The moment arms and lines of action of subscapularis after the Latarjet procedure

The Latarjet procedure is an established surgical treatment for recurrent glenohumeral joint instability with glenoid bone loss. Intraoperatively, the conjoint tendon and its attachement on the coracoid bone graft is routed through a split in subscapularis where the graft is fixed to and augments the anteroinferior glenoid. The objective of this in vitro study was to quantify the influence of glenohumeral joint position and conjoint tendon force on the lines of action and moment arms of subscapularis muscle sub‐regions after Latarjet surgery. Eight fresh‐frozen, entire upper extremities were mounted onto a testing apparatus, and a cable‐pulley system was used to apply physiological muscle loading to the major shoulder muscles. The lines of action and moment arms of four subregions of subscapularis (superior, mid‐superior, mid‐inferior, and inferior) were quantified radiographically with the conjoint tendon unloaded and loaded while the shoulder was in (i) 0° abduction (ii) 90° abduction (iii) 90° abduction and full external rotation (ABER), and (iv) the apprehension position, defined as ABER with 30° horizontal extension. Conjoint tendon loading after Latarjet surgery significantly increased the inferior inclination of the lines of action of the mid‐inferior and inferior subregions of subscapularis in the scapular plane in ABER and apprehension positions (p < 0.001), as well as decreased the horizontal flexion moment arm of the inferior subscapularis (p = 0.040). Increased subscapularis inferior inclination may ultimately increase inferior joint shear potential, while smaller horizontal flexion leverage may reduce joint flexion capacity. The findings have implications for Latarjet surgical planning and postoperative rehabilitation prescription.

2][3][4][5] The Latarjet procedure, first described in 1954, is one of the established bone block procedures for the treatment of recurrent anterior shoulder instability with glenoid bone loss in excess of 15%. 6The coracoid process is osteotomized and transferred to the anteroinferior glenoid neck to increase the surface area of the glenoid concavity. 7The conjoint tendon subsequently acts as a sling over the humerus, resisting anterior translation with the arm is abducted and external rotated, lowering the inferior subscapularis fibers and creating a hammock-effect that reinforces a weakened capsule and ligaments anteroinferiorly.
Early iterations of the Latarjet procedure completely detached the subscapularis to expose the anterior joint capsule; however, this was known to compromise the long-term function of this muscle and glenohumeral joint behavior, 8 since the subscapularis is critical to maintaining anterior shoulder stability and internal rotation force.The preferred Latarjet technique now employs a horizontal split through the subscapularis muscle fibers in which the conjoint tendon passes directly from the coracoid graft. 9Contact between the conjoint tendon and the subscapularis muscle fibers imparts force on these fibers as the conjoint tendon passes through the muscle split; however, the extent of this depends on the position and orientation of the humerus, since this directly influences tension in the conjoint tendon as well as the line of pull of the conjoint tendon on the subscapularis.While cadaveric studies have documented the sling effect of the Latarjet procedure on glenohumeral joint stability, 10,11 the extent to which the subscapularis muscle path is altered as a consequence of conjoint tendon force during shoulder joint motion remains poorly understood.
The function of a muscle about a joint in which it spans is dependent on the muscle's moment arm or leverage, for instance, as an elevator or depressor. 12,13The line of action of the muscle, which determines whether the force it produces at a joint is compressive or shear, also dictates function, including capacity to stabilize or destabilize the joint. 14Both the line of action and moment arm of a muscle vary with joint position due to the change in the spatial relationship between the muscle's origin and insertion, and muscletendon wrapping.This ultimately gives rise to a prominent task dependency of muscle function.The morphology and biomechanical function of subscapularis has been quantified in the native shoulder during activities of daily living, however, the extent to which Latarjet surgery alters the path of its fibers and therefore its moment arms and line of action about the glenohumeral joint has not been quantified to date.
