- Email: [email protected]
Coupling between 3D displacements and rotations at the glenohumeral joint during dynamic tasks in healthy participants
Clinical Biomechanics 29 (2014) 1048–1055
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Coupling between 3D displacements and rotations at the glenohumeral joint during dynamic tasks in healthy participants Fabien Dal Maso a,⁎, Maxime Raison b, Arne Lundberg c, Anton Arndt c,d, Mickaël Begon a a
Laboratoire de simulation et de modélisation du mouvement, Département de kinésiologie, Université de Montréal, 1700, rue Jacques Tétreault, Laval, QC H7N 0B6, Canada École Polytechnique de Montréal, 6079, Succursale, Centre Ville, Montréal, QC H3C 3A7, Canada Karolinska Institute, Stockholm, Sweden d The Swedish School of Sport and Health Sciences, Stockholm, Sweden b c
a r t i c l e
i n f o
Article history: Received 21 January 2014 Accepted 4 August 2014 Keywords: Intracortical pin Shoulder prosthesis Kinematic model 3D kinematics CT-scan
a b s t r a c t Background: Glenohumeral displacements assessment would help to design shoulder prostheses with physiological arthrokinematics and to establish more biofidelic musculoskeletal models. Though displacements were documented during static tasks, there is little information on their 3D coupling with glenohumeral angle during dynamic tasks. Our objective was to characterize the 3D glenohumeral displacement–rotation couplings during dynamic arm elevations and rotations. Methods: Glenohumeral displacements were measured from trajectories of reflective markers fitted on intracortical pins inserted into the scapula and humerus. Bone geometry was recorded using CT-scan. Only four participants were recruited to the experiment due to its invasiveness. Participants performed dynamic arm abduction, flexion and axial rotations. Linear regressions were performed between glenohumeral displacements and rotations. The pin of the scapula of one participant moved, his data were removed from analysis, and results are based on three participants. Findings: The measurement error of glenohumeral kinematics was less than 0.15 mm and 0.2°. Maximum glenohumeral displacements were measured along the longitudinal direction and reached up to + 12.4 mm for one participant. Significant couplings were reported especially between longitudinal displacement and rotation in abduction (adjusted R2 up to 0.94). Interpretation: The proposed method provides the potential to investigate glenohumeral kinematics during all kinds of movements. A linear increase of upward displacement during dynamic arm elevation was measured, which contrasts with results based on a series of static poses. The systematic investigation of glenohumeral displacements under dynamic condition may help to provide relevant recommendation for the design of shoulder prosthetic components and musculoskeletal models. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Though the glenohumeral joint is often simplified to a ball-andsocket in human movement analysis (Jackson et al., 2012; Wu et al., 2005), the humeral head undergoes displacements relative to the glenoid cavity, generally termed as glenohumeral translations (although translation should refer to the linear movement humerus with respect to the scapula along the helical axis). Displacements along the antero-posterior and longitudinal axes have been reported in in-vitro (Bryce et al., 2010; San Juan and Karduna, 2010; Sharkey and Marder, 1995) and in-vivo studies for both healthy subjects (Bey et al., 2008; Graichen et al., 2005; Massimini et al., 2012) and patients
⁎ Corresponding author at: Université de Montréal, 1700, rue Jacques Tétreault, Laval, QC H7N 0B6, Canada. E-mail address: [email protected] (F. Dal Maso).
http://dx.doi.org/10.1016/j.clinbiomech.2014.08.006 0268-0033/© 2014 Elsevier Ltd. All rights reserved.
(Bishop et al., 2009). Glenohumeral displacements of only a few millimeters are responsible for the loss of joint stability, which makes the glenohumeral joint potentially the most unstable joint in the body (Chan et al., 1996). The active co-contraction of rotator cuff and adductor muscles counteracts upward shear forces generated by deltoid muscles to reduce upward displacement (Graichen et al., 2000, 2005; Sharkey and Marder, 1995; von Eisenhart-Rothe et al., 2002). Though shoulder kinematics differ between static and dynamic tasks (Bey et al., 2006; Carey et al., 2000) and glenohumeral displacements are sensitive to muscular activity patterns (Graichen et al., 2000, 2005; Sharkey and Marder, 1995; von Eisenhart-Rothe et al., 2002), their measurements were often limited to a series of static poses (Bey et al., 2008; Giphart et al., 2013). Daily living activities correspond however to dynamic tasks (Bey et al., 2006). Consequently, recent studies (Bey et al., 2006; Massimini et al., 2012) agree that “shoulder biomechanics under dynamic conditions should be investigated” (Massimini et al., 2012).
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
While glenohumeral displacements increase joint instability, they reduce the prevalence of periprosthetic radiolucency in shoulder prostheses (Walch et al., 2002) and above all homogenize the stress at the implant fixation sites (Buchler et al., 2002; Karduna et al., 1998). Understanding the normal glenohumeral kinematics is therefore fundamental in orthopedics to design shoulder prostheses (Buchler et al., 2002). In another field of application, the replication of glenohumeral displacements in musculoskeletal computer simulation models may enable a better determination of the muscle moment arms, muscular activation patterns and activation level of stabilizing muscles (Bolsterlee et al., 2013; Favre et al., 2012). It may also better determine contact area between humeral head and glenoid cavity (Buchler et al., 2002). It may thus be crucial to characterize the normal coupling between glenohumeral displacements and rotations during upper-limb dynamic tasks. During dynamic tasks, the measurement of glenohumeral displacements with high accuracy is however challenging. Glenohumeral displacements were commonly measured using MRI (Graichen et al., 2005; Massimini et al., 2012), X-ray (San Juan and Karduna, 2010; Yamaguchi et al., 2000) and biplane X-ray recordings (Bey et al., 2008; Bishop et al., 2009). Imaging provides accurate measure of bone positions but is not suitable to record fast motion and is restricted to the volume of acquisition. Moreover, due to radiation during X-ray recordings, the number of movements acquired during the experimental protocol is limited. On the other hand, optoelectronic motion analysis systems with markers positioned on the skin are not limited in sampling rate and space but have limited accuracy because of soft tissues artifacts between skin and underlying bones (Karduna et al., 2001). This represents an issue to investigate glenohumeral displacements of a few millimeters (Bey et al., 2008; Graichen et al., 2000). To overcome these limitations, bone kinematics can be measured using the trajectories of reflective markers secured on intra-cortical pins implanted into the bones (Bourne et al., 2007; Liu et al., 2012; Ludewig et al., 2009). Since glenohumeral displacements are usually expressed in a glenoid-based coordinated system (GCS) (Bey et al., 2006; Graichen et al., 2000; Massimini et al., 2012), such a method would also require the acquisition of bone geometry but offers the perspective to investigate glenohumeral kinematics with no restrictions on the type of movements and on the number of repetitions. The purpose of this study was to establish a method that permits to characterize the three-dimensional (3D) glenohumeral displacement for dynamic movements without known restrictions in space and number of trials. This method was applied to a sample of participants and the secondary purpose was to provide preliminary results of coupling between glenohumeral displacements and rotations as well as preliminary values of glenohumeral displacements close to physiological limits during dynamic arm elevations and axial rotations in healthy participants. With regard to the sensitivity of glenohumeral displacements to muscular activity, we hypothesized that new patterns of glenohumeral displacements during dynamic tasks will be highlighted.
1049
hereafter, the data of only three participants were used in this study. Participants obtained a score lower than 10.5 at the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire (Hudak et al., 1996), indicating that they were asymptomatic at the left shoulder and without history of pain, injury or shoulder dysfunction. The experimental protocol was approved by the concerned local ethics committees of the Karolinska Institutet (Sweden) and the University of Montreal (Canada). 2.2. Instrumentation Intracortical pins were inserted by an experienced surgeon in an operating room under normal and sterile surgery conditions. After administration of local anesthesia (AstraZeneca, Södertälje, Sweden), one stainless steel self-drilling pin, 1.6 mm of diameter (Synthes, Bettlach, Switzerland) was inserted into the left scapular spine (Fig. 1A). A second pin (2.5 mm of diameter) was placed in the lateral aspect of the left humerus just distal to the deltoid attachment (Fig. 1A). Clusters of four and five reflective markers were secured on both pins (Fig. 1A). Pin locations were determined to avoid muscles so that the normal strategies of movement control remained unchanged. Pins avoided also nerves and blood vessels and were oriented to avoid contacts between markers and head or neck of the participants during movements. The rigidity of the montage was manually checked and insertion sites were cleaned, sterilized and covered with a sterile dressing. Pins were removed at the end of the experiment, and insertion sites were cleaned and covered again with new sterile dressings. Participants were provided with antibiotic and pain relief (AstraZeneca, Södertälje, Sweden) medication. No clinical complications occurred following the experiment. Six, nine and seven reflective skin markers were placed on the thorax, scapula and humerus, respectively (Jackson et al., 2012). Trajectories of pin and skin markers were collected at 300 Hz using an 18-camera VICON™ optoelectronic motion analysis system (Oxford Metrics Ltd., Oxford, UK). A CT-scan (General Electric, Milwaukee, USA) was used to obtain the 3D geometry of the scapula and humerus and their respective pin (Fig. 1B). For each participant, 226 tomographic slices (thickness: 0.61 mm; interslice space: 0.32 mm) were recorded. The field of view was set to 250 × 250 mm, defined by a matrix of 512 × 512 pixels. X-ray tube was adjusted at 120 kV and 110 μA, which corresponds to the usual settings (Meskers et al., 1997) to maximize contrast between cortical bone and soft tissues of living patients and also between markers and air. 2.3. Procedures
Pins were drilled into the bones of the participants to measure glenohumeral kinematics with high accuracy during dynamic tasks. Moreover, participants were exposed to radiation via computedtomography scanner (CT-scan). The proposed method is therefore invasive, which explains the small sample of participants.
First, a relaxed position trial and a series of setup functional movements, namely, maximum arm elevations, circumductions and arm sweeping were recorded to locate the glenohumeral joint center (Ehrig et al., 2007; Jackson et al., 2012). Secondly, 10 dynamic arm elevations in abduction and flexion with thumb pointing upward (with no specific instruction on the axial rotation of the glenohumeral and radioulnar joints) and 10 axial rotations in adducted position (0° of thoraco-humeral elevation) with elbow flexed at 90° were performed. Participant 2 performed only two trials of abduction and flexion. During each movement, participants were asked to move their arm at spontaneous speed and to reach their maximal range of motion.
