Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Med Sci Sports Exerc. 2021 Nov 1;53(11):2354–2362. doi: 10.1249/MSS.0000000000002708

Muscle Contraction Has a Reduced Effect on Increasing Glenohumeral Stability in the Apprehension Position

Constantine P Nicolozakes 1,2,3, Daniel Ludvig 1,2, Emma M Baillargeon 1,2,4, Eric J Perreault 1,2,5, Amee L Seitz 4
PMCID: PMC8516675  NIHMSID: NIHMS1706326  PMID: 34033623

Abstract

Purpose:

Glenohumeral instability accounts for 23% of all shoulder injuries among collegiate athletes. The apprehension position—combined shoulder abduction and external rotation—commonly reproduces symptoms in athletes with instability. Rehabilitation aims to increase glenohumeral stability by strengthening functional positions. However, it is unclear how much glenohumeral stability increases with muscle contraction in the apprehension position. The purpose of this study was to determine if the ability to increase translational glenohumeral stiffness, a quantitative measure of glenohumeral stability, with muscle contraction is reduced in the apprehension position.

Methods:

Seventeen asymptomatic adults participated. A precision-instrumented robotic system applied pseudorandom, anterior-posterior displacements to translate the humeral head within the glenoid fossa and measured the resultant forces as participants produced isometric shoulder torques. Measurements were made in neutral abduction (90° abduction/0° external rotation) and apprehension (90° abduction/90° external rotation) positions. Glenohumeral stiffness was estimated from the relationship between applied displacements and resultant forces. The ability to increase glenohumeral stiffness with increasing torque magnitude was compared between positions.

Results:

On average, participants increased glenohumeral stiffness from passive levels by 91% in the neutral abduction position and only 64% in the apprehension position while producing 10% of maximum torque production. The biggest decrease in the ability to modulate glenohumeral stiffness in the apprehension position was observed for torques generated in abduction (49% lower, P<0.001) and horizontal abduction (25% lower, P<0.001).

Conclusion:

Our results demonstrate that individuals are less able to increase glenohumeral stiffness with muscle contraction in the apprehension position compared to a neutral shoulder position. These results may help explain why individuals with shoulder instability more frequently experience symptoms in the apprehension position compared to neutral shoulder positions.

Keywords: shoulder instability, joint stiffness, neuromuscular control, overhead athletics

INTRODUCTION

Glenohumeral instability is broadly defined as pain and disability associated with an inability to keep the humeral head properly centered within the glenohumeral joint (1). Glenohumeral instability accounts for 23% of all shoulder injuries among collegiate athletes, affecting a wide variety of athletes participating in high-demand, upper-extremity sports (2). Athletes such as swimmers (3), gymnasts (4), weightlifters (5), and football players (6) are all at risk to experience glenohumeral instability due to the upper-extremity demands of their respective sports. Male athletes are over twice as likely to experience glenohumeral instability as female athletes (2, 7). Overall, less than a third of collegiate athletes who experience an in-season instability injury return to and finish their season without experiencing recurrent instability (8).

The apprehension position, defined as combined shoulder abduction and external rotation, commonly reproduces symptoms of glenohumeral instability (9). Female individuals are twice as likely as male individuals to experience symptoms of instability in this position (10). Overhead athletes complete many tasks in the apprehension position relevant to their respective sports (5, 11, 12), limiting their capacity to participate in sporting activity if they experience symptoms of instability. The goal of rehabilitation programs, which play in important role in the treatment for instability (13-15), is to reduce instability-related symptoms by augmenting the muscular contributions to glenohumeral stability (16). After first emphasizing strengthening in neutral shoulder positions, these programs progress to strengthening in more provocative positions (e.g., the apprehension position) that are necessary for function (14, 15). Such rehabilitation allows injured individuals to execute overhead tasks with fewer symptoms of instability (14, 15), which would facilitate an athlete’s return to sport participation following injury.

Clinical interpretations of glenohumeral stability can be quantified by estimating the translational stiffness of the glenohumeral joint, or the relationship between humeral head translation and the forces generated in response (17). Passive assessments of translational stiffness during physical examination of the shoulder are critical to both the diagnosis of shoulder instability and monitoring the progression of rehabilitation administered to reduce the recurrence of glenohumeral subluxation or dislocation (14, 18). Characterization of translational stiffness during muscle contraction would complement passive clinical assessments by evaluating how effective shoulder muscles are at actively modulating stability of the glenohumeral joint. This active modulation may be especially important in women, who have lower passive stiffness than men (19). Characterizing the modulation of translational stiffness from contracting shoulder muscles would also provide valuable insight that could be applied towards improving the specificity of rehabilitation protocols designed to treat athletes with instability (14, 15). However, only rotational stiffness, or the relationship between rotations applied to the glenohumeral joint and the torques generated in response, has been characterized in human shoulders during muscle contraction (20, 21). While rotational stiffness of the glenohumeral joint is clinically relevant to shoulder movement and postural control, rotational stiffness is less relevant to evaluating shoulder instability than translational stiffness. Quantification of translational stiffness during muscle contraction has yet to be accomplished, despite successful quantification of translational stiffness during passive conditions (22-24).

Changes to the translational stiffness of the glenohumeral joint can occur under passive conditions, with a change of shoulder position, and under active conditions, with the contraction of shoulder muscles. During passive conditions, when the shoulder muscles are relaxed, the apprehension position has increased translational stiffness compared to neutral shoulder positions (24). This increase in translational stiffness of the glenohumeral joint is attributed to increased tension on the glenohumeral ligaments that further restrain humeral head translation (25, 26). While no experimental data exist on how muscle contraction influences glenohumeral stiffness in the apprehension position, a modeling study has suggested that translational stability in the apprehension position will be reduced due to changes in how well shoulder muscles can compress, and thus stabilize, the glenohumeral joint (27). This modeled reduction in active stability may explain why athletes are vulnerable to experiencing symptoms of instability in the apprehension position. A reduced ability to increase stability of the glenohumeral joint also supports the emphasis on enhancing dynamic stability with strengthening and neuromuscular control exercises in the apprehension position during rehabilitation (14, 15). However, since this biomechanical model is limited in considering potential contributions to translational shoulder stability from intrinsic muscle mechanics and neuromuscular control (17), it remains unknown if stability under conditions of muscle contraction is reduced in the apprehension position in vivo.