The objective of this in vitro study was to evaluate the lines of action and moment arms of the subscapularis in the shoulder after Latarjet surgery, and quantify the influence of changes in joint position and conjoint tendon loading during abduction, abduction and external rotation (ABER), and in the position of shoulder apprehension.Changes in the fiber directions of subscapularis due to the conjoint tendon passing through the subscapularis muscle split are clinically relevant, since this may influence the capacity of this muscle to contribute to joint stability in the unstable shoulder.

| Specimen preparation
Eight fresh-frozen, entire upper extremities were obtained from human cadavers (four male, four female; mean age: 69.3 ± 8.2 years; mean weight: 78.4 ± 4.7 kg).6][17] All specimens were both radiographically and arthroscopically screened for macroscopic degenerative changes to the shoulder that included glenohumeral joint osteoarthritis, rotator cuff tears, fracture, and previous surgery, with care taken to verify the integrity of the subscapularis.Specimens were thawed at room temperature 24 h prior to dissecting and testing.
All skin and subcutaneous tissue proximal to the glenohumeral joint were removed by sharp dissection.Four functionally distinct subregions of the subscapularis were identified (superior, mid-superior, mid-inferior, inferior) 18 by marking the superior and inferior boundaries of the proximal attachment on the scapula and identifying four equal-sized muscle bundles.Also identified were the anterior deltoid (clavicular fibers), middle deltoid (acomial fibers), posterior deltoid (posterior scapular spine fibers), supraspinatus, infraspinatus, and teres minor.
Number-5 Ethibond suture (Ethicon) was secured to the tendons of insertion of each muscle-tendon unit while ensuring the proximal and distal muscle-tendon attachments remained intact.The wrist of each specimen was fused in extension, and the elbow in 90°of flexion using Steinmann pins, the latter of which was later used to manipulate the specimen.Ethical approval was granted from the Health Sciences Human Ethics Sub-Committee, University of Melbourne (0608480).

| Latarjet surgery
A simulated glenoid bone loss of 20% by area was applied to all specimens in the critical glenoid area involved in anterior instability. 19This bone loss was calculated from the best-fit-circle applied to the inferior two thirds of the entire glenoid surface area, which was assumed circular.
Area calculations were estimated based on glenoid width, which was measured with digital calipers.The osteotomy line along the defined glenoid bone loss area, which was parallel to the line between the supraand infra-glenoidal tubercles, was made with an oscillating saw.The Latarjet procedure was performed according to Walch and Boileau's 9 refinement of the original technique described by Latarjet. 20The coracoid was osteotomized at its base through the use of bent chisels.The glenoid neck was exposed through a horizontal split of the subscapularis muscle which was taken at the mid-level of muscle bundle as measured at the level of the glenoid margin (Figure 1).The harvested coracoid was positioned flush with the glenoid plane at the 2-to 5-o'clock position in a right shoulder and fixed with two bicortical cannulated screws and a drill guide from the Bristow-Latarjet Instability Shoulder System (Mitek, Depuy Synthes, Johnson & Johnson).The coracoacromial ligament was sutured to the most medial aspect of the incised capsule.