2.1. Participants
2.4. Data processing
Four voluntary males (age: 27, 44, 32 and 41 years; height: 1.65, 1.77, 1.72, and 1.82 m, mass: 57, 82, 80 and 115 kg for Participants 1, 2, 3 and 4, P1, P2, P3 and P4, respectively) signed an informed consent form and participated in the study. Due to technical problems described
An automatic segmentation of gray level of the CT-scan was obtained using the Seg3D2® software (Institute) and a manual correction of the potential segmentation errors was performed to identify pin markers and bone voxels (Landry et al., 1997). Thereafter,
2. Methods
1050
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
Fig. 1. A) Picture of participant fitted with intracortical pins inserted into the clavicle (unused in this study), scapula and humerus with their respective clusters of four to five pin markers. B) Back view of the CT-scan segmentation of the scapula and humerus bones with their respective clusters of markers. C) Representation of the CT-scan segmentation of the scapula. Circles represent the anterior, inferior and posterior reference points on the glenoid rim. The X-, Y- and Z-axes represent the antero-posterior, longitudinal and medio-lateral axes, respectively of the glenoid-based coordinate system in which glenohumeral displacements are expressed.
processing was performed using homemade and built-in Matlab® functions (R2013a version; The Mathworks, Natick, USA). CT-scan slices were piled using the iso2mesh MATLAB® toolbox (Fang and Boas, 2009) and 3D reconstructions were obtained for each bone. The anterior, inferior, posterior and superior reference points of the glenoid rim (Fig. 1C) were identified by three evaluators five times each. Inter-evaluator variability was 1.2 mm. An inferior GCS was defined from the anterior, inferior and posterior reference points (De Wilde et al., 2010). A nonlinear least-squares algorithm was applied to pin marker trajectories to reduce the error which arises from the optoelectronic motion analysis system to locate the markers (Monnet et al., 2012). Functional glenohumeral center of rotation was located using the Symmetrical Center of Rotation Estimation (SCoRE) method (Ehrig et al., 2006). The GCS was translated to the position of the glenohumeral center of rotation in the relaxed position trial (Massimini et al., 2012). Glenohumeral displacements corresponded to the vector from the origin of the translated GCS to the glenohumeral center of rotation along the medio-lateral, antero-posterior and longitudinal axes. A sphere fitting algorithm was applied to CT-scan data to locate the geometrical center of the humeral head. Glenohumeral displacements were also computed using this point. Both methods were compared on the data of abduction for all participants. Scapula and humerus skin markers were used to determine local coordinated systems following ISB recommendations (Wu et al., 2005). Glenohumeral rotations were expressed according to the abduction–flexion–axial rotation angle sequence (rotation around the X-axis, rotated Z-axis, twice rotated Y-axis, respectively) (Senk and Cheze, 2006). Note that internal and external axial rotations will produce negative and positive values, respectively, since the left side of the participants was recorded. The thorax skin markers were used to compute thoraco-humeral angles. Glenohumeral displacements and rotations were low-pass filtered at 10 Hz (4th-order zero-lag Butterworth filter). Skin markers were also used to compare the orientation of the pinand skin-based local coordinate systems at the beginning of each trial and to ensure that the orientation of the pins did not change relatively to the skin-based local coordinate system throughout the experiment. For P4, the orientation of the pin-based coordinate system of the scapula revealed a drift with regard to the orientation of the skin-based coordinate system. This indicates that the pin rotated during the trials. Consequently, the data of P4 were removed. The following analysis was performed on the three remaining participants, namely, Participants 1, 2 and 3.
2.5. Estimation of the uncertainty The uncertainty of glenohumeral kinematics was determined as follows. Pin-marker position artifacts were extracted from pinmarker raw trajectories with a 10 Hz high-pass filter (fourth-order zero-lag Butterworth filter). The computed glenohumeral displacements and rotations were low-pass filtered and the reverse kinematic was applied to obtain noiseless pin-marker trajectories. Noiseless pin-marker trajectories and extracted artifacts were summed to obtain artifact-simulated trajectories. Glenohumeral displacements and rotations were then re-computed to obtain artifact-simulated glenohumeral displacements and rotations. Finally, a root-mean-square (RMS) error was computed between lowpass filtered and artifact-simulated glenohumeral displacements and rotations. Using this method, uncertainty of glenohumeral displacements and rotations was estimated at less than 0.15 mm and 0.2°, respectively.
2.6. Analysis For each participant, Pearson's linear regression analysis was used to assess the coupling between glenohumeral displacements and rotations along and around the three axes. Regressions were performed using all data points recorded during abduction, flexion and axial rotation of each participant. The significance level was set at P b 0.05/12 due to Bonferroni correction for multiple testing. The maximum values of lateral, medial, anterior, posterior, upward and downward displacements were reported for each movement. Finally, medio-lateral displacements were expressed as a percentage of the humeral head diameter. Antero-posterior displacements were expressed as a percentage of the distance between the anterior and posterior reference points of the glenoid rim and longitudinal displacements as a percentage of the distance between the inferior and superior reference points of the glenoid rim. These values were however not compared statistically due to the small sample of participants (N = 3).
3. Results The duration and the maximum range of motion of each task are detailed in Table 1 for each participant. The elevations lasted between 1.5 and 2.6 s. The maximum thoracohumeral abduction angle was close to 135° and 130° during abduction and flexion, respectively. The
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
1051
Table 1 Participant-specific values of duration and maximum range of motion for each task. n.a.: not applicable. Tasks
Participants
Execution time (s)
Maximum angle (°) Abduction
Abduction
Flexion
Axial rotation
P1 P2 P3 P1 P2 P3 P1 P2 P3
2.6 1.5 2.4 2.5 1.7 1.7 2.3 1.4 1.4
axial rotations lasted between 1.4 and 2.3 s, and the amplitude of axial rotation ranged from 87° to 93°. 3.1. Effect of the reference location of the humeral head Table S1 (supplementary material) presents the difference between the location of the geometric center of the humeral head and the glenohumeral center of rotation. The 3D Euclidean distance between these two points was 4.0, 4.2 and 3.9 mm for P1, P2 and P3, respectively. Both methods lead to similar pattern of glenohumeral displacements for all participants during arm abductions (Fig. S2, supplementary material). However, using the humeral head center, medial displacements were larger than using the glenohumeral joint center, while less differences were found along the antero-posterior and longitudinal axes. Medial displacements were up to + 4 mm with the humeral head center while it was less than + 2 mm with the glenohumeral joint center of rotation. 3.2. Participant-specific glenohumeral displacements Only regressions between the glenohumeral displacements along the three axes and the glenohumeral rotation axis with which the
Axial rotation
Glenohumeral
Thoracohumeral
Internal
External
64 67 85 81 84 92 n.a. n.a. n.a.
135 121 149 129 117 143 n.a. n.a. n.a.
n.a. n.a. n.a. n.a. n.a. n.a. −23 −22 −25
n.a. n.a. n.a. n.a. n.a. n.a. 70 65 64
adjusted R2 was the greater are presented. In average, the adjusted R2 was the greatest with the regression between the rotations around the abduction and the axial rotation axis during elevation arm axial rotation, respectively. Thereby, Fig. 2 represents the displacements along the mediolateral, antero-posterior and longitudinal axes according to the rotations around the abduction axis during abduction and flexion and according to the axial rotation axis during axial rotations. Patterns of glenohumeral displacements according to the glenohumeral rotations were similar across participants in most situations (Fig. 2). However, values of measured displacements differed between participants: e.g. longitudinal displacement during abduction (Fig. 2C). Table 2 presents the maximum measured and normalized values of lateral, medial, anterior, posterior, upward and downward glenohumeral displacements during abduction, flexion and axial rotations for each participant. Table 3 summarizes constants, coefficients, adjusted R2 and F values of the regressions between glenohumeral displacements and rotations and for each participant. Medial and lateral displacements reached up to + 2.4 mm and −2.1 mm representing 5% and 4% of the humeral head, respectively (Table 2). These amplitudes were smaller than the values of displacements reached along antero-posterior and longitudinal axes.
Fig. 2. (A) Medio-lateral, (B) antero-posterior and (C) longitudinal displacements of the glenohumeral joint according to its abduction angle during abduction and flexion (first and second columns, respectively) and to axial rotation angle during axial rotation (third column). The dashed black thin lines represent 0 mm of displacement. From top to bottom, positive values of displacements indicate that the humeral head moves medially, forward and upward relatively to the glenoid. Positive and negative values of the glenohumeral axial rotations indicate external and internal rotations, respectively.
