). For alpha, the Tukey test shows a significant difference between Pan and Homo (p = 0.008) and between Pan and Pongo (p < 0.001). Furthermore, the alpha angle is significantly smaller in P. paniscus (30.7° ± 9.6°) compared to P. troglodytes (38.8° ± 4.8°; p = 0.03) (Figure S1).
For beta, the highest mean value (101.6° ± 6.2°) is found in Homo, followed by the mean value of Pongo (94.7° ± 4.2°) whilst Gorilla and Pan show similar mean values (90.0° ± 3.7° and 87.9° ± 5.0°). The Tukey test shows a significant difference between Homo and all the great apes (p < 0.001 for Gorilla and Pan, p = 0.002 for Pongo) and between Pongo and Pan (p = 0.002). Also for beta, a significant difference is found between P. paniscus (93.9° ± 4.6°) and P. troglodytes (86.9° ± 4.4°; p = 0.02) (Figure S1).
For delta, the lowest mean value is found in Homo (45.7° ± 3.9°) whilst Pan, Pongo, and Gorilla show similar means (54.5° ± 3.9°, 55.7° ± 3.7° and 56.4° ± 4.0°). Homo is significantly different from all the great apes (p < 0.001).
For the CGA (Figure 5), Pongo shows the lowest mean value (92.9° ± 11.4°) compared to the means of Homo, Gorilla and Pan (110.1° ± 8.3°, 114.9° ± 8.8° and 117.1° ± 8.3°). Pongo is significantly different from Homo, Gorilla, and Pan (p < 0.001). Additionally, for GCA a significant difference is found between G. gorilla (120.8° ± 8.4°) and G. beringei (110.2° ± 6.2°; p = 0.01) (Figure S2). In addition, the anterior and posterior tilt of the CGA was measured and compared between the genera. For the anterior tilt (AT CGA), the greatest mean value is found in Pan (67.2° ± 5.6°) whilst Pongo (44.6° ± 9.7°), Homo (48.9° ± 7.4°), and Gorilla (49.7° ± 11.8°) show similar means, with a significant difference between G. gorilla (57.4° ± 10.9°) and G. beringei (43.6° ± 8.9°; p = 0.01) (Figure S2). Pan is significantly different from the other genera (p < 0.001). In contrast, the posterior tilt (PT CGA) shows similar mean values in Pan (50.0° ± 7.1°) and Pongo (48.3° ± 8.6°), which are both significantly different from Homo (61.2° ± 6.8°) and Gorilla (65.1° ± 8.4°) (p < 0.001). This indicates that in Pan the coracoacromial complex is more anteriorly oriented relative to the scapular plane, whilst Gorilla and Homo show a posterior tilt of the coracoacromial complex. The coracoacromial complex of Pongo did not show a prominent anterior or posterior tilt.
FIGURE 5.
Overview of the average coracoid‐glenoid centre‐posterior acromial angle (CGA) and its anterior tilt (AT CGA) of each genus on a sagittal view of the right scapula. Pongo, which has a predominantly suspensory lifestyle, shows the smallest CGA, compared to Pan, Gorilla and Homo. Pan is the only genus showing a predominant anterior tilt of the coracoacromial complex. A representative three‐dimensional mesh of each genus is shown. Red (a) Pongo, green (b) Gorilla, blue (c) Pan, and purple (d) Homo.
4. DISCUSSION
In this study, we have quantified the 3D scapular morphology in humans and great apes using an image‐based methodology and identified differences between the hominid genera which we tried to correlate to differences in forelimb use and locomotion. We quantified acromial overhang, subacromial space, and coracoacromial space, but did not find an overall dichotomy in scapular morphology geared towards high stability (Pan/Gorilla) or high mobility (Homo/Pongo). Instead, for each morphological trait, a different clustering of genera was found. For some traits, Homo was most similar to Pongo (e.g., DFT alpha), whilst for other traits Homo was similar to Pan (e.g. AGA, CGA) and/or Gorilla (e.g., CGA). Also, despite the similarities in locomotor behaviour, the scapular morphology of Pan and Gorilla differed significantly in several aspects and we also found some inter‐specific differences within the genus Pan and Gorilla that cannot directly be linked to locomotor differences.