| Experimental protocol
Specimens were rigidly mounted to a testing apparatus using an external fixation device (De Puy Synthes, Johnson & Johnson, Warsaw, USA) with the plane of the scapula oriented vertically (Figure 2).Pins were screwed anteriorly through the lateral scapula spine and posteriorly through the medial scapula spine, and the two pins were then secured to four carbon-fiber shafts that were attached to the base of the testing apparatus using three-dimensional (3D) printed shaft supports.Radio-opaque braided cables were tied to the sutures of each of the four subscapularis muscle-tendon unit subregions and threaded through the approximate centroid of the muscle belly, while radiolucent nylon lines were tied to the sutures of the deltoid, supraspinatus, infraspinatus, and teres minor muscle-tendon units (Figure 3).To maintain constant tension in each muscle-tendon unit, and preserve each muscle's line of action, the cables and nylon lines were passed through 3 mm holes drilled at the centroid of tendon origin on the scapula.The lines then ran through nylon-lined holes in a backing plate to free-weights to simulate muscle-tendon tension.The weight used to load each individual muscle-tendon unit was calculated using an EMG-driven model as described previously. 22,23To simulate loading of the conjoint tendon, a nylon line was sutured to the distal end of the conjoint tendon after the brachialis was released from the radius.The nylon line passed through a pulley that was fastened to the radius, and a free weight attached, thus allowing load to be directly applied to the conjoint tendon along its physiological line of action.F I G U R E 2 Upper limb testing apparatus used for muscle line of action and moment arm measurements, and illustration of the abduction and external rotation position.Specimens were mounted to the apparatus frame using an ExFix external fixator, with two horizontal pins passing through the scapula spine, and the pins supported by four carbon fiber posts mounted to the rig frame using 3D printed shaft supports.Constant loads were applied to the shoulder muscle-tendon units using a cable-pulley-weight system that applied constant tension along each muscletendon unit's line of action.An additional pulley was attached to the distal humerus and used to apply conjoint tendon loading using a cable and free weight.To achieve this, nylon lines sutured to each tendon ran along the muscle to a hole passing through the centroid of the muscletendon unit origin, and the lines then passed through a perforated plate to a free-weight.The upper limb was supported by grasping an elbow fusion Steinman pin with a clamp.Each upper limb was passively positioned in (i) 0°abduction (ii) 90°abduction (iii) 90°ABER, and (iv) the apprehension position, which was defined as ABER with approximately 30°of horizontal extension.These joint positions were chosen to cover the range of motion in elevation, as well as ABER and apprehension, which are configurations associated with the conjoint tendon sling effect. 21ecimens were held in static equilibrium by clamping the elbow Steinmann pin to a rigid tripod, and the relevant physiological load in each muscle-tendon unit was calculated using the EMG-driven model.Joint angles were verified radiographically in the scapular plane and transverse planes using X-ray fluoroscopy (Fluoroscan InSight 2; Hologic Inc.).
At each joint position, the lines of action of the four subregions of the subscapularis were calculated radiographically from the path of the radio-opaque cables (Figure 4) and expressed relative to the glenoid in the scapular and transverse planes, as described previously 16 (Figure 5).This was achieved by digitizing landmarks on radiographic images to delineate each muscle's line of action and the plane of the glenoid.This included two points at the boundaries of the glenoid surface (Figure 4A), with the glenoid plane defined by the line adjoining these points (Figure 4B).Each muscle's line of action was defined as the unit vector between a distal and proximal via point on the muscle-tendon path, which were also digitized on radiographs.The distal via point was defined by the point at which the distal muscle-tendon unit loses contact with the humerus.The proximal via point was defined by the point in which the muscletendon unit begins to visibly change direction or "deform" as a result of contact with the conjoint tendon.When no perceivable change in muscle path occurs as a result of conjoint tendon contact, the proximal via point was defined as the approximate centroid of origin of the muscle's attachment area on the scapula.Muscle lines of action were expressed relative to a 2D glenoid coordinate system, which was established using the glenoid plane and the vector normal to this plane (Figure 4C).Vector projections of the line of action of each subscapularis subregion were subsequently calculated in both scapular and transverse planes.
The moment arm of each subscapularis muscle-tendon unit was calculated using the geometric method at each upper limb position.
To achieve this, a series of points on the outer boundary of the humeral head were digitized on each radiograph (Figure 4A).A geometric circle of best fit was then used to define the humeral head boundary, and the joint center was defined as the circle's center (Figure 4B).A muscle's moment arm in the scapular and transverse F I G U R E 3 Representative shoulder radiograph illustrating the four subregions of subscapularis, the superior, mid-superior, mid-inferior, inferior subregions, and how their lines of action were delineated from radio-opaque cables that threaded through the centroid of each muscle subregion.Shown are two points (red) used to define the line of action (red dashed line) of each muscle subregion when the conjoint tendon was unloaded.