The purpose of this study was to determine how translational stiffness of the glenohumeral joint was modulated by shoulder muscle contraction, and how that ability differed between a neutral abduction position and the apprehension position. Our primary hypothesis was that the ability to increase translational stiffness of the glenohumeral joint with muscle contraction would be reduced in the apprehension position compared to a neutral abduction position. We tested our hypothesis by quantifying translational stiffness of the glenohumeral joint in both neutral abduction and apprehension positions while individuals produced a wide range of isometric torques at their shoulder. We evaluated the ability to increase glenohumeral stiffness with muscle contraction separately in men and women to account for sex-specific differences that may exist. Finally, we explored the effect of shoulder position on passive glenohumeral stiffness and shoulder strength. Identifying if the apprehension position is susceptible to reduced glenohumeral stability with muscle contraction may be useful to refine rehabilitation protocols and design injury prevention programs for populations at risk for shoulder instability.

METHODS

Participants

Seventeen healthy adults (9 female, 8 male; mean age ± SD: 24.9 ± 2.6 years; mean body mass index ± SD: 23.5 ± 3.8 kg/m2) participated in this study. Participants gave written informed consent before participating in the study, which was approved by Northwestern University’s Institutional Review Board (STU00208382). Inclusion criteria were as follows: (i) functional and pain-free shoulder range-of-motion; (ii) pain-free cervical range-of-motion; and (iii) negative anterior apprehension test. All participants were right-hand dominant to accommodate for testing of the dominant arm in our robotic system. Participants were excluded if they had: (i) shoulder pain in the six months prior to testing that prevented participation in overhead activities or required treatment from a health professional; (ii) history of scapular or humeral fracture; (iii) history of shoulder dislocation; or (iv) prior shoulder surgery. All exclusion criteria were applied to both the dominant and non-dominant shoulder.

Experimental Setup for Estimation of Glenohumeral Stiffness

Translational stiffness of the glenohumeral joint was estimated using an instrumented robotic system to apply displacements and record forces at each participant’s glenohumeral joint. Each participant was seated in a Biodex chair (Biodex Medical Systems; Shirley, NY), and their dominant arm was attached midway between the acromion and the olecranon to the robotic system via a custom-made full-arm fiberglass cast. The arm was attached at 90° shoulder abduction and 20° horizontal adduction in two positions: 1) neutral abduction – 0° rotation, and 2) apprehension – 90° external rotation (Figure 1A-B). The cast maintained the arm at 90° elbow flexion and supported the weight of the arm while the robotic system applied displacements. Each participant’s scapula was stabilized with a form-fitting thermoplastic clamp over the acromion and posterior scapula to limit scapulothoracic movement and isolate displacements to the glenohumeral joint (Figure 1B). The computer-controlled robotic system (ThrustTube, Copley Controls Corporation; Canton, MA) applied anterior-posterior displacements to translate the humerus within the glenoid fossa and was instrumented with a linear encoder (RGH24, Renishaw; Gloucestershire, UK) to record the displacements. Both mechanical and electric safety stops were implemented to limit displacements to a safe translational range. A six degree-of-freedom load cell (67M25A3, JR3 Inc.; Woodland, CA) recorded forces and torques at the interface between the cast and the robotic system (Figure 1B). All force data were anti-alias filtered at 500 Hz, and all position and force data were then sampled at 2500 Hz (PCI-DAS1602/16 and PCI-QUAD04, Measurement Computing; Norton, MA). All data acquisition and control of the robotic system were performed using xPC Target (MathWorks; Natick, MA).

Figure 1.

Figure 1.

Open in a new tab

Displacements were applied by the robotic system at the glenohumeral joint with the participant positioned in one of two positions: neutral abduction (A) and apprehension (B). The glenohumeral joint was positioned in the scapular plane (20° horizontal adduction). The scapula was immobilized with a custom-fitting thermoplastic clamp. Displacements were applied in anterior and posterior directions from a neutral glenohumeral position. Participants produced isometric torque in one of six directions while the displacements were applied. Forces and torques were measured by a six degree-of-freedom load cell. Displacements were measured by a digital linear encoder. Participants were aided during all tasks with real-time visual feedback displayed on a computer monitor in their field of view (C). A cursor (red box) represented the magnitude and direction of torque produced at the shoulder. Ab/adduction torque translated the cursor up and down, horizontal ab/adduction torque translated the cursor right and left, and int/external rotation torque resulted in the growth of a triangle from the bottom or top of the cursor. During torque matching tasks, participants were instructed to move the cursor to a target box representing the desired torque for each trial.

Experimental Protocol

Prior to the experiment, all participants produced isometric maximum voluntary contractions (MVCs) in six directions (abduction/adduction, internal/external rotation, and horizontal abduction/adduction). MVCs were repeated with the arm in each of the test positions to scale the torque targets used throughout the experiment (Figure 1). Participants received standardized instructions on the MVC-testing procedures. Participants were aided by three-dimensional visual feedback (Figure 1C) of the generated shoulder torque to constrain their efforts to the desired direction. A sub-maximum (50%) effort trial was performed in each torque direction to familiarize participants to the procedure. Participants then performed the MVC with verbal encouragement until producing the two highest values within 10% torque magnitude of each other. Additional trials were recorded to ensure maximal effort was achieved in each torque direction. The peak value that was generated in each torque direction from each shoulder position was used to normalize submaximal torque production.