For the acromial overhang, indicated by AGA angle, we found that Gorilla has the largest acromial overhang, whilst the smallest acromial overhang is found in Homo. This is in contrast with the observations of Voisin et al. (2014) who reported an equally large overhang in both Homo and Gorilla. Different measurement techniques could explain the difference in results. We also took the orientation of the glenoid cavity into account when measuring the AGA. The glenoid is laterally oriented in Homo and more cranially in Gorilla (Voisin et al., 2014). This cranially‐oriented glenoid, in combination with a large acromial overhang, might lead to an AGA angle which is significantly larger in Gorilla than in Homo. The large acromial overhang might give protection to the large humeral head of Gorilla and provide greater stability. In addition, it could be related to a larger attachment site for the deltoid muscle and accordingly allows a greater force‐generating capacity (van Beesel et al., 2021). We do indeed see that the very muscular Gorilla is also significantly different from Pan despite their similar knuckle‐walking behaviour. In contrast, Pongo and Pan show similar values despite displaying distinct locomotor behaviour. Additional studies are needed to substantiate the functional implications of these morphological differences.
Our results show that the DFT, representing the subacromial space, is significantly different between humans and great apes. The dimensions of the subacromial space were also quantified by Potau et al. (2007), but they found that the subacromial space only differed significantly between humans and great apes when comparing absolute dimensions. When scaling for body mass, the subacromial space dimensions were no longer significantly different. In our study, we used angles rather than linear measurements, avoiding the need for scaling. So whilst the subacromial space might not differ in relative dimensions between humans and great apes, we demonstrate that its shape is significantly different, with a lower space in humans and a higher space in great apes.
As hypothesized, the DFT of Pongo is most similar to that of Homo and is characterized by a relatively low subacromial space with a small alpha and large beta angle. Previous literature showed that a narrowed subacromial space in modern humans, combined with other morphological features, probably contributes to a high prevalence of degenerative rotator cuff tears (RCT; Lee et al., 2020; Voisin et al., 2014). The observation that the highly suspensory orangutan has an equally low subacromial space, questions the relation between subacromial space and RCT.
Pan showed the highest subacromial space, with a large alpha and small beta angle. Although Pan and Gorilla are both knuckle‐walkers, Gorilla showed an in‐between configuration with the same small beta angle as found in Pan but a lesser alpha angle as shown in Homo. When comparing the geometry of the DFT, it must be noted that in all great apes the apex of the subacromial roof (beta angle) is oriented more posteriorly than in humans.
The width of the coracoacromial complex (CGA) was significantly smaller in Pongo than Gorilla, Pan, and Homo, all showing similar dimensions. We assume that this CGA is negatively correlated to humeral mobility in the sagittal plane, as the highest humeral mobility is found in orangutans amongst hominids (Zihlman et al., 2011).
In addition to variation in the angles discussed above, there is also an important variation in the tilt of the coracoid/acromion in humans (Lee et al., 2020). Therefore, we also examined the anterior and posterior tilt of the coracoacromial complex in this study (AT and PT CGA). We found that Pan has a significantly greater anterior tilt of the coracoacromial complex compared to Pongo, Gorilla and Homo, which all show similar values of the AT CGA angle. Gorilla showed the largest posterior tilt of the coracoacromial complex, being significantly different from Pongo and Pan but not from Homo. Thus, Pan demonstrates a prominent anterior tilt of the coracoacromial complex, whilst Gorilla and Homo display a posterior tilted complex. Pongo showed similar values for AT CGA and PT CGA angles, indicating that the coracoacromial complex is not tilted anteriorly nor posteriorly.