F I G U R E 4 Scapular-plane radiographs illustrating the muscle line of action and moment arm calculation workflow used in the present study, which included (A) digitization of each subscapularis line of action (red dots), the superior and inferior margins of the glenoid (yellow dots), and boundary of the humeral head (cyan dots), (B) the geometric fit of a circle to the humeral head boundary (cyan) and identification of the joint center defined by the center of the circle fitted to the humeral head boundary (cyan dot), the line between the superior and inferior margins of the glenoid defining the glenoid plane (yellow) and a line defining each subscapularis subregion line of action (red), (C) the line of action vector (red arrow) expressed relative to the glenoid coordinate system, which was defined by the vector parallel to the glenoid plane (yellow arrow) and its perpendicular (green arrow), and (D) moment arm calculation using the perpendicular distance between the subscapularis subregion line of action and the humeral head center (cyan).
planes was defined as the perpendicular distance between its line of action and the joint center of rotation in the given plane (Figure 4D).
In the scapular plane for all joint positions, a positive moment arm represents abduction capacity, while a negative moment arm represents adduction capacity.In the transverse plane at 0°a bduction, a positive moment arm represents external rotation capacity while a negative moment arm internal rotation capacity; for 90°abduction, ABER, and the apprehension position, a positive moment arm represents horizontal extension while a negative moment arm represents horizontal flexion capacity.The joint center of rotation in each plane was approximated as the center of a circle fitted to the humeral head.The glenohumeral joint was maintained congruent and centered at each joint position through application of the muscle-tendon loading and subsequent immobilization of the humerus using the elbow Steinman pin clamp.Joint congruency was verified radiographically at each joint position.All subscapularis subregion lines of action and moment arms were calculated for each joint position with the conjoint tendon load set to 0, 20, and 40 N.This range of forces was estimated from a musculoskeletal model for activities of daily living, 24 and is similar to that adopted in a previous analysis of conjoint tendon biomechanical function. 25l images were digitized and analyzed by a single investigator.
To examine repeatability of the digitization and analysis procedures, forty random data points across all specimens, loads, and subscapularis muscle regions were selected and re-digitized by the single investigator.The mean difference and 95% limits of agreement were calculated between the repeat measurements for estimated moment

| Muscle lines of action
Conjoint tendon loading after Latarjet surgery affected the scapular plane lines of action of the mid-inferior and inferior subregions of subscapularis (p < 0.001) (Table 1), and this occurred with the shoulder in ABER and the apprehension position (Figure 6).During ABER, the mid-inferior subregion of subscapularis was significantly more inferiorly inclined after application of 20 N (mean difference: Bone-embedded scapular reference frame used in this study (A) and the projected angles of muscle force vectors calculated with respect to the scapular reference frame defined the scapular plane (B) and the transverse plane (C).The negative x axis corresponds to the medial (compressive) direction, the y axis to the anterior direction, and the z axis to the superior direction.Directions of all muscle force vectors were measured anti-clockwise from the mediolateral (x) axis.Adapted from Ackland et al., 16 with permission from Elsevier.
transverse plane lines of action were observed as a consequence of conjoint tendon loading (p > 0.05) (Table 2).

| DISCUSSION
The present study showed that conjoint tendon loading following Latarjet surgery had a significant influence on the scapular plane lines of action of the subscapularis muscle subregions inferior to the subscapularis muscle split.This was most evident when the humerus was placed at the upper limits of shoulder motion range (ABER and apprehension positions), which significantly distorted the inferior and mid-inferior fibers of subscapularis downward as a consequence of the sling effect of the conjoint tendon in these joint positions.The overall effect of this was a net increase in the inferior inclination of the lines of action of these subregions.Conjoint tendon loading also caused a decrease in the horizontal flexion moment arm of the inferior subregion of subscapularis when the arm was positioned in the neutral position.The magnitude of changes in the lines of action and moment arms of the subscapularis below the level of the split increased with conjoint tendon force.