The experiment was designed to estimate the translational stiffness of the glenohumeral joint while participants generated muscle contractions by producing isometric torque (active conditions). During active conditions, participants were instructed to produce torque in one of the six torque directions (Figure 1). Different directions of torque production were used to generate different patterns of muscle activity (21). Participants held their shoulder torque constant at one of two torque levels (5% or 10% MVC) for the duration of each trial. This torque-matching task was aided by visual feedback (Figure 1C). Trials were also collected during passive conditions while participants were instructed to relax fully and ignore all applied displacements (0% MVC). Trials in each position (neutral abduction and apprehension) were tested separately and in a random order across participants. The torque magnitudes and directions were randomized for every participant within each shoulder position.

During each trial, the robotic system applied displacement perturbations in the anterior-posterior direction. Perturbations were applied to the casted arm at the point of attachment to the robotic system between the acromion and the olecranon. Each perturbation was a pseudorandom binary sequence with a peak-to-peak amplitude of 7mm and a switching interval of 0.15s (Figure 2A). A peak-to-peak amplitude of 7mm was chosen to ensure estimates of translational stiffness were not affected by non-linearities due to short-range muscle stiffness (28). All perturbations were centered about a neutral glenohumeral position. Anterior-posterior resistance forces were recorded continuously throughout the trial (Figure 2B). Each trial lasted 55 seconds, a duration chosen to minimize experimental time while still providing sufficient data to estimate glenohumeral stiffness. Data from the first 5 seconds of each trial were discarded to eliminate transient behaviors associated with the onset of the perturbation. Participants rested between trials as needed to prevent fatigue and discomfort from the pressure used to stabilize the scapula. Three trials were completed for each condition, totaling 78 trials per participant (72 active trials: 6 torque directions x 2 torque levels x 2 shoulder positions x 3 repetitions; 6 passive trials: 1 torque level x 2 positions x 3 repetitions).

Figure 2.

Figure 2.

Open in a new tab

The dynamic relationship between an applied displacement (A) and the force produced in response (B) is defined as impedance (C). Pseudorandom binary sequence perturbations were defined by random switching at a fixed switching interval between two fixed amplitudes (A). Glenohumeral stiffness was estimated by averaging the low-frequency (0.2-1.0 Hz) magnitude of the impedance frequency response function (C). Coherence values that approach 1.0 in the low-frequency region of interest represent an appropriate assumption of linearity in the calculation of impedance (D). Glenohumeral stiffness was estimated for each trial, providing three estimates of stiffness at each torque level for each torque direction (E). Examples from three trails during a single direction of torque production are shown above: one from a passive trial and two from active trials (5% and 10% MVC).

Data Processing

Translational stiffness of the glenohumeral joint was the primary outcome measure of our experiment. Translational stiffness, which is defined in the glenohumeral joint as the force required to produce humeral head displacement, is typically assessed by clinicians to evaluate glenohumeral stability (17). We computed glenohumeral stiffness from estimates of shoulder impedance, the dynamic relationship between the applied displacements and the forces generated in response, using non-parametric system identification techniques (29). The estimated impedance of the glenohumeral joint was approximately constant at frequencies below 1.0 Hz, as is indicative of a mechanical system with spring-like properties (Figure 2C) (21, 30, 31). Therefore, we estimated glenohumeral stiffness, the static component of glenohumeral impedance, as the average magnitude of impedance between 0.2-1.0 Hz for all conditions. The goodness of fit of the impedance estimations, calculated from the percentage of variance accounted for (%VAF) for all trials, was high across all participants (%VAF ± 95% confidence limits: 86 ± 2%). Additionally, we computed mean squared coherence at all frequencies of interest to assess how well the linear approximations of impedance fit the data (Figure 2D). Coherence was high within this frequency range (0.98 ± 0.004 across all participants out of a maximum of 1.0), indicating that our characterization of the glenohumeral mechanics described the experimental data particularly well in the frequency range used to estimate stiffness.

Statistical Analysis

All data were analyzed using MATLAB statistical packages (version R2020a, MathWorks; Natick, MA). Our primary hypothesis was that the ability to increase glenohumeral stiffness during shoulder torque production would be reduced in the apprehension position compared to a neutral abduction position. We tested this hypothesis by comparing the slope of the relationship between torque magnitude and glenohumeral stiffness for the neutral abduction and apprehension positions (Figure 2E). We defined this slope as the torque-dependent stiffness modulation. This slope was estimated using a linear mixed effects model. Glenohumeral stiffness was the dependent variable in this analysis. The independent factors were torque magnitude (continuous based on magnitude at 0%, 5%, and 10% MVC), torque direction (fixed with six levels: abduction/adduction, internal/external rotation, horizontal abduction/adduction), shoulder position (fixed with two levels: neutral abduction, apprehension), and sex (fixed with two levels: male and female). Participant was treated as a random factor. A stepwise regression with backwards elimination was applied to simplify the model. We tested our primary hypothesis by comparing the effect of torque on glenohumeral stiffness (torque-dependent stiffness modulation) between the two shoulder positions. Separate comparisons were made for each torque direction, resulting in six comparisons.

Although our primary purpose was to compare torque-dependent stiffness modulation between shoulder positions, common clinical tests assess glenohumeral stiffness during passive conditions (32). Therefore, we also evaluated the effect of shoulder position on glenohumeral stiffness in the absence of torque production to determine if shoulder position altered the passive stiffness of the joint. We compared passive stiffness (dependent variable) between shoulder positions (independent fixed factor) with a linear mixed effects model. Participant was treated as a random factor. We included sex as an independent fixed factor and compared passive stiffness between shoulder positions separately for each sex based on differences in glenohumeral joint laxity between male and female individuals (19).

Finally, we compared the maximum torque between shoulder positions in each torque direction, as reduced maximum torque in the apprehension position would limit the maximum glenohumeral stiffness that could be achieved. We compared maximum torque magnitude (dependent variable) between positions (fixed independent factor) for each torque direction (fixed independent factor) with a linear mixed effects model. Participant was treated as a random factor. We included sex as an independent fixed factor and we evaluated each comparison separately for each sex based on differences in isometric shoulder strength between male and female individuals (33).