The 3D measurements (i.e. AGA, DFT, CGA, AT and PT CGA) used in this study are derived from the orthopaedic practice and were previously shown to differ significantly between the scapula of healthy humans and of patients with shoulder pathologies, such as degenerative RCT (Moor et al., 2013; Naidoo et al., 2017; Van Parys et al., 2021). Whilst great apes do possess a true rotator cuff tendon (Sonnabend & Young, 2009), degenerative RCT has not been observed (Craik et al., 2014; Potau et al., 2007). However, it remains unclear whether degenerative RCT pathology does not develop in great apes or if it just remains undetected or unreported. Detection is only possible in great apes living in captivity but even their RCT pathology could be overlooked as this is not something that is routinely checked. Other possible reason for the absence of RCT in great apes could be the difference in age expectancy, however, degenerative changes in the scapula have been described in nonhuman primates (e.g. vervet monkeys; Plate et al., 2013, chimpanzees, pers. obs.).
In humans, it has been found that the critical shoulder angle, which combines measurements of glenoid inclination and acromial overhang, can be used to differentiate healthy volunteers from patients with RCT and osteoarthritis (Moor et al., 2013). A greater acromial overhang was found in patients with degenerative RCTs, whilst patients with glenohumeral osteoarthritis had less acromial overhang compared to the healthy control group (Moor et al., 2013). The acromial overhang was also a distinguishing parameter in our hominid sample, with Gorilla displaying the largest acromial overhang. However, here there does not seem to be a link with pathology but rather with relative development of the deltoid muscle.
The DFT, which is a measure of subacromial space, also differs significantly between healthy volunteers and patients with RCT (Naidoo et al., 2017), with patients displaying a more narrow subacromial space. A narrow subacromial space results in a reduced area for tendon movement, which can eventually cause more friction on the tendons and an increased risk of pathology, such as RCTs. In our study, Pongo was found to have a low subacromial space, comparable to that found in humans, whilst Gorilla but especially Pan is characterized by a high subacromial space. Given the absence of reported cases of RCT in hominids, there does not seem to be a link with pathology. However, it would be interesting to further investigate the functional implications of a low subacromial space as Pongo combines this with high glenohumeral mobility.
To investigate the effect of specific morphological traits on shoulder mechanics we need to examine the shoulder girdle in its entirety. The shoulder musculature, such as the transverse force couple, as well as the kinematics of the shoulder girdle during different activities, should be included. In addition, future research should include the clavicle and humerus along with the scapula to assess the variation in the 3D configuration of these three bones across hominids. Together this would allow us to obtain a better understanding of morphological correlates of shoulder pathology in humans.
Whilst this study includes a unique and highly valuable dataset of 49 great ape scapulae, the sample size remains small and heterogeneous, including captive and wild apes of different ages. We know that, in humans, the glenoid version is affected by physical activity and it would be reasonable to assume the same holds true for nonhuman primates. To evaluate this effect on scapular morphology we would need a large sample size consisting of a similar number of captive and wild specimens in each genus. Previous analyses investigating the effect of captivity on scapular shape have, however, concluded that scapulae from captive hominoid primates cannot be distinguished from those of wild‐caught hominoids (Bello‐Hellegouarch et al., 2013). Furthermore, age could also affect some of our parameters (Plate et al., 2013), but this effect has been largely avoided by only including adult specimens as controlling for age was not possible. In our sample, we randomly selected the left or right side, as handedness was unknown for all specimens. In future analyses, it might be interesting to investigate if the angles that quantify 3D scapular morphology differ between the dominant and non‐dominant sides in humans. Such analysis is less relevant for great apes as a left‐ or right‐handed dominance on species‐ or genus‐level is not observed.
5. CONCLUSIONS
Using our medical image‐based methodology, we were able to quantify the 3D morphology of the scapula. This allowed us to identify morphological features which are significantly different between humans and great apes but also pointed to differences between great ape genera and species. Further research is needed to conclude the functional implications of these differences in scapular morphology and the development of degenerative RCT in humans.