The increase in the inferior inclination of the subscapularis subregions below the level of the split has an effect of increasing the inferior shear potential of subscapularis, which may ultimately increase the contribution of these fibers to inferior glenohumeral joint shear.This is because a muscle's contribution to compression or shear at a joint is equal to its line of action multiplied by its muscle force, the latter of which may be associated with active or passive force-generation but cannot typically be measured noninvasively.Subscapularis, as well as being an internal rotator of the humerus, provides antagonistic function by counteracting the superior shear potential of the deltoid via a scapular-plane force couple.If subscapularis muscle force does not substantially decrease following Latarjet surgery, inferior glenohumeral shear force is likely to be increased as a consequence of the sling effect.This may ultimately increase resistance to the upward pull of the deltoid on the humerus. 26,27While weakness of subscapularis may occur postoperatively, 28,29 resistive internal rotation exercises are likely to strengthen this muscle and increase its overall contribution to glenohumeral joint compression.
In the shoulder following Latarjet surgery, we showed that by increasing conjoint tendon force, a significant decrease in the internal rotation moment arms of the inferior subregion of subscapularis in the neutral shoulder position occurred.As a consequence, capacity of these muscle fibers to produce an internal rotation moment decreases due to the sling effect.Since the subscapularis works in concert with the posterior rotator cuff muscles to produce glenohumeral joint compression via a transverse-plane force couple, its ability to produce an antagonistic force and moment to oppose the posterior rotator cuff is reduced.Clinically, such a change to one of four subregions of the subscapularis is unlikely to alter joint function T A B L E 1 Scapular plane lines of action (°) for the subscapularis, including its superior, mid-superior, mid-inferior and inferior subregions.Note: Data are provided for 0°abduction, 90°abduction, ABER, and apprehension for the shoulder after Latarjet surgery when 0, 20, and 40 N of conjoint tendon loading is applied.Mean and standard deviation are provided, and significant differences in lines of action as a result of joint position, conjoint tendon force, and interaction effects are also indicated.
substantially; however, the deformation of the muscle fibers inferior to the muscle split that we observed may adversely affect normal muscle force production, which may further reduce glenohumeral joint compressive function.This has implications for the location of the subscapularis split, suggesting that a more superior split location may impact more of the subscapularis muscle fibers through the downward pull of the conjoint tendon sling.
[17] Furthermore, moment arms and lines of action were shown to vary across different subregions of subscapularis, and across all joint positions and load configurations (Tables 1-4), particularly in the scapular plane.In general, the scapular plane lines of action of subscapularis became progressively more inferiorly inclined from the superior subregion to the inferior subregion, reflecting the line of pull of the subregions across the subscapularis fossa on the anterior scapula.The lines of action of subscapularis were at their minimal inferior inclination at 0°of abduction and maximum inferior inclination during ABER.Collectively, these findings suggest that both elevation angle and axial rotation of the humerus significantly Postoperative muscle line of action angles for the superior, mid-superior, mid-inferior and inferior subregions of subscapularis defined in the scapular plane (A) and transverse plane (B).Data are provided when the shoulder is positioned in 0°abduction, 90°abduction, abduction and external rotation (ABER) and the apprehension position, as well as when the conjoint tendon is unloaded (light gray boxes), in 20 N of tension (dark gray boxes) and 40 N of tension (black boxes).Scapular plane lines of action above 180°are superiorly inclined, while scapular plane lines of action below 180°are inferiorly inclined.Transverse plane lines of action above 180°are posteriorly inclined, while transverse plane lines of action below 180°are anteriorly inclined.
influence the spatial position of the distal tendonous insertions of subscapularis relative to the scapula and wrapping of the subscapularis muscle-tendon unit.This is clinically relevant, as this influences the ability of the muscle to generate a joint moment and create shear and compressive glenohumeral joint force.