All confidence limits reported in the text reflect 95% confidence intervals (α = 0.05) unless otherwise noted. Bonferroni corrections were used to control for multiple comparisons within each model. We used the Wald t-test statistic with a Satterthwaite approximation to estimate P-values (34). An a priori power analysis was performed to determine the number of participants needed to test the primary hypothesis. The study was powered to detect a 30% reduction in torque-dependent stiffness modulation between neutral abduction and apprehension positions based on modeled changes of force magnitude and orientation in the apprehension position (27). A pooled standard deviation of 33% was assumed based on collected pilot data. Using this estimate of effect size, (d = 0.9, α = 0.05), 16 participants would be required to detect differences between positions in the primary outcome variable with greater than 80% power.

RESULTS

Effects of Shoulder Position on Torque-Dependent Stiffness Modulation

Our primary hypothesis was that the ability to increase glenohumeral stiffness during shoulder torque production, which we defined as torque-dependent stiffness modulation, would be reduced in the apprehension position compared to a neutral abduction position. Overall, torque-dependent stiffness modulation was reduced by 13% in the apprehension position compared to the neutral abduction position (Δ ± 95% confidence limits: −0.04 ± 0.03 N/mm per newton-meter, P = 0.02). We found that torque-dependent stiffness modulation was decreased in the apprehension position while participants produced certain directions of shoulder torque (Figure 3A). Torque-dependent stiffness modulation decreased the most in the apprehension position while participants produced abduction (−0.14 ± 0.03 N/mm per newton-meter, P < 0.001), horizontal abduction (−0.11 ± 0.04 N/mm per newton-meter, P < 0.001), and internal rotation (−0.06 ± 0.07 N/mm per newton-meter, P = 0.06) torque. These effects of shoulder position reflected 49%, 25%, and 15% reductions in torque-dependent stiffness modulation, respectively. In contrast, torque-dependent stiffness modulation increased 19% while participants produced horizontal adduction torques (+0.04 ± 0.03 N/mm per newton-meter, P = 0.02) in the apprehension position.

Figure 3.

Figure 3.

Open in a new tab

Torque-dependent stiffness modulation group results in each torque direction stratified by shoulder position (A). Points that share the same letter across the six torque directions do not differ within each position at α=0.05/15 (a-cneutral abduction; y-zapprehension). Passive stiffness (B) was higher in the apprehension position compared to the neutral abduction position in female participants only. Error bars represent the standard error for torque-dependent stiffness modulation and passive stiffness estimates in both panels.

Overall, participants were able to increase the translational stiffness of their glenohumeral joint with muscle contraction. Glenohumeral stiffness increased linearly with torque magnitude in all directions of voluntary torque (whole model: R2 = 0.82) in both shoulder positions. Participants could increase glenohumeral stiffness by a mean of 0.30 ± 0.06 N/mm for every newton-meter of torque they produced across all directions (P < 0.001). Across all torque directions and shoulder positions, torque-dependent stiffness modulation ranged from an increase of 0.15 N/mm to an increase of 0.44 N/mm for every newton-meter of torque produced (all P < 0.001; Figure 3A). Male participants were able to increase their glenohumeral stiffness less than female participants for a given increase in torque magnitude (−0.10 ± 0.12 N/mm per newton-meter, P = 0.11). This effect reflects a 30% decrease in torque-dependent stiffness modulation among male participants compared to female participants. Torque-dependent stiffness modulation can be observed for a typical female participant in Figure 4.

Figure 4.

Figure 4.

Open in a new tab

All trials of glenohumeral stiffness estimation are displayed for a typical participant. Glenohumeral stiffness was estimated in neutral abduction (A) and apprehension (B) positions. Torque-dependent stiffness modulation was defined as the linear relationship between glenohumeral stiffness and torque magnitude in each of six torque directions. Each slope was estimated from a linear model fit to data collected from this individual participant. Passive stiffness was determined experimentally as the glenohumeral stiffness at 0% MVC torque magnitude. In this participant, the difference in torque-dependent stiffness modulation can be visualized as the difference of the six slopes between torque directions and shoulder positions. For example, in a neutral abduction position, torque-dependent stiffness modulation for this participant was markedly greater while producing horizontal abduction (green) and internal rotation (orange) torques than for torques in the other four directions. Additionally, torque-dependent stiffness modulation was lower in the apprehension position while this participant produced certain directions of shoulder torque. For example, torque-dependent stiffness modulation was markedly lower in the apprehension position while this participant generated torque in horizontal abduction (green) and internal rotation (orange) directions.

Effects of Shoulder Position on Passive Stiffness

Passive measures of glenohumeral stiffness were lower in the neutral abduction position among female participants only. In female participants, passive stiffness was approximately 17% lower in the neutral abduction position than in the apprehension position (−0.18 ± 0.08 N/mm, P < 0.001; Figure 3B). On the contrary, passive stiffness was slightly increased in the neutral abduction position compared to the apprehension position in male participants (+0.06 ± 0.09 N/mm, P = 0.20). Comparing between sexes, passive stiffness was approximately 23% lower in female participants than in male participants in the neutral abduction position (−0.27 ± 0.23 N/mm, P = 0.03), but was similar between female and male participants in the apprehension position (−0.03 ± 0.23 N/mm, P = 0.79).

Effects of Shoulder Position on Maximum Torque

Changing the position of the shoulder impacted the maximum torque participants could produce. Participants were substantially weaker in the apprehension position compared to the neutral abduction position in two of the six torque directions (descriptive results in Table 1): horizontal adduction (female: −10 ± 6 N·m, P = 0.006; male: −26 ± 17 N·m, P = 0.01) and external rotation (female: −5 ± 1 N·m, P < 0.001; male: −18 ± 10 N·m, P = 0.004). In horizontal adduction, these effects reflect a 29% and 38% decrease in isometric strength among female and male participants, respectively. In external rotation, these effects reflect a 36% and 55% decrease in isometric strength, respectively. Differences in strength between shoulder positions were smaller in the other four torque directions, ranging from a 14% decrease in isometric strength to a 25% increase in the apprehension position (all P = 0.24-0.87). Overall, male participants were stronger than female participants in all torque directions and both shoulder positions (Table 1).

Table 1.