AUTHOR CONTRIBUTIONS
Valérie Vermeulen, Evie E. Vereecke, Alexander Van Tongel and Lieven De Wilde made contributions to the concept and design of the study; Valérie Vermeulen, Arne Delsupehe, Elaine Kozma, Pieter Cornillie and Emmelie Stock were involved in data acquisition; Valérie Vermeulen, Evie E. Vereecke, Alexander Van Tongel and Lieven De Wilde took part in data analysis and interpretation, participated in the drafting of the manuscript and made a critical revision of the manuscript; all authors approved the final version of the manuscript.
Supporting information
Table S1.
Table S2
Table S3
Figure S1
Figure S2
ACKNOWLEDGMENTS
The authors would like to express their appreciation to Jimmy Saunders for his authorisation to scan the great ape cadavers at the faculty of Veterinary Medicine, University of Ghent. In addition, we would like to thank the collaborating zoos (Dierenpark Amersfoort, the Netherlands; Royal Zoological Society Antwerp, Belgium; Pairi Daiza, Belgium; ZooParc de Beauval, France) for providing access to their deceased primates or to CT scan images of their primates. Particular thanks go to Dr. Emmanuel Gilissen from the Royal Museum for Central Africa, Belgium, for providing access to the invaluable collection of primates of the institute.
Vermeulen, V. , Kozma, E. , Delsupehe, A. , Cornillie, P. , Stock, E. & Van Tongel, A. et al. (2023) Scapular morphology of great apes and humans: A three‐dimensional computed tomography‐based comparative study. Journal of Anatomy, 242, 164–173. Available from: 10.1111/joa.13784
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- Astúa, D. (2009) Evolution of scapula size and shape in didelphid marsupials (Didelphimorphia: Didelphidae). Evolution, 63(9), 2438–2456. Available from: 10.1111/j.1558-5646.2009.00720.x [DOI] [PubMed] [Google Scholar]
- Bello‐Hellegouarch, G. , Potau, J.M. , Arias‐Martorell, J. , Pastor, J.F. & Pérez‐Pérez, A. (2013) Brief communication: morphological effects of captivity: a geometric morphometric analysis of the dorsal side of the scapula in captive‐bred and wild‐caught hominoidea. American Journal of Physical Anthropology, 152(2), 306–310. Available from: 10.1002/AJPA.22356 [DOI] [PubMed] [Google Scholar]
- Corruccini, R.S. & Ciochon, R.L. (1976) Morphometric affinities of the human shoulder. American Journal of Physical Anthropology, 45(1), 19–37. Available from: 10.1002/AJPA.1330450104 [DOI] [PubMed] [Google Scholar]
- Craik, J.D. , Mallina, R. , Ramasamy, V. & Little, N.J. (2014) Human evolution and tears of the rotator cuff. International Orthopaedics, 38(3), 547–552. Available from: 10.1007/s00264-013-2204-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- De Wilde, L.F. , Verstraeten, T. , Speeckaert, W. & Karelse, A. (2010) Reliability of the glenoid plane. Journal of Shoulder and Elbow Surgery, 19(3), 414–422. Available from: 10.1016/J.JSE.2009.10.005 [DOI] [PubMed] [Google Scholar]
- Doyle, W.J. , Siegel, M.I. & Kimes, K.R. (1980) Scapular correlates of muscle morphology in Macaca mulatta . Cells, Tissues, Organs, 106(4), 493–501. Available from: 10.1159/000145219 [DOI] [PubMed] [Google Scholar]
- Green, D.J. (2013) Ontogeny of the hominoid scapula: the influence of locomotion on morphology. American Journal of Physical Anthropology, 152(2), 239–260. Available from: 10.1002/AJPA.22353 [DOI] [PubMed] [Google Scholar]
- Green, D.J. , Serrins, J.D. , Seitelman, B. , Martiny, A.R. & Gunz, P. (2015) Geometric morphometrics of hominoid infraspinous fossa shape. The Anatomical Record, 298(1), 180–194. Available from: 10.1002/AR.23071 [DOI] [PubMed] [Google Scholar]