The biomechanical effect of conjoint tendon loading following the Latarjet procedure observed in this study is in general agreement with previous biomechanical and clinical studies. 9,10,30The Latarjet procedure produces two stabilizing mechanisms: The "sling effect" and the "bone-block effect." 9,31While the latter is a purely static stabilizing function (i.e., increasing glenoid surface area), the sling effect is created by the conjoint tendon re-orientating the subscapularis fibers inferior to the muscle split and pushing the humeral head posteriorly.This is a dynamic process with a primary stabilizing mechanism in end-range arm positions such as ABER. 10 We observed in ABER and the apprehension arm positions a net increase in the inferior inclination of the lines of action of the subscapularis muscle subregions inferior to the subscapularis muscle split.However, the success of the Latarjet procedure in restoring glenohumeral stability by altering the line of pull of subscapularis fibers inferior to the muscle split is matter of concern regarding the potential loss of glenohumeral rotation and strength.Recent clinical studies have shown that the Latarjet procedure results in loss of active internal and external rotation capacity and strength. 29,32A B L E 2 Transverse plane lines of action (°) for the subscapularis, including its superior, mid-superior, mid-inferior and inferior subregions.Note: Data are provided for 0°abduction, 90°abduction, ABER and apprehension for the shoulder after Latarjet surgery when 0, 20, and 40 N of conjoint tendon loading is applied.Mean and standard deviation are provided, and significant differences in lines of action as a result of joint position, conjoint tendon force, and interaction effects are also indicated.
T A B L E 3 Scapular plane abduction moment arms (mm) for the subscapularis, including its superior, mid-superior, mid-inferior and inferior subregions.Note: Data are provided for 0°abduction, 90°abduction, ABER and apprehension for the shoulder after Latarjet surgery when 0, 20, and 40 N of conjoint tendon loading is applied.Mean and standard deviation are provided, and significant differences in moment arms as a result of joint position, conjoint tendon force, and interaction effects are also indicated.Positive moment arms represent abduction, while negative moment arms represent adduction.
Although we did not study the effect of conjoint loading following the Latarjet procedure on external rotation, we found a significant decrease in the internal rotation moment arms as a consequence of the sling effect of the inferior subregion of the subscapularis.
We expect that different subscapularis split locations will influence changes in muscle moment arms and lines of action postoperatively.For instance, a more superiorly positioned split is likely to create a larger sling, but deform more muscle belly fibers under the inferior-directed pull of the conjoint tendon.While a larger sling may confer greater joint stability, this may ultimately increase the extent of contact-induced impairment through the muscle.In contrast, an inferiorly positioned split will reduce the sling effect, but may result in larger muscle deformation across fewer muscle fibers, since less of the muscle belly is available to resist conjoint tendon loading.Future studies ought to investigate the effect of split location on subscapularis muscle mechanics after the Latarjet procedure.
There are a number of limitations of this study that ought to be considered.First, the glenohumeral joint was held congruent and centered during testing where possible, however, some joint translations may have influenced calculations of muscle lines of action and moment arms, particularly for ABER and apprehension where the glenohumeral joint had a greater tendency to sublux anteriorly and inferiorly.Also, the glenohumeral joint center of rotation was assumed coincident with the geometric center of a circle fitted to the humeral head in each plane, and in practice during dynamic movements, the axis of rotation changes dynamically with muscle action and scapula position, which may influence calculation of muscle moment arms.It is likely that errors in joint center of rotation position may lead to a commensurate change in moment arm magnitudes.Nonetheless, Veeger et al. 33 found that a sphere fitted to the humeral head produced small errors for joint center estimation compared to an instantaneous helical axis approach (<0.4 mm).The same authors also showed that identification of the glenohumeral joint center in three-dimensions, using the instantaneous helical axis or sphere center approach, resulted in similar joint center location compared with 2D geometric circle fit to the humeral head in radiographic images. 34cond, morphological variability in subscapularis anatomy In conclusion, the present study showed that increasing conjoint tendon loading after Latarjet surgery mechanically deforms the subscapularis muscle fibers inferior to the muscle split, which impacts the lines of action and moment arms of this region of the muscle.