Maximum voluntary contraction torque magnitudes in Newton-meters stratified by torque direction, shoulder position, and sex.

Maximum Voluntary Contraction Torque
[mean ± SD Nm]
Sex Male Female
Position Neutral Apprehension Neutral Apprehension
Torque Direction Abduction 54 ± 20 56 ± 25 26 ± 10 28 ± 7
Adduction 59 ± 23 51 ± 19 30 ± 9 25 ± 11
External Rotation 33 ± 13* 15 ± 8* 14 ± 1* 9 ± 2*
Internal Rotation 24 ± 9 30 ± 12 11 ± 3 11 ± 1
Horizontal Adduction 68 ± 24^ 42 ± 10^ 35 ± 7^ 25 ± 6^
Horizontal Abduction 45 ± 16 38 ± 13 20 ± 7 20 ± 4
Open in a new tab
*

P < 0.05/12

^

P < 0.05 comparing maximum voluntary contraction torque between shoulder positions within each sex

Male participants were stronger than female participants in all combinations of torque direction and shoulder position (abduction/apprehension: P = 0.01; external rotation/apprehension: P = 0.07; all other combinations: P = 0.0003-0.006).

DISCUSSION

The goal of this study was to determine if the ability to increase translational stiffness of the glenohumeral joint with muscle contraction differed with shoulder position. We compared the modulation of glenohumeral stiffness provided by muscle contraction between neutral abduction and apprehension positions. Glenohumeral stiffness increased linearly with the magnitude of shoulder torque. Thus, the contraction of shoulder muscles during torque generation increased the forces needed to translate or partially subluxate the humeral head. This ability to increase glenohumeral stiffness was reduced in the apprehension position compared to the neutral abduction position, consistent with our primary hypothesis. These results suggest that the capacity of contracting shoulder muscles to stabilize the humeral head is lower in the apprehension position. Additionally, passive stiffness in female participants was increased in the apprehension position, demonstrating that passive structures of the shoulder may increase the stability of the glenohumeral joint in the apprehension position for women.

Effects of Shoulder Position on Torque-Dependent Stiffness Modulation

One explanation for reduced active modulation of glenohumeral stiffness in the apprehension position is that differences in muscles’ lines of action limit their ability to compress the humeral head into the glenoid fossa. These changes in lines of action would affect the modulation of glenohumeral stiffness independently from changes in muscle coordination. Cadaveric studies have demonstrated that compressing the convex humeral head into the concave glenoid fossa through simulated muscle contraction can increase the stability of the glenohumeral joint and help prevent subluxation (35-38). The rotator cuff muscles are thought to modulate glenohumeral stiffness through this mechanism because their lines of action are oriented in compression, while other shoulder muscles may produce forces oriented less in compression and more in shear (39). A prior musculoskeletal model predicted that the modulation of glenohumeral stiffness would decrease in the apprehension position due to changes in the orientation of muscle forces from compression to shear (27). In our study, this change in the orientation may explain why the modulation of glenohumeral stiffness decreased in the apprehension position during certain directions of torque production. For example, the orientation of force produced by the subscapularis muscle increases its shear component in the apprehension position (27). This change in orientation of muscle force may explain why decreased torque-dependent stiffness modulation was observed during horizontal abduction, a direction in which the subscapularis is active at 90° of abduction (40). Prior estimates of how the deltoid and supraspinatus muscles’ lines of action change between the neutral abduction and apprehension positions present more conflicting interpretations. On one hand, a decrease in compressional forces produced by the supraspinatus and deltoid muscles in the apprehension position compared to a neutral abduction position (39) may explain the 49% reduction of torque-dependent stiffness modulation observed during abduction in our study. The supraspinatus and deltoid muscles would be active during abduction, and any decrease in compressional forces should similarly result in reduced modulation of glenohumeral stiffness. On the other hand, other estimates of the deltoid and supraspinatus lines of action were unchanged in the apprehension position compared a neutral abduction position (27), suggesting no change in stiffness modulation should be observed between shoulder positions. The current theory that increasing shear lines of action contributes to reductions in glenohumeral stability in the apprehension position remains unclear.

Overall, translational stiffness of the glenohumeral joint increased with voluntary torque production, similar to the torque-dependent modulation of rotational stiffness for the glenohumeral joint (20, 21). On average, the magnitude of torque-dependent stiffness modulation observed in our study indicated that participants could increase their glenohumeral stiffness from passive levels by 80% while only producing 10% of their maximum torque. Based the linear relationship between rotational joint stiffness and torque magnitude observed beyond 10% MVC in the shoulder (20), we extrapolate that shoulder torques at 50% of maximum strength would result in a fourfold increase in glenohumeral stiffness relative to passive levels. Co-contraction would yield even larger increases for a given level of torque (41).

Our results provide in vivo evidence of the active modulation of translational stiffness that has been primarily modeled in cadaveric studies (25, 35, 38), which are unable to consider the mechanical properties of muscle that change with activation (42) and may also influence glenohumeral joint stiffness. Further, these results provide the first quantitative estimates of active modulation of glenohumeral stiffness, in contrast to the passive manual assessments of glenohumeral displacements commonly made by clinicians as part of the examination for patients with shoulder instability (32). While glenohumeral laxity tests can be helpful to diagnose shoulder instability, methods to evaluate the change in laxity or increase in stiffness while patients activate their shoulder musculature are lacking. An important limitation to the assessment of glenohumeral stiffness during muscle contraction with available clinical techniques is that voluntary contractions produced in response to slowly-applied displacements may bias stiffness estimates (30). This limitation may have similarly biased previous attempts to quantitatively estimate active translational stiffness with slow, predictable displacements (43, 44). In our study, to overcome any confounding effects from reactionary torque production, the displacements applied to the humeral head were instead applied briskly and at random intervals. Such unbiased estimates of translational stiffness during torque production may provide a tool to quantify active glenohumeral stability in individuals with shoulder instability, in which impairments are hypothesized (45) and have been demonstrated for rotational stiffness (46).