1052
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
Table 2 Participant-specific maximum measured and normalized values of medial, lateral, anterior, posterior, upward and downward glenohumeral displacements during abduction, flexion and axial rotations. Tasks
Abduction
Flexion
Axial rotation
Participants
P1 P2 P3 P1 P2 P3 P1 P2 P3
Medial
Lateral
Anterior
Posterior
Upward
Downward
mm
%
mm
%
mm
%
mm
%
mm
%
mm
%
1.0 2.0 2.4 0.4 0.9 1.4 0.5 0.7 1.1
3 4 5 1 2 3 1 1 2
−0.8 −2.0 −1.2 −1.4 −1.5 −2.1 −1.0 −1.3 −1.2
2 4 2 3 3 4 2 2 2
1.8 2.4 1.1 1.8 2.0 0.8 2.2 2.8 0.5
7 8 4 7 7 3 9 9 2
−1.0 −1.6 −3.1 −3.9 −1.0 −3.6 −1.2 −2.8 −1.0
4 5 12 16 3 14 5 9 4
8.3 2.1 12.4 5.8 6.2 8.8 0.9 2.7 2.3
31 6 39 22 18 28 4 8 7
−0.5 −1.4 −1.0 −0.2 −1.1 −1.1 −3.0 −3.0 −1.0
2 4 3 1 3 4 11 9 3
Anterior displacement reached up to +2.8 mm, which corresponded to 9% of the distance between the anterior and posterior points of the glenoid rim (Table 2). Maximum values of posterior displacement were larger than anterior displacement and reached up to −3.9 mm, which corresponded to 16% of the distance between the anterior and posterior points of the glenoid rim (Table 2). Maximum values of posterior displacements were measured close to maximum elevation angle in P1 and P3 and close to maximum external rotation in P2 (Fig. 2B). For P2 only, the greatest value of adjusted R2, 0.94, was observed during axial rotations for the regression between the glenohumeral axial rotation and glenohumeral displacement along antero-posterior axis (Table 3). With external rotation, the humeral head moved posteriorly with regard to the relaxed trial. Upward displacement represents the largest measured displacements with maximum value up to +12.4 mm, which corresponded to 39% of the distance between the superior and inferior reference points of the glenoid rim (Table 2). Maximum upward displacement was measured close to maximal abduction in P1 and P3, and during flexion in P2 (Fig. 2C). For P1 and P3 during abduction and flexion, values of adjusted R2 between displacement along longitudinal axis and rotation around abduction-axis ranged from 0.60 to 0.94 (Table 3); upward
displacement increased with glenohumeral abduction rotation. In P2, during abduction, values of upward displacement remained close to 0 mm; during flexion, values of upward increased up to 55° of glenohumeral abduction angle and decreased up to maximum arm elevation. For P2, the displacement along the longitudinal axis was highly correlated to rotation around the axial rotation axis (adjusted R2 = 0.72) during axial rotation. Downward displacement reached up to −3.0 mm, which corresponded to 11% of the distance between the superior and inferior reference points of the glenoid rim. Maximum values of downward displacement were measured close to maximum external rotation during axial rotations (Fig. 2C). 4. Discussion Glenohumeral displacements measurement was often limited to a series of static poses. Since internal forces differ from static to dynamic tasks and that muscular pattern affects glenohumeral displacements, a protocol based on CT-scan, intra-cortical pins and a motion analysis system was developed to measure 3D glenohumeral kinematics during dynamic tasks. The present study based on a small sample of participants provides preliminary values of normal range of glenohumeral
Table 3 Participant-specific results of Pearson's linear regressions (displacement (mm) = coefficient × glenohumeral angle (°) + constant (mm)) between, glenohumeral abduction angle during arm elevations or glenohumeral axial rotation angle during arm rotations and glenohumeral displacements along medio-lateral, antero-posterior and longitudinal axes. n.s.: nonsignificant. Tasks
Linear displacements
Participants
Constant
Coefficient
Adjusted R2
F
Abduction
Medio-lateral
P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3
0.4 0.5 −0.7 0.7 1.4 0.9 0.0 0.2 −0.4 −0.3 −1.2 −0.1 0.9 1.6 −0.5 0.9 n.s. 0.1 −0.5 −0.5 0.0 1.0 1.5 −0.2 −0.3 1.5 1.0
0.00 −0.02 0.01 −0.01 −0.01 −0.03 0.09 0.00 0.12 0.00 0.02 0.01 −0.02 −0.02 −0.01 0.02 n.s. 0.08 0.01 0.01 0.00 −0.02 −0.04 0.00 −0.01 −0.05 −0.02
0.07 0.26 0.35 0.07 0.17 0.73 0.83 0.02 0.94 0.19 0.49 0.16 0.61 0.42 0.15 0.60 n.s. 0.91 0.30 0.51 0.20 0.68 0.94 0.22 0.14 0.72 0.45
619.11 320.29 4301.35 602.05 186.40 21612.88 37062.02 17.70 126440.18 1954.02 966.92 943.73 13020.61 726.13 877.85 12508.08 n.s. 50182.00 887.81 1742.58 514.09 4555.60 25875.02 579.47 335.28 4224.87 1687.86
Antero-posterior
Longitudinal
Flexion
Medio-lateral
Antero-posterior
Longitudinal
Axial rotation
Medio-lateral
Antero-posterior
Longitudinal
* * * * * * * * * * * * * * * * * * * * * * * * * *
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
displacements during dynamic movements and their normal coupling with rotations. Glenohumeral displacements have been a controversial topic since their ranges of only a few millimeters were below the accuracy of electromagnetic or optoelectronic systems with skin markers (Veeger and van der Helm, 2007). However, measurement based on intracortical pin marker trajectories using optoelectronic system provided a measure of the linear displacement of the glenohumeral joint with an uncertainty lower than 0.15 mm. This is more accurate than techniques based on biplane fluoroscopic measurements (Bey et al., 2008). Therefore, measured displacements lower than 1 mm are considered as reliable. Moreover, glenohumeral displacements were measured from the linear displacement of the glenohumeral center of rotation located using a functional method contrarily to previous studies that were based on the geometrical center of the humeral head which is assumed to be the glenohumeral center of rotation. Indeed, glenohumeral joint center is often assumed to be the center of the humeral head and used to measure glenohumeral displacements. However, since the glenohumeral joint has six degrees of freedom, there is not a fixed center of rotation, but rather instantaneous centers of rotation during glenohumeral movements. The geometric center of the humeral head determined using sphere fitting does not necessarily take into account the mechanical properties of the glenohumeral joint. On the contrary, the center of rotation determined using the functional SCoRE method relies on the assumption that the glenohumeral joint is a quasi-perfect ball-andsocket joint. This method finds an optimal location which corresponds to a point with least displacements throughout the range of motion. Though the patterns of glenohumeral displacements were similar for the two methods, medial displacements were smaller with the glenohumeral center of rotation. From an anatomical point of view, 2 mm of medial displacement obtained using the method based on the glenohumeral joint center of rotation seems to be more representative of the humeral head displacements than 4 mm obtained using humeral head center given the narrow spacing between the humeral head and the glenoid. The proposed method may thus provide more realistic values of glenohumeral displacements. Though patterns of glenohumeral displacements were in some cases different between participants, the results showed that the patterns of glenohumeral displacements were similar across the repetitions of all the tasks for each subject. This result suggests that glenohumeral displacements are highly reproducible between trials of the same experimental condition. This result reinforces also that our method is accurate enough to measure glenohumeral displacements that are altered of only few tenths of millimeters between trials. Thus, using our method, the acquisition of a limited number of trials (n = 10) may be enough to measure representative kinematics of the glenohumeral joint. Whereas our study only focused on nearly-pure movements of elevation and rotation, shoulder degrees-of-freedom can interact, such that shoulder range of motion covers almost 65% of a sphere (Engin and Chen, 1986). Thus, our method that is not restricted in space and sampling rate could provide a strong benefit when assessing joint displacements during sports activities associated to anterior instability, daily living tasks and occupational activities. The experimental protocol is only restricted to the duration of the anesthesia, 2 h in our case, during which a large number of experimental conditions can be acquired. Up to one hundred trials have been collected, only a fraction is presented in the present study. The method is however highly invasive and the small sample of participants is the main limitation of our study and other similar studies based on pins inserted into the bones (Arndt et al., 2004; Braman et al., 2010; Houck et al., 2004). Our findings need to be corroborated by other studies based on larger sample of participants and on various movements before direct clinical applications. Nevertheless the results suggest that new pattern of glenohumeral displacements could be evidenced during dynamic tasks, especially such as coupling between displacements and rotations at the glenohumeral
1053
joint. Moreover, the heterogeneity in terms of age of our sample gave certainly more representative values of the living population compared to in-vitro studies where average age exceeds 70 years (San Juan and Karduna, 2010). As evidenced by the measure of the displacements along the mediolateral axis, the distance between the glenohumeral joint center of rotation and the glenoid cavity changes during movements. Medial and lateral displacements observed can be due to non-homogeneous curvature of the humeral head (Carey et al., 2000; Harrold and Wigderowitz, 2013). Medial displacements did not exceed + 2.4 mm which corresponds to the range already measured during arm static elevation (Bryce et al., 2010; Massimini et al., 2012) and rotation (Massimini et al., 2012). Medial displacements may also be due to compression of the labrum (Carey et al., 2000) provoked by compressive forces generated by rotator cuff muscles and tendons arranged in a half circle around the humeral head (Blasier et al., 1997; Wilk et al., 1997). Displacements in antero-posterior direction are of interest due to glenohumeral dislocation occurring in anterior direction in 95% of the cases (Cutts et al., 2009). The results showed that the humeral head moved maximally from + 2.8 mm forward to − 3.9 mm backward during dynamic elevations and rotations, which is in the range of displacements already measured during static poses (Bryce et al., 2010; Massimini et al., 2012). This result suggests that anteroposterior displacements may not be sensitive to the dynamic of the movement. Humeral head tended to displace anteriorly during abduction and posteriorly during flexion, which agrees with the alteration of antero-posterior shear forces with the plane of elevation during elevations (Kontaxis and Johnson, 2009). Further studies should consider other planes of elevation under dynamic tasks to determine the plane of elevation that minimize antero-posterior displacements and thereby limit antero-posterior instability for elevation rehabilitation exercises. Glenohumeral displacements along the longitudinal axis increased at low arm elevation angles before returning close to its initial position at maximum elevation angle for P2. This pattern corresponds to the pattern of longitudinal displacements measured during static (Massimini et al., 2012) and slow dynamic poses (Bey et al., 2008; Bishop et al., 2009). For P1 and P3, contrarily to the literature, values of upward displacements increase linearly with abduction angle during elevations with maximum values at maximum elevation. For these two participants, upward displacements were the largest measured displacements. This result obtained during dynamic tasks is of particular importance since upward displacements are one of the major extrinsic causes of the shoulder impingement syndrome (Deutsch et al., 1996; Sharkey and Marder, 1995; Williams et al., 2001). It suggests that even during dynamic daily-living tasks such as touching the head where the arm elevation can represent up to 77% of the maximum arm elevation (Lovern et al., 2010), large values of upward displacements may occur and reduce the subacromial space. This may also be the case during overhead handling tasks which could explain the high prevalence of supraspinatus tendon tear (Milgrom et al., 1995) since the acromiohumeral distance may be reduced. The results also suggest that longitudinal displacements are sensitive to the dynamic of the movement due to the increase of the ratio between the contribution of deltoids and rotator cuff muscles (Graichen et al., 2005; Reddy et al., 2000; Sharkey and Marder, 1995; Teyhen et al., 2008; Yamaguchi et al., 2000). Noteworthy, for P1 and P3, upward displacements were larger during abduction than during flexion. Further studies should consider other planes of elevation to determine the plane of elevation that limit upward displacements to reduce the risk of shoulder impingement for rehabilitation exercises. Maximum values of downward displacements were measured at maximum external rotation. This result reinforces the relevancy of this exercise to strengthen rotator cuff muscle while reducing the risk of subacromial impingement. Our method permitted also to evidence that glenohumeral displacements along the longitudinal axis were coupled with glenohumeral