This was most evident in ABER and apprehension, with the conjoint tendon sling effect on the humerus causing a net increase in the inferior inclination of the lines of action of the inferior fibers of subscapularis.An increase in inferior shear potential may increase capacity for the subscapularis to resist the superior glenohumeral joint shear generated by the deltoid.A loss in horizontal flexion moment arm of the inferior subregion of subscapularis observed in T A B L E 4 Horizontal abduction moment arms (mm) for the subscapularis, including its superior, mid-superior, mid-inferior and inferior subregions.Note: Data are provided for 0°abduction, 90°abduction, ABER, and apprehension for the shoulder after Latarjet surgery when 0, 20, and 40 N of conjoint tendon loading is applied.Mean and standard deviation are provided, and significant differences in moment arms as a result of joint position, conjoint tendon force, and interaction effects are also indicated.Positive moment arms represent horizontal abduction (anteriorly).

F I G U R E 1
Illustration of the subscapularis in the native shoulder during abduction and external rotation (ABER) (A), the Latarjet in the neutral arm position, showing the coracoid bone block fixed to the glenoid (B), the creation of a sling beneath the anteroinferior capsule in the ABER position, and the resulting inferior displacement of the lower half of subscapularis by the conjoint tendon (C).Adapted from Mattern et al., 21 with permission from AME Publishing Company.
Two-way repeated values analysis of variance was performed for lines of action and moment arms of each muscle tendon unit after the Latarjet procedure.The independent variables were joint position (i.e., 0°abduction, 90°abduction, ABER, and the apprehension position) and conjoint tendon force (0, 20, and 40 N), while the dependent variables were muscle lines of action and muscle moment arms in each plane.Interactions between the independent variables were also quantified, and standard deviation and 95% confidence intervals (CIs) were used as a measure of the dispersion of the results.Games-Howell post-hoc tests were employed for nonparametric data, and Shapiro-Wilk tests was used to assess normality of data.Level of statistical significance was defined as p < 0.05.

6 .
7°, 95% CI: [0.7, 12.8], p = 0.034) and 40 N of conjoint tendon force (mean difference: 10.0°, 95% CI: [−0.7, 19.3], p = 0.038) compared to that in the unloaded tendon state.The inferior subregion of subscapularis was similarly significantly more inferiorly inclined after application of 20 N (mean difference: 7.7°, 95% CI: [3.1, 12.3], p = 0.006) and 40 N of conjoint tendon force (mean difference: 11.2°, 95% CI: [6.6, 15.7], p = 0.001).The changes in scapular plane lines of action as a result of conjoint tendon force were significant but less substantial in the apprehension position, with inferiorization of the mid-inferior line of action after application of 20 N (mean difference: 2.9°, 95%CI: [0.9, 4.9], p = 0.013), and 40 N (mean difference: 3.6°, 95% CI: [0.5, 6.8], p = 0.030).No statistically significant changes in may have increased dispersion of muscle line of action and moment arm data; however, we minimized this source of error by standardizing the definition of muscle subregions across each specimen, ensuring joint congruency, and expressing data relative to a common joint coordinate system.Third, while statistically significant differences in muscle moment arms and lines of action were detected, further research is required to elucidate clinically relevant changes in muscle and joint function as a consequence of muscle lines of action and moment arms.Finally, the present study considered a variety of conjoint tendon loading scenarios; however, the actual force generated in this tendon remains unknown and is likely to vary depending on shoulder joint position and the nature of external forces applied to the upper limb.Nonetheless, this study provides a preliminary assessment of the effects of conjoint tendon unloading and loading through a range of physiologically relevant loads for each upper limb configuration.