Effects of Shoulder Position on Passive Stiffness

Passive glenohumeral stiffness was decreased in the neutral abduction position compared to the apprehension position in female participants. Higher passive glenohumeral stiffness in the apprehension position may be due to increased contributions of passive glenohumeral ligaments to resist translation when the arm is abducted and externally rotated. In a cadaveric model, the anterior band and sling of the inferior glenohumeral ligament are drawn tightly across the bottom half of the anterior humeral head at 90° of shoulder abduction. In the apprehension position, the sling of the inferior glenohumeral ligament continues to slide anteriorly and superiorly to fully cover the mid-portion of the humeral head (26). Such rearrangement of the inferior glenohumeral ligament in the apprehension position would prevent anterior translation and increase passive stiffness, consistent with the results of our study in female participants. Decreased passive stiffness in the neutral abduction position of female participants may be explained by increased generalized joint laxity in females, which has been observed in prior studies (female mean ± SD Beighton score: 2.9 ± 2.1; male: 1.0 ± 1.7) (19) and among participants in our study (female mean ± SD Beighton score: 2.8 ± 2.2; male: 0.9 ± 1.1). A Beighton score ≥2 has been associated with a history of glenohumeral instability (47). Further, increased anterior glenohumeral laxity in healthy women compared to men may explain the differences in passive stiffness between sexes that we observed in the neutral abduction position (19).

Effects of Shoulder Position on Maximum Torque

Shoulder strength was decreased in the apprehension posture while participants produced horizontal adduction and external rotation torque. Decreased isometric strength in horizontal adduction or external rotation would limit the maximum glenohumeral stiffness that an individual could achieve when producing torque in these two directions despite small increases in torque-dependent stiffness modulation observed between shoulder positions (+0.04 N/mm per newton-meter and +0.02 N/mm per newton-meter, respectively). For example, torque-dependent stiffness modulation during horizontal adduction torque production was 19% greater in the apprehension position than in the neutral abduction position, but the decreased strength in this position means that an average individual could still increase their glenohumeral stiffness to a greater maximum level in the neutral abduction position than in the apprehension position. Additionally, although male participants were able to increase their glenohumeral stiffness less than female participants for given increases in torque magnitude, since male participants were stronger, the overall increases in glenohumeral stiffness for matched torque magnitudes would be greater.

Clinical Implications

Since overhead athletes are believed to rely on active stabilizing mechanisms to compensate for increased glenohumeral laxity (48), the ability to increase glenohumeral stiffness with muscle contraction may be important to healthy shoulder function. Further, many overhead sporting tasks require torque production in the apprehension position in directions where the ability to increase glenohumeral stiffness with muscle contraction was reduced in our study. For example, in baseball, the acceleration phase of pitching requires forced internal rotation from the apprehension position (11). In weightlifting, military presses require abduction in the apprehension position to lift barbells or dumbbells (5). Both activities may induce symptoms of pain or fear of instability because athletes are placing large loads on their glenohumeral joint while not sufficiently increasing glenohumeral stiffness with muscle contraction. While our study reveals that modulation of glenohumeral stiffness may be reduced in the apprehension position during certain torque directions in an asymptomatic population, whether similar deficits exist in individuals with shoulder instability remains unknown.

Shoulder strengthening in the apprehension position is a common component of shoulder instability rehabilitation and shoulder injury prevention programs (14, 15, 49). Strengthening in the apprehension position is typically performed with internal and external rotation torque generation based on the expected contributions of rotator cuff muscles to enhance dynamic glenohumeral stability (16). Based on the results of our study, rehabilitation and injury prevention programs could also emphasize shoulder strengthening with other torque directions (e.g., abduction and horizontal abduction) which are the least efficient at increasing glenohumeral stiffness in the apprehension position compared to more neutral shoulder positions. Given the slightly reduced torque-dependent stiffness modulation we observed for males, targeted shoulder strengthening for male athletes, especially in the apprehension position, may be warranted to decrease the incidence of instability events.

Limitations

We were unable to collect shoulder muscle activity with electromyography due to the restraints used to minimize scapular motion. Electromyography collected in neutral abduction and apprehension positions could have clarified potential differences in muscle coordination between positions, providing additional rationale for decreased modulation of glenohumeral stiffness in the apprehension position. Additionally, while glenohumeral stiffness was estimated well in this study, it may not represent the true translational stiffness of the glenohumeral joint because displacements and forces were measured externally at each participant’s arm. True humeral head translation, and subsequently true glenohumeral stiffness, could not be confirmed with ultrasound (44) or electromagnetic sensors (22) because the scapular stabilization and full-arm casting prevented their use. However, our experiment was designed to minimize the improper measurement of soft tissue displacement in lieu of humeral head displacement by applying displacements through a tight-fitting cast that interfaced with bony prominences on the humerus. Finally, the nature of the humeral head position at which perturbations were applied in our study could not discern between anterior and posterior glenohumeral stiffness. Instead, our results reflect an appropriate estimation of the average glenohumeral stiffness in a neutral glenohumeral position.

CONCLUSIONS

In summary, we found that the ability to increase translational stiffness of the glenohumeral joint with muscle contraction is reduced in the apprehension position compared to a position of neutral abduction. This outcome reveals potentially compromised glenohumeral stability in the apprehension position during multiple directions of torque production. The results of this study may help explain why individuals more frequently experience symptoms of shoulder instability in the apprehension position compared to neutral positions. Future work is warranted in populations with shoulder instability to determine if decreased modulation of glenohumeral stiffness is associated with painful symptoms or shoulder dysfunction. Such findings may be useful to refine rehabilitation protocols or design prevention programs for shoulder instability.

ACKNOWLEDGEMENTS

This work was supported in part by the National Institutes of Health (NIAMS F31AR074288, NIA F31AG057137, NIGMS T32GM008152, NCATS UL1TR001422) and Northwestern University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors would like to acknowledge Timothy M. Haswell for his assistance with designing the experimental setup and collecting data. The authors would also like to acknowledge Margaret Coats-Thomas for her assistance with collecting and analyzing data. The authors have no conflicts of interest to disclose. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of this study do not constitute endorsement by the American College of Sports Medicine.