1054
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
rotation around the abduction axis. This result provides a new insight of glenohumeral kinematics that could find an application for the design of complete glenohumeral joint prostheses as well as for musculoskeletal computer simulation models. 5. Conclusions In summary, the method to acquire glenohumeral kinematics permitted to measure glenohumeral displacements with an uncertainty lower than 0.15 mm. Our results reaffirm that kinematics of healthy glenohumeral joint is a combination of displacements and rotations rather than pure rotations during dynamic tasks. During arm elevations and axial rotations, the glenohumeral displacements along longitudinal axis were linearly coupled with glenohumeral rotations around abduction axis. The largest displacements were observed at maximum elevation in the upward direction. Compared to previous results in the literature during static tasks, this result reinforces that glenohumeral displacements are sensitive to muscular dynamics, and further studies are needed to better characterize the kinematics of the glenohumeral joint during dynamic tasks. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiomech.2014.08.006. Acknowledgments This work was supported by the RRSSTQ (grant no. 101220). In addition, the MEDITIS training program in biomedical technologies (NSERC, CREATE program) supported the postdoctoral fellowship of the first author. None of these programs played a role in the present investigation. References Arndt, A., Westblad, P., Winson, I., Hashimoto, T., Lundberg, A., 2004. Ankle and subtalar kinematics measured with intracortical pins during the stance phase of walking. Foot Ankle Int. 25, 357–364. Bey, M.J., Zauel, R., Brock, S.K., Tashman, S., 2006. Validation of a new model-based tracking technique for measuring three-dimensional, in vivo glenohumeral joint kinematics. J. Biomech. Eng. 128, 604–609. Bey, M.J., Kline, S.K., Zauel, R., Lock, T.R., Kolowich, P.A., 2008. Measuring dynamic in-vivo glenohumeral joint kinematics: technique and preliminary results. J. Biomech. 41, 711–714. Bishop, J.L., Kline, S.K., Aalderink, K.J., Zauel, R., Bey, M.J., 2009. Glenoid inclination: In vivo measures in rotator cuff tear patients and associations with superior glenohumeral joint translation. J. Shoulder Elb. Surg. 18, 231–236. Blasier, R.B., Soslowsky, L.J., Malicky, D.M., Palmer, M.L., 1997. Posterior glenohumeral subluxation: active and passive stabilization in a biomechanical model. J. Bone Joint Surg. Am. 79, 433–440. Bolsterlee, B., Veeger, D.H., Chadwick, E.K., 2013. Clinical applications of musculoskeletal modelling for the shoulder and upper limb. Med. Biol. Eng. Comput. 51, 953–963. Bourne, D.A., Choo, A.M., Regan, W.D., MacIntyre, D.L., Oxland, T.R., 2007. Threedimensional rotation of the scapula during functional movements: an in vivo study in healthy volunteers. J. Shoulder Elbow Surg. 16, 150–162. Braman, J., Thomas, B., LaPrade, R., Phadke, V., Ludewig, P., 2010. Three-dimensional in vivo kinematics of an osteoarthritic shoulder before and after total shoulder arthroplasty. Knee Surg. Sports Traumatol. Arthrosc. 18, 1774–1778. Bryce, C.D., Davison, A.C., Okita, N., Lewis, G.S., Sharkey, N.A., Armstrong, A.D., 2010. A biomechanical study of posterior glenoid bone loss and humeral head translation. J. Shoulder Elbow Surg. 19, 994–1002. Buchler, P., Ramaniraka, N.A., Rakotomanana, L.R., Iannotti, J.P., Farron, A., 2002. A finite element model of the shoulder: application to the comparison of normal and osteoarthritic joints. Clin. Biomech. 17, 630–639. Carey, J., Small, C.F., Pichora, D.R., 2000. In situ compressive properties of the glenoid labrum. J. Biomed. Mater. Res. 51, 711–716. Chan, K.M., Maffulli, N., Nobuhara, M., Wu, J.J., 1996. Shoulder instability in athletes. The Asian perspective. Clin. Orthop. Relat. Res. 106–112. Cutts, S., Prempeh, M., Drew, S., 2009. Anterior shoulder dislocation. Ann. R. Coll. Surg. Engl. 91, 2–7. De Wilde, L.F., Verstraeten, T., Speeckaert, W., Karelse, A., 2010. Reliability of the glenoid plane. J. Shoulder Elbow Surg. 19, 414–422. Deutsch, A., Altchek, D.W., Schwartz, E., Otis, J.C., Warren, R.F., 1996. Radiologic measurement of superior displacement of the humeral head in the impingement syndrome. J. Shoulder Elb. Surg. 5, 186–193. Ehrig, R.M., Taylor, W.R., Duda, G.N., Heller, M.O., 2006. A survey of formal methods for determining the centre of rotation of ball joints. J. Biomech. 39, 2798–2809. Ehrig, R.M., Taylor, W.R., Duda, G.N., Heller, M.O., 2007. A survey of formal methods for determining functional joint axes. J. Biomech. 40, 2150–2157.
Engin, A.E., Chen, S.M., 1986. Statistical data base for the biomechanical properties of the human shoulder complex—I: kinematics of the shoulder complex. J. Biomech. Eng. 108, 215–221. Fang, Q., Boas, D.A., 2009. Tetrahedral mesh generation from volumetric binary and grayscale images. Biomedical Imaging: From Nano to Macro, 2009. ISBI '09. IEEE International Symposium on, pp. 1142–1145. Favre, P., Senteler, M., Hipp, J., Scherrer, S., Gerber, C., Snedeker, J.G., 2012. An integrated model of active glenohumeral stability. J. Biomech. 45, 2248–2255. Giphart, J.E., Brunkhorst, J.P., Horn, N.H., Shelburne, K.B., Torry, M.R., Millett, P.J., 2013. Effect of plane of arm elevation on glenohumeral kinematics: a normative biplane fluoroscopy study. J. Bone Joint Surg. Am. 95, 238–245. Graichen, H., Stammberger, T., Bonel, H., Karl-Hans, E., Reiser, M., Eckstein, F., 2000. Glenohumeral translation during active and passive elevation of the shoulder — a 3D open-MRI study. J. Biomech. 33, 609–613. Graichen, H., Hinterwimmer, S., Eisenhart-Rothe, R.v., Vogl, T., Englmeier, K.H., Eckstein, F., 2005. Effect of abducting and adducting muscle activity on glenohumeral translation, scapular kinematics and subacromial space width in vivo. J. Biomech. 38, 755–760. Harrold, F., Wigderowitz, C., 2013. Humeral head arthroplasty and its ability to restore original humeral head geometry. J. Shoulder Elbow Surg. 22, 115–121. Houck, J., Yack, H.J., Cuddeford, T., 2004. Validity and comparisons of tibiofemoral orientations and displacement using a femoral tracking device during early to mid stance of walking. Gait Posture 19, 76–84. Hudak, P.L., Amadio, P.C., Bombardier, C., 1996. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG). Am. J. Ind. Med. 29, 602–608. Institute, S C.a.I., 2014. “Seg3D” Volumetric Image Segmentation and Visualization. Scientific Computing and Imaging Institute (SCI). Jackson, M., Michaud, B., Tetreault, P., Begon, M., 2012. Improvements in measuring shoulder joint kinematics. J. Biomech. 45, 2180–2183. Karduna, A.R., Williams, G.R., Iannotti, J.P., Williams, J.L., 1998. Total shoulder arthroplasty biomechanics: a study of the forces and strains at the glenoid component. J. Biomech. Eng. 120, 92–99. Karduna, A.R., McClure, P.W., Michener, L.A., Sennett, B., 2001. Dynamic measurements of three-dimensional scapular kinematics: a validation study. J. Biomech. Eng. 123, 184–190. Kontaxis, A., Johnson, G.R., 2009. The biomechanics of reverse anatomy shoulder replacement — a modelling study. Clin. Biomech. 24, 254–260. Landry, C., De Guise, J.A., Dansereau, J., Labelle, H., Skalli, W., Zeller, R., et al., 1997. Computer graphic analysis of the three dimensional deformities of scoliotic vertebrae. Ann. Chir. 51, 868–874. Liu, A., Nester, C.J., Jones, R.K., Lundgren, P., Lundberg, A., Arndt, A., et al., 2012. Effect of an antipronation foot orthosis on ankle and subtalar kinematics. Med. Sci. Sports Exerc. 44, 2384–2391. Lovern, B., Stroud, L.A., Ferran, N.A., Evans, S.L., Evans, R.O., Holt, C.A., 2010. Motion analysis of the glenohumeral joint during activities of daily living. Comput. Methods Biomech. Biomed. Eng. 13, 803–809. Ludewig, P.M., Phadke, V., Braman, J.P., Hassett, D.R., Cieminski, C.J., LaPrade, R.F., 2009. Motion of the shoulder complex during multiplanar humeral elevation. J. Bone Joint Surg. 91, 378–389. Massimini, D.F., Boyer, P.J., Papannagari, R., Gill, T.J., Warner, J.P., Li, G., 2012. In-vivo glenohumeral translation and ligament elongation during abduction and abduction with internal and external rotation. J. Orthop. Surg. Res. 7–29. Meskers, C.G.M., van der Helm, F.C.T., Rozendaal, L.A., Rozing, P.M., 1997. In vivo estimation of the glenohumeral joint rotation center from scapular bony landmarks by linear regression. J. Biomech. 31, 93–96. Milgrom, C., Schaffler, M., Gilbert, S., van Holsbeeck, M., 1995. Rotator-cuff changes in asymptomatic adults. The effect of age, hand dominance and gender. J. Bone Joint Surg. (Br.) 77, 296–298. Monnet, T., Thouze, A., Pain, M.T., Begon, M., 2012. Assessment of reproducibility of thigh marker ranking during walking and landing tasks. Med. Eng. Phys. 34, 1200–1208. Reddy, A.S., Mohr, K.J., Pink, M.M., Jobe, F.W., 2000. Electromyographic analysis of the deltoid and rotator cuff muscles in persons with subacromial impingement. J. Shoulder Elbow Surg. 9, 519–523. San Juan, J.G., Karduna, A.R., 2010. Measuring humeral head translation using fluoroscopy: a validation study. J. Biomech. 43, 771–774. Senk, M., Cheze, L., 2006. Rotation sequence as an important factor in shoulder kinematics. Clin. Biomech. 21 (Suppl. 1), S3–S8. Sharkey, N.A., Marder, R.A., 1995. The rotator cuff opposes superior translation of the humeral head. Am. J. Sports Med. 23, 270–275. Teyhen, D.S., Miller, J.M., Middag, T.R., Kane, E.J., 2008. Rotator cuff fatigue and glenohumeral kinematics in participants without shoulder dysfunction. J. Athl. Train. 43, 352–358. Veeger, H.E., van der Helm, F.C., 2007. Shoulder function: the perfect compromise between mobility and stability. J. Biomech. 40, 2119–2129. von Eisenhart-Rothe, R.M., Jager, A., Englmeier, K.H., Vogl, T.J., Graichen, H., 2002. Relevance of arm position and muscle activity on three-dimensional glenohumeral translation in patients with traumatic and atraumatic shoulder instability. Am. J. Sports Med. 30, 514–522. Walch, G., Edwards, T.B., Boulahia, A., Boileau, P., Mole, D., Adeleine, P., 2002. The influence of glenohumeral prosthetic mismatch on glenoid radiolucent lines results of a multicenter study. J. Bone Joint Surg. 84, 2186–2191. Wilk, K.E., Andrews, J.R., Arrigo, C.A., 1997. The physical examination of the glenohumeral joint: emphasis on the stabilizing structures. J. Orthop. Sports Phys. Ther. 25, 380–389. Williams Jr., G.R., Wong, K.L., Pepe, M.D., Tan, V., Silverberg, D., Ramsey, M.L., et al., 2001. The effect of articular malposition after total shoulder arthroplasty on glenohumeral
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055 translations, range of motion, and subacromial impingement. J. Shoulder Elbow Surg. 10, 399–409. Wu, G., van der Helm, F.C., Veeger, H.E., Makhsous, M., Van Roy, P., Anglin, C., et al., 2005. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion—part II: shoulder, elbow, wrist and hand. J. Biomech. 38, 981–992.
1055
Yamaguchi, K., Sher, J.S., Andersen, W.K., Garretson, R., Uribe, J.W., Hechtman, K., et al., 2000. Glenohumeral motion in patients with rotator cuff tears: a comparison of asymptomatic and symptomatic shoulders. J. Shoulder Elbow Surg. 9, 6–11.