- Jacxsens, M. , Van Tongel, A. , Henninger, H.B. , De Coninck, B. , Mueller, A.M. & De Wilde, L. (2016) A three‐dimensional comparative study on the scapulohumeral relationship in normal and osteoarthritic shoulders. Journal of Shoulder and Elbow Surgery, 25(10), 1607–1615. Available from: 10.1016/J.JSE.2016.02.035 [DOI] [PubMed] [Google Scholar]
- Lee, E.C.S. , Roach, N.T. , Clouthier, A.L. , Bicknell, R.T. , Bey, M.J. , Young, N.M. et al. (2020) Three‐dimensional scapular morphology is associated with rotator cuff tears and alters the abduction moment arm of the supraspinatus. Clinical biomechanics, 78(January), 105091. Available from: 10.1016/j.clinbiomech.2020.105091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matsuo, T. , Morita, F. , Tani, D. , Nakamura, H. , Higurashi, Y. , Ohgi, J. et al. (2019) Anatomical variation of habitat‐related changes in scapular morphology. Journal of Veterinary Medicine Series C: Anatomia Histologia Embryologia, 48(3), 218–227. Available from: 10.1111/AHE.12426 [DOI] [PubMed] [Google Scholar]
- Monteiro, L.R. & Abe, A.S. (1999) Functional and historical determinants of shape in the scapula of Xenarthran mammals: evolution of a complex morphological structure. Journal of Morphology, 241, 251–263. Available from: 10.1002/(SICI)1097-4687(199909)241:3 [DOI] [PubMed] [Google Scholar]
- Moor, B.K. , Bouaicha, S. , Rothenfluh, D.A. , Sukthankar, A. & Gerber, C. (2013) Is there an association between the individual anatomy of the scapula and the development of rotator cuff tears or osteoarthritis of the glenohumeral joint? a radiological study of the critical shoulder angle. The Bone & Joint Journal, 95‐B(7), 935–941. Available from: 10.1302/0301-620X.95B7.31028 [DOI] [PubMed] [Google Scholar]
- Naidoo, N. , Lazarus, L. , Satyapal, K.S. , De Wilde, L. & Van Tongel, A. (2017) The morphometric anatomy of the delto‐fulcral triangle: a 3D CT‐based reconstruction study. Journal of Orthopaedics, 14(1), 62–67. Available from: 10.1016/j.jor.2016.10.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Oxnard, C.E. (1968) The architecture of the shoulder in some mammals. Journal of Morphology, 126(3), 249–290. Available from: 10.1002/JMOR.1051260302 [DOI] [PubMed] [Google Scholar]
- Plate, J.F. , Bates, C.M. , Mannava, S. , Smith, T.L. , Jorgensen, M.J. , Register, T.C. et al. (2013) Age‐related degenerative functional, radiographic, and histological changes of the shoulder in nonhuman primates. Journal of Shoulder and Elbow Surgery, 22(8), 1019–1029. Available from: 10.1016/j.jse.2012.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Potau, J.M. , Bardina, X. & Ciurana, N. (2007) Subacromial space in African great apes and subacromial impingement syndrome in humans. International Journal of Primatology, 28(4), 865–880. Available from: 10.1007/s10764-007-9167-z [DOI] [Google Scholar]
- Preuschoft, H. , Hohn, B. , Scherf, H. , Schmidt, M. , Krause, C. & Witzel, U. (2010) Functional analysis of the primate shoulder. International Journal of Primatology, 31(2), 301–320. Available from: 10.1007/S10764-010-9399-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Püschel, T.A. & Sellers, W.I. (2016) Standing on the shoulders of apes: analyzing the form and function of the hominoid scapula using geometric morphometrics and finite element analysis. American Journal of Physical Anthropology, 159(2), 325–341. Available from: 10.1002/AJPA.22882 [DOI] [PubMed] [Google Scholar]