REFERENCES

  • 1.Matsen FA 3rd, Lippitt SB, Sidles JA, Harryman DT 2nd. Practical evaluation and management of the shoulder. WB Saunders Company; 1994. [Google Scholar]
  • 2.Owens BD, Agel J, Mountcastle SB, Cameron KL, Nelson BJ. Incidence of glenohumeral instability in collegiate athletics. Am J Sports Med. 2009;37(9):1750–4. [DOI] [PubMed] [Google Scholar]
  • 3.Bak K, Fauno P. Clinical findings in competitive swimmers with shoulder pain. Am J Sports Med. 1997;25(2):254–60. [DOI] [PubMed] [Google Scholar]
  • 4.Caplan J, Julien TP, Michelson J, Neviaser RJ. Multidirectional instability of the shoulder in elite female gymnasts. Am J Orthop (Belle Mead NJ). 2007;36(12):660–5. [PubMed] [Google Scholar]
  • 5.Gross ML, Brenner SL, Esformes I, Sonzogni JJ. Anterior shoulder instability in weight lifters. Am J Sports Med. 1993;21(4):599–603. [DOI] [PubMed] [Google Scholar]
  • 6.Kaplan LD, Flanigan DC, Norwig J, Jost P, Bradley J. Prevalence and variance of shoulder injuries in elite collegiate football players. Am J Sports Med. 2005;33(8):1142–6. [DOI] [PubMed] [Google Scholar]
  • 7.Trojan JD, Meyer LE, Edgar CM, Brown SM, Mulcahey MK. Epidemiology of Shoulder Instability Injuries in Collision Collegiate Sports From 2009 to 2014. Arthroscopy. 2020;36(1):36–43. [DOI] [PubMed] [Google Scholar]
  • 8.Dickens JF, Owens BD, Cameron KL et al. Return to play and recurrent instability after in-season anterior shoulder instability: a prospective multicenter study. Am J Sports Med. 2014;42(12):2842–50. [DOI] [PubMed] [Google Scholar]
  • 9.Rowe CR, Zarins B. Recurrent transient subluxation of the shoulder. J Bone Joint Surg Am. 1981;63(6):863–72. [PubMed] [Google Scholar]
  • 10.Kaipel M, Reichetseder J, Schuetzenberger S, Hertz H, Majewski M. Sex-related outcome differences after arthroscopic shoulder stabilization. Orthopedics. 2010;33(3). [DOI] [PubMed] [Google Scholar]
  • 11.Braatz JH, Gogia PP. The mechanics of pitching. J Orthop Sports Phys Ther. 1987;9(3):56–69. [DOI] [PubMed] [Google Scholar]
  • 12.Kennedy JC, Hawkins R, Krissoff WB. Orthopaedic manifestations of swimming. Am J Sports Med. 1978;6(6):309–22. [DOI] [PubMed] [Google Scholar]
  • 13.Shanley E, Thigpen C, Brooks J et al. Return to Sport as an Outcome Measure for Shoulder Instability: Surprising Findings in Nonoperative Management in a High School Athlete Population. Am J Sports Med. 2019;47(5):1062–7. [DOI] [PubMed] [Google Scholar]
  • 14.Eshoj HR, Rasmussen S, Frich LH et al. Neuromuscular Exercises Improve Shoulder Function More Than Standard Care Exercises in Patients With a Traumatic Anterior Shoulder Dislocation: A Randomized Controlled Trial. Orthop J Sports Med. 2020;8(1):2325967119896102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Warby SA, Ford JJ, Hahne AJ et al. Comparison of 2 Exercise Rehabilitation Programs for Multidirectional Instability of the Glenohumeral Joint: A Randomized Controlled Trial. Am J Sports Med. 2018;46(1):87–97. [DOI] [PubMed] [Google Scholar]
  • 16.Cools AM, Borms D, Castelein B, Vanderstukken F, Johansson FR. Evidence-based rehabilitation of athletes with glenohumeral instability. Knee Surg Sports Traumatol Arthrosc. 2016;24(2):382–9. [DOI] [PubMed] [Google Scholar]
  • 17.Veeger HE, van der Helm FC. Shoulder function: the perfect compromise between mobility and stability. J Biomech. 2007;40(10):2119–29. [DOI] [PubMed] [Google Scholar]
  • 18.Tzannes A, Murrell GA. Clinical examination of the unstable shoulder. Sports Med. 2002;32(7):447–57. [DOI] [PubMed] [Google Scholar]
  • 19.Borsa PA, Sauers EL, Herling DE. Patterns of glenohumeral joint laxity and stiffness in healthy men and women. Med Sci Sports Exerc. 2000;32(10):1685–90. [DOI] [PubMed] [Google Scholar]
  • 20.Zhang LQ, Portland GH, Wang G et al. Stiffness, viscosity, and upper-limb inertia about the glenohumeral abduction axis. J Orthop Res. 2000;18(1):94–100. [DOI] [PubMed] [Google Scholar]
  • 21.Lipps DB, Baillargeon EM, Ludvig D, Perreault EJ. Quantifying the Multidimensional Impedance of the Shoulder During Volitional Contractions. Ann Biomed Eng. 2020; 48(9):2354–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Borsa PA, Sauers EL, Herling DE, Manzour WF. In vivo quantification of capsular end-point in the nonimpaired glenohumeral joint using an instrumented measurement system. J Orthop Sports Phys Ther. 2001;31(8):419–26; discussion 27-31. [DOI] [PubMed] [Google Scholar]
  • 23.Borsa PA, Sauers EL, Herling DE. Glenohumeral Stiffness Response Between Men and Women for Anterior, Posterior, and Inferior Translation. J Athl Train. 2002;37(3):240–5. [PMC free article] [PubMed] [Google Scholar]
  • 24.McQuade KJ, Shelley I, Cvitkovic J. Patterns of stiffness during clinical examination of the glenohumeral joint. Clin Biomech (Bristol, Avon). 1999;14(9):620–7. [DOI] [PubMed] [Google Scholar]
  • 25.Blasier RB, Guldberg RE, Rothman ED. Anterior shoulder stability: Contributions of rotator cuff forces and the capsular ligaments in a cadaver model. J Shoulder Elbow Surg. 1992;1(3):140–50. [DOI] [PubMed] [Google Scholar]