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Coupling between 3D displacements and rotations at the glenohumeral joint during dynamic tasks in healthy participants Fabien Dal Maso a,⁎, Maxime Raison b, Arne Lundberg c, Anton Arndt c,d, Mickaël Begon a a
Laboratoire de simulation et de modélisation du mouvement, Département de kinésiologie, Université de Montréal, 1700, rue Jacques Tétreault, Laval, QC H7N 0B6, Canada École Polytechnique de Montréal, 6079, Succursale, Centre Ville, Montréal, QC H3C 3A7, Canada Karolinska Institute, Stockholm, Sweden d The Swedish School of Sport and Health Sciences, Stockholm, Sweden b c
a r t i c l e
i n f o
Article history: Received 21 January 2014 Accepted 4 August 2014 Keywords: Intracortical pin Shoulder prosthesis Kinematic model 3D kinematics CT-scan
a b s t r a c t Background: Glenohumeral displacements assessment would help to design shoulder prostheses with physiological arthrokinematics and to establish more biofidelic musculoskeletal models. Though displacements were documented during static tasks, there is little information on their 3D coupling with glenohumeral angle during dynamic tasks. Our objective was to characterize the 3D glenohumeral displacement–rotation couplings during dynamic arm elevations and rotations. Methods: Glenohumeral displacements were measured from trajectories of reflective markers fitted on intracortical pins inserted into the scapula and humerus. Bone geometry was recorded using CT-scan. Only four participants were recruited to the experiment due to its invasiveness. Participants performed dynamic arm abduction, flexion and axial rotations. Linear regressions were performed between glenohumeral displacements and rotations. The pin of the scapula of one participant moved, his data were removed from analysis, and results are based on three participants. Findings: The measurement error of glenohumeral kinematics was less than 0.15 mm and 0.2°. Maximum glenohumeral displacements were measured along the longitudinal direction and reached up to + 12.4 mm for one participant. Significant couplings were reported especially between longitudinal displacement and rotation in abduction (adjusted R2 up to 0.94). Interpretation: The proposed method provides the potential to investigate glenohumeral kinematics during all kinds of movements. A linear increase of upward displacement during dynamic arm elevation was measured, which contrasts with results based on a series of static poses. The systematic investigation of glenohumeral displacements under dynamic condition may help to provide relevant recommendation for the design of shoulder prosthetic components and musculoskeletal models. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Though the glenohumeral joint is often simplified to a ball-andsocket in human movement analysis (Jackson et al., 2012; Wu et al., 2005), the humeral head undergoes displacements relative to the glenoid cavity, generally termed as glenohumeral translations (although translation should refer to the linear movement humerus with respect to the scapula along the helical axis). Displacements along the antero-posterior and longitudinal axes have been reported in in-vitro (Bryce et al., 2010; San Juan and Karduna, 2010; Sharkey and Marder, 1995) and in-vivo studies for both healthy subjects (Bey et al., 2008; Graichen et al., 2005; Massimini et al., 2012) and patients
⁎ Corresponding author at: Université de Montréal, 1700, rue Jacques Tétreault, Laval, QC H7N 0B6, Canada. E-mail address: [email protected] (F. Dal Maso).
http://dx.doi.org/10.1016/j.clinbiomech.2014.08.006 0268-0033/© 2014 Elsevier Ltd. All rights reserved.
(Bishop et al., 2009). Glenohumeral displacements of only a few millimeters are responsible for the loss of joint stability, which makes the glenohumeral joint potentially the most unstable joint in the body (Chan et al., 1996). The active co-contraction of rotator cuff and adductor muscles counteracts upward shear forces generated by deltoid muscles to reduce upward displacement (Graichen et al., 2000, 2005; Sharkey and Marder, 1995; von Eisenhart-Rothe et al., 2002). Though shoulder kinematics differ between static and dynamic tasks (Bey et al., 2006; Carey et al., 2000) and glenohumeral displacements are sensitive to muscular activity patterns (Graichen et al., 2000, 2005; Sharkey and Marder, 1995; von Eisenhart-Rothe et al., 2002), their measurements were often limited to a series of static poses (Bey et al., 2008; Giphart et al., 2013). Daily living activities correspond however to dynamic tasks (Bey et al., 2006). Consequently, recent studies (Bey et al., 2006; Massimini et al., 2012) agree that “shoulder biomechanics under dynamic conditions should be investigated” (Massimini et al., 2012).
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
While glenohumeral displacements increase joint instability, they reduce the prevalence of periprosthetic radiolucency in shoulder prostheses (Walch et al., 2002) and above all homogenize the stress at the implant fixation sites (Buchler et al., 2002; Karduna et al., 1998). Understanding the normal glenohumeral kinematics is therefore fundamental in orthopedics to design shoulder prostheses (Buchler et al., 2002). In another field of application, the replication of glenohumeral displacements in musculoskeletal computer simulation models may enable a better determination of the muscle moment arms, muscular activation patterns and activation level of stabilizing muscles (Bolsterlee et al., 2013; Favre et al., 2012). It may also better determine contact area between humeral head and glenoid cavity (Buchler et al., 2002). It may thus be crucial to characterize the normal coupling between glenohumeral displacements and rotations during upper-limb dynamic tasks. During dynamic tasks, the measurement of glenohumeral displacements with high accuracy is however challenging. Glenohumeral displacements were commonly measured using MRI (Graichen et al., 2005; Massimini et al., 2012), X-ray (San Juan and Karduna, 2010; Yamaguchi et al., 2000) and biplane X-ray recordings (Bey et al., 2008; Bishop et al., 2009). Imaging provides accurate measure of bone positions but is not suitable to record fast motion and is restricted to the volume of acquisition. Moreover, due to radiation during X-ray recordings, the number of movements acquired during the experimental protocol is limited. On the other hand, optoelectronic motion analysis systems with markers positioned on the skin are not limited in sampling rate and space but have limited accuracy because of soft tissues artifacts between skin and underlying bones (Karduna et al., 2001). This represents an issue to investigate glenohumeral displacements of a few millimeters (Bey et al., 2008; Graichen et al., 2000). To overcome these limitations, bone kinematics can be measured using the trajectories of reflective markers secured on intra-cortical pins implanted into the bones (Bourne et al., 2007; Liu et al., 2012; Ludewig et al., 2009). Since glenohumeral displacements are usually expressed in a glenoid-based coordinated system (GCS) (Bey et al., 2006; Graichen et al., 2000; Massimini et al., 2012), such a method would also require the acquisition of bone geometry but offers the perspective to investigate glenohumeral kinematics with no restrictions on the type of movements and on the number of repetitions. The purpose of this study was to establish a method that permits to characterize the three-dimensional (3D) glenohumeral displacement for dynamic movements without known restrictions in space and number of trials. This method was applied to a sample of participants and the secondary purpose was to provide preliminary results of coupling between glenohumeral displacements and rotations as well as preliminary values of glenohumeral displacements close to physiological limits during dynamic arm elevations and axial rotations in healthy participants. With regard to the sensitivity of glenohumeral displacements to muscular activity, we hypothesized that new patterns of glenohumeral displacements during dynamic tasks will be highlighted.
1049
hereafter, the data of only three participants were used in this study. Participants obtained a score lower than 10.5 at the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire (Hudak et al., 1996), indicating that they were asymptomatic at the left shoulder and without history of pain, injury or shoulder dysfunction. The experimental protocol was approved by the concerned local ethics committees of the Karolinska Institutet (Sweden) and the University of Montreal (Canada). 2.2. Instrumentation Intracortical pins were inserted by an experienced surgeon in an operating room under normal and sterile surgery conditions. After administration of local anesthesia (AstraZeneca, Södertälje, Sweden), one stainless steel self-drilling pin, 1.6 mm of diameter (Synthes, Bettlach, Switzerland) was inserted into the left scapular spine (Fig. 1A). A second pin (2.5 mm of diameter) was placed in the lateral aspect of the left humerus just distal to the deltoid attachment (Fig. 1A). Clusters of four and five reflective markers were secured on both pins (Fig. 1A). Pin locations were determined to avoid muscles so that the normal strategies of movement control remained unchanged. Pins avoided also nerves and blood vessels and were oriented to avoid contacts between markers and head or neck of the participants during movements. The rigidity of the montage was manually checked and insertion sites were cleaned, sterilized and covered with a sterile dressing. Pins were removed at the end of the experiment, and insertion sites were cleaned and covered again with new sterile dressings. Participants were provided with antibiotic and pain relief (AstraZeneca, Södertälje, Sweden) medication. No clinical complications occurred following the experiment. Six, nine and seven reflective skin markers were placed on the thorax, scapula and humerus, respectively (Jackson et al., 2012). Trajectories of pin and skin markers were collected at 300 Hz using an 18-camera VICON™ optoelectronic motion analysis system (Oxford Metrics Ltd., Oxford, UK). A CT-scan (General Electric, Milwaukee, USA) was used to obtain the 3D geometry of the scapula and humerus and their respective pin (Fig. 1B). For each participant, 226 tomographic slices (thickness: 0.61 mm; interslice space: 0.32 mm) were recorded. The field of view was set to 250 × 250 mm, defined by a matrix of 512 × 512 pixels. X-ray tube was adjusted at 120 kV and 110 μA, which corresponds to the usual settings (Meskers et al., 1997) to maximize contrast between cortical bone and soft tissues of living patients and also between markers and air. 2.3. Procedures
Pins were drilled into the bones of the participants to measure glenohumeral kinematics with high accuracy during dynamic tasks. Moreover, participants were exposed to radiation via computedtomography scanner (CT-scan). The proposed method is therefore invasive, which explains the small sample of participants.
First, a relaxed position trial and a series of setup functional movements, namely, maximum arm elevations, circumductions and arm sweeping were recorded to locate the glenohumeral joint center (Ehrig et al., 2007; Jackson et al., 2012). Secondly, 10 dynamic arm elevations in abduction and flexion with thumb pointing upward (with no specific instruction on the axial rotation of the glenohumeral and radioulnar joints) and 10 axial rotations in adducted position (0° of thoraco-humeral elevation) with elbow flexed at 90° were performed. Participant 2 performed only two trials of abduction and flexion. During each movement, participants were asked to move their arm at spontaneous speed and to reach their maximal range of motion.