- Schmidt, M. (2008) Forelimb proportions and kinematics: how are small primates different from other small mammals? Journal of Experimental Biology, 211(24), 3775–3789. Available from: 10.1242/JEB.019802 [DOI] [PubMed] [Google Scholar]
- Selby, M.S. & Lovejoy, C.O. (2017) Evolution of the hominoid scapula and its implications for earliest hominid locomotion. American Journal of Physical Anthropology, 162(4), 682–700. Available from: 10.1002/ajpa.23158 [DOI] [PubMed] [Google Scholar]
- Shea, B.T. (1986) Scapula form and locomotion in chimpanzee evolution. American Journal of Physical Anthropology, 70(4), 475–488. Available from: 10.1002/AJPA.1330700407 [DOI] [Google Scholar]
- Sonnabend, D.H. & Young, A.A. (2009) Comparative anatomy of the rotator cuff. Journal of Bone and Joint Surgery ‐ Series B, 91(12), 1631–1637. Available from: 10.1302/0301-620X.91B12.22370 [DOI] [PubMed] [Google Scholar]
- Taylor, A.B. (1997) Scapula form and biomechanics in gorillas. Journal of Human Evolution, 33(5), 529–553. Available from: 10.1006/JHEV.1997.0147 [DOI] [PubMed] [Google Scholar]
- van Beesel, J. , Hutchinson, J.R. , Hublin, J. & Melillo, S.M. (2021) Exploring the functional morphology of the gorilla shoulder through musculoskeletal modelling. Journal of Anatomy, 239(1), 207–227. Available from: 10.1111/joa.13412 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Parys, M. , Alkiar, O. , Naidoo, N. , Van Tongel, A. & De Wilde, L. (2021) Three‐dimensional evaluation of scapular morphology in primary glenohumeral arthritis, rotator cuff arthropathy, and asymptomatic shoulders. Journal of Shoulder and Elbow Surgery, 30(8), 1803–1810. Available from: 10.1016/J.JSE.2020.10.027 [DOI] [PubMed] [Google Scholar]
- Verstraeten, T.R.G.M. , Deschepper, E. , Jacxsens, M. , Walravens, S. , De Coninck, B. , Pouliart, N. et al. (2013) Determination of a reference system for the three‐dimensional study of the glenohumeral relationship. Skeletal Radiology, 42, 1061–1071. Available from: 10.1007/s00256-013-1572-0 [DOI] [PubMed] [Google Scholar]
- Voisin, J.L. , Ropars, M. & Thomazeau, H. (2014) The human acromion viewed from an evolutionary perspective. Orthopaedics and Traumatology: Surgery and Research, 100(8), S355–S360. Available from: 10.1016/j.otsr.2014.09.011 [DOI] [PubMed] [Google Scholar]
- Young, N.M. (2006) Function, ontogeny and canalization of shape variance in the primate scapula. Journal of Anatomy, 209(5), 623–636. Available from: 10.1111/j.1469-7580.2006.00639.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Young, N.M. (2008) A comparison of the ontogeny of shape variation in the anthropoid scapula: functional and phylogenetic signal. American Journal of Physical Anthropology, 136(3), 247–264. Available from: 10.1002/AJPA.20799 [DOI] [PubMed] [Google Scholar]
- Zhang, K.Y. , Wiktorowicz‐Conroy, A. , Hutchinson, J.R. , Doube, M. , Klosowski, M. , Shefelbine, S.J. et al. (2012) 3D morphometric and posture study of felid scapulae using statistical shape modelling. PLoS One, 7(4), e34619. Available from: 10.1371/JOURNAL.PONE.0034619 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zihlman, A.L. , McFarland, R.K. & Underwood, C.E. (2011) Functional anatomy and adaptation of male gorillas (Gorilla gorilla gorilla) with comparison to male orangutans (Pongo pygmaeus). The Anatomical Record: Advances in Integrative Anatomy and Evolutionary Biology, 294(11), 1842–1855. Available from: 10.1002/AR.21449 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1.
Table S2
Table S3
Figure S1
Figure S2
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.