  • 26.Turkel SJ, Panio MW, Marshall JL, Girgis FG. Stabilizing mechanisms preventing anterior dislocation of the glenohumeral joint. J Bone Joint Surg Am. 1981;63(8):1208–17. [PubMed] [Google Scholar]
  • 27.Labriola JE, Jolly JT, McMahon PJ, Debski RE. Active stability of the glenohumeral joint decreases in the apprehension position. Clin Biomech (Bristol, Avon). 2004;19(8):801–9. [DOI] [PubMed] [Google Scholar]
  • 28.Rack PM, Westbury DR. The short range stiffness of active mammalian muscle and its effect on mechanical properties. J Physiol. 1974;240(2):331–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Perreault EJ, Kirsch RF, Acosta AM. Multiple-input, multiple-output system identification for characterization of limb stiffness dynamics. Biol Cybern. 1999;80(5):327–37. [DOI] [PubMed] [Google Scholar]
  • 30.Kearney RE, Hunter IW. System identification of human joint dynamics. Crit Rev Biomed Eng. 1990;18(1):55–87. [PubMed] [Google Scholar]
  • 31.Hunter IW, Kearney RE. Dynamics of human ankle stiffness: variation with mean ankle torque. J Biomech. 1982;15(10):747–52. [DOI] [PubMed] [Google Scholar]
  • 32.Gerber C, Ganz R. Clinical assessment of instability of the shoulder. With special reference to anterior and posterior drawer tests. J Bone Joint Surg Br. 1984;66(4):551–6. [DOI] [PubMed] [Google Scholar]
  • 33.Westrick RB, Duffey ML, Cameron KL, Gerber JP, Owens BD. Isometric shoulder strength reference values for physically active collegiate males and females. Sports Health. 2013;5(1):17–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Luke SG. Evaluating significance in linear mixed-effects models in R. Behav Res Methods. 2017;49(4):1494–502. [DOI] [PubMed] [Google Scholar]
  • 35.Wuelker N, Korell M, Thren K. Dynamic glenohumeral joint stability. J Shoulder Elbow Surg. 1998;7(1):43–52. [DOI] [PubMed] [Google Scholar]
  • 36.Kawano Y, Matsumura N, Murai A et al. Evaluation of the Translation Distance of the Glenohumeral Joint and the Function of the Rotator Cuff on Its Translation: A Cadaveric Study. Arthroscopy. 2018;34(6):1776–84. [DOI] [PubMed] [Google Scholar]
  • 37.Lippitt SB, Vanderhooft JE, Harris SL, Sidles JA, Harryman DT 2nd, Matsen FA 3rd. Glenohumeral stability from concavity-compression: A quantitative analysis. J Shoulder Elbow Surg. 1993;2(1):27–35. [DOI] [PubMed] [Google Scholar]
  • 38.Halder AM, Kuhl SG, Zobitz ME, Larson D, An KN. Effects of the glenoid labrum and glenohumeral abduction on stability of the shoulder joint through concavity-compression : an in vitro study. J Bone Joint Surg Am. 2001;83(7):1062–9. [DOI] [PubMed] [Google Scholar]
  • 39.Lee SB, An KN. Dynamic glenohumeral stability provided by three heads of the deltoid muscle. Clin Orthop Relat Res. 2002;(400):40–7. [DOI] [PubMed] [Google Scholar]
  • 40.Kronberg M, Nemeth G, Brostrom LA. Muscle activity and coordination in the normal shoulder. An electromyographic study. Clin Orthop Relat Res. 1990;(257):76–85. [PubMed] [Google Scholar]
  • 41.Sangwan S, Green RA, Taylor NF. Characteristics of stabilizer muscles: a systematic review. Physiother Can. 2014;66(4):348–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Morgan DL. Separation of active and passive components of short-range stiffness of muscle. Am J Physiol. 1977;232(1):C45–9. [DOI] [PubMed] [Google Scholar]
  • 43.Rathi S, Taylor NF, Soo B, Green RA. Glenohumeral joint translation and muscle activity in patients with symptomatic rotator cuff pathology: An ultrasonographic and electromyographic study with age-matched controls. J Sci Med Sport. 2018;21(9):885–9. [DOI] [PubMed] [Google Scholar]
  • 44.Rathi S, Taylor NF, Green RA. The effect of in vivo rotator cuff muscle contraction on glenohumeral joint translation: An ultrasonographic and electromyographic study. J Biomech. 2016;49(16):3840–7. [DOI] [PubMed] [Google Scholar]
  • 45.Gaskill TR, Taylor DC, Millett PJ. Management of multidirectional instability of the shoulder. J Am Acad Orthop Surg. 2011;19(12):758–67. [DOI] [PubMed] [Google Scholar]
  • 46.Olds M, McNair P, Nordez A, Cornu C. Active stiffness and strength in people with unilateral anterior shoulder instability: a bilateral comparison. J Athl Train. 2011;46(6):642–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Cameron KL, Duffey ML, DeBerardino TM, Stoneman PD, Jones CJ, Owens BD. Association of generalized joint hypermobility with a history of glenohumeral joint instability. J Athl Train. 2010;45(3):253–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Wilk KE, Meister K, Andrews JR. Current concepts in the rehabilitation of the overhead throwing athlete. Am J Sports Med. 2002;30(1):136–51. [DOI] [PubMed] [Google Scholar]
  • 49.Wilk KE, Yenchak AJ, Arrigo CA, Andrews JR. The Advanced Throwers Ten Exercise Program: a new exercise series for enhanced dynamic shoulder control in the overhead throwing athlete. Phys Sportsmed. 2011;39(4):90–7. [DOI] [PubMed] [Google Scholar]

RESOURCES