2.1. Participants
2.4. Data processing
Four voluntary males (age: 27, 44, 32 and 41 years; height: 1.65, 1.77, 1.72, and 1.82 m, mass: 57, 82, 80 and 115 kg for Participants 1, 2, 3 and 4, P1, P2, P3 and P4, respectively) signed an informed consent form and participated in the study. Due to technical problems described
An automatic segmentation of gray level of the CT-scan was obtained using the Seg3D2® software (Institute) and a manual correction of the potential segmentation errors was performed to identify pin markers and bone voxels (Landry et al., 1997). Thereafter,
2. Methods
1050
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
Fig. 1. A) Picture of participant fitted with intracortical pins inserted into the clavicle (unused in this study), scapula and humerus with their respective clusters of four to five pin markers. B) Back view of the CT-scan segmentation of the scapula and humerus bones with their respective clusters of markers. C) Representation of the CT-scan segmentation of the scapula. Circles represent the anterior, inferior and posterior reference points on the glenoid rim. The X-, Y- and Z-axes represent the antero-posterior, longitudinal and medio-lateral axes, respectively of the glenoid-based coordinate system in which glenohumeral displacements are expressed.
processing was performed using homemade and built-in Matlab® functions (R2013a version; The Mathworks, Natick, USA). CT-scan slices were piled using the iso2mesh MATLAB® toolbox (Fang and Boas, 2009) and 3D reconstructions were obtained for each bone. The anterior, inferior, posterior and superior reference points of the glenoid rim (Fig. 1C) were identified by three evaluators five times each. Inter-evaluator variability was 1.2 mm. An inferior GCS was defined from the anterior, inferior and posterior reference points (De Wilde et al., 2010). A nonlinear least-squares algorithm was applied to pin marker trajectories to reduce the error which arises from the optoelectronic motion analysis system to locate the markers (Monnet et al., 2012). Functional glenohumeral center of rotation was located using the Symmetrical Center of Rotation Estimation (SCoRE) method (Ehrig et al., 2006). The GCS was translated to the position of the glenohumeral center of rotation in the relaxed position trial (Massimini et al., 2012). Glenohumeral displacements corresponded to the vector from the origin of the translated GCS to the glenohumeral center of rotation along the medio-lateral, antero-posterior and longitudinal axes. A sphere fitting algorithm was applied to CT-scan data to locate the geometrical center of the humeral head. Glenohumeral displacements were also computed using this point. Both methods were compared on the data of abduction for all participants. Scapula and humerus skin markers were used to determine local coordinated systems following ISB recommendations (Wu et al., 2005). Glenohumeral rotations were expressed according to the abduction–flexion–axial rotation angle sequence (rotation around the X-axis, rotated Z-axis, twice rotated Y-axis, respectively) (Senk and Cheze, 2006). Note that internal and external axial rotations will produce negative and positive values, respectively, since the left side of the participants was recorded. The thorax skin markers were used to compute thoraco-humeral angles. Glenohumeral displacements and rotations were low-pass filtered at 10 Hz (4th-order zero-lag Butterworth filter). Skin markers were also used to compare the orientation of the pinand skin-based local coordinate systems at the beginning of each trial and to ensure that the orientation of the pins did not change relatively to the skin-based local coordinate system throughout the experiment. For P4, the orientation of the pin-based coordinate system of the scapula revealed a drift with regard to the orientation of the skin-based coordinate system. This indicates that the pin rotated during the trials. Consequently, the data of P4 were removed. The following analysis was performed on the three remaining participants, namely, Participants 1, 2 and 3.
2.5. Estimation of the uncertainty The uncertainty of glenohumeral kinematics was determined as follows. Pin-marker position artifacts were extracted from pinmarker raw trajectories with a 10 Hz high-pass filter (fourth-order zero-lag Butterworth filter). The computed glenohumeral displacements and rotations were low-pass filtered and the reverse kinematic was applied to obtain noiseless pin-marker trajectories. Noiseless pin-marker trajectories and extracted artifacts were summed to obtain artifact-simulated trajectories. Glenohumeral displacements and rotations were then re-computed to obtain artifact-simulated glenohumeral displacements and rotations. Finally, a root-mean-square (RMS) error was computed between lowpass filtered and artifact-simulated glenohumeral displacements and rotations. Using this method, uncertainty of glenohumeral displacements and rotations was estimated at less than 0.15 mm and 0.2°, respectively.
2.6. Analysis For each participant, Pearson's linear regression analysis was used to assess the coupling between glenohumeral displacements and rotations along and around the three axes. Regressions were performed using all data points recorded during abduction, flexion and axial rotation of each participant. The significance level was set at P b 0.05/12 due to Bonferroni correction for multiple testing. The maximum values of lateral, medial, anterior, posterior, upward and downward displacements were reported for each movement. Finally, medio-lateral displacements were expressed as a percentage of the humeral head diameter. Antero-posterior displacements were expressed as a percentage of the distance between the anterior and posterior reference points of the glenoid rim and longitudinal displacements as a percentage of the distance between the inferior and superior reference points of the glenoid rim. These values were however not compared statistically due to the small sample of participants (N = 3).
3. Results The duration and the maximum range of motion of each task are detailed in Table 1 for each participant. The elevations lasted between 1.5 and 2.6 s. The maximum thoracohumeral abduction angle was close to 135° and 130° during abduction and flexion, respectively. The
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
1051
Table 1 Participant-specific values of duration and maximum range of motion for each task. n.a.: not applicable. Tasks
Participants
Execution time (s)
Maximum angle (°) Abduction
Abduction
Flexion
Axial rotation
P1 P2 P3 P1 P2 P3 P1 P2 P3
2.6 1.5 2.4 2.5 1.7 1.7 2.3 1.4 1.4
axial rotations lasted between 1.4 and 2.3 s, and the amplitude of axial rotation ranged from 87° to 93°. 3.1. Effect of the reference location of the humeral head Table S1 (supplementary material) presents the difference between the location of the geometric center of the humeral head and the glenohumeral center of rotation. The 3D Euclidean distance between these two points was 4.0, 4.2 and 3.9 mm for P1, P2 and P3, respectively. Both methods lead to similar pattern of glenohumeral displacements for all participants during arm abductions (Fig. S2, supplementary material). However, using the humeral head center, medial displacements were larger than using the glenohumeral joint center, while less differences were found along the antero-posterior and longitudinal axes. Medial displacements were up to + 4 mm with the humeral head center while it was less than + 2 mm with the glenohumeral joint center of rotation. 3.2. Participant-specific glenohumeral displacements Only regressions between the glenohumeral displacements along the three axes and the glenohumeral rotation axis with which the
Axial rotation
Glenohumeral
Thoracohumeral
Internal
External
64 67 85 81 84 92 n.a. n.a. n.a.
135 121 149 129 117 143 n.a. n.a. n.a.
n.a. n.a. n.a. n.a. n.a. n.a. −23 −22 −25
n.a. n.a. n.a. n.a. n.a. n.a. 70 65 64
adjusted R2 was the greater are presented. In average, the adjusted R2 was the greatest with the regression between the rotations around the abduction and the axial rotation axis during elevation arm axial rotation, respectively. Thereby, Fig. 2 represents the displacements along the mediolateral, antero-posterior and longitudinal axes according to the rotations around the abduction axis during abduction and flexion and according to the axial rotation axis during axial rotations. Patterns of glenohumeral displacements according to the glenohumeral rotations were similar across participants in most situations (Fig. 2). However, values of measured displacements differed between participants: e.g. longitudinal displacement during abduction (Fig. 2C). Table 2 presents the maximum measured and normalized values of lateral, medial, anterior, posterior, upward and downward glenohumeral displacements during abduction, flexion and axial rotations for each participant. Table 3 summarizes constants, coefficients, adjusted R2 and F values of the regressions between glenohumeral displacements and rotations and for each participant. Medial and lateral displacements reached up to + 2.4 mm and −2.1 mm representing 5% and 4% of the humeral head, respectively (Table 2). These amplitudes were smaller than the values of displacements reached along antero-posterior and longitudinal axes.
Fig. 2. (A) Medio-lateral, (B) antero-posterior and (C) longitudinal displacements of the glenohumeral joint according to its abduction angle during abduction and flexion (first and second columns, respectively) and to axial rotation angle during axial rotation (third column). The dashed black thin lines represent 0 mm of displacement. From top to bottom, positive values of displacements indicate that the humeral head moves medially, forward and upward relatively to the glenoid. Positive and negative values of the glenohumeral axial rotations indicate external and internal rotations, respectively.
1052
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
Table 2 Participant-specific maximum measured and normalized values of medial, lateral, anterior, posterior, upward and downward glenohumeral displacements during abduction, flexion and axial rotations. Tasks
Abduction
Flexion
Axial rotation
Participants
P1 P2 P3 P1 P2 P3 P1 P2 P3
Medial
Lateral
Anterior
Posterior
Upward
Downward
mm
%
mm
%
mm
%
mm
%
mm
%
mm
%
1.0 2.0 2.4 0.4 0.9 1.4 0.5 0.7 1.1
3 4 5 1 2 3 1 1 2
−0.8 −2.0 −1.2 −1.4 −1.5 −2.1 −1.0 −1.3 −1.2
2 4 2 3 3 4 2 2 2
1.8 2.4 1.1 1.8 2.0 0.8 2.2 2.8 0.5
7 8 4 7 7 3 9 9 2
−1.0 −1.6 −3.1 −3.9 −1.0 −3.6 −1.2 −2.8 −1.0
4 5 12 16 3 14 5 9 4
8.3 2.1 12.4 5.8 6.2 8.8 0.9 2.7 2.3
31 6 39 22 18 28 4 8 7
−0.5 −1.4 −1.0 −0.2 −1.1 −1.1 −3.0 −3.0 −1.0
2 4 3 1 3 4 11 9 3
Anterior displacement reached up to +2.8 mm, which corresponded to 9% of the distance between the anterior and posterior points of the glenoid rim (Table 2). Maximum values of posterior displacement were larger than anterior displacement and reached up to −3.9 mm, which corresponded to 16% of the distance between the anterior and posterior points of the glenoid rim (Table 2). Maximum values of posterior displacements were measured close to maximum elevation angle in P1 and P3 and close to maximum external rotation in P2 (Fig. 2B). For P2 only, the greatest value of adjusted R2, 0.94, was observed during axial rotations for the regression between the glenohumeral axial rotation and glenohumeral displacement along antero-posterior axis (Table 3). With external rotation, the humeral head moved posteriorly with regard to the relaxed trial. Upward displacement represents the largest measured displacements with maximum value up to +12.4 mm, which corresponded to 39% of the distance between the superior and inferior reference points of the glenoid rim (Table 2). Maximum upward displacement was measured close to maximal abduction in P1 and P3, and during flexion in P2 (Fig. 2C). For P1 and P3 during abduction and flexion, values of adjusted R2 between displacement along longitudinal axis and rotation around abduction-axis ranged from 0.60 to 0.94 (Table 3); upward
displacement increased with glenohumeral abduction rotation. In P2, during abduction, values of upward displacement remained close to 0 mm; during flexion, values of upward increased up to 55° of glenohumeral abduction angle and decreased up to maximum arm elevation. For P2, the displacement along the longitudinal axis was highly correlated to rotation around the axial rotation axis (adjusted R2 = 0.72) during axial rotation. Downward displacement reached up to −3.0 mm, which corresponded to 11% of the distance between the superior and inferior reference points of the glenoid rim. Maximum values of downward displacement were measured close to maximum external rotation during axial rotations (Fig. 2C). 4. Discussion Glenohumeral displacements measurement was often limited to a series of static poses. Since internal forces differ from static to dynamic tasks and that muscular pattern affects glenohumeral displacements, a protocol based on CT-scan, intra-cortical pins and a motion analysis system was developed to measure 3D glenohumeral kinematics during dynamic tasks. The present study based on a small sample of participants provides preliminary values of normal range of glenohumeral
Table 3 Participant-specific results of Pearson's linear regressions (displacement (mm) = coefficient × glenohumeral angle (°) + constant (mm)) between, glenohumeral abduction angle during arm elevations or glenohumeral axial rotation angle during arm rotations and glenohumeral displacements along medio-lateral, antero-posterior and longitudinal axes. n.s.: nonsignificant. Tasks
Linear displacements
Participants
Constant
Coefficient
Adjusted R2
F
Abduction
Medio-lateral
P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3 P1 P2 P3
0.4 0.5 −0.7 0.7 1.4 0.9 0.0 0.2 −0.4 −0.3 −1.2 −0.1 0.9 1.6 −0.5 0.9 n.s. 0.1 −0.5 −0.5 0.0 1.0 1.5 −0.2 −0.3 1.5 1.0
0.00 −0.02 0.01 −0.01 −0.01 −0.03 0.09 0.00 0.12 0.00 0.02 0.01 −0.02 −0.02 −0.01 0.02 n.s. 0.08 0.01 0.01 0.00 −0.02 −0.04 0.00 −0.01 −0.05 −0.02
0.07 0.26 0.35 0.07 0.17 0.73 0.83 0.02 0.94 0.19 0.49 0.16 0.61 0.42 0.15 0.60 n.s. 0.91 0.30 0.51 0.20 0.68 0.94 0.22 0.14 0.72 0.45
619.11 320.29 4301.35 602.05 186.40 21612.88 37062.02 17.70 126440.18 1954.02 966.92 943.73 13020.61 726.13 877.85 12508.08 n.s. 50182.00 887.81 1742.58 514.09 4555.60 25875.02 579.47 335.28 4224.87 1687.86
Antero-posterior
Longitudinal
Flexion
Medio-lateral
Antero-posterior
Longitudinal
Axial rotation
Medio-lateral
Antero-posterior
Longitudinal
* * * * * * * * * * * * * * * * * * * * * * * * * *
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
displacements during dynamic movements and their normal coupling with rotations. Glenohumeral displacements have been a controversial topic since their ranges of only a few millimeters were below the accuracy of electromagnetic or optoelectronic systems with skin markers (Veeger and van der Helm, 2007). However, measurement based on intracortical pin marker trajectories using optoelectronic system provided a measure of the linear displacement of the glenohumeral joint with an uncertainty lower than 0.15 mm. This is more accurate than techniques based on biplane fluoroscopic measurements (Bey et al., 2008). Therefore, measured displacements lower than 1 mm are considered as reliable. Moreover, glenohumeral displacements were measured from the linear displacement of the glenohumeral center of rotation located using a functional method contrarily to previous studies that were based on the geometrical center of the humeral head which is assumed to be the glenohumeral center of rotation. Indeed, glenohumeral joint center is often assumed to be the center of the humeral head and used to measure glenohumeral displacements. However, since the glenohumeral joint has six degrees of freedom, there is not a fixed center of rotation, but rather instantaneous centers of rotation during glenohumeral movements. The geometric center of the humeral head determined using sphere fitting does not necessarily take into account the mechanical properties of the glenohumeral joint. On the contrary, the center of rotation determined using the functional SCoRE method relies on the assumption that the glenohumeral joint is a quasi-perfect ball-andsocket joint. This method finds an optimal location which corresponds to a point with least displacements throughout the range of motion. Though the patterns of glenohumeral displacements were similar for the two methods, medial displacements were smaller with the glenohumeral center of rotation. From an anatomical point of view, 2 mm of medial displacement obtained using the method based on the glenohumeral joint center of rotation seems to be more representative of the humeral head displacements than 4 mm obtained using humeral head center given the narrow spacing between the humeral head and the glenoid. The proposed method may thus provide more realistic values of glenohumeral displacements. Though patterns of glenohumeral displacements were in some cases different between participants, the results showed that the patterns of glenohumeral displacements were similar across the repetitions of all the tasks for each subject. This result suggests that glenohumeral displacements are highly reproducible between trials of the same experimental condition. This result reinforces also that our method is accurate enough to measure glenohumeral displacements that are altered of only few tenths of millimeters between trials. Thus, using our method, the acquisition of a limited number of trials (n = 10) may be enough to measure representative kinematics of the glenohumeral joint. Whereas our study only focused on nearly-pure movements of elevation and rotation, shoulder degrees-of-freedom can interact, such that shoulder range of motion covers almost 65% of a sphere (Engin and Chen, 1986). Thus, our method that is not restricted in space and sampling rate could provide a strong benefit when assessing joint displacements during sports activities associated to anterior instability, daily living tasks and occupational activities. The experimental protocol is only restricted to the duration of the anesthesia, 2 h in our case, during which a large number of experimental conditions can be acquired. Up to one hundred trials have been collected, only a fraction is presented in the present study. The method is however highly invasive and the small sample of participants is the main limitation of our study and other similar studies based on pins inserted into the bones (Arndt et al., 2004; Braman et al., 2010; Houck et al., 2004). Our findings need to be corroborated by other studies based on larger sample of participants and on various movements before direct clinical applications. Nevertheless the results suggest that new pattern of glenohumeral displacements could be evidenced during dynamic tasks, especially such as coupling between displacements and rotations at the glenohumeral
1053
joint. Moreover, the heterogeneity in terms of age of our sample gave certainly more representative values of the living population compared to in-vitro studies where average age exceeds 70 years (San Juan and Karduna, 2010). As evidenced by the measure of the displacements along the mediolateral axis, the distance between the glenohumeral joint center of rotation and the glenoid cavity changes during movements. Medial and lateral displacements observed can be due to non-homogeneous curvature of the humeral head (Carey et al., 2000; Harrold and Wigderowitz, 2013). Medial displacements did not exceed + 2.4 mm which corresponds to the range already measured during arm static elevation (Bryce et al., 2010; Massimini et al., 2012) and rotation (Massimini et al., 2012). Medial displacements may also be due to compression of the labrum (Carey et al., 2000) provoked by compressive forces generated by rotator cuff muscles and tendons arranged in a half circle around the humeral head (Blasier et al., 1997; Wilk et al., 1997). Displacements in antero-posterior direction are of interest due to glenohumeral dislocation occurring in anterior direction in 95% of the cases (Cutts et al., 2009). The results showed that the humeral head moved maximally from + 2.8 mm forward to − 3.9 mm backward during dynamic elevations and rotations, which is in the range of displacements already measured during static poses (Bryce et al., 2010; Massimini et al., 2012). This result suggests that anteroposterior displacements may not be sensitive to the dynamic of the movement. Humeral head tended to displace anteriorly during abduction and posteriorly during flexion, which agrees with the alteration of antero-posterior shear forces with the plane of elevation during elevations (Kontaxis and Johnson, 2009). Further studies should consider other planes of elevation under dynamic tasks to determine the plane of elevation that minimize antero-posterior displacements and thereby limit antero-posterior instability for elevation rehabilitation exercises. Glenohumeral displacements along the longitudinal axis increased at low arm elevation angles before returning close to its initial position at maximum elevation angle for P2. This pattern corresponds to the pattern of longitudinal displacements measured during static (Massimini et al., 2012) and slow dynamic poses (Bey et al., 2008; Bishop et al., 2009). For P1 and P3, contrarily to the literature, values of upward displacements increase linearly with abduction angle during elevations with maximum values at maximum elevation. For these two participants, upward displacements were the largest measured displacements. This result obtained during dynamic tasks is of particular importance since upward displacements are one of the major extrinsic causes of the shoulder impingement syndrome (Deutsch et al., 1996; Sharkey and Marder, 1995; Williams et al., 2001). It suggests that even during dynamic daily-living tasks such as touching the head where the arm elevation can represent up to 77% of the maximum arm elevation (Lovern et al., 2010), large values of upward displacements may occur and reduce the subacromial space. This may also be the case during overhead handling tasks which could explain the high prevalence of supraspinatus tendon tear (Milgrom et al., 1995) since the acromiohumeral distance may be reduced. The results also suggest that longitudinal displacements are sensitive to the dynamic of the movement due to the increase of the ratio between the contribution of deltoids and rotator cuff muscles (Graichen et al., 2005; Reddy et al., 2000; Sharkey and Marder, 1995; Teyhen et al., 2008; Yamaguchi et al., 2000). Noteworthy, for P1 and P3, upward displacements were larger during abduction than during flexion. Further studies should consider other planes of elevation to determine the plane of elevation that limit upward displacements to reduce the risk of shoulder impingement for rehabilitation exercises. Maximum values of downward displacements were measured at maximum external rotation. This result reinforces the relevancy of this exercise to strengthen rotator cuff muscle while reducing the risk of subacromial impingement. Our method permitted also to evidence that glenohumeral displacements along the longitudinal axis were coupled with glenohumeral
1054
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055
rotation around the abduction axis. This result provides a new insight of glenohumeral kinematics that could find an application for the design of complete glenohumeral joint prostheses as well as for musculoskeletal computer simulation models. 5. Conclusions In summary, the method to acquire glenohumeral kinematics permitted to measure glenohumeral displacements with an uncertainty lower than 0.15 mm. Our results reaffirm that kinematics of healthy glenohumeral joint is a combination of displacements and rotations rather than pure rotations during dynamic tasks. During arm elevations and axial rotations, the glenohumeral displacements along longitudinal axis were linearly coupled with glenohumeral rotations around abduction axis. The largest displacements were observed at maximum elevation in the upward direction. Compared to previous results in the literature during static tasks, this result reinforces that glenohumeral displacements are sensitive to muscular dynamics, and further studies are needed to better characterize the kinematics of the glenohumeral joint during dynamic tasks. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiomech.2014.08.006. Acknowledgments This work was supported by the RRSSTQ (grant no. 101220). In addition, the MEDITIS training program in biomedical technologies (NSERC, CREATE program) supported the postdoctoral fellowship of the first author. None of these programs played a role in the present investigation. References Arndt, A., Westblad, P., Winson, I., Hashimoto, T., Lundberg, A., 2004. Ankle and subtalar kinematics measured with intracortical pins during the stance phase of walking. Foot Ankle Int. 25, 357–364. Bey, M.J., Zauel, R., Brock, S.K., Tashman, S., 2006. Validation of a new model-based tracking technique for measuring three-dimensional, in vivo glenohumeral joint kinematics. J. Biomech. Eng. 128, 604–609. Bey, M.J., Kline, S.K., Zauel, R., Lock, T.R., Kolowich, P.A., 2008. Measuring dynamic in-vivo glenohumeral joint kinematics: technique and preliminary results. J. Biomech. 41, 711–714. Bishop, J.L., Kline, S.K., Aalderink, K.J., Zauel, R., Bey, M.J., 2009. Glenoid inclination: In vivo measures in rotator cuff tear patients and associations with superior glenohumeral joint translation. J. Shoulder Elb. Surg. 18, 231–236. Blasier, R.B., Soslowsky, L.J., Malicky, D.M., Palmer, M.L., 1997. Posterior glenohumeral subluxation: active and passive stabilization in a biomechanical model. J. Bone Joint Surg. Am. 79, 433–440. Bolsterlee, B., Veeger, D.H., Chadwick, E.K., 2013. Clinical applications of musculoskeletal modelling for the shoulder and upper limb. Med. Biol. Eng. Comput. 51, 953–963. Bourne, D.A., Choo, A.M., Regan, W.D., MacIntyre, D.L., Oxland, T.R., 2007. Threedimensional rotation of the scapula during functional movements: an in vivo study in healthy volunteers. J. Shoulder Elbow Surg. 16, 150–162. Braman, J., Thomas, B., LaPrade, R., Phadke, V., Ludewig, P., 2010. Three-dimensional in vivo kinematics of an osteoarthritic shoulder before and after total shoulder arthroplasty. Knee Surg. Sports Traumatol. Arthrosc. 18, 1774–1778. Bryce, C.D., Davison, A.C., Okita, N., Lewis, G.S., Sharkey, N.A., Armstrong, A.D., 2010. A biomechanical study of posterior glenoid bone loss and humeral head translation. J. Shoulder Elbow Surg. 19, 994–1002. Buchler, P., Ramaniraka, N.A., Rakotomanana, L.R., Iannotti, J.P., Farron, A., 2002. A finite element model of the shoulder: application to the comparison of normal and osteoarthritic joints. Clin. Biomech. 17, 630–639. Carey, J., Small, C.F., Pichora, D.R., 2000. In situ compressive properties of the glenoid labrum. J. Biomed. Mater. Res. 51, 711–716. Chan, K.M., Maffulli, N., Nobuhara, M., Wu, J.J., 1996. Shoulder instability in athletes. The Asian perspective. Clin. Orthop. Relat. Res. 106–112. Cutts, S., Prempeh, M., Drew, S., 2009. Anterior shoulder dislocation. Ann. R. Coll. Surg. Engl. 91, 2–7. De Wilde, L.F., Verstraeten, T., Speeckaert, W., Karelse, A., 2010. Reliability of the glenoid plane. J. Shoulder Elbow Surg. 19, 414–422. Deutsch, A., Altchek, D.W., Schwartz, E., Otis, J.C., Warren, R.F., 1996. Radiologic measurement of superior displacement of the humeral head in the impingement syndrome. J. Shoulder Elb. Surg. 5, 186–193. Ehrig, R.M., Taylor, W.R., Duda, G.N., Heller, M.O., 2006. A survey of formal methods for determining the centre of rotation of ball joints. J. Biomech. 39, 2798–2809. Ehrig, R.M., Taylor, W.R., Duda, G.N., Heller, M.O., 2007. A survey of formal methods for determining functional joint axes. J. Biomech. 40, 2150–2157.
Engin, A.E., Chen, S.M., 1986. Statistical data base for the biomechanical properties of the human shoulder complex—I: kinematics of the shoulder complex. J. Biomech. Eng. 108, 215–221. Fang, Q., Boas, D.A., 2009. Tetrahedral mesh generation from volumetric binary and grayscale images. Biomedical Imaging: From Nano to Macro, 2009. ISBI '09. IEEE International Symposium on, pp. 1142–1145. Favre, P., Senteler, M., Hipp, J., Scherrer, S., Gerber, C., Snedeker, J.G., 2012. An integrated model of active glenohumeral stability. J. Biomech. 45, 2248–2255. Giphart, J.E., Brunkhorst, J.P., Horn, N.H., Shelburne, K.B., Torry, M.R., Millett, P.J., 2013. Effect of plane of arm elevation on glenohumeral kinematics: a normative biplane fluoroscopy study. J. Bone Joint Surg. Am. 95, 238–245. Graichen, H., Stammberger, T., Bonel, H., Karl-Hans, E., Reiser, M., Eckstein, F., 2000. Glenohumeral translation during active and passive elevation of the shoulder — a 3D open-MRI study. J. Biomech. 33, 609–613. Graichen, H., Hinterwimmer, S., Eisenhart-Rothe, R.v., Vogl, T., Englmeier, K.H., Eckstein, F., 2005. Effect of abducting and adducting muscle activity on glenohumeral translation, scapular kinematics and subacromial space width in vivo. J. Biomech. 38, 755–760. Harrold, F., Wigderowitz, C., 2013. Humeral head arthroplasty and its ability to restore original humeral head geometry. J. Shoulder Elbow Surg. 22, 115–121. Houck, J., Yack, H.J., Cuddeford, T., 2004. Validity and comparisons of tibiofemoral orientations and displacement using a femoral tracking device during early to mid stance of walking. Gait Posture 19, 76–84. Hudak, P.L., Amadio, P.C., Bombardier, C., 1996. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG). Am. J. Ind. Med. 29, 602–608. Institute, S C.a.I., 2014. “Seg3D” Volumetric Image Segmentation and Visualization. Scientific Computing and Imaging Institute (SCI). Jackson, M., Michaud, B., Tetreault, P., Begon, M., 2012. Improvements in measuring shoulder joint kinematics. J. Biomech. 45, 2180–2183. Karduna, A.R., Williams, G.R., Iannotti, J.P., Williams, J.L., 1998. Total shoulder arthroplasty biomechanics: a study of the forces and strains at the glenoid component. J. Biomech. Eng. 120, 92–99. Karduna, A.R., McClure, P.W., Michener, L.A., Sennett, B., 2001. Dynamic measurements of three-dimensional scapular kinematics: a validation study. J. Biomech. Eng. 123, 184–190. Kontaxis, A., Johnson, G.R., 2009. The biomechanics of reverse anatomy shoulder replacement — a modelling study. Clin. Biomech. 24, 254–260. Landry, C., De Guise, J.A., Dansereau, J., Labelle, H., Skalli, W., Zeller, R., et al., 1997. Computer graphic analysis of the three dimensional deformities of scoliotic vertebrae. Ann. Chir. 51, 868–874. Liu, A., Nester, C.J., Jones, R.K., Lundgren, P., Lundberg, A., Arndt, A., et al., 2012. Effect of an antipronation foot orthosis on ankle and subtalar kinematics. Med. Sci. Sports Exerc. 44, 2384–2391. Lovern, B., Stroud, L.A., Ferran, N.A., Evans, S.L., Evans, R.O., Holt, C.A., 2010. Motion analysis of the glenohumeral joint during activities of daily living. Comput. Methods Biomech. Biomed. Eng. 13, 803–809. Ludewig, P.M., Phadke, V., Braman, J.P., Hassett, D.R., Cieminski, C.J., LaPrade, R.F., 2009. Motion of the shoulder complex during multiplanar humeral elevation. J. Bone Joint Surg. 91, 378–389. Massimini, D.F., Boyer, P.J., Papannagari, R., Gill, T.J., Warner, J.P., Li, G., 2012. In-vivo glenohumeral translation and ligament elongation during abduction and abduction with internal and external rotation. J. Orthop. Surg. Res. 7–29. Meskers, C.G.M., van der Helm, F.C.T., Rozendaal, L.A., Rozing, P.M., 1997. In vivo estimation of the glenohumeral joint rotation center from scapular bony landmarks by linear regression. J. Biomech. 31, 93–96. Milgrom, C., Schaffler, M., Gilbert, S., van Holsbeeck, M., 1995. Rotator-cuff changes in asymptomatic adults. The effect of age, hand dominance and gender. J. Bone Joint Surg. (Br.) 77, 296–298. Monnet, T., Thouze, A., Pain, M.T., Begon, M., 2012. Assessment of reproducibility of thigh marker ranking during walking and landing tasks. Med. Eng. Phys. 34, 1200–1208. Reddy, A.S., Mohr, K.J., Pink, M.M., Jobe, F.W., 2000. Electromyographic analysis of the deltoid and rotator cuff muscles in persons with subacromial impingement. J. Shoulder Elbow Surg. 9, 519–523. San Juan, J.G., Karduna, A.R., 2010. Measuring humeral head translation using fluoroscopy: a validation study. J. Biomech. 43, 771–774. Senk, M., Cheze, L., 2006. Rotation sequence as an important factor in shoulder kinematics. Clin. Biomech. 21 (Suppl. 1), S3–S8. Sharkey, N.A., Marder, R.A., 1995. The rotator cuff opposes superior translation of the humeral head. Am. J. Sports Med. 23, 270–275. Teyhen, D.S., Miller, J.M., Middag, T.R., Kane, E.J., 2008. Rotator cuff fatigue and glenohumeral kinematics in participants without shoulder dysfunction. J. Athl. Train. 43, 352–358. Veeger, H.E., van der Helm, F.C., 2007. Shoulder function: the perfect compromise between mobility and stability. J. Biomech. 40, 2119–2129. von Eisenhart-Rothe, R.M., Jager, A., Englmeier, K.H., Vogl, T.J., Graichen, H., 2002. Relevance of arm position and muscle activity on three-dimensional glenohumeral translation in patients with traumatic and atraumatic shoulder instability. Am. J. Sports Med. 30, 514–522. Walch, G., Edwards, T.B., Boulahia, A., Boileau, P., Mole, D., Adeleine, P., 2002. The influence of glenohumeral prosthetic mismatch on glenoid radiolucent lines results of a multicenter study. J. Bone Joint Surg. 84, 2186–2191. Wilk, K.E., Andrews, J.R., Arrigo, C.A., 1997. The physical examination of the glenohumeral joint: emphasis on the stabilizing structures. J. Orthop. Sports Phys. Ther. 25, 380–389. Williams Jr., G.R., Wong, K.L., Pepe, M.D., Tan, V., Silverberg, D., Ramsey, M.L., et al., 2001. The effect of articular malposition after total shoulder arthroplasty on glenohumeral
F. Dal Maso et al. / Clinical Biomechanics 29 (2014) 1048–1055 translations, range of motion, and subacromial impingement. J. Shoulder Elbow Surg. 10, 399–409. Wu, G., van der Helm, F.C., Veeger, H.E., Makhsous, M., Van Roy, P., Anglin, C., et al., 2005. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion—part II: shoulder, elbow, wrist and hand. J. Biomech. 38, 981–992.
1055
Yamaguchi, K., Sher, J.S., Andersen, W.K., Garretson, R., Uribe, J.W., Hechtman, K., et al., 2000. Glenohumeral motion in patients with rotator cuff tears: a comparison of asymptomatic and symptomatic shoulders. J. Shoulder Elbow Surg. 9, 6–11.